1
0
mirror of https://github.com/RIOT-OS/RIOT.git synced 2024-12-29 04:50:03 +01:00
RIOT/drivers/include/cc110x.h

657 lines
28 KiB
C
Raw Normal View History

drivers/cc110x: Rewrite of the cc110x driver The cc110x driver has been re-written from scratch to overcome the limitations of the old driver. The main motivation of the rewrite was to achieve better maintainability by a detailed documentation, reduce the complexity and the overhead of the SPI communication with the device, and to allow to simultaneously use transceivers with different configuration regarding the used base band, the channel bandwidth, the modulation rate, and the channel map. Features of this driver include: - Support for the CC1100, CC1101, and the CC1100e sub-gigahertz transceivers. - Detailed documentation of every aspect of this driver. - An easy to use configuration API that allows setting the transceiver configuration (modulation rate, channel bandwidth, base frequency) and the channel map. - Fast channel hopping by pre-calibration of the channels during device configuration (so that no calibration is needed during hopping). - Simplified SPI communication: Only during start-up the MCU has to wait for the transceiver to be ready (for the power regulators and the crystal to stabilize). The old driver did this for every SPI transfer, which resulted in complex communication code. This driver will wait on start up for the transceiver to power up and then use RIOT's SPI API like every other driver. (Not only the data sheet states that this is fine, it also proved to be reliable in practise.) - Greatly reduced latency: The RTT on the old driver (@150 kbps data rate) was about 16ms, the new driver (@250 kbps data rate) has as RTT of ~3ms (depending on SPI clock and on CPU performance) (measured with ping6). - Increased reliability: The preamble size and the sync word size have been doubled compared to the old driver (preamble: 8 bytes instead of 4, sync word: 4 byte instead of 2). The new values are the once recommended by the data sheet for reliable communication. - Basic diagnostic during driver initialization to detect common issues as SPI communication issues and GDO pin configuration/wiring issues. - TX power configuration with netdev_driver_t::set() API-integration - Calls to netdev_driver_t::send() block until the transmission has completed to ease the use of the API (implemented without busy waiting, so that the MCU can enter lower power states or other threads can be executed).
2018-11-08 17:37:07 +01:00
/*
* Copyright (C) 2013 INRIA
* 2014 Freie Universität Berlin
* 2015 Kaspar Schleiser <kaspar@schleiser.de>
* 2018,2019 Otto-von-Guericke-Universität Magdeburg
*
* This file is subject to the terms and conditions of the GNU Lesser General
* Public License v2.1. See the file LICENSE in the top level directory for more
* details.
*/
/**
* @defgroup drivers_cc110x CC1100/CC1100e/CC1101 Sub-GHz transceiver driver
* @ingroup drivers_netdev
*
* This module contains the driver for the TI CC1100/CC110e/CC1101 Sub-GHz
* transceivers.
*
* @warning How the CC1100/CC1101 operate can be configured quite sophistically.
* This has the drawback, that configurations breaking laws and rules
* are complete possible. Please make sure that the configured output
* power, duty cycle, frequency range, etc. conform to the rules,
* standards and laws that apply in your use case.
*
*
* Supported Hardware and how to obtain
* ====================================
*
* This driver has been developed for the CC1101 and the older CC1100
* transceiver and tested for both. However, it should work with the CC1100e
* as well - but this has *NOT* been tested at all.
*
* It is suggested to go for the CC1101 when considering to buy one of the
* supported transceivers. The easiest way is to obtain CC1101 break out boards
* with a complete antenna circuit & antenna that can be connected via jumper
* wires using an 8 pin DIP pin header. These are sold in various flavours start
* from less than 2 at quantity one at your favourite Far East store. Beware
* that while the CC1101 chip can operate a various base frequencies, the
* antenna circuit will only work for a single frequency band. Most break out
* boards will operate at 433 MHz, which is license free in many countries (but
* verify that for your country before buying!). EU citizens might prefer the
* 868 MHz band over the 433 MHz, as more license free bandwidth is available
* in the 868 MHz band in the EU. (But when deploying only a few dozens of
* devices, the 433 MHz band is also fine for EU citizens.) US citizens should
* go for the 900 MHz band (as 868 MHz is not license free in the USA), which
* even contains more bandwidth than the 868 MHz band. (However, the 900 MHz
* band has not been tested, as using it would be illegal in the EU.)
*
*
* Packet Format
* =============
*
* ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
* 0 1 2 3
* 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
* +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
* | Preamble (4 bytes, handled by hardware, see MDMCFG1) |
* +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
* | Sync Word (4 bytes, handled by hardware, see MDMCFG2) |
* +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
* | Length Field | Destination | Source | Payload...
* +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
* ... (see Length Field) | CRC (handled by hardware) |
* +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
* ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
*
* | Field | Description |
* |--------------|------------------------------------------------------------|
* | Preamble | 4 bytes, handled by hardware |
* | Sync Word | 4 bytes, handled by hardware |
* | Length Field | Handled by software & hardware, length of payload + 2 |
* | Destination | Handled by software & hardware, destination MAC address |
* | Source | Handled by software only, source MAC address |
* | Payload | Handled by software only, the payload to send |
* | CRC | 2 bytes, handled by hardware |
*
* The Length Field contains the length of the driver supplied data in bytes,
* not counting the Length Field. Thus, it contains the length of the payload
* plus the length of the Destination and Source address.
*
*
* Layer-2 Addresses
* -----------------
*
* The layer 2 addresses of the CC110x transceivers is a single byte long and
* the special value `0x00` for the destination address is used in broadcast
* transmissions. The transceiver is configured by this driver to ignore all
* packets unless the destination address matches the address of the transceiver
* or the destination address is `0x00`.
*
* Please note that the layer 2 address by default is derived from the CPU ID.
* Due to the birthday paradox with only 20 devices the probability of a
* collision is already bigger than 50%. Thus, manual address assignment is
* supported by providing an implementation of @ref cc1xxx_eui_get.
drivers/cc110x: Rewrite of the cc110x driver The cc110x driver has been re-written from scratch to overcome the limitations of the old driver. The main motivation of the rewrite was to achieve better maintainability by a detailed documentation, reduce the complexity and the overhead of the SPI communication with the device, and to allow to simultaneously use transceivers with different configuration regarding the used base band, the channel bandwidth, the modulation rate, and the channel map. Features of this driver include: - Support for the CC1100, CC1101, and the CC1100e sub-gigahertz transceivers. - Detailed documentation of every aspect of this driver. - An easy to use configuration API that allows setting the transceiver configuration (modulation rate, channel bandwidth, base frequency) and the channel map. - Fast channel hopping by pre-calibration of the channels during device configuration (so that no calibration is needed during hopping). - Simplified SPI communication: Only during start-up the MCU has to wait for the transceiver to be ready (for the power regulators and the crystal to stabilize). The old driver did this for every SPI transfer, which resulted in complex communication code. This driver will wait on start up for the transceiver to power up and then use RIOT's SPI API like every other driver. (Not only the data sheet states that this is fine, it also proved to be reliable in practise.) - Greatly reduced latency: The RTT on the old driver (@150 kbps data rate) was about 16ms, the new driver (@250 kbps data rate) has as RTT of ~3ms (depending on SPI clock and on CPU performance) (measured with ping6). - Increased reliability: The preamble size and the sync word size have been doubled compared to the old driver (preamble: 8 bytes instead of 4, sync word: 4 byte instead of 2). The new values are the once recommended by the data sheet for reliable communication. - Basic diagnostic during driver initialization to detect common issues as SPI communication issues and GDO pin configuration/wiring issues. - TX power configuration with netdev_driver_t::set() API-integration - Calls to netdev_driver_t::send() block until the transmission has completed to ease the use of the API (implemented without busy waiting, so that the MCU can enter lower power states or other threads can be executed).
2018-11-08 17:37:07 +01:00
*
*
* Base Band, Data Rate, Channel Bandwidth and Channel Map Configuration
* =====================================================================
*
* This driver allows to configure the base band, the data rate and the channel
* bandwidth using an @ref cc110x_config_t data structure. Default
* configurations are supplied and name using the following scheme:
* `cc110x_config_<BASE-BAND>_<DATA-RATE>_<CHANNEL-BANDWIDTH>`. (E.g.
* @ref cc110x_config_868mhz_250kbps_300khz is the default configuration used by
* the MSB-A2 and the MSB-IoT boards.)
*
* Using the @ref cc110x_chanmap_t data structure the channel layout can be
* defined. This map contains 8 entries, each defines the offset from the base
* frequency defined in the @ref cc110x_config_t data structure for each
* channel in steps of 50kHz. E.g. @ref cc110x_chanmap_868mhz_lora provides
* the LoRa channels 10 to 17 in the 868MHz band. (The RIOT channel numbers
* will always start from 0, and currently only up to eight channels are
* supported. A special value of 255 as offset from the base frequency in the
* channel map is used mark the channel as disabled. This can be used if less
* than 8 non-overlapping channels are possible in the license free band.)
*
* Please note that the channel map (@ref cc110x_chanmap_t) must match the
* base configuration (@ref cc110x_config_t), as the channel map is relative
* to the configured base frequency. Also, the distance between the channels
* in the channel map should match the channel bandwidth of the configuration,
* as otherwise channels could overlap.
*
* Both configuration and matching channel map can be applied using
* @ref cc110x_apply_config. Please consider this as a slow operation, as the
* transceiver needs to be calibrated for each channel in the channel map.
*
*
* Calibration of the Frequency Generator
* ======================================
*
* The CC110x transceivers use a voltage controlled oscillator (VCO) and a
* phase locked loop (PLL) for frequency generation. However, they need to be
* calibrated to work correctly with the given supply voltage and the current
* temperature. The driver will perform this calibration during startup, but
* when the supply voltage or the temperature is not stable, a recalibration is
* required whenever the supply voltage of temperature has changed too much since
* the last calibration. This can be done by calling
* @ref cc110x_full_calibration. It is left to the application developer to
* perform this calibration when needed. During a test of about 2 hours of
* operation in an in-door environment with a stable temperature the CC1101 has
* worked reliable without any calibration at all (except for the automatic
* calibration at start up). So there are use cases which do not require any
* recalibration at all.
*
*
* Troubleshooting
* ===============
*
* The Driver Does Not Initialize Properly
* ---------------------------------------
* Set `ENABLE_DEBUG` in `cc110x_netdev.c` to `1` to get debug output, which
* will likely tell you what is going wrong. There are basically two things
* that can fail:
*
* Upon initialization the driver will read out the part number and version of
* the transceiver. If those do not match the ones expected for the CC1100,
* CC1100E, or the CC1101 the driver will refuse to initialize. If this fails,
* most likely incorrect values are read out and the SPI communication does not
* work correctly. However, future revisions of the CC110X transceivers might
* be produced and might have different values for the part number or version.
* If this should happen and they remain compatible with the driver, their
* part number & revision needs to be added to the driver.
*
* After uploading the configuration, the driver will read back the
* configuration to verify it. If the SPI communication is not reliable (e.g.
* sporadically bits flip), this will fail from time to time. E.g. on the
* MSB-IoT boards this is the case when the SPI interface operates at a clock of
* 5MHz, but it becomes reliable when clocked at 1MHz.
*
* The Driver Initializes, but Communication Is Impossible
* -------------------------------------------------------
* If two transceivers are too close to each other and TX power is at maximum,
* the signal is just too strong to be received correctly. Reducing TX power
* or increasing the distance (about half a meter should be fine) will solve
* this issue.
*
* While the chips can operate at any base frequency offered by the driver,
* the circuit the chip is connected to and the antenna are build for a single
* base band. Check if your configuration matches the frequency range the
* board is build for. E.g. most break out boards operate at 433MHz, but there
* are also boards for 868MHz.
*
* @{
*
* @file
* @brief Interface definition for the CC1100/CC1101 driver
*
* @author Marian Buschsieweke <marian.buschsieweke@ovgu.de>
* @author Oliver Hahm <oliver.hahm@inria.fr>
* @author Fabian Nack <nack@inf.fu-berlin.de>
* @author Kaspar Schleiser <kaspar@schleiser.de>
*/
#ifndef CC110X_H
#define CC110X_H
#include <stdint.h>
#include "cc1xxx_common.h"
#include "mutex.h"
#include "net/gnrc/nettype.h"
#include "net/netdev.h"
#include "periph/adc.h"
#include "periph/gpio.h"
#include "periph/spi.h"
#ifdef __cplusplus
extern "C" {
#endif
/**
* @brief Length of a layer 2 frame
drivers/cc110x: Rewrite of the cc110x driver The cc110x driver has been re-written from scratch to overcome the limitations of the old driver. The main motivation of the rewrite was to achieve better maintainability by a detailed documentation, reduce the complexity and the overhead of the SPI communication with the device, and to allow to simultaneously use transceivers with different configuration regarding the used base band, the channel bandwidth, the modulation rate, and the channel map. Features of this driver include: - Support for the CC1100, CC1101, and the CC1100e sub-gigahertz transceivers. - Detailed documentation of every aspect of this driver. - An easy to use configuration API that allows setting the transceiver configuration (modulation rate, channel bandwidth, base frequency) and the channel map. - Fast channel hopping by pre-calibration of the channels during device configuration (so that no calibration is needed during hopping). - Simplified SPI communication: Only during start-up the MCU has to wait for the transceiver to be ready (for the power regulators and the crystal to stabilize). The old driver did this for every SPI transfer, which resulted in complex communication code. This driver will wait on start up for the transceiver to power up and then use RIOT's SPI API like every other driver. (Not only the data sheet states that this is fine, it also proved to be reliable in practise.) - Greatly reduced latency: The RTT on the old driver (@150 kbps data rate) was about 16ms, the new driver (@250 kbps data rate) has as RTT of ~3ms (depending on SPI clock and on CPU performance) (measured with ping6). - Increased reliability: The preamble size and the sync word size have been doubled compared to the old driver (preamble: 8 bytes instead of 4, sync word: 4 byte instead of 2). The new values are the once recommended by the data sheet for reliable communication. - Basic diagnostic during driver initialization to detect common issues as SPI communication issues and GDO pin configuration/wiring issues. - TX power configuration with netdev_driver_t::set() API-integration - Calls to netdev_driver_t::send() block until the transmission has completed to ease the use of the API (implemented without busy waiting, so that the MCU can enter lower power states or other threads can be executed).
2018-11-08 17:37:07 +01:00
*
* This does not include the preamble, sync word, CRC field, and length field.
*/
#define CC110X_MAX_FRAME_SIZE 0xFF
/**
* @brief Maximum (layer 2) payload size supported by the driver
drivers/cc110x: Rewrite of the cc110x driver The cc110x driver has been re-written from scratch to overcome the limitations of the old driver. The main motivation of the rewrite was to achieve better maintainability by a detailed documentation, reduce the complexity and the overhead of the SPI communication with the device, and to allow to simultaneously use transceivers with different configuration regarding the used base band, the channel bandwidth, the modulation rate, and the channel map. Features of this driver include: - Support for the CC1100, CC1101, and the CC1100e sub-gigahertz transceivers. - Detailed documentation of every aspect of this driver. - An easy to use configuration API that allows setting the transceiver configuration (modulation rate, channel bandwidth, base frequency) and the channel map. - Fast channel hopping by pre-calibration of the channels during device configuration (so that no calibration is needed during hopping). - Simplified SPI communication: Only during start-up the MCU has to wait for the transceiver to be ready (for the power regulators and the crystal to stabilize). The old driver did this for every SPI transfer, which resulted in complex communication code. This driver will wait on start up for the transceiver to power up and then use RIOT's SPI API like every other driver. (Not only the data sheet states that this is fine, it also proved to be reliable in practise.) - Greatly reduced latency: The RTT on the old driver (@150 kbps data rate) was about 16ms, the new driver (@250 kbps data rate) has as RTT of ~3ms (depending on SPI clock and on CPU performance) (measured with ping6). - Increased reliability: The preamble size and the sync word size have been doubled compared to the old driver (preamble: 8 bytes instead of 4, sync word: 4 byte instead of 2). The new values are the once recommended by the data sheet for reliable communication. - Basic diagnostic during driver initialization to detect common issues as SPI communication issues and GDO pin configuration/wiring issues. - TX power configuration with netdev_driver_t::set() API-integration - Calls to netdev_driver_t::send() block until the transmission has completed to ease the use of the API (implemented without busy waiting, so that the MCU can enter lower power states or other threads can be executed).
2018-11-08 17:37:07 +01:00
*/
#define CC110X_MAX_PAYLOAD_SIZE (CC110X_MAX_FRAME_SIZE - CC1XXX_HEADER_SIZE)
/**
* @brief Maximum number of channels supported by the driver
drivers/cc110x: Rewrite of the cc110x driver The cc110x driver has been re-written from scratch to overcome the limitations of the old driver. The main motivation of the rewrite was to achieve better maintainability by a detailed documentation, reduce the complexity and the overhead of the SPI communication with the device, and to allow to simultaneously use transceivers with different configuration regarding the used base band, the channel bandwidth, the modulation rate, and the channel map. Features of this driver include: - Support for the CC1100, CC1101, and the CC1100e sub-gigahertz transceivers. - Detailed documentation of every aspect of this driver. - An easy to use configuration API that allows setting the transceiver configuration (modulation rate, channel bandwidth, base frequency) and the channel map. - Fast channel hopping by pre-calibration of the channels during device configuration (so that no calibration is needed during hopping). - Simplified SPI communication: Only during start-up the MCU has to wait for the transceiver to be ready (for the power regulators and the crystal to stabilize). The old driver did this for every SPI transfer, which resulted in complex communication code. This driver will wait on start up for the transceiver to power up and then use RIOT's SPI API like every other driver. (Not only the data sheet states that this is fine, it also proved to be reliable in practise.) - Greatly reduced latency: The RTT on the old driver (@150 kbps data rate) was about 16ms, the new driver (@250 kbps data rate) has as RTT of ~3ms (depending on SPI clock and on CPU performance) (measured with ping6). - Increased reliability: The preamble size and the sync word size have been doubled compared to the old driver (preamble: 8 bytes instead of 4, sync word: 4 byte instead of 2). The new values are the once recommended by the data sheet for reliable communication. - Basic diagnostic during driver initialization to detect common issues as SPI communication issues and GDO pin configuration/wiring issues. - TX power configuration with netdev_driver_t::set() API-integration - Calls to netdev_driver_t::send() block until the transmission has completed to ease the use of the API (implemented without busy waiting, so that the MCU can enter lower power states or other threads can be executed).
2018-11-08 17:37:07 +01:00
*/
#define CC110X_MAX_CHANNELS 8
/**
* @brief Default protocol for data that is coming in
drivers/cc110x: Rewrite of the cc110x driver The cc110x driver has been re-written from scratch to overcome the limitations of the old driver. The main motivation of the rewrite was to achieve better maintainability by a detailed documentation, reduce the complexity and the overhead of the SPI communication with the device, and to allow to simultaneously use transceivers with different configuration regarding the used base band, the channel bandwidth, the modulation rate, and the channel map. Features of this driver include: - Support for the CC1100, CC1101, and the CC1100e sub-gigahertz transceivers. - Detailed documentation of every aspect of this driver. - An easy to use configuration API that allows setting the transceiver configuration (modulation rate, channel bandwidth, base frequency) and the channel map. - Fast channel hopping by pre-calibration of the channels during device configuration (so that no calibration is needed during hopping). - Simplified SPI communication: Only during start-up the MCU has to wait for the transceiver to be ready (for the power regulators and the crystal to stabilize). The old driver did this for every SPI transfer, which resulted in complex communication code. This driver will wait on start up for the transceiver to power up and then use RIOT's SPI API like every other driver. (Not only the data sheet states that this is fine, it also proved to be reliable in practise.) - Greatly reduced latency: The RTT on the old driver (@150 kbps data rate) was about 16ms, the new driver (@250 kbps data rate) has as RTT of ~3ms (depending on SPI clock and on CPU performance) (measured with ping6). - Increased reliability: The preamble size and the sync word size have been doubled compared to the old driver (preamble: 8 bytes instead of 4, sync word: 4 byte instead of 2). The new values are the once recommended by the data sheet for reliable communication. - Basic diagnostic during driver initialization to detect common issues as SPI communication issues and GDO pin configuration/wiring issues. - TX power configuration with netdev_driver_t::set() API-integration - Calls to netdev_driver_t::send() block until the transmission has completed to ease the use of the API (implemented without busy waiting, so that the MCU can enter lower power states or other threads can be executed).
2018-11-08 17:37:07 +01:00
*/
#ifdef MODULE_GNRC_SIXLOWPAN
#define CC110X_DEFAULT_PROTOCOL (GNRC_NETTYPE_SIXLOWPAN)
#else
#define CC110X_DEFAULT_PROTOCOL (GNRC_NETTYPE_UNDEF)
#endif
/**
2021-03-25 13:56:22 +01:00
* @defgroup drivers_cc110x_config CC1100/CC1100e/CC1101 Sub-GHz transceiver driver
* compile time configuration
* @ingroup config_drivers_netdev
* @{
*/
/**
* @brief The default channel to set up after initializing the device
*/
#ifndef CONFIG_CC110X_DEFAULT_CHANNEL
#define CONFIG_CC110X_DEFAULT_CHANNEL (0U)
#endif
/** @} */
drivers/cc110x: Rewrite of the cc110x driver The cc110x driver has been re-written from scratch to overcome the limitations of the old driver. The main motivation of the rewrite was to achieve better maintainability by a detailed documentation, reduce the complexity and the overhead of the SPI communication with the device, and to allow to simultaneously use transceivers with different configuration regarding the used base band, the channel bandwidth, the modulation rate, and the channel map. Features of this driver include: - Support for the CC1100, CC1101, and the CC1100e sub-gigahertz transceivers. - Detailed documentation of every aspect of this driver. - An easy to use configuration API that allows setting the transceiver configuration (modulation rate, channel bandwidth, base frequency) and the channel map. - Fast channel hopping by pre-calibration of the channels during device configuration (so that no calibration is needed during hopping). - Simplified SPI communication: Only during start-up the MCU has to wait for the transceiver to be ready (for the power regulators and the crystal to stabilize). The old driver did this for every SPI transfer, which resulted in complex communication code. This driver will wait on start up for the transceiver to power up and then use RIOT's SPI API like every other driver. (Not only the data sheet states that this is fine, it also proved to be reliable in practise.) - Greatly reduced latency: The RTT on the old driver (@150 kbps data rate) was about 16ms, the new driver (@250 kbps data rate) has as RTT of ~3ms (depending on SPI clock and on CPU performance) (measured with ping6). - Increased reliability: The preamble size and the sync word size have been doubled compared to the old driver (preamble: 8 bytes instead of 4, sync word: 4 byte instead of 2). The new values are the once recommended by the data sheet for reliable communication. - Basic diagnostic during driver initialization to detect common issues as SPI communication issues and GDO pin configuration/wiring issues. - TX power configuration with netdev_driver_t::set() API-integration - Calls to netdev_driver_t::send() block until the transmission has completed to ease the use of the API (implemented without busy waiting, so that the MCU can enter lower power states or other threads can be executed).
2018-11-08 17:37:07 +01:00
/**
* @brief The state of the CC1100/CC1101 transceiver
drivers/cc110x: Rewrite of the cc110x driver The cc110x driver has been re-written from scratch to overcome the limitations of the old driver. The main motivation of the rewrite was to achieve better maintainability by a detailed documentation, reduce the complexity and the overhead of the SPI communication with the device, and to allow to simultaneously use transceivers with different configuration regarding the used base band, the channel bandwidth, the modulation rate, and the channel map. Features of this driver include: - Support for the CC1100, CC1101, and the CC1100e sub-gigahertz transceivers. - Detailed documentation of every aspect of this driver. - An easy to use configuration API that allows setting the transceiver configuration (modulation rate, channel bandwidth, base frequency) and the channel map. - Fast channel hopping by pre-calibration of the channels during device configuration (so that no calibration is needed during hopping). - Simplified SPI communication: Only during start-up the MCU has to wait for the transceiver to be ready (for the power regulators and the crystal to stabilize). The old driver did this for every SPI transfer, which resulted in complex communication code. This driver will wait on start up for the transceiver to power up and then use RIOT's SPI API like every other driver. (Not only the data sheet states that this is fine, it also proved to be reliable in practise.) - Greatly reduced latency: The RTT on the old driver (@150 kbps data rate) was about 16ms, the new driver (@250 kbps data rate) has as RTT of ~3ms (depending on SPI clock and on CPU performance) (measured with ping6). - Increased reliability: The preamble size and the sync word size have been doubled compared to the old driver (preamble: 8 bytes instead of 4, sync word: 4 byte instead of 2). The new values are the once recommended by the data sheet for reliable communication. - Basic diagnostic during driver initialization to detect common issues as SPI communication issues and GDO pin configuration/wiring issues. - TX power configuration with netdev_driver_t::set() API-integration - Calls to netdev_driver_t::send() block until the transmission has completed to ease the use of the API (implemented without busy waiting, so that the MCU can enter lower power states or other threads can be executed).
2018-11-08 17:37:07 +01:00
*
* The three least significant bytes match the representation of the matching
* transceiver state given in the status byte of the hardware. See Table 32 on
* page 31 in the data sheet for the possible states in the status byte.
*/
typedef enum {
CC110X_STATE_IDLE = 0x00, /**< IDLE state */
drivers/cc110x: Rewrite of the cc110x driver The cc110x driver has been re-written from scratch to overcome the limitations of the old driver. The main motivation of the rewrite was to achieve better maintainability by a detailed documentation, reduce the complexity and the overhead of the SPI communication with the device, and to allow to simultaneously use transceivers with different configuration regarding the used base band, the channel bandwidth, the modulation rate, and the channel map. Features of this driver include: - Support for the CC1100, CC1101, and the CC1100e sub-gigahertz transceivers. - Detailed documentation of every aspect of this driver. - An easy to use configuration API that allows setting the transceiver configuration (modulation rate, channel bandwidth, base frequency) and the channel map. - Fast channel hopping by pre-calibration of the channels during device configuration (so that no calibration is needed during hopping). - Simplified SPI communication: Only during start-up the MCU has to wait for the transceiver to be ready (for the power regulators and the crystal to stabilize). The old driver did this for every SPI transfer, which resulted in complex communication code. This driver will wait on start up for the transceiver to power up and then use RIOT's SPI API like every other driver. (Not only the data sheet states that this is fine, it also proved to be reliable in practise.) - Greatly reduced latency: The RTT on the old driver (@150 kbps data rate) was about 16ms, the new driver (@250 kbps data rate) has as RTT of ~3ms (depending on SPI clock and on CPU performance) (measured with ping6). - Increased reliability: The preamble size and the sync word size have been doubled compared to the old driver (preamble: 8 bytes instead of 4, sync word: 4 byte instead of 2). The new values are the once recommended by the data sheet for reliable communication. - Basic diagnostic during driver initialization to detect common issues as SPI communication issues and GDO pin configuration/wiring issues. - TX power configuration with netdev_driver_t::set() API-integration - Calls to netdev_driver_t::send() block until the transmission has completed to ease the use of the API (implemented without busy waiting, so that the MCU can enter lower power states or other threads can be executed).
2018-11-08 17:37:07 +01:00
/**
* @brief Frame received, waiting for upper layer to retrieve it
drivers/cc110x: Rewrite of the cc110x driver The cc110x driver has been re-written from scratch to overcome the limitations of the old driver. The main motivation of the rewrite was to achieve better maintainability by a detailed documentation, reduce the complexity and the overhead of the SPI communication with the device, and to allow to simultaneously use transceivers with different configuration regarding the used base band, the channel bandwidth, the modulation rate, and the channel map. Features of this driver include: - Support for the CC1100, CC1101, and the CC1100e sub-gigahertz transceivers. - Detailed documentation of every aspect of this driver. - An easy to use configuration API that allows setting the transceiver configuration (modulation rate, channel bandwidth, base frequency) and the channel map. - Fast channel hopping by pre-calibration of the channels during device configuration (so that no calibration is needed during hopping). - Simplified SPI communication: Only during start-up the MCU has to wait for the transceiver to be ready (for the power regulators and the crystal to stabilize). The old driver did this for every SPI transfer, which resulted in complex communication code. This driver will wait on start up for the transceiver to power up and then use RIOT's SPI API like every other driver. (Not only the data sheet states that this is fine, it also proved to be reliable in practise.) - Greatly reduced latency: The RTT on the old driver (@150 kbps data rate) was about 16ms, the new driver (@250 kbps data rate) has as RTT of ~3ms (depending on SPI clock and on CPU performance) (measured with ping6). - Increased reliability: The preamble size and the sync word size have been doubled compared to the old driver (preamble: 8 bytes instead of 4, sync word: 4 byte instead of 2). The new values are the once recommended by the data sheet for reliable communication. - Basic diagnostic during driver initialization to detect common issues as SPI communication issues and GDO pin configuration/wiring issues. - TX power configuration with netdev_driver_t::set() API-integration - Calls to netdev_driver_t::send() block until the transmission has completed to ease the use of the API (implemented without busy waiting, so that the MCU can enter lower power states or other threads can be executed).
2018-11-08 17:37:07 +01:00
*
* Transceiver is in IDLE state.
*/
CC110X_STATE_FRAME_READY = 0x08,
drivers/cc110x: Rewrite of the cc110x driver The cc110x driver has been re-written from scratch to overcome the limitations of the old driver. The main motivation of the rewrite was to achieve better maintainability by a detailed documentation, reduce the complexity and the overhead of the SPI communication with the device, and to allow to simultaneously use transceivers with different configuration regarding the used base band, the channel bandwidth, the modulation rate, and the channel map. Features of this driver include: - Support for the CC1100, CC1101, and the CC1100e sub-gigahertz transceivers. - Detailed documentation of every aspect of this driver. - An easy to use configuration API that allows setting the transceiver configuration (modulation rate, channel bandwidth, base frequency) and the channel map. - Fast channel hopping by pre-calibration of the channels during device configuration (so that no calibration is needed during hopping). - Simplified SPI communication: Only during start-up the MCU has to wait for the transceiver to be ready (for the power regulators and the crystal to stabilize). The old driver did this for every SPI transfer, which resulted in complex communication code. This driver will wait on start up for the transceiver to power up and then use RIOT's SPI API like every other driver. (Not only the data sheet states that this is fine, it also proved to be reliable in practise.) - Greatly reduced latency: The RTT on the old driver (@150 kbps data rate) was about 16ms, the new driver (@250 kbps data rate) has as RTT of ~3ms (depending on SPI clock and on CPU performance) (measured with ping6). - Increased reliability: The preamble size and the sync word size have been doubled compared to the old driver (preamble: 8 bytes instead of 4, sync word: 4 byte instead of 2). The new values are the once recommended by the data sheet for reliable communication. - Basic diagnostic during driver initialization to detect common issues as SPI communication issues and GDO pin configuration/wiring issues. - TX power configuration with netdev_driver_t::set() API-integration - Calls to netdev_driver_t::send() block until the transmission has completed to ease the use of the API (implemented without busy waiting, so that the MCU can enter lower power states or other threads can be executed).
2018-11-08 17:37:07 +01:00
/**
* @brief Devices is powered down
drivers/cc110x: Rewrite of the cc110x driver The cc110x driver has been re-written from scratch to overcome the limitations of the old driver. The main motivation of the rewrite was to achieve better maintainability by a detailed documentation, reduce the complexity and the overhead of the SPI communication with the device, and to allow to simultaneously use transceivers with different configuration regarding the used base band, the channel bandwidth, the modulation rate, and the channel map. Features of this driver include: - Support for the CC1100, CC1101, and the CC1100e sub-gigahertz transceivers. - Detailed documentation of every aspect of this driver. - An easy to use configuration API that allows setting the transceiver configuration (modulation rate, channel bandwidth, base frequency) and the channel map. - Fast channel hopping by pre-calibration of the channels during device configuration (so that no calibration is needed during hopping). - Simplified SPI communication: Only during start-up the MCU has to wait for the transceiver to be ready (for the power regulators and the crystal to stabilize). The old driver did this for every SPI transfer, which resulted in complex communication code. This driver will wait on start up for the transceiver to power up and then use RIOT's SPI API like every other driver. (Not only the data sheet states that this is fine, it also proved to be reliable in practise.) - Greatly reduced latency: The RTT on the old driver (@150 kbps data rate) was about 16ms, the new driver (@250 kbps data rate) has as RTT of ~3ms (depending on SPI clock and on CPU performance) (measured with ping6). - Increased reliability: The preamble size and the sync word size have been doubled compared to the old driver (preamble: 8 bytes instead of 4, sync word: 4 byte instead of 2). The new values are the once recommended by the data sheet for reliable communication. - Basic diagnostic during driver initialization to detect common issues as SPI communication issues and GDO pin configuration/wiring issues. - TX power configuration with netdev_driver_t::set() API-integration - Calls to netdev_driver_t::send() block until the transmission has completed to ease the use of the API (implemented without busy waiting, so that the MCU can enter lower power states or other threads can be executed).
2018-11-08 17:37:07 +01:00
*
* Transceiver is in SLEEP state. There is no matching representation in the
* status byte, as reading the status byte will power up the transceiver in
* bring it in the IDLE state. Thus, we set the three least significant bits
* to the IDLE state
*/
CC110X_STATE_OFF = 0x10,
CC110X_STATE_RX_MODE = 0x01, /**< Listening for frames */
drivers/cc110x: Rewrite of the cc110x driver The cc110x driver has been re-written from scratch to overcome the limitations of the old driver. The main motivation of the rewrite was to achieve better maintainability by a detailed documentation, reduce the complexity and the overhead of the SPI communication with the device, and to allow to simultaneously use transceivers with different configuration regarding the used base band, the channel bandwidth, the modulation rate, and the channel map. Features of this driver include: - Support for the CC1100, CC1101, and the CC1100e sub-gigahertz transceivers. - Detailed documentation of every aspect of this driver. - An easy to use configuration API that allows setting the transceiver configuration (modulation rate, channel bandwidth, base frequency) and the channel map. - Fast channel hopping by pre-calibration of the channels during device configuration (so that no calibration is needed during hopping). - Simplified SPI communication: Only during start-up the MCU has to wait for the transceiver to be ready (for the power regulators and the crystal to stabilize). The old driver did this for every SPI transfer, which resulted in complex communication code. This driver will wait on start up for the transceiver to power up and then use RIOT's SPI API like every other driver. (Not only the data sheet states that this is fine, it also proved to be reliable in practise.) - Greatly reduced latency: The RTT on the old driver (@150 kbps data rate) was about 16ms, the new driver (@250 kbps data rate) has as RTT of ~3ms (depending on SPI clock and on CPU performance) (measured with ping6). - Increased reliability: The preamble size and the sync word size have been doubled compared to the old driver (preamble: 8 bytes instead of 4, sync word: 4 byte instead of 2). The new values are the once recommended by the data sheet for reliable communication. - Basic diagnostic during driver initialization to detect common issues as SPI communication issues and GDO pin configuration/wiring issues. - TX power configuration with netdev_driver_t::set() API-integration - Calls to netdev_driver_t::send() block until the transmission has completed to ease the use of the API (implemented without busy waiting, so that the MCU can enter lower power states or other threads can be executed).
2018-11-08 17:37:07 +01:00
/**
* @brief Receiving a frame just now
drivers/cc110x: Rewrite of the cc110x driver The cc110x driver has been re-written from scratch to overcome the limitations of the old driver. The main motivation of the rewrite was to achieve better maintainability by a detailed documentation, reduce the complexity and the overhead of the SPI communication with the device, and to allow to simultaneously use transceivers with different configuration regarding the used base band, the channel bandwidth, the modulation rate, and the channel map. Features of this driver include: - Support for the CC1100, CC1101, and the CC1100e sub-gigahertz transceivers. - Detailed documentation of every aspect of this driver. - An easy to use configuration API that allows setting the transceiver configuration (modulation rate, channel bandwidth, base frequency) and the channel map. - Fast channel hopping by pre-calibration of the channels during device configuration (so that no calibration is needed during hopping). - Simplified SPI communication: Only during start-up the MCU has to wait for the transceiver to be ready (for the power regulators and the crystal to stabilize). The old driver did this for every SPI transfer, which resulted in complex communication code. This driver will wait on start up for the transceiver to power up and then use RIOT's SPI API like every other driver. (Not only the data sheet states that this is fine, it also proved to be reliable in practise.) - Greatly reduced latency: The RTT on the old driver (@150 kbps data rate) was about 16ms, the new driver (@250 kbps data rate) has as RTT of ~3ms (depending on SPI clock and on CPU performance) (measured with ping6). - Increased reliability: The preamble size and the sync word size have been doubled compared to the old driver (preamble: 8 bytes instead of 4, sync word: 4 byte instead of 2). The new values are the once recommended by the data sheet for reliable communication. - Basic diagnostic during driver initialization to detect common issues as SPI communication issues and GDO pin configuration/wiring issues. - TX power configuration with netdev_driver_t::set() API-integration - Calls to netdev_driver_t::send() block until the transmission has completed to ease the use of the API (implemented without busy waiting, so that the MCU can enter lower power states or other threads can be executed).
2018-11-08 17:37:07 +01:00
*
* Transceiver is in RX state.
*/
CC110X_STATE_RECEIVING = 0x09,
CC110X_STATE_TX_MODE = 0x02, /**< Transmit mode */
drivers/cc110x: Rewrite of the cc110x driver The cc110x driver has been re-written from scratch to overcome the limitations of the old driver. The main motivation of the rewrite was to achieve better maintainability by a detailed documentation, reduce the complexity and the overhead of the SPI communication with the device, and to allow to simultaneously use transceivers with different configuration regarding the used base band, the channel bandwidth, the modulation rate, and the channel map. Features of this driver include: - Support for the CC1100, CC1101, and the CC1100e sub-gigahertz transceivers. - Detailed documentation of every aspect of this driver. - An easy to use configuration API that allows setting the transceiver configuration (modulation rate, channel bandwidth, base frequency) and the channel map. - Fast channel hopping by pre-calibration of the channels during device configuration (so that no calibration is needed during hopping). - Simplified SPI communication: Only during start-up the MCU has to wait for the transceiver to be ready (for the power regulators and the crystal to stabilize). The old driver did this for every SPI transfer, which resulted in complex communication code. This driver will wait on start up for the transceiver to power up and then use RIOT's SPI API like every other driver. (Not only the data sheet states that this is fine, it also proved to be reliable in practise.) - Greatly reduced latency: The RTT on the old driver (@150 kbps data rate) was about 16ms, the new driver (@250 kbps data rate) has as RTT of ~3ms (depending on SPI clock and on CPU performance) (measured with ping6). - Increased reliability: The preamble size and the sync word size have been doubled compared to the old driver (preamble: 8 bytes instead of 4, sync word: 4 byte instead of 2). The new values are the once recommended by the data sheet for reliable communication. - Basic diagnostic during driver initialization to detect common issues as SPI communication issues and GDO pin configuration/wiring issues. - TX power configuration with netdev_driver_t::set() API-integration - Calls to netdev_driver_t::send() block until the transmission has completed to ease the use of the API (implemented without busy waiting, so that the MCU can enter lower power states or other threads can be executed).
2018-11-08 17:37:07 +01:00
/**
* @brief Waiting for transceiver to complete outgoing transmission
drivers/cc110x: Rewrite of the cc110x driver The cc110x driver has been re-written from scratch to overcome the limitations of the old driver. The main motivation of the rewrite was to achieve better maintainability by a detailed documentation, reduce the complexity and the overhead of the SPI communication with the device, and to allow to simultaneously use transceivers with different configuration regarding the used base band, the channel bandwidth, the modulation rate, and the channel map. Features of this driver include: - Support for the CC1100, CC1101, and the CC1100e sub-gigahertz transceivers. - Detailed documentation of every aspect of this driver. - An easy to use configuration API that allows setting the transceiver configuration (modulation rate, channel bandwidth, base frequency) and the channel map. - Fast channel hopping by pre-calibration of the channels during device configuration (so that no calibration is needed during hopping). - Simplified SPI communication: Only during start-up the MCU has to wait for the transceiver to be ready (for the power regulators and the crystal to stabilize). The old driver did this for every SPI transfer, which resulted in complex communication code. This driver will wait on start up for the transceiver to power up and then use RIOT's SPI API like every other driver. (Not only the data sheet states that this is fine, it also proved to be reliable in practise.) - Greatly reduced latency: The RTT on the old driver (@150 kbps data rate) was about 16ms, the new driver (@250 kbps data rate) has as RTT of ~3ms (depending on SPI clock and on CPU performance) (measured with ping6). - Increased reliability: The preamble size and the sync word size have been doubled compared to the old driver (preamble: 8 bytes instead of 4, sync word: 4 byte instead of 2). The new values are the once recommended by the data sheet for reliable communication. - Basic diagnostic during driver initialization to detect common issues as SPI communication issues and GDO pin configuration/wiring issues. - TX power configuration with netdev_driver_t::set() API-integration - Calls to netdev_driver_t::send() block until the transmission has completed to ease the use of the API (implemented without busy waiting, so that the MCU can enter lower power states or other threads can be executed).
2018-11-08 17:37:07 +01:00
*
* Transceiver is in TX state
*/
CC110X_STATE_TX_COMPLETING = 0x0A,
CC110X_STATE_FSTXON = 0x03, /**< Fast TX ready */
CC110X_STATE_CALIBRATE = 0x04, /**< Device is calibrating */
CC110X_STATE_SETTLING = 0x05, /**< PLL is settling */
CC110X_STATE_RXFIFO_OVERFLOW = 0x06, /**< RX FIFO overflown */
CC110X_STATE_TXFIFO_UNDERFLOW = 0x07, /**< TX FIFO underflown */
drivers/cc110x: Rewrite of the cc110x driver The cc110x driver has been re-written from scratch to overcome the limitations of the old driver. The main motivation of the rewrite was to achieve better maintainability by a detailed documentation, reduce the complexity and the overhead of the SPI communication with the device, and to allow to simultaneously use transceivers with different configuration regarding the used base band, the channel bandwidth, the modulation rate, and the channel map. Features of this driver include: - Support for the CC1100, CC1101, and the CC1100e sub-gigahertz transceivers. - Detailed documentation of every aspect of this driver. - An easy to use configuration API that allows setting the transceiver configuration (modulation rate, channel bandwidth, base frequency) and the channel map. - Fast channel hopping by pre-calibration of the channels during device configuration (so that no calibration is needed during hopping). - Simplified SPI communication: Only during start-up the MCU has to wait for the transceiver to be ready (for the power regulators and the crystal to stabilize). The old driver did this for every SPI transfer, which resulted in complex communication code. This driver will wait on start up for the transceiver to power up and then use RIOT's SPI API like every other driver. (Not only the data sheet states that this is fine, it also proved to be reliable in practise.) - Greatly reduced latency: The RTT on the old driver (@150 kbps data rate) was about 16ms, the new driver (@250 kbps data rate) has as RTT of ~3ms (depending on SPI clock and on CPU performance) (measured with ping6). - Increased reliability: The preamble size and the sync word size have been doubled compared to the old driver (preamble: 8 bytes instead of 4, sync word: 4 byte instead of 2). The new values are the once recommended by the data sheet for reliable communication. - Basic diagnostic during driver initialization to detect common issues as SPI communication issues and GDO pin configuration/wiring issues. - TX power configuration with netdev_driver_t::set() API-integration - Calls to netdev_driver_t::send() block until the transmission has completed to ease the use of the API (implemented without busy waiting, so that the MCU can enter lower power states or other threads can be executed).
2018-11-08 17:37:07 +01:00
} cc110x_state_t;
/**
* @brief Enumeration over the possible TX power settings the driver offers
*/
typedef enum {
CC110X_TX_POWER_MINUS_30_DBM, /**< -30 dBm */
CC110X_TX_POWER_MINUS_20_DBM, /**< -20 dBm */
CC110X_TX_POWER_MINUS_15_DBM, /**< -15 dBm */
CC110X_TX_POWER_MINUS_10_DBM, /**< -10 dBm */
CC110X_TX_POWER_0_DBM, /**< 0 dBm */
CC110X_TX_POWER_PLUS_5_DBM, /**< 5 dBm */
CC110X_TX_POWER_PLUS_7_DBM, /**< 7 dBm */
2020-01-23 14:10:40 +01:00
CC110X_TX_POWER_PLUS_10_DBM, /**< 10 dBm */
drivers/cc110x: Rewrite of the cc110x driver The cc110x driver has been re-written from scratch to overcome the limitations of the old driver. The main motivation of the rewrite was to achieve better maintainability by a detailed documentation, reduce the complexity and the overhead of the SPI communication with the device, and to allow to simultaneously use transceivers with different configuration regarding the used base band, the channel bandwidth, the modulation rate, and the channel map. Features of this driver include: - Support for the CC1100, CC1101, and the CC1100e sub-gigahertz transceivers. - Detailed documentation of every aspect of this driver. - An easy to use configuration API that allows setting the transceiver configuration (modulation rate, channel bandwidth, base frequency) and the channel map. - Fast channel hopping by pre-calibration of the channels during device configuration (so that no calibration is needed during hopping). - Simplified SPI communication: Only during start-up the MCU has to wait for the transceiver to be ready (for the power regulators and the crystal to stabilize). The old driver did this for every SPI transfer, which resulted in complex communication code. This driver will wait on start up for the transceiver to power up and then use RIOT's SPI API like every other driver. (Not only the data sheet states that this is fine, it also proved to be reliable in practise.) - Greatly reduced latency: The RTT on the old driver (@150 kbps data rate) was about 16ms, the new driver (@250 kbps data rate) has as RTT of ~3ms (depending on SPI clock and on CPU performance) (measured with ping6). - Increased reliability: The preamble size and the sync word size have been doubled compared to the old driver (preamble: 8 bytes instead of 4, sync word: 4 byte instead of 2). The new values are the once recommended by the data sheet for reliable communication. - Basic diagnostic during driver initialization to detect common issues as SPI communication issues and GDO pin configuration/wiring issues. - TX power configuration with netdev_driver_t::set() API-integration - Calls to netdev_driver_t::send() block until the transmission has completed to ease the use of the API (implemented without busy waiting, so that the MCU can enter lower power states or other threads can be executed).
2018-11-08 17:37:07 +01:00
CC110X_TX_POWER_NUMOF, /**< Number of TX power options */
} cc110x_tx_power_t;
/**
* @brief Structure that holds the PATABLE, which allows to configure the
* 8 available output power levels using a magic number for each level.
drivers/cc110x: Rewrite of the cc110x driver The cc110x driver has been re-written from scratch to overcome the limitations of the old driver. The main motivation of the rewrite was to achieve better maintainability by a detailed documentation, reduce the complexity and the overhead of the SPI communication with the device, and to allow to simultaneously use transceivers with different configuration regarding the used base band, the channel bandwidth, the modulation rate, and the channel map. Features of this driver include: - Support for the CC1100, CC1101, and the CC1100e sub-gigahertz transceivers. - Detailed documentation of every aspect of this driver. - An easy to use configuration API that allows setting the transceiver configuration (modulation rate, channel bandwidth, base frequency) and the channel map. - Fast channel hopping by pre-calibration of the channels during device configuration (so that no calibration is needed during hopping). - Simplified SPI communication: Only during start-up the MCU has to wait for the transceiver to be ready (for the power regulators and the crystal to stabilize). The old driver did this for every SPI transfer, which resulted in complex communication code. This driver will wait on start up for the transceiver to power up and then use RIOT's SPI API like every other driver. (Not only the data sheet states that this is fine, it also proved to be reliable in practise.) - Greatly reduced latency: The RTT on the old driver (@150 kbps data rate) was about 16ms, the new driver (@250 kbps data rate) has as RTT of ~3ms (depending on SPI clock and on CPU performance) (measured with ping6). - Increased reliability: The preamble size and the sync word size have been doubled compared to the old driver (preamble: 8 bytes instead of 4, sync word: 4 byte instead of 2). The new values are the once recommended by the data sheet for reliable communication. - Basic diagnostic during driver initialization to detect common issues as SPI communication issues and GDO pin configuration/wiring issues. - TX power configuration with netdev_driver_t::set() API-integration - Calls to netdev_driver_t::send() block until the transmission has completed to ease the use of the API (implemented without busy waiting, so that the MCU can enter lower power states or other threads can be executed).
2018-11-08 17:37:07 +01:00
*
* See Section "24 Output Power Programming" on page 59ff in the data sheet.
* The values suggested in Table 39 on page 60 in the data sheet are already
* available by this driver, but will only be linked in (8 bytes of ROM size)
* when they are referenced.
*
* @see cc110x_patable_433mhz
* @see cc110x_patable_868mhz
* @see cc110x_patable_915mhz
*/
typedef struct {
uint8_t data[8]; /**< Magic number to store in the configuration register */
} cc110x_patable_t;
/**
* @brief Configuration of the transceiver to use
drivers/cc110x: Rewrite of the cc110x driver The cc110x driver has been re-written from scratch to overcome the limitations of the old driver. The main motivation of the rewrite was to achieve better maintainability by a detailed documentation, reduce the complexity and the overhead of the SPI communication with the device, and to allow to simultaneously use transceivers with different configuration regarding the used base band, the channel bandwidth, the modulation rate, and the channel map. Features of this driver include: - Support for the CC1100, CC1101, and the CC1100e sub-gigahertz transceivers. - Detailed documentation of every aspect of this driver. - An easy to use configuration API that allows setting the transceiver configuration (modulation rate, channel bandwidth, base frequency) and the channel map. - Fast channel hopping by pre-calibration of the channels during device configuration (so that no calibration is needed during hopping). - Simplified SPI communication: Only during start-up the MCU has to wait for the transceiver to be ready (for the power regulators and the crystal to stabilize). The old driver did this for every SPI transfer, which resulted in complex communication code. This driver will wait on start up for the transceiver to power up and then use RIOT's SPI API like every other driver. (Not only the data sheet states that this is fine, it also proved to be reliable in practise.) - Greatly reduced latency: The RTT on the old driver (@150 kbps data rate) was about 16ms, the new driver (@250 kbps data rate) has as RTT of ~3ms (depending on SPI clock and on CPU performance) (measured with ping6). - Increased reliability: The preamble size and the sync word size have been doubled compared to the old driver (preamble: 8 bytes instead of 4, sync word: 4 byte instead of 2). The new values are the once recommended by the data sheet for reliable communication. - Basic diagnostic during driver initialization to detect common issues as SPI communication issues and GDO pin configuration/wiring issues. - TX power configuration with netdev_driver_t::set() API-integration - Calls to netdev_driver_t::send() block until the transmission has completed to ease the use of the API (implemented without busy waiting, so that the MCU can enter lower power states or other threads can be executed).
2018-11-08 17:37:07 +01:00
*
* @warning Two transceivers with different configurations will be unable
* to communicate.
*
* The data uploaded into configuration registers are stored in
* @ref cc110x_conf. Most of them cannot be changed, as the driver relies on
* their values. However, the base frequency, the symbol rate (which equals
* the bit rate for the chosen modulation and error correction) and the
* channel bandwidth can be configured using this data structure.
*
* Please note that while the CC1100/CC1101 chip is compatible with a huge
* frequency range (300 MHz - 928 MHz), the complete circuit is optimized to
* a narrow frequency band. So make sure the configured base frequency is within
* that frequency band that is compatible with that circuit. (Most break out
* board will operate at the 433 MHz band. In the EU the 868 MHz band would be
* more interesting, but 433 MHz is license free as well. In the USA the 915 MHz
* band is license free.
*
* Please verify that the driver is configured in a way that allows legal
* operation according to rules and laws that apply for you.
*/
typedef struct {
uint8_t base_freq[3]; /**< Base frequency to use */
/**
* @brief FSCTRL1 configuration register value that affects the
* intermediate frequency of the transceiver to use
* @note The 4 most significant bits have to be 0.
drivers/cc110x: Rewrite of the cc110x driver The cc110x driver has been re-written from scratch to overcome the limitations of the old driver. The main motivation of the rewrite was to achieve better maintainability by a detailed documentation, reduce the complexity and the overhead of the SPI communication with the device, and to allow to simultaneously use transceivers with different configuration regarding the used base band, the channel bandwidth, the modulation rate, and the channel map. Features of this driver include: - Support for the CC1100, CC1101, and the CC1100e sub-gigahertz transceivers. - Detailed documentation of every aspect of this driver. - An easy to use configuration API that allows setting the transceiver configuration (modulation rate, channel bandwidth, base frequency) and the channel map. - Fast channel hopping by pre-calibration of the channels during device configuration (so that no calibration is needed during hopping). - Simplified SPI communication: Only during start-up the MCU has to wait for the transceiver to be ready (for the power regulators and the crystal to stabilize). The old driver did this for every SPI transfer, which resulted in complex communication code. This driver will wait on start up for the transceiver to power up and then use RIOT's SPI API like every other driver. (Not only the data sheet states that this is fine, it also proved to be reliable in practise.) - Greatly reduced latency: The RTT on the old driver (@150 kbps data rate) was about 16ms, the new driver (@250 kbps data rate) has as RTT of ~3ms (depending on SPI clock and on CPU performance) (measured with ping6). - Increased reliability: The preamble size and the sync word size have been doubled compared to the old driver (preamble: 8 bytes instead of 4, sync word: 4 byte instead of 2). The new values are the once recommended by the data sheet for reliable communication. - Basic diagnostic during driver initialization to detect common issues as SPI communication issues and GDO pin configuration/wiring issues. - TX power configuration with netdev_driver_t::set() API-integration - Calls to netdev_driver_t::send() block until the transmission has completed to ease the use of the API (implemented without busy waiting, so that the MCU can enter lower power states or other threads can be executed).
2018-11-08 17:37:07 +01:00
*
* Assuming a 26 MHz crystal the IF is calculated as follows (in kHz):
*
* ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
* double intermediate_frequency = 26000 / 1024 * fsctrl1;
* ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
*/
uint8_t fsctrl1;
/**
* @brief MDMCFG4 configuration register value that affects channel filter
* bandwidth and the data rate
drivers/cc110x: Rewrite of the cc110x driver The cc110x driver has been re-written from scratch to overcome the limitations of the old driver. The main motivation of the rewrite was to achieve better maintainability by a detailed documentation, reduce the complexity and the overhead of the SPI communication with the device, and to allow to simultaneously use transceivers with different configuration regarding the used base band, the channel bandwidth, the modulation rate, and the channel map. Features of this driver include: - Support for the CC1100, CC1101, and the CC1100e sub-gigahertz transceivers. - Detailed documentation of every aspect of this driver. - An easy to use configuration API that allows setting the transceiver configuration (modulation rate, channel bandwidth, base frequency) and the channel map. - Fast channel hopping by pre-calibration of the channels during device configuration (so that no calibration is needed during hopping). - Simplified SPI communication: Only during start-up the MCU has to wait for the transceiver to be ready (for the power regulators and the crystal to stabilize). The old driver did this for every SPI transfer, which resulted in complex communication code. This driver will wait on start up for the transceiver to power up and then use RIOT's SPI API like every other driver. (Not only the data sheet states that this is fine, it also proved to be reliable in practise.) - Greatly reduced latency: The RTT on the old driver (@150 kbps data rate) was about 16ms, the new driver (@250 kbps data rate) has as RTT of ~3ms (depending on SPI clock and on CPU performance) (measured with ping6). - Increased reliability: The preamble size and the sync word size have been doubled compared to the old driver (preamble: 8 bytes instead of 4, sync word: 4 byte instead of 2). The new values are the once recommended by the data sheet for reliable communication. - Basic diagnostic during driver initialization to detect common issues as SPI communication issues and GDO pin configuration/wiring issues. - TX power configuration with netdev_driver_t::set() API-integration - Calls to netdev_driver_t::send() block until the transmission has completed to ease the use of the API (implemented without busy waiting, so that the MCU can enter lower power states or other threads can be executed).
2018-11-08 17:37:07 +01:00
*
* See page 76 in the data sheet.
*
* Assuming a 26 MHz crystal the channel filter bandwidth is calculated
* as follows (in kHz):
*
* ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
* uint8_t exponent = mdmcfg4 >> 6;
* uint8_t mantissa = (mdmcfg4 >> 4) & 0x03;
* double bandwidth = 26000.0 / (8 * (4 + mantissa) * (1L << exponent));
* ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
*/
uint8_t mdmcfg4;
/**
* @brief MDMCFG3 configuration register value that affects the data rate
drivers/cc110x: Rewrite of the cc110x driver The cc110x driver has been re-written from scratch to overcome the limitations of the old driver. The main motivation of the rewrite was to achieve better maintainability by a detailed documentation, reduce the complexity and the overhead of the SPI communication with the device, and to allow to simultaneously use transceivers with different configuration regarding the used base band, the channel bandwidth, the modulation rate, and the channel map. Features of this driver include: - Support for the CC1100, CC1101, and the CC1100e sub-gigahertz transceivers. - Detailed documentation of every aspect of this driver. - An easy to use configuration API that allows setting the transceiver configuration (modulation rate, channel bandwidth, base frequency) and the channel map. - Fast channel hopping by pre-calibration of the channels during device configuration (so that no calibration is needed during hopping). - Simplified SPI communication: Only during start-up the MCU has to wait for the transceiver to be ready (for the power regulators and the crystal to stabilize). The old driver did this for every SPI transfer, which resulted in complex communication code. This driver will wait on start up for the transceiver to power up and then use RIOT's SPI API like every other driver. (Not only the data sheet states that this is fine, it also proved to be reliable in practise.) - Greatly reduced latency: The RTT on the old driver (@150 kbps data rate) was about 16ms, the new driver (@250 kbps data rate) has as RTT of ~3ms (depending on SPI clock and on CPU performance) (measured with ping6). - Increased reliability: The preamble size and the sync word size have been doubled compared to the old driver (preamble: 8 bytes instead of 4, sync word: 4 byte instead of 2). The new values are the once recommended by the data sheet for reliable communication. - Basic diagnostic during driver initialization to detect common issues as SPI communication issues and GDO pin configuration/wiring issues. - TX power configuration with netdev_driver_t::set() API-integration - Calls to netdev_driver_t::send() block until the transmission has completed to ease the use of the API (implemented without busy waiting, so that the MCU can enter lower power states or other threads can be executed).
2018-11-08 17:37:07 +01:00
*
* @see cc110x_config_t::mdmcfg4
drivers/cc110x: Rewrite of the cc110x driver The cc110x driver has been re-written from scratch to overcome the limitations of the old driver. The main motivation of the rewrite was to achieve better maintainability by a detailed documentation, reduce the complexity and the overhead of the SPI communication with the device, and to allow to simultaneously use transceivers with different configuration regarding the used base band, the channel bandwidth, the modulation rate, and the channel map. Features of this driver include: - Support for the CC1100, CC1101, and the CC1100e sub-gigahertz transceivers. - Detailed documentation of every aspect of this driver. - An easy to use configuration API that allows setting the transceiver configuration (modulation rate, channel bandwidth, base frequency) and the channel map. - Fast channel hopping by pre-calibration of the channels during device configuration (so that no calibration is needed during hopping). - Simplified SPI communication: Only during start-up the MCU has to wait for the transceiver to be ready (for the power regulators and the crystal to stabilize). The old driver did this for every SPI transfer, which resulted in complex communication code. This driver will wait on start up for the transceiver to power up and then use RIOT's SPI API like every other driver. (Not only the data sheet states that this is fine, it also proved to be reliable in practise.) - Greatly reduced latency: The RTT on the old driver (@150 kbps data rate) was about 16ms, the new driver (@250 kbps data rate) has as RTT of ~3ms (depending on SPI clock and on CPU performance) (measured with ping6). - Increased reliability: The preamble size and the sync word size have been doubled compared to the old driver (preamble: 8 bytes instead of 4, sync word: 4 byte instead of 2). The new values are the once recommended by the data sheet for reliable communication. - Basic diagnostic during driver initialization to detect common issues as SPI communication issues and GDO pin configuration/wiring issues. - TX power configuration with netdev_driver_t::set() API-integration - Calls to netdev_driver_t::send() block until the transmission has completed to ease the use of the API (implemented without busy waiting, so that the MCU can enter lower power states or other threads can be executed).
2018-11-08 17:37:07 +01:00
*
* See page 76 in the data sheet.
*
* Assuming a 26 MHz crystal the symbol rate of the transceiver is calculated
2019-10-23 21:25:51 +02:00
* as follows (in kBaud):
drivers/cc110x: Rewrite of the cc110x driver The cc110x driver has been re-written from scratch to overcome the limitations of the old driver. The main motivation of the rewrite was to achieve better maintainability by a detailed documentation, reduce the complexity and the overhead of the SPI communication with the device, and to allow to simultaneously use transceivers with different configuration regarding the used base band, the channel bandwidth, the modulation rate, and the channel map. Features of this driver include: - Support for the CC1100, CC1101, and the CC1100e sub-gigahertz transceivers. - Detailed documentation of every aspect of this driver. - An easy to use configuration API that allows setting the transceiver configuration (modulation rate, channel bandwidth, base frequency) and the channel map. - Fast channel hopping by pre-calibration of the channels during device configuration (so that no calibration is needed during hopping). - Simplified SPI communication: Only during start-up the MCU has to wait for the transceiver to be ready (for the power regulators and the crystal to stabilize). The old driver did this for every SPI transfer, which resulted in complex communication code. This driver will wait on start up for the transceiver to power up and then use RIOT's SPI API like every other driver. (Not only the data sheet states that this is fine, it also proved to be reliable in practise.) - Greatly reduced latency: The RTT on the old driver (@150 kbps data rate) was about 16ms, the new driver (@250 kbps data rate) has as RTT of ~3ms (depending on SPI clock and on CPU performance) (measured with ping6). - Increased reliability: The preamble size and the sync word size have been doubled compared to the old driver (preamble: 8 bytes instead of 4, sync word: 4 byte instead of 2). The new values are the once recommended by the data sheet for reliable communication. - Basic diagnostic during driver initialization to detect common issues as SPI communication issues and GDO pin configuration/wiring issues. - TX power configuration with netdev_driver_t::set() API-integration - Calls to netdev_driver_t::send() block until the transmission has completed to ease the use of the API (implemented without busy waiting, so that the MCU can enter lower power states or other threads can be executed).
2018-11-08 17:37:07 +01:00
* ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
* uint8_t exponent = mdmcfg4 & 0x0f;
* int32_t mantissa = mdmcfg3;
* double baudrate = (256 + mantissa) * 26000.0 / (1L << (28 - exponent));
* ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
*/
uint8_t mdmcfg3;
/**
* @brief DEVIANT configuration register that affects the amount by which
* the radio frequency is shifted in FSK/GFSK modulation
drivers/cc110x: Rewrite of the cc110x driver The cc110x driver has been re-written from scratch to overcome the limitations of the old driver. The main motivation of the rewrite was to achieve better maintainability by a detailed documentation, reduce the complexity and the overhead of the SPI communication with the device, and to allow to simultaneously use transceivers with different configuration regarding the used base band, the channel bandwidth, the modulation rate, and the channel map. Features of this driver include: - Support for the CC1100, CC1101, and the CC1100e sub-gigahertz transceivers. - Detailed documentation of every aspect of this driver. - An easy to use configuration API that allows setting the transceiver configuration (modulation rate, channel bandwidth, base frequency) and the channel map. - Fast channel hopping by pre-calibration of the channels during device configuration (so that no calibration is needed during hopping). - Simplified SPI communication: Only during start-up the MCU has to wait for the transceiver to be ready (for the power regulators and the crystal to stabilize). The old driver did this for every SPI transfer, which resulted in complex communication code. This driver will wait on start up for the transceiver to power up and then use RIOT's SPI API like every other driver. (Not only the data sheet states that this is fine, it also proved to be reliable in practise.) - Greatly reduced latency: The RTT on the old driver (@150 kbps data rate) was about 16ms, the new driver (@250 kbps data rate) has as RTT of ~3ms (depending on SPI clock and on CPU performance) (measured with ping6). - Increased reliability: The preamble size and the sync word size have been doubled compared to the old driver (preamble: 8 bytes instead of 4, sync word: 4 byte instead of 2). The new values are the once recommended by the data sheet for reliable communication. - Basic diagnostic during driver initialization to detect common issues as SPI communication issues and GDO pin configuration/wiring issues. - TX power configuration with netdev_driver_t::set() API-integration - Calls to netdev_driver_t::send() block until the transmission has completed to ease the use of the API (implemented without busy waiting, so that the MCU can enter lower power states or other threads can be executed).
2018-11-08 17:37:07 +01:00
*
* @see cc110x_config_t::mdmcfg4
*
* See page 79 in the data sheet.
*
* In an ideal world the channel bandwidth would be twice the channel
* deviation. In the real world the used channel bandwidth is higher.
* Assuming a 26 MHz crystal and GFSK modulation (the driver will configure
* the transceiver to use GFSK) the deviation
*
* ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
* uint8_t exponent = (deviatn >> 4) & 0x07;
* int32_t mantissa = deviatn & 0x07;
* double deviation = (8 + mantissa) * 26000.0 / (1L << (17 - exponent));
* ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
*
* For reliable operation at high symbol rates, the deviation has to be
* increased as well.
*/
uint8_t deviatn;
} cc110x_config_t;
/**
* @brief Structure to hold mapping between virtual and physical channel numbers
drivers/cc110x: Rewrite of the cc110x driver The cc110x driver has been re-written from scratch to overcome the limitations of the old driver. The main motivation of the rewrite was to achieve better maintainability by a detailed documentation, reduce the complexity and the overhead of the SPI communication with the device, and to allow to simultaneously use transceivers with different configuration regarding the used base band, the channel bandwidth, the modulation rate, and the channel map. Features of this driver include: - Support for the CC1100, CC1101, and the CC1100e sub-gigahertz transceivers. - Detailed documentation of every aspect of this driver. - An easy to use configuration API that allows setting the transceiver configuration (modulation rate, channel bandwidth, base frequency) and the channel map. - Fast channel hopping by pre-calibration of the channels during device configuration (so that no calibration is needed during hopping). - Simplified SPI communication: Only during start-up the MCU has to wait for the transceiver to be ready (for the power regulators and the crystal to stabilize). The old driver did this for every SPI transfer, which resulted in complex communication code. This driver will wait on start up for the transceiver to power up and then use RIOT's SPI API like every other driver. (Not only the data sheet states that this is fine, it also proved to be reliable in practise.) - Greatly reduced latency: The RTT on the old driver (@150 kbps data rate) was about 16ms, the new driver (@250 kbps data rate) has as RTT of ~3ms (depending on SPI clock and on CPU performance) (measured with ping6). - Increased reliability: The preamble size and the sync word size have been doubled compared to the old driver (preamble: 8 bytes instead of 4, sync word: 4 byte instead of 2). The new values are the once recommended by the data sheet for reliable communication. - Basic diagnostic during driver initialization to detect common issues as SPI communication issues and GDO pin configuration/wiring issues. - TX power configuration with netdev_driver_t::set() API-integration - Calls to netdev_driver_t::send() block until the transmission has completed to ease the use of the API (implemented without busy waiting, so that the MCU can enter lower power states or other threads can be executed).
2018-11-08 17:37:07 +01:00
*
* This driver will provide "virtual" channel numbers 0 to 7, which will be
* translated to "physical" channel numbers before being send to the
* transceiver. This is used to overcome the following limitations:
*
* - The transceiver does not support channel maps with varying distance between
* channels. However, e.g. the LoRa channels 10 - 16 in the 868 MHz band have
* a distance of 300 kHz, but channel 16 and 17 have a distance of 1 MHz.
* - The transceiver does not supports channel distances higher than 405.46 kHz.
*
* This mapping overcomes both limitations be using 50kHz physical channel
* spacing and use the map to translate to the correct physical channel. This
* also allows to keep the same MDMCFG1 and MDMCFG0 configuration register
* values for all channel layouts. Finally, different channel sets can be
* used by different groups of IoT device in the same environment to limit
* collisions between those groups - assuming that enough non-overlapping
* channels are available.
*
* The "virtual" channel (the channel number presented to RIOT) will be used
* as index in @ref cc110x_chanmap_t::map, the value in there will give the
* corresponding "physical" channel number, or 255 if this virtual channel
* number is not available.
*/
typedef struct {
uint8_t map[CC110X_MAX_CHANNELS]; /**< "Physical" channel numbers */
} cc110x_chanmap_t;
/**
* @brief Structure holding all parameter for driver initialization
drivers/cc110x: Rewrite of the cc110x driver The cc110x driver has been re-written from scratch to overcome the limitations of the old driver. The main motivation of the rewrite was to achieve better maintainability by a detailed documentation, reduce the complexity and the overhead of the SPI communication with the device, and to allow to simultaneously use transceivers with different configuration regarding the used base band, the channel bandwidth, the modulation rate, and the channel map. Features of this driver include: - Support for the CC1100, CC1101, and the CC1100e sub-gigahertz transceivers. - Detailed documentation of every aspect of this driver. - An easy to use configuration API that allows setting the transceiver configuration (modulation rate, channel bandwidth, base frequency) and the channel map. - Fast channel hopping by pre-calibration of the channels during device configuration (so that no calibration is needed during hopping). - Simplified SPI communication: Only during start-up the MCU has to wait for the transceiver to be ready (for the power regulators and the crystal to stabilize). The old driver did this for every SPI transfer, which resulted in complex communication code. This driver will wait on start up for the transceiver to power up and then use RIOT's SPI API like every other driver. (Not only the data sheet states that this is fine, it also proved to be reliable in practise.) - Greatly reduced latency: The RTT on the old driver (@150 kbps data rate) was about 16ms, the new driver (@250 kbps data rate) has as RTT of ~3ms (depending on SPI clock and on CPU performance) (measured with ping6). - Increased reliability: The preamble size and the sync word size have been doubled compared to the old driver (preamble: 8 bytes instead of 4, sync word: 4 byte instead of 2). The new values are the once recommended by the data sheet for reliable communication. - Basic diagnostic during driver initialization to detect common issues as SPI communication issues and GDO pin configuration/wiring issues. - TX power configuration with netdev_driver_t::set() API-integration - Calls to netdev_driver_t::send() block until the transmission has completed to ease the use of the API (implemented without busy waiting, so that the MCU can enter lower power states or other threads can be executed).
2018-11-08 17:37:07 +01:00
*/
typedef struct {
const cc110x_patable_t *patable; /**< Pointer to the PATABLE to use */
/**
* @brief Pointer to the configuration of the base frequency, data rate
* and channel bandwidth; or `NULL` to keep the default.
drivers/cc110x: Rewrite of the cc110x driver The cc110x driver has been re-written from scratch to overcome the limitations of the old driver. The main motivation of the rewrite was to achieve better maintainability by a detailed documentation, reduce the complexity and the overhead of the SPI communication with the device, and to allow to simultaneously use transceivers with different configuration regarding the used base band, the channel bandwidth, the modulation rate, and the channel map. Features of this driver include: - Support for the CC1100, CC1101, and the CC1100e sub-gigahertz transceivers. - Detailed documentation of every aspect of this driver. - An easy to use configuration API that allows setting the transceiver configuration (modulation rate, channel bandwidth, base frequency) and the channel map. - Fast channel hopping by pre-calibration of the channels during device configuration (so that no calibration is needed during hopping). - Simplified SPI communication: Only during start-up the MCU has to wait for the transceiver to be ready (for the power regulators and the crystal to stabilize). The old driver did this for every SPI transfer, which resulted in complex communication code. This driver will wait on start up for the transceiver to power up and then use RIOT's SPI API like every other driver. (Not only the data sheet states that this is fine, it also proved to be reliable in practise.) - Greatly reduced latency: The RTT on the old driver (@150 kbps data rate) was about 16ms, the new driver (@250 kbps data rate) has as RTT of ~3ms (depending on SPI clock and on CPU performance) (measured with ping6). - Increased reliability: The preamble size and the sync word size have been doubled compared to the old driver (preamble: 8 bytes instead of 4, sync word: 4 byte instead of 2). The new values are the once recommended by the data sheet for reliable communication. - Basic diagnostic during driver initialization to detect common issues as SPI communication issues and GDO pin configuration/wiring issues. - TX power configuration with netdev_driver_t::set() API-integration - Calls to netdev_driver_t::send() block until the transmission has completed to ease the use of the API (implemented without busy waiting, so that the MCU can enter lower power states or other threads can be executed).
2018-11-08 17:37:07 +01:00
*/
const cc110x_config_t *config;
const cc110x_chanmap_t *channels; /**< Pointer to the default channel map */
spi_t spi; /**< SPI bus connected to the device */
spi_clk_t spi_clk; /**< SPI clock to use (max 6.5 MHz) */
spi_cs_t cs; /**< GPIO pin connected to chip select */
gpio_t gdo0; /**< GPIO pin connected to GDO0 */
gpio_t gdo2; /**< GPIO pin connected to GDO2 */
} cc110x_params_t;
/**
* @brief Structure holding the calibration data of the frequency synthesizer
drivers/cc110x: Rewrite of the cc110x driver The cc110x driver has been re-written from scratch to overcome the limitations of the old driver. The main motivation of the rewrite was to achieve better maintainability by a detailed documentation, reduce the complexity and the overhead of the SPI communication with the device, and to allow to simultaneously use transceivers with different configuration regarding the used base band, the channel bandwidth, the modulation rate, and the channel map. Features of this driver include: - Support for the CC1100, CC1101, and the CC1100e sub-gigahertz transceivers. - Detailed documentation of every aspect of this driver. - An easy to use configuration API that allows setting the transceiver configuration (modulation rate, channel bandwidth, base frequency) and the channel map. - Fast channel hopping by pre-calibration of the channels during device configuration (so that no calibration is needed during hopping). - Simplified SPI communication: Only during start-up the MCU has to wait for the transceiver to be ready (for the power regulators and the crystal to stabilize). The old driver did this for every SPI transfer, which resulted in complex communication code. This driver will wait on start up for the transceiver to power up and then use RIOT's SPI API like every other driver. (Not only the data sheet states that this is fine, it also proved to be reliable in practise.) - Greatly reduced latency: The RTT on the old driver (@150 kbps data rate) was about 16ms, the new driver (@250 kbps data rate) has as RTT of ~3ms (depending on SPI clock and on CPU performance) (measured with ping6). - Increased reliability: The preamble size and the sync word size have been doubled compared to the old driver (preamble: 8 bytes instead of 4, sync word: 4 byte instead of 2). The new values are the once recommended by the data sheet for reliable communication. - Basic diagnostic during driver initialization to detect common issues as SPI communication issues and GDO pin configuration/wiring issues. - TX power configuration with netdev_driver_t::set() API-integration - Calls to netdev_driver_t::send() block until the transmission has completed to ease the use of the API (implemented without busy waiting, so that the MCU can enter lower power states or other threads can be executed).
2018-11-08 17:37:07 +01:00
*/
typedef struct {
/**
* @brief VCO capacitance calibration, which depends on the frequency and,
* thus, has to be stored for each channel
drivers/cc110x: Rewrite of the cc110x driver The cc110x driver has been re-written from scratch to overcome the limitations of the old driver. The main motivation of the rewrite was to achieve better maintainability by a detailed documentation, reduce the complexity and the overhead of the SPI communication with the device, and to allow to simultaneously use transceivers with different configuration regarding the used base band, the channel bandwidth, the modulation rate, and the channel map. Features of this driver include: - Support for the CC1100, CC1101, and the CC1100e sub-gigahertz transceivers. - Detailed documentation of every aspect of this driver. - An easy to use configuration API that allows setting the transceiver configuration (modulation rate, channel bandwidth, base frequency) and the channel map. - Fast channel hopping by pre-calibration of the channels during device configuration (so that no calibration is needed during hopping). - Simplified SPI communication: Only during start-up the MCU has to wait for the transceiver to be ready (for the power regulators and the crystal to stabilize). The old driver did this for every SPI transfer, which resulted in complex communication code. This driver will wait on start up for the transceiver to power up and then use RIOT's SPI API like every other driver. (Not only the data sheet states that this is fine, it also proved to be reliable in practise.) - Greatly reduced latency: The RTT on the old driver (@150 kbps data rate) was about 16ms, the new driver (@250 kbps data rate) has as RTT of ~3ms (depending on SPI clock and on CPU performance) (measured with ping6). - Increased reliability: The preamble size and the sync word size have been doubled compared to the old driver (preamble: 8 bytes instead of 4, sync word: 4 byte instead of 2). The new values are the once recommended by the data sheet for reliable communication. - Basic diagnostic during driver initialization to detect common issues as SPI communication issues and GDO pin configuration/wiring issues. - TX power configuration with netdev_driver_t::set() API-integration - Calls to netdev_driver_t::send() block until the transmission has completed to ease the use of the API (implemented without busy waiting, so that the MCU can enter lower power states or other threads can be executed).
2018-11-08 17:37:07 +01:00
*/
char fscal1[CC110X_MAX_CHANNELS];
char fscal2; /**< VCO current calibration, independent of channel */
char fscal3; /**< charge pump current calibration, independent of channel */
} cc110x_fs_calibration_t;
/**
* @brief Buffer to temporary store incoming/outgoing packet
drivers/cc110x: Rewrite of the cc110x driver The cc110x driver has been re-written from scratch to overcome the limitations of the old driver. The main motivation of the rewrite was to achieve better maintainability by a detailed documentation, reduce the complexity and the overhead of the SPI communication with the device, and to allow to simultaneously use transceivers with different configuration regarding the used base band, the channel bandwidth, the modulation rate, and the channel map. Features of this driver include: - Support for the CC1100, CC1101, and the CC1100e sub-gigahertz transceivers. - Detailed documentation of every aspect of this driver. - An easy to use configuration API that allows setting the transceiver configuration (modulation rate, channel bandwidth, base frequency) and the channel map. - Fast channel hopping by pre-calibration of the channels during device configuration (so that no calibration is needed during hopping). - Simplified SPI communication: Only during start-up the MCU has to wait for the transceiver to be ready (for the power regulators and the crystal to stabilize). The old driver did this for every SPI transfer, which resulted in complex communication code. This driver will wait on start up for the transceiver to power up and then use RIOT's SPI API like every other driver. (Not only the data sheet states that this is fine, it also proved to be reliable in practise.) - Greatly reduced latency: The RTT on the old driver (@150 kbps data rate) was about 16ms, the new driver (@250 kbps data rate) has as RTT of ~3ms (depending on SPI clock and on CPU performance) (measured with ping6). - Increased reliability: The preamble size and the sync word size have been doubled compared to the old driver (preamble: 8 bytes instead of 4, sync word: 4 byte instead of 2). The new values are the once recommended by the data sheet for reliable communication. - Basic diagnostic during driver initialization to detect common issues as SPI communication issues and GDO pin configuration/wiring issues. - TX power configuration with netdev_driver_t::set() API-integration - Calls to netdev_driver_t::send() block until the transmission has completed to ease the use of the API (implemented without busy waiting, so that the MCU can enter lower power states or other threads can be executed).
2018-11-08 17:37:07 +01:00
*
* The CC1100/CC1101 transceiver's FIFO sadly is only 64 bytes in size. To
* support frames bigger than that, chunks of the frame have to be
* transferred between the MCU and the CC1100/CC1101 transceiver while the
* frame is in transit.
*/
typedef struct __attribute__((packed)) {
uint8_t len; /**< Length of the frame in bytes */
/**
* @brief The payload data of the frame
drivers/cc110x: Rewrite of the cc110x driver The cc110x driver has been re-written from scratch to overcome the limitations of the old driver. The main motivation of the rewrite was to achieve better maintainability by a detailed documentation, reduce the complexity and the overhead of the SPI communication with the device, and to allow to simultaneously use transceivers with different configuration regarding the used base band, the channel bandwidth, the modulation rate, and the channel map. Features of this driver include: - Support for the CC1100, CC1101, and the CC1100e sub-gigahertz transceivers. - Detailed documentation of every aspect of this driver. - An easy to use configuration API that allows setting the transceiver configuration (modulation rate, channel bandwidth, base frequency) and the channel map. - Fast channel hopping by pre-calibration of the channels during device configuration (so that no calibration is needed during hopping). - Simplified SPI communication: Only during start-up the MCU has to wait for the transceiver to be ready (for the power regulators and the crystal to stabilize). The old driver did this for every SPI transfer, which resulted in complex communication code. This driver will wait on start up for the transceiver to power up and then use RIOT's SPI API like every other driver. (Not only the data sheet states that this is fine, it also proved to be reliable in practise.) - Greatly reduced latency: The RTT on the old driver (@150 kbps data rate) was about 16ms, the new driver (@250 kbps data rate) has as RTT of ~3ms (depending on SPI clock and on CPU performance) (measured with ping6). - Increased reliability: The preamble size and the sync word size have been doubled compared to the old driver (preamble: 8 bytes instead of 4, sync word: 4 byte instead of 2). The new values are the once recommended by the data sheet for reliable communication. - Basic diagnostic during driver initialization to detect common issues as SPI communication issues and GDO pin configuration/wiring issues. - TX power configuration with netdev_driver_t::set() API-integration - Calls to netdev_driver_t::send() block until the transmission has completed to ease the use of the API (implemented without busy waiting, so that the MCU can enter lower power states or other threads can be executed).
2018-11-08 17:37:07 +01:00
*/
uint8_t data[CC110X_MAX_FRAME_SIZE];
/**
* @brief Index of the next @ref cc110x_framebuf_t::data element to
* transfer
drivers/cc110x: Rewrite of the cc110x driver The cc110x driver has been re-written from scratch to overcome the limitations of the old driver. The main motivation of the rewrite was to achieve better maintainability by a detailed documentation, reduce the complexity and the overhead of the SPI communication with the device, and to allow to simultaneously use transceivers with different configuration regarding the used base band, the channel bandwidth, the modulation rate, and the channel map. Features of this driver include: - Support for the CC1100, CC1101, and the CC1100e sub-gigahertz transceivers. - Detailed documentation of every aspect of this driver. - An easy to use configuration API that allows setting the transceiver configuration (modulation rate, channel bandwidth, base frequency) and the channel map. - Fast channel hopping by pre-calibration of the channels during device configuration (so that no calibration is needed during hopping). - Simplified SPI communication: Only during start-up the MCU has to wait for the transceiver to be ready (for the power regulators and the crystal to stabilize). The old driver did this for every SPI transfer, which resulted in complex communication code. This driver will wait on start up for the transceiver to power up and then use RIOT's SPI API like every other driver. (Not only the data sheet states that this is fine, it also proved to be reliable in practise.) - Greatly reduced latency: The RTT on the old driver (@150 kbps data rate) was about 16ms, the new driver (@250 kbps data rate) has as RTT of ~3ms (depending on SPI clock and on CPU performance) (measured with ping6). - Increased reliability: The preamble size and the sync word size have been doubled compared to the old driver (preamble: 8 bytes instead of 4, sync word: 4 byte instead of 2). The new values are the once recommended by the data sheet for reliable communication. - Basic diagnostic during driver initialization to detect common issues as SPI communication issues and GDO pin configuration/wiring issues. - TX power configuration with netdev_driver_t::set() API-integration - Calls to netdev_driver_t::send() block until the transmission has completed to ease the use of the API (implemented without busy waiting, so that the MCU can enter lower power states or other threads can be executed).
2018-11-08 17:37:07 +01:00
*
* In RX mode: Index of the next @ref cc110x_framebuf_t::data element to
* store data read from the RX-FIFO into.
drivers/cc110x: Rewrite of the cc110x driver The cc110x driver has been re-written from scratch to overcome the limitations of the old driver. The main motivation of the rewrite was to achieve better maintainability by a detailed documentation, reduce the complexity and the overhead of the SPI communication with the device, and to allow to simultaneously use transceivers with different configuration regarding the used base band, the channel bandwidth, the modulation rate, and the channel map. Features of this driver include: - Support for the CC1100, CC1101, and the CC1100e sub-gigahertz transceivers. - Detailed documentation of every aspect of this driver. - An easy to use configuration API that allows setting the transceiver configuration (modulation rate, channel bandwidth, base frequency) and the channel map. - Fast channel hopping by pre-calibration of the channels during device configuration (so that no calibration is needed during hopping). - Simplified SPI communication: Only during start-up the MCU has to wait for the transceiver to be ready (for the power regulators and the crystal to stabilize). The old driver did this for every SPI transfer, which resulted in complex communication code. This driver will wait on start up for the transceiver to power up and then use RIOT's SPI API like every other driver. (Not only the data sheet states that this is fine, it also proved to be reliable in practise.) - Greatly reduced latency: The RTT on the old driver (@150 kbps data rate) was about 16ms, the new driver (@250 kbps data rate) has as RTT of ~3ms (depending on SPI clock and on CPU performance) (measured with ping6). - Increased reliability: The preamble size and the sync word size have been doubled compared to the old driver (preamble: 8 bytes instead of 4, sync word: 4 byte instead of 2). The new values are the once recommended by the data sheet for reliable communication. - Basic diagnostic during driver initialization to detect common issues as SPI communication issues and GDO pin configuration/wiring issues. - TX power configuration with netdev_driver_t::set() API-integration - Calls to netdev_driver_t::send() block until the transmission has completed to ease the use of the API (implemented without busy waiting, so that the MCU can enter lower power states or other threads can be executed).
2018-11-08 17:37:07 +01:00
*
* In TX mode: Index of the next @ref cc110x_framebuf_t::data element to
* write to the TX-FIFO.
drivers/cc110x: Rewrite of the cc110x driver The cc110x driver has been re-written from scratch to overcome the limitations of the old driver. The main motivation of the rewrite was to achieve better maintainability by a detailed documentation, reduce the complexity and the overhead of the SPI communication with the device, and to allow to simultaneously use transceivers with different configuration regarding the used base band, the channel bandwidth, the modulation rate, and the channel map. Features of this driver include: - Support for the CC1100, CC1101, and the CC1100e sub-gigahertz transceivers. - Detailed documentation of every aspect of this driver. - An easy to use configuration API that allows setting the transceiver configuration (modulation rate, channel bandwidth, base frequency) and the channel map. - Fast channel hopping by pre-calibration of the channels during device configuration (so that no calibration is needed during hopping). - Simplified SPI communication: Only during start-up the MCU has to wait for the transceiver to be ready (for the power regulators and the crystal to stabilize). The old driver did this for every SPI transfer, which resulted in complex communication code. This driver will wait on start up for the transceiver to power up and then use RIOT's SPI API like every other driver. (Not only the data sheet states that this is fine, it also proved to be reliable in practise.) - Greatly reduced latency: The RTT on the old driver (@150 kbps data rate) was about 16ms, the new driver (@250 kbps data rate) has as RTT of ~3ms (depending on SPI clock and on CPU performance) (measured with ping6). - Increased reliability: The preamble size and the sync word size have been doubled compared to the old driver (preamble: 8 bytes instead of 4, sync word: 4 byte instead of 2). The new values are the once recommended by the data sheet for reliable communication. - Basic diagnostic during driver initialization to detect common issues as SPI communication issues and GDO pin configuration/wiring issues. - TX power configuration with netdev_driver_t::set() API-integration - Calls to netdev_driver_t::send() block until the transmission has completed to ease the use of the API (implemented without busy waiting, so that the MCU can enter lower power states or other threads can be executed).
2018-11-08 17:37:07 +01:00
*/
uint8_t pos;
} cc110x_framebuf_t;
/**
* @brief Device descriptor for CC1100/CC1101 transceivers
drivers/cc110x: Rewrite of the cc110x driver The cc110x driver has been re-written from scratch to overcome the limitations of the old driver. The main motivation of the rewrite was to achieve better maintainability by a detailed documentation, reduce the complexity and the overhead of the SPI communication with the device, and to allow to simultaneously use transceivers with different configuration regarding the used base band, the channel bandwidth, the modulation rate, and the channel map. Features of this driver include: - Support for the CC1100, CC1101, and the CC1100e sub-gigahertz transceivers. - Detailed documentation of every aspect of this driver. - An easy to use configuration API that allows setting the transceiver configuration (modulation rate, channel bandwidth, base frequency) and the channel map. - Fast channel hopping by pre-calibration of the channels during device configuration (so that no calibration is needed during hopping). - Simplified SPI communication: Only during start-up the MCU has to wait for the transceiver to be ready (for the power regulators and the crystal to stabilize). The old driver did this for every SPI transfer, which resulted in complex communication code. This driver will wait on start up for the transceiver to power up and then use RIOT's SPI API like every other driver. (Not only the data sheet states that this is fine, it also proved to be reliable in practise.) - Greatly reduced latency: The RTT on the old driver (@150 kbps data rate) was about 16ms, the new driver (@250 kbps data rate) has as RTT of ~3ms (depending on SPI clock and on CPU performance) (measured with ping6). - Increased reliability: The preamble size and the sync word size have been doubled compared to the old driver (preamble: 8 bytes instead of 4, sync word: 4 byte instead of 2). The new values are the once recommended by the data sheet for reliable communication. - Basic diagnostic during driver initialization to detect common issues as SPI communication issues and GDO pin configuration/wiring issues. - TX power configuration with netdev_driver_t::set() API-integration - Calls to netdev_driver_t::send() block until the transmission has completed to ease the use of the API (implemented without busy waiting, so that the MCU can enter lower power states or other threads can be executed).
2018-11-08 17:37:07 +01:00
*/
typedef struct {
netdev_t netdev; /**< RIOT's interface to this driver */
uint8_t addr; /**< Layer 2 address of this device */
/* Keep above in sync with cc1xx_t members, as they must overlap! */
cc110x_state_t state; /**< State of the transceiver */
cc110x_tx_power_t tx_power; /**< TX power of the receiver */
uint8_t channel; /**< Currently tuned (virtual) channel */
/* Struct packing: addr, state, tx_power and channel add up to 32 bit */
const cc110x_chanmap_t *channels; /**< Pointer to the channel map to use. */
cc110x_params_t params; /**< Configuration of the driver */
cc110x_framebuf_t buf; /**< Temporary frame buffer */
/**
* @brief RSSI and LQI of the last received frame
drivers/cc110x: Rewrite of the cc110x driver The cc110x driver has been re-written from scratch to overcome the limitations of the old driver. The main motivation of the rewrite was to achieve better maintainability by a detailed documentation, reduce the complexity and the overhead of the SPI communication with the device, and to allow to simultaneously use transceivers with different configuration regarding the used base band, the channel bandwidth, the modulation rate, and the channel map. Features of this driver include: - Support for the CC1100, CC1101, and the CC1100e sub-gigahertz transceivers. - Detailed documentation of every aspect of this driver. - An easy to use configuration API that allows setting the transceiver configuration (modulation rate, channel bandwidth, base frequency) and the channel map. - Fast channel hopping by pre-calibration of the channels during device configuration (so that no calibration is needed during hopping). - Simplified SPI communication: Only during start-up the MCU has to wait for the transceiver to be ready (for the power regulators and the crystal to stabilize). The old driver did this for every SPI transfer, which resulted in complex communication code. This driver will wait on start up for the transceiver to power up and then use RIOT's SPI API like every other driver. (Not only the data sheet states that this is fine, it also proved to be reliable in practise.) - Greatly reduced latency: The RTT on the old driver (@150 kbps data rate) was about 16ms, the new driver (@250 kbps data rate) has as RTT of ~3ms (depending on SPI clock and on CPU performance) (measured with ping6). - Increased reliability: The preamble size and the sync word size have been doubled compared to the old driver (preamble: 8 bytes instead of 4, sync word: 4 byte instead of 2). The new values are the once recommended by the data sheet for reliable communication. - Basic diagnostic during driver initialization to detect common issues as SPI communication issues and GDO pin configuration/wiring issues. - TX power configuration with netdev_driver_t::set() API-integration - Calls to netdev_driver_t::send() block until the transmission has completed to ease the use of the API (implemented without busy waiting, so that the MCU can enter lower power states or other threads can be executed).
2018-11-08 17:37:07 +01:00
*/
cc1xxx_rx_info_t rx_info;
/**
* @brief Frequency synthesizer calibration data
drivers/cc110x: Rewrite of the cc110x driver The cc110x driver has been re-written from scratch to overcome the limitations of the old driver. The main motivation of the rewrite was to achieve better maintainability by a detailed documentation, reduce the complexity and the overhead of the SPI communication with the device, and to allow to simultaneously use transceivers with different configuration regarding the used base band, the channel bandwidth, the modulation rate, and the channel map. Features of this driver include: - Support for the CC1100, CC1101, and the CC1100e sub-gigahertz transceivers. - Detailed documentation of every aspect of this driver. - An easy to use configuration API that allows setting the transceiver configuration (modulation rate, channel bandwidth, base frequency) and the channel map. - Fast channel hopping by pre-calibration of the channels during device configuration (so that no calibration is needed during hopping). - Simplified SPI communication: Only during start-up the MCU has to wait for the transceiver to be ready (for the power regulators and the crystal to stabilize). The old driver did this for every SPI transfer, which resulted in complex communication code. This driver will wait on start up for the transceiver to power up and then use RIOT's SPI API like every other driver. (Not only the data sheet states that this is fine, it also proved to be reliable in practise.) - Greatly reduced latency: The RTT on the old driver (@150 kbps data rate) was about 16ms, the new driver (@250 kbps data rate) has as RTT of ~3ms (depending on SPI clock and on CPU performance) (measured with ping6). - Increased reliability: The preamble size and the sync word size have been doubled compared to the old driver (preamble: 8 bytes instead of 4, sync word: 4 byte instead of 2). The new values are the once recommended by the data sheet for reliable communication. - Basic diagnostic during driver initialization to detect common issues as SPI communication issues and GDO pin configuration/wiring issues. - TX power configuration with netdev_driver_t::set() API-integration - Calls to netdev_driver_t::send() block until the transmission has completed to ease the use of the API (implemented without busy waiting, so that the MCU can enter lower power states or other threads can be executed).
2018-11-08 17:37:07 +01:00
*/
cc110x_fs_calibration_t fscal;
/**
* @brief Use mutex to block during TX and unblock from ISR when ISR
* needs to be handled from thread-context
*
* Blocking during TX within the driver prevents the upper layers from
* calling @ref netdev_driver_t::send while already transmitting a frame.
*/
mutex_t isr_signal;
uint8_t rssi_offset; /**< dBm to subtract from raw RSSI data */
} cc110x_t;
/**
* @brief Setup the CC1100/CC1101 driver, but perform no initialization
drivers/cc110x: Rewrite of the cc110x driver The cc110x driver has been re-written from scratch to overcome the limitations of the old driver. The main motivation of the rewrite was to achieve better maintainability by a detailed documentation, reduce the complexity and the overhead of the SPI communication with the device, and to allow to simultaneously use transceivers with different configuration regarding the used base band, the channel bandwidth, the modulation rate, and the channel map. Features of this driver include: - Support for the CC1100, CC1101, and the CC1100e sub-gigahertz transceivers. - Detailed documentation of every aspect of this driver. - An easy to use configuration API that allows setting the transceiver configuration (modulation rate, channel bandwidth, base frequency) and the channel map. - Fast channel hopping by pre-calibration of the channels during device configuration (so that no calibration is needed during hopping). - Simplified SPI communication: Only during start-up the MCU has to wait for the transceiver to be ready (for the power regulators and the crystal to stabilize). The old driver did this for every SPI transfer, which resulted in complex communication code. This driver will wait on start up for the transceiver to power up and then use RIOT's SPI API like every other driver. (Not only the data sheet states that this is fine, it also proved to be reliable in practise.) - Greatly reduced latency: The RTT on the old driver (@150 kbps data rate) was about 16ms, the new driver (@250 kbps data rate) has as RTT of ~3ms (depending on SPI clock and on CPU performance) (measured with ping6). - Increased reliability: The preamble size and the sync word size have been doubled compared to the old driver (preamble: 8 bytes instead of 4, sync word: 4 byte instead of 2). The new values are the once recommended by the data sheet for reliable communication. - Basic diagnostic during driver initialization to detect common issues as SPI communication issues and GDO pin configuration/wiring issues. - TX power configuration with netdev_driver_t::set() API-integration - Calls to netdev_driver_t::send() block until the transmission has completed to ease the use of the API (implemented without busy waiting, so that the MCU can enter lower power states or other threads can be executed).
2018-11-08 17:37:07 +01:00
*
* @ref netdev_driver_t::init can be used after this call to initialize the
* transceiver.
*
* @param dev Device descriptor to use
* @param params Parameter of the device to setup
* @param index Index of @p params in a global parameter struct array.
* If initialized manually, pass a unique identifier instead.
drivers/cc110x: Rewrite of the cc110x driver The cc110x driver has been re-written from scratch to overcome the limitations of the old driver. The main motivation of the rewrite was to achieve better maintainability by a detailed documentation, reduce the complexity and the overhead of the SPI communication with the device, and to allow to simultaneously use transceivers with different configuration regarding the used base band, the channel bandwidth, the modulation rate, and the channel map. Features of this driver include: - Support for the CC1100, CC1101, and the CC1100e sub-gigahertz transceivers. - Detailed documentation of every aspect of this driver. - An easy to use configuration API that allows setting the transceiver configuration (modulation rate, channel bandwidth, base frequency) and the channel map. - Fast channel hopping by pre-calibration of the channels during device configuration (so that no calibration is needed during hopping). - Simplified SPI communication: Only during start-up the MCU has to wait for the transceiver to be ready (for the power regulators and the crystal to stabilize). The old driver did this for every SPI transfer, which resulted in complex communication code. This driver will wait on start up for the transceiver to power up and then use RIOT's SPI API like every other driver. (Not only the data sheet states that this is fine, it also proved to be reliable in practise.) - Greatly reduced latency: The RTT on the old driver (@150 kbps data rate) was about 16ms, the new driver (@250 kbps data rate) has as RTT of ~3ms (depending on SPI clock and on CPU performance) (measured with ping6). - Increased reliability: The preamble size and the sync word size have been doubled compared to the old driver (preamble: 8 bytes instead of 4, sync word: 4 byte instead of 2). The new values are the once recommended by the data sheet for reliable communication. - Basic diagnostic during driver initialization to detect common issues as SPI communication issues and GDO pin configuration/wiring issues. - TX power configuration with netdev_driver_t::set() API-integration - Calls to netdev_driver_t::send() block until the transmission has completed to ease the use of the API (implemented without busy waiting, so that the MCU can enter lower power states or other threads can be executed).
2018-11-08 17:37:07 +01:00
*
* @retval 0 Device successfully set up
* @retval -EINVAL @p dev or @p params is `NULL`, or @p params is invalid
drivers/cc110x: Rewrite of the cc110x driver The cc110x driver has been re-written from scratch to overcome the limitations of the old driver. The main motivation of the rewrite was to achieve better maintainability by a detailed documentation, reduce the complexity and the overhead of the SPI communication with the device, and to allow to simultaneously use transceivers with different configuration regarding the used base band, the channel bandwidth, the modulation rate, and the channel map. Features of this driver include: - Support for the CC1100, CC1101, and the CC1100e sub-gigahertz transceivers. - Detailed documentation of every aspect of this driver. - An easy to use configuration API that allows setting the transceiver configuration (modulation rate, channel bandwidth, base frequency) and the channel map. - Fast channel hopping by pre-calibration of the channels during device configuration (so that no calibration is needed during hopping). - Simplified SPI communication: Only during start-up the MCU has to wait for the transceiver to be ready (for the power regulators and the crystal to stabilize). The old driver did this for every SPI transfer, which resulted in complex communication code. This driver will wait on start up for the transceiver to power up and then use RIOT's SPI API like every other driver. (Not only the data sheet states that this is fine, it also proved to be reliable in practise.) - Greatly reduced latency: The RTT on the old driver (@150 kbps data rate) was about 16ms, the new driver (@250 kbps data rate) has as RTT of ~3ms (depending on SPI clock and on CPU performance) (measured with ping6). - Increased reliability: The preamble size and the sync word size have been doubled compared to the old driver (preamble: 8 bytes instead of 4, sync word: 4 byte instead of 2). The new values are the once recommended by the data sheet for reliable communication. - Basic diagnostic during driver initialization to detect common issues as SPI communication issues and GDO pin configuration/wiring issues. - TX power configuration with netdev_driver_t::set() API-integration - Calls to netdev_driver_t::send() block until the transmission has completed to ease the use of the API (implemented without busy waiting, so that the MCU can enter lower power states or other threads can be executed).
2018-11-08 17:37:07 +01:00
*/
int cc110x_setup(cc110x_t *dev, const cc110x_params_t *params, uint8_t index);
drivers/cc110x: Rewrite of the cc110x driver The cc110x driver has been re-written from scratch to overcome the limitations of the old driver. The main motivation of the rewrite was to achieve better maintainability by a detailed documentation, reduce the complexity and the overhead of the SPI communication with the device, and to allow to simultaneously use transceivers with different configuration regarding the used base band, the channel bandwidth, the modulation rate, and the channel map. Features of this driver include: - Support for the CC1100, CC1101, and the CC1100e sub-gigahertz transceivers. - Detailed documentation of every aspect of this driver. - An easy to use configuration API that allows setting the transceiver configuration (modulation rate, channel bandwidth, base frequency) and the channel map. - Fast channel hopping by pre-calibration of the channels during device configuration (so that no calibration is needed during hopping). - Simplified SPI communication: Only during start-up the MCU has to wait for the transceiver to be ready (for the power regulators and the crystal to stabilize). The old driver did this for every SPI transfer, which resulted in complex communication code. This driver will wait on start up for the transceiver to power up and then use RIOT's SPI API like every other driver. (Not only the data sheet states that this is fine, it also proved to be reliable in practise.) - Greatly reduced latency: The RTT on the old driver (@150 kbps data rate) was about 16ms, the new driver (@250 kbps data rate) has as RTT of ~3ms (depending on SPI clock and on CPU performance) (measured with ping6). - Increased reliability: The preamble size and the sync word size have been doubled compared to the old driver (preamble: 8 bytes instead of 4, sync word: 4 byte instead of 2). The new values are the once recommended by the data sheet for reliable communication. - Basic diagnostic during driver initialization to detect common issues as SPI communication issues and GDO pin configuration/wiring issues. - TX power configuration with netdev_driver_t::set() API-integration - Calls to netdev_driver_t::send() block until the transmission has completed to ease the use of the API (implemented without busy waiting, so that the MCU can enter lower power states or other threads can be executed).
2018-11-08 17:37:07 +01:00
/**
* @brief Apply the given configuration and the given channel map and performs
* a recalibration
*
* @param dev Device descriptor of the transceiver
* @param conf Configuration to apply or `NULL` to only change channel map
* @param chanmap Channel map to apply (must be compatible with @p conf)
* @param channel The channel to tune in after applying the config
drivers/cc110x: Rewrite of the cc110x driver The cc110x driver has been re-written from scratch to overcome the limitations of the old driver. The main motivation of the rewrite was to achieve better maintainability by a detailed documentation, reduce the complexity and the overhead of the SPI communication with the device, and to allow to simultaneously use transceivers with different configuration regarding the used base band, the channel bandwidth, the modulation rate, and the channel map. Features of this driver include: - Support for the CC1100, CC1101, and the CC1100e sub-gigahertz transceivers. - Detailed documentation of every aspect of this driver. - An easy to use configuration API that allows setting the transceiver configuration (modulation rate, channel bandwidth, base frequency) and the channel map. - Fast channel hopping by pre-calibration of the channels during device configuration (so that no calibration is needed during hopping). - Simplified SPI communication: Only during start-up the MCU has to wait for the transceiver to be ready (for the power regulators and the crystal to stabilize). The old driver did this for every SPI transfer, which resulted in complex communication code. This driver will wait on start up for the transceiver to power up and then use RIOT's SPI API like every other driver. (Not only the data sheet states that this is fine, it also proved to be reliable in practise.) - Greatly reduced latency: The RTT on the old driver (@150 kbps data rate) was about 16ms, the new driver (@250 kbps data rate) has as RTT of ~3ms (depending on SPI clock and on CPU performance) (measured with ping6). - Increased reliability: The preamble size and the sync word size have been doubled compared to the old driver (preamble: 8 bytes instead of 4, sync word: 4 byte instead of 2). The new values are the once recommended by the data sheet for reliable communication. - Basic diagnostic during driver initialization to detect common issues as SPI communication issues and GDO pin configuration/wiring issues. - TX power configuration with netdev_driver_t::set() API-integration - Calls to netdev_driver_t::send() block until the transmission has completed to ease the use of the API (implemented without busy waiting, so that the MCU can enter lower power states or other threads can be executed).
2018-11-08 17:37:07 +01:00
*
* @retval 0 Success
* @retval -EINVAL Called with invalid argument
* @retval -EIO Communication with the transceiver failed
* @retval -ERANGE Channel out of range or not supported by channel map
drivers/cc110x: Rewrite of the cc110x driver The cc110x driver has been re-written from scratch to overcome the limitations of the old driver. The main motivation of the rewrite was to achieve better maintainability by a detailed documentation, reduce the complexity and the overhead of the SPI communication with the device, and to allow to simultaneously use transceivers with different configuration regarding the used base band, the channel bandwidth, the modulation rate, and the channel map. Features of this driver include: - Support for the CC1100, CC1101, and the CC1100e sub-gigahertz transceivers. - Detailed documentation of every aspect of this driver. - An easy to use configuration API that allows setting the transceiver configuration (modulation rate, channel bandwidth, base frequency) and the channel map. - Fast channel hopping by pre-calibration of the channels during device configuration (so that no calibration is needed during hopping). - Simplified SPI communication: Only during start-up the MCU has to wait for the transceiver to be ready (for the power regulators and the crystal to stabilize). The old driver did this for every SPI transfer, which resulted in complex communication code. This driver will wait on start up for the transceiver to power up and then use RIOT's SPI API like every other driver. (Not only the data sheet states that this is fine, it also proved to be reliable in practise.) - Greatly reduced latency: The RTT on the old driver (@150 kbps data rate) was about 16ms, the new driver (@250 kbps data rate) has as RTT of ~3ms (depending on SPI clock and on CPU performance) (measured with ping6). - Increased reliability: The preamble size and the sync word size have been doubled compared to the old driver (preamble: 8 bytes instead of 4, sync word: 4 byte instead of 2). The new values are the once recommended by the data sheet for reliable communication. - Basic diagnostic during driver initialization to detect common issues as SPI communication issues and GDO pin configuration/wiring issues. - TX power configuration with netdev_driver_t::set() API-integration - Calls to netdev_driver_t::send() block until the transmission has completed to ease the use of the API (implemented without busy waiting, so that the MCU can enter lower power states or other threads can be executed).
2018-11-08 17:37:07 +01:00
*
* @pre The application developer checked in the documentation that the
* channel map in @p chanmap is compatible with the configuration in
* @p conf
drivers/cc110x: Rewrite of the cc110x driver The cc110x driver has been re-written from scratch to overcome the limitations of the old driver. The main motivation of the rewrite was to achieve better maintainability by a detailed documentation, reduce the complexity and the overhead of the SPI communication with the device, and to allow to simultaneously use transceivers with different configuration regarding the used base band, the channel bandwidth, the modulation rate, and the channel map. Features of this driver include: - Support for the CC1100, CC1101, and the CC1100e sub-gigahertz transceivers. - Detailed documentation of every aspect of this driver. - An easy to use configuration API that allows setting the transceiver configuration (modulation rate, channel bandwidth, base frequency) and the channel map. - Fast channel hopping by pre-calibration of the channels during device configuration (so that no calibration is needed during hopping). - Simplified SPI communication: Only during start-up the MCU has to wait for the transceiver to be ready (for the power regulators and the crystal to stabilize). The old driver did this for every SPI transfer, which resulted in complex communication code. This driver will wait on start up for the transceiver to power up and then use RIOT's SPI API like every other driver. (Not only the data sheet states that this is fine, it also proved to be reliable in practise.) - Greatly reduced latency: The RTT on the old driver (@150 kbps data rate) was about 16ms, the new driver (@250 kbps data rate) has as RTT of ~3ms (depending on SPI clock and on CPU performance) (measured with ping6). - Increased reliability: The preamble size and the sync word size have been doubled compared to the old driver (preamble: 8 bytes instead of 4, sync word: 4 byte instead of 2). The new values are the once recommended by the data sheet for reliable communication. - Basic diagnostic during driver initialization to detect common issues as SPI communication issues and GDO pin configuration/wiring issues. - TX power configuration with netdev_driver_t::set() API-integration - Calls to netdev_driver_t::send() block until the transmission has completed to ease the use of the API (implemented without busy waiting, so that the MCU can enter lower power states or other threads can be executed).
2018-11-08 17:37:07 +01:00
*
* Because the configuration (potentially) changes the channel bandwidth, the
* old channel map is rendered invalid. This API therefore asks for both to make
* sure an application developer does not forget to update the channel map.
* Because the old calibration data is also rendered invalid,
* @ref cc110x_full_calibration is called to update it.
*/
int cc110x_apply_config(cc110x_t *dev, const cc110x_config_t *conf,
const cc110x_chanmap_t *chanmap, uint8_t channel);
drivers/cc110x: Rewrite of the cc110x driver The cc110x driver has been re-written from scratch to overcome the limitations of the old driver. The main motivation of the rewrite was to achieve better maintainability by a detailed documentation, reduce the complexity and the overhead of the SPI communication with the device, and to allow to simultaneously use transceivers with different configuration regarding the used base band, the channel bandwidth, the modulation rate, and the channel map. Features of this driver include: - Support for the CC1100, CC1101, and the CC1100e sub-gigahertz transceivers. - Detailed documentation of every aspect of this driver. - An easy to use configuration API that allows setting the transceiver configuration (modulation rate, channel bandwidth, base frequency) and the channel map. - Fast channel hopping by pre-calibration of the channels during device configuration (so that no calibration is needed during hopping). - Simplified SPI communication: Only during start-up the MCU has to wait for the transceiver to be ready (for the power regulators and the crystal to stabilize). The old driver did this for every SPI transfer, which resulted in complex communication code. This driver will wait on start up for the transceiver to power up and then use RIOT's SPI API like every other driver. (Not only the data sheet states that this is fine, it also proved to be reliable in practise.) - Greatly reduced latency: The RTT on the old driver (@150 kbps data rate) was about 16ms, the new driver (@250 kbps data rate) has as RTT of ~3ms (depending on SPI clock and on CPU performance) (measured with ping6). - Increased reliability: The preamble size and the sync word size have been doubled compared to the old driver (preamble: 8 bytes instead of 4, sync word: 4 byte instead of 2). The new values are the once recommended by the data sheet for reliable communication. - Basic diagnostic during driver initialization to detect common issues as SPI communication issues and GDO pin configuration/wiring issues. - TX power configuration with netdev_driver_t::set() API-integration - Calls to netdev_driver_t::send() block until the transmission has completed to ease the use of the API (implemented without busy waiting, so that the MCU can enter lower power states or other threads can be executed).
2018-11-08 17:37:07 +01:00
/**
* @brief Perform a calibration of the frequency generator for each supported
* channel
drivers/cc110x: Rewrite of the cc110x driver The cc110x driver has been re-written from scratch to overcome the limitations of the old driver. The main motivation of the rewrite was to achieve better maintainability by a detailed documentation, reduce the complexity and the overhead of the SPI communication with the device, and to allow to simultaneously use transceivers with different configuration regarding the used base band, the channel bandwidth, the modulation rate, and the channel map. Features of this driver include: - Support for the CC1100, CC1101, and the CC1100e sub-gigahertz transceivers. - Detailed documentation of every aspect of this driver. - An easy to use configuration API that allows setting the transceiver configuration (modulation rate, channel bandwidth, base frequency) and the channel map. - Fast channel hopping by pre-calibration of the channels during device configuration (so that no calibration is needed during hopping). - Simplified SPI communication: Only during start-up the MCU has to wait for the transceiver to be ready (for the power regulators and the crystal to stabilize). The old driver did this for every SPI transfer, which resulted in complex communication code. This driver will wait on start up for the transceiver to power up and then use RIOT's SPI API like every other driver. (Not only the data sheet states that this is fine, it also proved to be reliable in practise.) - Greatly reduced latency: The RTT on the old driver (@150 kbps data rate) was about 16ms, the new driver (@250 kbps data rate) has as RTT of ~3ms (depending on SPI clock and on CPU performance) (measured with ping6). - Increased reliability: The preamble size and the sync word size have been doubled compared to the old driver (preamble: 8 bytes instead of 4, sync word: 4 byte instead of 2). The new values are the once recommended by the data sheet for reliable communication. - Basic diagnostic during driver initialization to detect common issues as SPI communication issues and GDO pin configuration/wiring issues. - TX power configuration with netdev_driver_t::set() API-integration - Calls to netdev_driver_t::send() block until the transmission has completed to ease the use of the API (implemented without busy waiting, so that the MCU can enter lower power states or other threads can be executed).
2018-11-08 17:37:07 +01:00
*
* @param dev Device descriptor of the transceiver
drivers/cc110x: Rewrite of the cc110x driver The cc110x driver has been re-written from scratch to overcome the limitations of the old driver. The main motivation of the rewrite was to achieve better maintainability by a detailed documentation, reduce the complexity and the overhead of the SPI communication with the device, and to allow to simultaneously use transceivers with different configuration regarding the used base band, the channel bandwidth, the modulation rate, and the channel map. Features of this driver include: - Support for the CC1100, CC1101, and the CC1100e sub-gigahertz transceivers. - Detailed documentation of every aspect of this driver. - An easy to use configuration API that allows setting the transceiver configuration (modulation rate, channel bandwidth, base frequency) and the channel map. - Fast channel hopping by pre-calibration of the channels during device configuration (so that no calibration is needed during hopping). - Simplified SPI communication: Only during start-up the MCU has to wait for the transceiver to be ready (for the power regulators and the crystal to stabilize). The old driver did this for every SPI transfer, which resulted in complex communication code. This driver will wait on start up for the transceiver to power up and then use RIOT's SPI API like every other driver. (Not only the data sheet states that this is fine, it also proved to be reliable in practise.) - Greatly reduced latency: The RTT on the old driver (@150 kbps data rate) was about 16ms, the new driver (@250 kbps data rate) has as RTT of ~3ms (depending on SPI clock and on CPU performance) (measured with ping6). - Increased reliability: The preamble size and the sync word size have been doubled compared to the old driver (preamble: 8 bytes instead of 4, sync word: 4 byte instead of 2). The new values are the once recommended by the data sheet for reliable communication. - Basic diagnostic during driver initialization to detect common issues as SPI communication issues and GDO pin configuration/wiring issues. - TX power configuration with netdev_driver_t::set() API-integration - Calls to netdev_driver_t::send() block until the transmission has completed to ease the use of the API (implemented without busy waiting, so that the MCU can enter lower power states or other threads can be executed).
2018-11-08 17:37:07 +01:00
*
* @retval 0 Success
* @retval -EINVAL Called with invalid argument
* @retval -EAGAIN Current state prevents deliberate calibration
* @retval -EIO Communication with the transceiver failed
drivers/cc110x: Rewrite of the cc110x driver The cc110x driver has been re-written from scratch to overcome the limitations of the old driver. The main motivation of the rewrite was to achieve better maintainability by a detailed documentation, reduce the complexity and the overhead of the SPI communication with the device, and to allow to simultaneously use transceivers with different configuration regarding the used base band, the channel bandwidth, the modulation rate, and the channel map. Features of this driver include: - Support for the CC1100, CC1101, and the CC1100e sub-gigahertz transceivers. - Detailed documentation of every aspect of this driver. - An easy to use configuration API that allows setting the transceiver configuration (modulation rate, channel bandwidth, base frequency) and the channel map. - Fast channel hopping by pre-calibration of the channels during device configuration (so that no calibration is needed during hopping). - Simplified SPI communication: Only during start-up the MCU has to wait for the transceiver to be ready (for the power regulators and the crystal to stabilize). The old driver did this for every SPI transfer, which resulted in complex communication code. This driver will wait on start up for the transceiver to power up and then use RIOT's SPI API like every other driver. (Not only the data sheet states that this is fine, it also proved to be reliable in practise.) - Greatly reduced latency: The RTT on the old driver (@150 kbps data rate) was about 16ms, the new driver (@250 kbps data rate) has as RTT of ~3ms (depending on SPI clock and on CPU performance) (measured with ping6). - Increased reliability: The preamble size and the sync word size have been doubled compared to the old driver (preamble: 8 bytes instead of 4, sync word: 4 byte instead of 2). The new values are the once recommended by the data sheet for reliable communication. - Basic diagnostic during driver initialization to detect common issues as SPI communication issues and GDO pin configuration/wiring issues. - TX power configuration with netdev_driver_t::set() API-integration - Calls to netdev_driver_t::send() block until the transmission has completed to ease the use of the API (implemented without busy waiting, so that the MCU can enter lower power states or other threads can be executed).
2018-11-08 17:37:07 +01:00
*
* Tunes in each supported channel and calibrates the transceiver. The
* calibration data is stored so that @ref cc110x_set_channel can skip the
* calibration phase and use the stored calibration data instead.
*/
int cc110x_full_calibration(cc110x_t *dev);
/**
* @brief Hops to the specified channel
*
* @param dev Device descriptor of the transceiver
* @param channel Channel to hop to
drivers/cc110x: Rewrite of the cc110x driver The cc110x driver has been re-written from scratch to overcome the limitations of the old driver. The main motivation of the rewrite was to achieve better maintainability by a detailed documentation, reduce the complexity and the overhead of the SPI communication with the device, and to allow to simultaneously use transceivers with different configuration regarding the used base band, the channel bandwidth, the modulation rate, and the channel map. Features of this driver include: - Support for the CC1100, CC1101, and the CC1100e sub-gigahertz transceivers. - Detailed documentation of every aspect of this driver. - An easy to use configuration API that allows setting the transceiver configuration (modulation rate, channel bandwidth, base frequency) and the channel map. - Fast channel hopping by pre-calibration of the channels during device configuration (so that no calibration is needed during hopping). - Simplified SPI communication: Only during start-up the MCU has to wait for the transceiver to be ready (for the power regulators and the crystal to stabilize). The old driver did this for every SPI transfer, which resulted in complex communication code. This driver will wait on start up for the transceiver to power up and then use RIOT's SPI API like every other driver. (Not only the data sheet states that this is fine, it also proved to be reliable in practise.) - Greatly reduced latency: The RTT on the old driver (@150 kbps data rate) was about 16ms, the new driver (@250 kbps data rate) has as RTT of ~3ms (depending on SPI clock and on CPU performance) (measured with ping6). - Increased reliability: The preamble size and the sync word size have been doubled compared to the old driver (preamble: 8 bytes instead of 4, sync word: 4 byte instead of 2). The new values are the once recommended by the data sheet for reliable communication. - Basic diagnostic during driver initialization to detect common issues as SPI communication issues and GDO pin configuration/wiring issues. - TX power configuration with netdev_driver_t::set() API-integration - Calls to netdev_driver_t::send() block until the transmission has completed to ease the use of the API (implemented without busy waiting, so that the MCU can enter lower power states or other threads can be executed).
2018-11-08 17:37:07 +01:00
*
* @retval 0 Success
* @retval -EINVAL Called with `NULL` as @p dev
* @retval -ERANGE Channel out of range or not supported by channel map
* @retval -EAGAIN Currently in a state that does not allow hopping, e.g.
* sending/receiving a packet, calibrating or handling
* transmission errors
* @retval -EIO Communication with the transceiver failed
drivers/cc110x: Rewrite of the cc110x driver The cc110x driver has been re-written from scratch to overcome the limitations of the old driver. The main motivation of the rewrite was to achieve better maintainability by a detailed documentation, reduce the complexity and the overhead of the SPI communication with the device, and to allow to simultaneously use transceivers with different configuration regarding the used base band, the channel bandwidth, the modulation rate, and the channel map. Features of this driver include: - Support for the CC1100, CC1101, and the CC1100e sub-gigahertz transceivers. - Detailed documentation of every aspect of this driver. - An easy to use configuration API that allows setting the transceiver configuration (modulation rate, channel bandwidth, base frequency) and the channel map. - Fast channel hopping by pre-calibration of the channels during device configuration (so that no calibration is needed during hopping). - Simplified SPI communication: Only during start-up the MCU has to wait for the transceiver to be ready (for the power regulators and the crystal to stabilize). The old driver did this for every SPI transfer, which resulted in complex communication code. This driver will wait on start up for the transceiver to power up and then use RIOT's SPI API like every other driver. (Not only the data sheet states that this is fine, it also proved to be reliable in practise.) - Greatly reduced latency: The RTT on the old driver (@150 kbps data rate) was about 16ms, the new driver (@250 kbps data rate) has as RTT of ~3ms (depending on SPI clock and on CPU performance) (measured with ping6). - Increased reliability: The preamble size and the sync word size have been doubled compared to the old driver (preamble: 8 bytes instead of 4, sync word: 4 byte instead of 2). The new values are the once recommended by the data sheet for reliable communication. - Basic diagnostic during driver initialization to detect common issues as SPI communication issues and GDO pin configuration/wiring issues. - TX power configuration with netdev_driver_t::set() API-integration - Calls to netdev_driver_t::send() block until the transmission has completed to ease the use of the API (implemented without busy waiting, so that the MCU can enter lower power states or other threads can be executed).
2018-11-08 17:37:07 +01:00
*
* This function implements the fact channel hopping approach outlined in
* section 28.2 on page 64 in the data sheet, which skips the calibration phase
* by storing the calibration date for each channel in the driver.
*/
int cc110x_set_channel(cc110x_t *dev, uint8_t channel);
/**
* @brief Set the TX power to the specified value
drivers/cc110x: Rewrite of the cc110x driver The cc110x driver has been re-written from scratch to overcome the limitations of the old driver. The main motivation of the rewrite was to achieve better maintainability by a detailed documentation, reduce the complexity and the overhead of the SPI communication with the device, and to allow to simultaneously use transceivers with different configuration regarding the used base band, the channel bandwidth, the modulation rate, and the channel map. Features of this driver include: - Support for the CC1100, CC1101, and the CC1100e sub-gigahertz transceivers. - Detailed documentation of every aspect of this driver. - An easy to use configuration API that allows setting the transceiver configuration (modulation rate, channel bandwidth, base frequency) and the channel map. - Fast channel hopping by pre-calibration of the channels during device configuration (so that no calibration is needed during hopping). - Simplified SPI communication: Only during start-up the MCU has to wait for the transceiver to be ready (for the power regulators and the crystal to stabilize). The old driver did this for every SPI transfer, which resulted in complex communication code. This driver will wait on start up for the transceiver to power up and then use RIOT's SPI API like every other driver. (Not only the data sheet states that this is fine, it also proved to be reliable in practise.) - Greatly reduced latency: The RTT on the old driver (@150 kbps data rate) was about 16ms, the new driver (@250 kbps data rate) has as RTT of ~3ms (depending on SPI clock and on CPU performance) (measured with ping6). - Increased reliability: The preamble size and the sync word size have been doubled compared to the old driver (preamble: 8 bytes instead of 4, sync word: 4 byte instead of 2). The new values are the once recommended by the data sheet for reliable communication. - Basic diagnostic during driver initialization to detect common issues as SPI communication issues and GDO pin configuration/wiring issues. - TX power configuration with netdev_driver_t::set() API-integration - Calls to netdev_driver_t::send() block until the transmission has completed to ease the use of the API (implemented without busy waiting, so that the MCU can enter lower power states or other threads can be executed).
2018-11-08 17:37:07 +01:00
*
* @param dev Device descriptor of the transceiver
* @param power Output power to apply
drivers/cc110x: Rewrite of the cc110x driver The cc110x driver has been re-written from scratch to overcome the limitations of the old driver. The main motivation of the rewrite was to achieve better maintainability by a detailed documentation, reduce the complexity and the overhead of the SPI communication with the device, and to allow to simultaneously use transceivers with different configuration regarding the used base band, the channel bandwidth, the modulation rate, and the channel map. Features of this driver include: - Support for the CC1100, CC1101, and the CC1100e sub-gigahertz transceivers. - Detailed documentation of every aspect of this driver. - An easy to use configuration API that allows setting the transceiver configuration (modulation rate, channel bandwidth, base frequency) and the channel map. - Fast channel hopping by pre-calibration of the channels during device configuration (so that no calibration is needed during hopping). - Simplified SPI communication: Only during start-up the MCU has to wait for the transceiver to be ready (for the power regulators and the crystal to stabilize). The old driver did this for every SPI transfer, which resulted in complex communication code. This driver will wait on start up for the transceiver to power up and then use RIOT's SPI API like every other driver. (Not only the data sheet states that this is fine, it also proved to be reliable in practise.) - Greatly reduced latency: The RTT on the old driver (@150 kbps data rate) was about 16ms, the new driver (@250 kbps data rate) has as RTT of ~3ms (depending on SPI clock and on CPU performance) (measured with ping6). - Increased reliability: The preamble size and the sync word size have been doubled compared to the old driver (preamble: 8 bytes instead of 4, sync word: 4 byte instead of 2). The new values are the once recommended by the data sheet for reliable communication. - Basic diagnostic during driver initialization to detect common issues as SPI communication issues and GDO pin configuration/wiring issues. - TX power configuration with netdev_driver_t::set() API-integration - Calls to netdev_driver_t::send() block until the transmission has completed to ease the use of the API (implemented without busy waiting, so that the MCU can enter lower power states or other threads can be executed).
2018-11-08 17:37:07 +01:00
*
* @retval 0 Success
* @retval -EINVAL Called with `NULL` as @p dev
* @retval -ERANGE Called with an invalid value for @p power
* @retval -EAGAIN Changing the TX power is in the current state not possible
* @retval -EIO Communication with the transceiver failed
drivers/cc110x: Rewrite of the cc110x driver The cc110x driver has been re-written from scratch to overcome the limitations of the old driver. The main motivation of the rewrite was to achieve better maintainability by a detailed documentation, reduce the complexity and the overhead of the SPI communication with the device, and to allow to simultaneously use transceivers with different configuration regarding the used base band, the channel bandwidth, the modulation rate, and the channel map. Features of this driver include: - Support for the CC1100, CC1101, and the CC1100e sub-gigahertz transceivers. - Detailed documentation of every aspect of this driver. - An easy to use configuration API that allows setting the transceiver configuration (modulation rate, channel bandwidth, base frequency) and the channel map. - Fast channel hopping by pre-calibration of the channels during device configuration (so that no calibration is needed during hopping). - Simplified SPI communication: Only during start-up the MCU has to wait for the transceiver to be ready (for the power regulators and the crystal to stabilize). The old driver did this for every SPI transfer, which resulted in complex communication code. This driver will wait on start up for the transceiver to power up and then use RIOT's SPI API like every other driver. (Not only the data sheet states that this is fine, it also proved to be reliable in practise.) - Greatly reduced latency: The RTT on the old driver (@150 kbps data rate) was about 16ms, the new driver (@250 kbps data rate) has as RTT of ~3ms (depending on SPI clock and on CPU performance) (measured with ping6). - Increased reliability: The preamble size and the sync word size have been doubled compared to the old driver (preamble: 8 bytes instead of 4, sync word: 4 byte instead of 2). The new values are the once recommended by the data sheet for reliable communication. - Basic diagnostic during driver initialization to detect common issues as SPI communication issues and GDO pin configuration/wiring issues. - TX power configuration with netdev_driver_t::set() API-integration - Calls to netdev_driver_t::send() block until the transmission has completed to ease the use of the API (implemented without busy waiting, so that the MCU can enter lower power states or other threads can be executed).
2018-11-08 17:37:07 +01:00
*/
int cc110x_set_tx_power(cc110x_t *dev, cc110x_tx_power_t power);
/**
* @brief Wakes the transceiver from SLEEP mode and enters RX mode
*
* @retval 0 Success
* @retval -EIO Communication with the transceiver failed
*/
int cc110x_wakeup(cc110x_t *dev);
/**
* @brief Sets the transceiver into SLEEP mode.
*
* Only @ref cc110x_wakeup can awake the device again.
*/
void cc110x_sleep(cc110x_t *dev);
drivers/cc110x: Rewrite of the cc110x driver The cc110x driver has been re-written from scratch to overcome the limitations of the old driver. The main motivation of the rewrite was to achieve better maintainability by a detailed documentation, reduce the complexity and the overhead of the SPI communication with the device, and to allow to simultaneously use transceivers with different configuration regarding the used base band, the channel bandwidth, the modulation rate, and the channel map. Features of this driver include: - Support for the CC1100, CC1101, and the CC1100e sub-gigahertz transceivers. - Detailed documentation of every aspect of this driver. - An easy to use configuration API that allows setting the transceiver configuration (modulation rate, channel bandwidth, base frequency) and the channel map. - Fast channel hopping by pre-calibration of the channels during device configuration (so that no calibration is needed during hopping). - Simplified SPI communication: Only during start-up the MCU has to wait for the transceiver to be ready (for the power regulators and the crystal to stabilize). The old driver did this for every SPI transfer, which resulted in complex communication code. This driver will wait on start up for the transceiver to power up and then use RIOT's SPI API like every other driver. (Not only the data sheet states that this is fine, it also proved to be reliable in practise.) - Greatly reduced latency: The RTT on the old driver (@150 kbps data rate) was about 16ms, the new driver (@250 kbps data rate) has as RTT of ~3ms (depending on SPI clock and on CPU performance) (measured with ping6). - Increased reliability: The preamble size and the sync word size have been doubled compared to the old driver (preamble: 8 bytes instead of 4, sync word: 4 byte instead of 2). The new values are the once recommended by the data sheet for reliable communication. - Basic diagnostic during driver initialization to detect common issues as SPI communication issues and GDO pin configuration/wiring issues. - TX power configuration with netdev_driver_t::set() API-integration - Calls to netdev_driver_t::send() block until the transmission has completed to ease the use of the API (implemented without busy waiting, so that the MCU can enter lower power states or other threads can be executed).
2018-11-08 17:37:07 +01:00
#ifdef __cplusplus
}
#endif
#endif /* CC110X_H */
/** @} */