LoraWAN-in-C library, adapted to run under the Arduino environment
#157 Compare This branch is 416 commits ahead, 10 commits behind things-nyc:master.
Latest commit 714a7d6 Nov 20, 2018

README.md

Arduino-LMIC library

This repository contains the IBM LMIC (LoraMAC-in-C) library, slightly modified to run in the Arduino environment, allowing using the SX1272, SX1276 transceivers and compatible modules (such as some HopeRF RFM9x modules and the Murata LoRa modules).

This library mostly exposes the functions defined by LMIC, it makes no attempt to wrap them in a higher level API that is more in the Arduino style. To find out how to use the library itself, see the examples, or see the PDF file in the doc subdirectory.

The MCCI arduino-lorawan library provides a higher level, more Arduino-like wrapper which may be useful.

This library requires Arduino IDE version 1.6.6 or above, since it requires C99 mode to be enabled by default.

GitHub release GitHub commits Build Status

Contents:

Installing

To install this library:

  • install it using the Arduino Library manager ("Sketch" -> "Include Library" -> "Manage Libraries..."), or
  • download a zipfile from github using the "Download ZIP" button and install it using the IDE ("Sketch" -> "Include Library" -> "Add .ZIP Library..."
  • clone this git repository into your sketchbook/libraries folder.

For more info, see https://www.arduino.cc/en/Guide/Libraries

Features

The LMIC library provides a fairly complete LoRaWAN Class A and Class B implementation, supporting the EU-868, US-915, AU-921, AS-923, and IN-866 bands. Only a limited number of features was tested using this port on Arduino hardware, so be careful when using any of the untested features.

The library has only been tested with LoRaWAN 1.0.2 networks and does not have the separated key structure defined by LoRaWAN 1.1.

What certainly works:

  • Sending packets uplink, taking into account duty cycling.
  • Encryption and message integrity checking.
  • Receiving downlink packets in the RX2 window.
  • Custom frequencies and datarate settings.
  • Over-the-air activation (OTAA / joining).
  • Receiving downlink packets in the RX1 and RX2 windows.
  • Some MAC command processing.

What has not been tested:

  • Receiving and processing all MAC commands.
  • Class B operation.
  • FSK has not been extensively tested.

If you try one of these untested features and it works, be sure to let us know (creating a github issue is probably the best way for that).

Additional Documentation

PDF/Word Documentation

The doc directory contains LMiC-v2.3.pdf, which documents the library APIs and use. It's based on the original IBM documentation, but has been adapted for this version of the library. However, as this library is used for more than Arduino, that document is supplemented by practical details in this document.

Adding Bandplans

There is a general framework for adding new region support. HOWTO-ADD-REGION.md has step-by-step instructions for adding a region.

Known bugs and issues

See the list of bugs at mcci-catena/arduino-lmic.

Configuration

A number of features can be enabled or disabled at compile time. This is done by adding the desired settings to the file project_settings/lmic_project_config.h. The project_settings directory is the only directory that contains files that you should edit to match your project; we organize things this way so that your local changes are more clearly separated from the distribution files. The Arduino environment doesn't give us a better way to do this, unless you change BOARDS.txt.

Unlike other ports of the LMIC code, in this port, you should not edit src/lmic/config.h to configure this package.

The following configuration variables are available.

Selecting the LoRaWAN Region Configuration

The library supports the following regions:

-D variable CFG region name CFG region value LoRa Spec Reference Frequency
-D CFG_eu868 LMIC_REGION_eu868 1 2.1 EU 863-870 MHz ISM
-D CFG_us915 LMIC_REGION_us915 2 2.2 US 902-928 MHz ISM
-D CFG_au921 LMIC_REGION_au921 5 2.5 Australia 915-928 MHz ISM
-D CFG_as923 LMIC_REGION_as923 7 2.7 Asia 923 MHz ISM
-D CFG_as923jp LMIC_REGION_as923 and LMIC_COUNTRY_CODE_JP 7 2.7 Asia 923 MHz ISM with Japan listen-before-talk (LBT) rules
-D CFG_in866 LMIC_REGION_in866 9 2.9 India 865-867 MHz ISM

You should define exactly one of CFG_... variables. If you don't, the library assumes CFG_eu868. The library changes configuration pretty substantially according to the region. Some of the differences are listed below.

eu868, as923, in866

If the library is configured for EU868, AS923, or IN866 operation, we make the following changes:

  • Add the API LMIC_setupBand().
  • Add the constants MAX_CHANNELS, MAX_BANDS, LIMIT_CHANNELS, BAND_MILLI, BAND_CENTI, BAND_DECI, and BAND_AUX.

us915, au921

If the library is configured for US915 operation, we make the following changes:

  • Add the APIs LMIC_enableChannel(), LMIC_enableSubBand(), LMIC_disableSubBand(), and LMIC_selectSubBand().
  • Add the constants MAX_XCHANNELS.
  • Add a number of additional DR_... symbols.

Selecting the target radio transceiver

You should define one of the following variables. If you don't, the library assumes sx1276. There is a runtime check to make sure the actual transceiver matches the library configuration.

#define CFG_sx1272_radio 1

Configures the library for use with an sx1272 transceiver.

#define CFG_sx1276_radio 1

Configures the library for use with an sx1276 transceiver.

Controlling use of interrupts

#define LMIC_USE_INTERRUPTS

If defined, configures the library to use interrupts for detecting events from the transceiver. If left undefined, the library will poll for events from the transceiver. See Timing for more info.

Disabling PING

#define DISABLE_PING

If defined, removes all code needed for Class B downlink during ping slots (PING). Removes the APIs LMIC_setPingable() and LMIC_stopPingable(). Class A devices don't support PING, so defining DISABLE_PING is often a good idea.

By default, PING support is included in the library.

Disabling Beacons

#define DISABLE_BEACONS

If defined, removes all code needed for handling beacons. Removes the APIs LMIC_enableTracking() and LMIC_disableTracking().

Enabling beacon handling allows tracking of network time, and is required if you want to enable downlink during ping slots. However, many networks don't support Class B devices. Class A devices don't support tracking beacons, so defining DISABLE_BEACONS might be a good idea.

By default, beacon support is included in the library.

Enabling Network Time Support

#define LMIC_ENABLE_DeviceTimeReq number /* boolean: 0 or non-zero */

Disable or enable support for device network-time requests (LoRaWAN MAC request 0x0D). If zero, support is disabled. If non-zero, support is enabled.

If disabled, stub routines are provided that will return failure (so you don't need conditional compiles in client code).

Rarely changed variables

The remaining variables are rarely used, but we list them here for completeness.

Changing debug output

#define LMIC_PRINTF_TO SerialLikeObject

This variable should be set to the name of a Serial-like object, used for printing messages. If not defined, Serial is assumed.

Getting debug from the RF library

#define LMIC_DEBUG_LEVEL number /* 0, 1, or 2 */

This variable determines the amount of debug output to be produced by the library. The default is 0.

If LMIC_DEBUG_LEVEL is zero, no output is produced. If 1, limited output is produced. If 2, more extensive output is produced. If non-zero, printf() is used, and the Arduino environment must be configured to support it, otherwise the sketch will crash at runtime.

Selecting the AES library

The library comes with two AES implementations. The original implementation is better on ARM processors becasue it's faster, but it's larger. For smaller AVR8 processors, a second library ("IDEETRON") is provided that has a smaller code footprint. You may define one of the following variables to choose the AES implementation. If you don't, the library uses the IDEETRON version.

#define USE_ORIGINAL_AES

If defined, the original AES implementation is used.

#define USE_IDEETRON_AES

If defined, the IDEETRON AES implementation is used.

Defining the OS Tick Frequency

#define US_PER_OSTICK_EXPONENT number

This variable should be set to the base-2 logarithm of the number of microseconds per OS tick. The default is 4, which indicates that each tick corresponds to 16 microseconds (because 16 == 2^4).

Setting the SPI-bus frequency

#define LMIC_SPI_FREQ floatNumber

This variable sets the default frequency for the SPI bus connection to the transceiver. The default is 1E6, meaning 1 MHz. However, this can be overridden by the contents of the lmic_pinmap structure, and we recommend that you use that approach rather than editing the project_settings/lmic_project_config.h file.

Changing handling of runtime assertion failures

The variables LMIC_FAILURE_TO and DISABLE_LMIC_FAILURE_TO control the handling of runtime assertion failures. By default, assertion messages are displayed using the Serial object. You can define LMIC_FAILURE_TO to be the name of some other Print-like obect. You can also define DISABLE_LMIC_FAILURE_TO to any value, in which case assert failures will silently halt execution.

Disabling JOIN

#define DISABLE_JOIN

If defined, removes code needed for OTAA activation. Removes the APIs LMIC_startJoining() and LMIC_tryRejoin().

Disabling Class A MAC commands

DISABLE_MCMD_DCAP_REQ, DISABLE_MCMD_DN2P_SET, and DISABLE_MCMD_SNCH_REQ respectively disable code for various Class A MAC commands.

Disabling Class B MAC commands

DISABLE_MCMD_PING_SET disables the PING_SET MAC commands. It's implied by DISABLE_PING.

DISABLE_MCMD_BCNI_ANS disables the next-beacon start command. It's implied by DISABLE_BEACON

Special purpose

#define DISABLE_INVERT_IQ_ON_RX disables the inverted Q-I polarity on RX. If this is defined, end-devices will be able to receive messages from each other, but will not be able to hear the gateway.

Supported hardware

This library is intended to be used with plain LoRa transceivers, connecting to them using SPI. In particular, the SX1272 and SX1276 families are supported (which should include SX1273, SX1277, SX1278 and SX1279 which only differ in the available frequencies, bandwidths and spreading factors). It has been tested with both SX1272 and SX1276 chips, using the Semtech SX1272 evaluation board and the HopeRF RFM92 and RFM95 boards (which supposedly contain an SX1272 and SX1276 chip respectively).

This library contains a full LoRaWAN stack and is intended to drive these Transceivers directly. It is not intended to be used with full-stack devices like the Microchip RN2483 and the Embit LR1272E. These contain a transceiver and microcontroller that implements the LoRaWAN stack and exposes a high-level serial interface instead of the low-level SPI transceiver interface.

This library is intended to be used inside the Arduino environment. It should be architecture-independent, so it should run on "normal" AVR arduinos, but also on the ARM-based ones, and some success has been seen running on the ESP8266 board as well. It was tested on the Arduino Uno, Pinoccio Scout, Teensy LC and 3.x, ESP8266, Arduino 101, Adafruit Feather M0 LoRa 900. It has been tested on the Lattice RISC-V CPU soft core running in an iCE40 UltraPlus, and also on the Murata LoRaWAN module on the MCCI Catena 4551.

This library an be quite heavy on small systems, especially if the fairly small ATmega 328p (such as in the Arduino Uno) is used. In the default configuration, the available 32K flash space is nearly filled up (this includes some debug output overhead, though). By disabling some features in project_settings/lmic_project_config.h (like beacon tracking and ping slots, which are not needed for Class A devices), some space can be freed up.

Connections

To make this library work, your Arduino (or whatever Arduino-compatible board you are using) should be connected to the transceiver. In some cases (such as the Adafruit Feather series and Murata-based boards such as the MCCI Catena 4551), the settings are fixed by the board, and you won't have to worry about many of these details. However, you'll need to find the configuration that's suitable for your board.

To help you know if you have to worry, we'll call such boards "pre-integrated" and prefix each section with suitable guidance.

The exact connections are a bit dependent on the transceiver board and Arduino used, so this section tries to explain what each connection is for and in what cases it is (not) required.

Note that the SX1272 module runs at 3.3V and likely does not like 5V on its pins (though the datasheet is not say anything about this, and my transceiver did not obviously break after accidentally using 5V I/O for a few hours). To be safe, make sure to use a level shifter, or an Arduino running at 3.3V. The Semtech evaluation board has 100 ohm resistors in series with all data lines that might prevent damage, but I would not count on that.

Power

If you're using a pre-integrated board, you can skip this section.

The SX127x transceivers need a supply voltage between 1.8V and 3.9V. Using a 3.3V supply is typical. Some modules have a single power pin (like the HopeRF modules, labeled 3.3V) but others expose multiple power pins for different parts (like the Semtech evaluation board that has VDD_RF, VDD_ANA and VDD_FEM), which can all be connected together. Any GND pins need to be connected to the Arduino GND pin(s).

SPI

If you're using a pre-integrated board, you can skip this section, and instead refer to your board's documentation on the pins to be used.

The primary way of communicating with the transceiver is through SPI (Serial Peripheral Interface). This uses four pins: MOSI, MISO, SCK and SS. The former three need to be directly connected: so MOSI to MOSI, MISO to MISO, SCK to SCK. Where these pins are located on your Arduino varies, see for example the "Connections" section of the Arduino SPI documentation.

The SS (slave select) connection is a bit more flexible. On the SPI slave side (the transceiver), this must be connect to the pin (typically) labeled NSS. On the SPI master (Arduino) side, this pin can connect to any I/O pin. Most Arduinos also have a pin labeled "SS", but this is only relevant when the Arduino works as an SPI slave, which is not the case here. Whatever pin you pick, you need to tell the library what pin you used through the pin mapping (see below).

DIO pins

If you're using a pre-integrated board, you can ignore this section; refer to your board's documentation for information on what DIO pins need to be used.

The DIO (digital I/O) pins on the SX127x can be configured for various functions. The LMIC library uses them to get instant status information from the transceiver. For example, when a LoRa transmission starts, the DIO0 pin is configured as a TxDone output. When the transmission is complete, the DIO0 pin is made high by the transceiver, which can be detected by the LMIC library.

The LMIC library needs only access to DIO0, DIO1 and DIO2, the other DIOx pins can be left disconnected. On the Arduino side, they can connect to any I/O pin. If interrupts are used, the accuracy of timing will be improved, particularly the rest of your loop() function has lengthy calculations; but in that case, the enabled DIO pins must all support rising-edge interrupts. See the Timing section below.

In LoRa mode the DIO pins are used as follows:

  • DIO0: TxDone and RxDone
  • DIO1: RxTimeout

In FSK mode they are used as follows::

  • DIO0: PayloadReady and PacketSent
  • DIO2: TimeOut

Both modes need only 2 pins, but the transceiver does not allow mapping them in such a way that all needed interrupts map to the same 2 pins. So, if both LoRa and FSK modes are used, all three pins must be connected.

The pins used on the Arduino side should be configured in the pin mapping in your sketch, by setting the values of lmic_pinmap::dio[0], [1], and [2] (see below).

Reset

If you're using a pre-configured module, refer to the documentation for your board.

The transceiver has a reset pin that can be used to explicitly reset it. The LMIC library uses this to ensure the chip is in a consistent state at startup. In practice, this pin can be left disconnected, since the transceiver will already be in a sane state on power-on, but connecting it might prevent problems in some cases.

On the Arduino side, any I/O pin can be used. The pin number used must be configured in the pin mapping lmic_pinmap::rst field (see below).

RXTX

If you're using a pre-configured module, refer to the documentation for your board.

The transceiver contains two separate antenna connections: One for RX and one for TX. A typical transceiver board contains an antenna switch chip, that allows switching a single antenna between these RX and TX connections. Such a antenna switcher can typically be told what position it should be through an input pin, often labeled RXTX.

The easiest way to control the antenna switch is to use the RXTX pin on the SX127x transceiver. This pin is automatically set high during TX and low during RX. For example, the HopeRF boards seem to have this connection in place, so they do not expose any RXTX pins and the pin can be marked as unused in the pin mapping.

Some boards do expose the antenna switcher pin, and sometimes also the SX127x RXTX pin. For example, the SX1272 evaluation board calls the former FEM_CTX and the latter RXTX. Again, simply connecting these together with a jumper wire is the easiest solution.

Alternatively, or if the SX127x RXTX pin is not available, LMIC can be configured to control the antenna switch. Connect the antenna switch control pin (e.g. FEM_CTX on the Semtech evaluation board) to any I/O pin on the Arduino side, and configure the pin used in the pin map (see below).

The configuration entry lmic_pinmap::rxtx configures the pin to be used for the RXTX control function, in terms of the Arduino wire.h digital pin number. If set to LMIC_UNUSED_PIN, then the library assumes that software does not need to control the antenna switch.

RXTX Polarity

If an external switch is used, you also must specify the polarity. Some modules want RXTX to be high for transmit, low for receive; Others want it to be low for transmit, high for receive. The Murata module, for example, requires that RXTX be high for receive, low for transmit.

The configuration entry lmic_pinmap::rxtx_rx_active should be set to the state to be written to the RXTX pin to make the receiver active. The opposite state is written to make the transmitter active. If lmic_pinmap::rxtx is LMIC_UNUSED_PIN, then the value of lmic_pinmap::rxtx_rx_active is ignored.

Pin mapping

For pre-configured boards, refer to the documentation on your board for the required settings. See the following:

If you don't have the board documentation, you need to provide your own lmic_pinmap values. As described above, a variety of configurations are possible. To tell the LMIC library how your board is configured, you must declare a variable containing a pin mapping struct in the sketch file.

For example, this could look like this:

  lmic_pinmap lmic_pins = {
    .nss = 6,
    .rxtx = LMIC_UNUSED_PIN,
    .rst = 5,
    .dio = {2, 3, 4},
    // optional: set polarity of rxtx pin.
    .rxtx_rx_active = 0,
    // optional: set RSSI cal for listen-before-talk
    // this value is in dB, and is added to RSSI
    // measured prior to decision.
    // Must include noise guardband! Ignored in US,
    // EU, IN, other markets where LBT is not required.
    .rssi_cal = 0,
    // optional: override LMIC_SPI_FREQ if non-zero
    .spi_freq = 0,
  };

The names refer to the pins on the transceiver side, the numbers refer to the Arduino pin numbers (to use the analog pins, use constants like A0). For the DIO pins, the three numbers refer to DIO0, DIO1 and DIO2 respectively. Any pins that are not needed should be specified as LMIC_UNUSED_PIN. The nss and dio0 pin is required, the others can potentially left out (depending on the environments and requirements, see the notes above for when a pin can or cannot be left out).

The name of the variable containing this struct must always be lmic_pins, which is a special name recognized by the library.

Adafruit Feather M0 LoRa

See Adafruit's Feather M0 LoRa product page. This board uses the following pin mapping, as shown in the various "...-feather" sketches.

DIO0 is hard-wired by Adafruit to Arduino D3, but DIO1 is not connected to any Arduino pin (it comes to JP1 pin 1, but is not otherwise connected). This pin table assumes that you have manually wired JP1 pin 1 to Arduino JP3 pin 9 (Arduino D6).

DIO2 is not connected.

const lmic_pinmap lmic_pins = {
    .nss = 8,
    .rxtx = LMIC_UNUSED_PIN,
    .rst = 4,
    .dio = {3, 6, LMIC_UNUSED_PIN},
};

Adafruit Feather 32u4 LoRa

See Adafruit's Feather 32u4 LoRa product page. This board is supported by the ttn-otaa-feather-us915.ino example sketch. It uses the same pin mapping as the Feather M0 LoRa.

LoRa Nexus by Ideetron

This board uses the following pin mapping:

  const lmic_pinmap lmic_pins = {
      .nss = 10,
      .rxtx = LMIC_UNUSED_PIN,
      .rst = LMIC_UNUSED_PIN, // hardwired to AtMega RESET
      .dio = {4, 5, 7},
  };

MCCI Catena 4450/4460

See MCCI Catena 4450 and MCCI Catena 4460.

These modules are based on the Feather M0 LoRa. Since they include an extra Feather wing for the sensors, the Feather wing includes the trace connecting DIO1 to Arduino D6. No user wiring is needed on the Feather M0.

const lmic_pinmap lmic_pins = {
    .nss = 8,
    .rxtx = LMIC_UNUSED_PIN,
    .rst = 4,
    .dio = {3, 6, LMIC_UNUSED_PIN},
};

MCCI Catena 4551

See MCCI Catena 4551. This board uses a Murata LoRa module and has the following pin mapping:

const lmic_pinmap lmic_pins = {
    .nss = 7,
    .rxtx = 29,
    .rst = 8,
    .dio = {25, 26, 27},
    // the Murata module needs D29 high for RX, low for TX.
    .rxtx_rx_active = 1,
    // the Murata module is direct-wired, we can use 8 MHz for SPI.
    .spi_freq = 8000000
};

Example Sketches

This library provides several examples.

  • ttn-otaa.ino shows a basic transmission of a "Hello, world!" message using the LoRaWAN protocol. It contains some frequency settings and encryption keys intended for use with The Things Network, but these also correspond to the default settings of most gateways, so it should work with other networks and gateways as well. The example uses over-the-air activation (OTAA) to first join the network to establish a session and security keys. This was tested with The Things Network, but should also work (perhaps with some changes) for other networks. OTAA is the preferred way to work with production LoRaWAN networks.

  • ttn-otaa-feather-us915.ino is a version of ttn-otaa.ino that has been configured for use with the Feather M0 LoRa, on the US915 bandplan, with The Things Network. Remember that you may also have to change config.h from defaults. This sketch also works with the MCCI Catena family of products as well as with the Feather 32u4 LoRa.

  • ttn-otaa-feather-us915-dht22.ino is a further refinement of ttn-otaa-feather-us915.ino. It measures and transmits temperature and relative humidity using a DHT22 sensor. It's only been tested with Feather M0-family products.

  • raw.ino shows how to access the radio on a somewhat low level, and allows to send raw (non-LoRaWAN) packets between nodes directly. This is useful to verify basic connectivity, and when no gateway is available, but this example also bypasses duty cycle checks, so be careful when changing the settings.

  • raw-feather.ino is a version of raw.ino that is completely configured for the Adafruit Feather M0 LoRa, and for a variety of other MCCI products.

  • ttn-abp.ino shows a basic transmission of a "Hello, world!" message using the LoRaWAN protocol. This example uses activation-by-personalization (ABP, preconfiguring a device address and encryption keys), and does not employ over-the-air activation.

    ABP should not be used if you have access to a production gateway and network; it's not compliant with LoRaWAN standards, it's not FCC compliant, and it's uses spectrum in a way that's unfair to other users. However, it's often the most economical way to get your feet wet with this technology. It's possible to do ABP compliantly with the LMIC framework, but you need to have FRAM storage and a framework that saves uplink and downlink counts across reboots and resets. See, for example, Catena-Arduino-Platform.

  • ttn-abp-feather-us915-dht22.ino refines ttn-abp.ino by configuring for use with the Feather M0 LoRa on the US915 bandplan, with a single-channel gateway on The Things Network; it measures and transmits temperature and relative humidity using a DHT22 sensor. It's only been tested with Feather M0-family products.

    ABP should not be used if you have access to a production gateway and network; it's not compliant with LoRaWAN standards, it's not FCC compliant, and it's uses spectrum in a way that's unfair to other users. However, it's often the most economical way to get your feet wet with this technology. It's possible to do ABP compliantly with the LMIC framework, but you need to have FRAM storage and a framework that saves uplink and downlink counts across reboots and resets. See, for example, Catena-Arduino-Platform.

  • header_test.ino just tests the header files; it's used for regression testing.

Timing

The library is responsible for keeping track of time of certain network events, and scheduling other events relative to those events. In particular, the library must note when a packet finishes transmitting, so it can open up the RX1 and RX2 receive windows at a fixed time after the end of transmission. The library does this by watching for rising edges on the DIO0 output of the SX127x, and noting the time.

The library observes and processes rising edges on the pins as part of os_runloop() processing. This can be configured in one of two ways (see Controlling use of interrupts).

By default, the routine hal_io_check() polls the enabled pins to determine whether an event has occured. This approach allows use of any CPU pin to sense the DIOs, and makes no assummptions about interrupts. However, it means that the end-of-transmit event is not observed (and time-stamped) until os_runloop() is called.

Optionally, you can configure the LMIC library to use interrupts. The interrupt handlers capture the time of the event. Actual processing is done the next time that os_runloop() is called, using the captured time. However, this requires that the DIO pins be wired to Arduino pins that support rising-edge interrupts.

Fortunately, LoRa is a fairly slow protocol and the timing of the receive windows is not super critical. To synchronize transmitter and receiver, a preamble is first transmitted. Using LoRaWAN, this preamble consists of 8 symbols, of which the receiver needs to see 4 symbols to lock on. The current implementation tries to enable the receiver for 5 symbol times at 1.5 symbol after the start of the receive window, meaning that a inaccuracy of plus or minus 2.5 symbol times should be acceptable.

The HAL bases all timing on the Arduino micros() timer, which has a platform-specific granularity, and is based on the primary microcontroller clock.

At the fastest LoRa setting supported by the SX127x (SF5BW500) a single preamble symbol takes 64 microseconds, so the receive window timing should be accurate within 160 microseconds (for LoRaWAN this is SF7BW250, needing accuracy within 1280μs). This accuracy should also be feasible with the polling approach used, provided that the LMIC loop is run often enough.

If using an internal oscillator (which is 1% - 10% accurate, depending on calibration), or if your other loop() processing is time consuming, you may have to use LMIC_setClockError() to cause the library to leave the radio on longer.

An even more accurate solution could be to use a dedicated timer with an input capture unit, that can store the timestamp of a change on the DIO0 pin (the only one that is timing-critical) entirely in hardware. Experience shows that this is not normally required, so we leave this as a customization to be performed on a platform-by-platfom basis. We provide a special API, radio_irq_handler_v2(u1_t dio, ostime_t tEvent). This API allows you to supply a hardware-captured time for extra accuracy.

The practical consequence of inaccurate timing is reduced battery life; the LMIC must turn on the reciever earlier in order to be sure to capture downlink packets.

LMIC_setClockError()

You may call this routine during intialization to infom the LMIC code about the timing accuracy of your system.

enum { MAX_CLOCK_ERROR = 65535 };

void LMIC_setClockError(
    u2_t error
);

This function sets the anticipated relative clock error. MAX_CLOCK_ERROR represents +/- 100%, and 0 represents no additional clock compensation. To allow for an error of 20%, you would call

LMIC_setClockError(MAX_CLOCK_ERROR * 20 / 100);

Setting a high clock error causes the RX windows to be opened earlier than it otherwise would be. This causes more power to be consumed. For Class A devices, this extra power is not substantial, but for Class B devices, this can be significant.

This clock error is not reset by LMIC_reset().

Downlink datarate

Note that the datarate used for downlink packets in the RX2 window varies by region. Consult your network's manual for any divergences from the LoRaWAN Regional Parameters. This library assumes that the network follows the regional default.

Some networks use different values than the specification. For example, in Europe, the specification default is DR0 (SF12, 125 kHz bandwidth). However, iot.semtech.com and The Things Network both used SF9 / 125 kHz or DR3). If using over-the-air activation (OTAA), the network will download RX2 parameters as part of the JoinAccept message; the LMIC will honor the downloaded parameters.

However, when using personalized activate (ABP), it is your responsibility to set the right settings, e.g. by adding this to your sketch (after calling LMIC_setSession). ttn-abp.ino already does this.

LMIC.dn2Dr = DR_SF9;

Encoding Utilities

It is generally important to make LoRaWAN messages as small as practical. Extra bytes mean extra transmit time, which wastes battery power and interferes with other nodes on the network.

To simplify coding, the Arduino header file <lmic.h> defines some data encoding utility functions to encode floating-point data into uint16_t values using sflt16 or uflt16 bit layout. For even more efficiency, there are versions that use only the bottom 12 bits of the uint16_t, allowing for other bits to be carried in the top 4 bits, or for two values to be crammed into three bytes.

  • uint16_t LMIC_f2sflt16(float) converts a floating point number to a sflt16-encoded uint16_t.
  • uint16_t LMIC_f2uflt16(float) converts a floating-point number to a uflt16-encoded uint16_t.
  • uint16_t LMIC_f2sflt12(float) converts a floating-point number to a sflt12-encoded uint16_t, leaving the top four bits of the result set to zero.
  • uint16_t LMIC_f2uflt12(float) converts a floating-point number to a uflt12-encoded uint16_t, leaving the top four bits of the result set to zero.

JavaScript code for decoding the data can be found in the following sections.

sflt16

A sflt16 datum represents an unsigned floating point number in the range [0, 1.0), transmitted as a 16-bit field. The encoded field is interpreted as follows:

bits description
15 Sign bit
14..11 binary exponent b
10..0 fraction f

The corresponding floating point value is computed by computing f/2048 * 2^(b-15). Note that this format is deliberately not IEEE-compliant; it's intended to be easy to decode by hand and not overwhelmingly sophisticated. However, it is similar to IEEE format in that it uses sign-magnitude rather than twos-complement for negative values.

For example, if the data value is 0x8D, 0x55, the equivalent floating point number is found as follows.

  1. The full 16-bit number is 0x8D55.
  2. Bit 15 is 1, so this is a negative value.
  3. b is 1, and b-15 is -14. 2^-14 is 1/16384
  4. f is 0x555. 0x555/2048 = 1365/2048 is 0.667
  5. f * 2^(b-15) is therefore 0.667/16384 or 0.00004068
  6. Since the number is negative, the value is -0.00004068

Floating point mavens will immediately recognize:

  • This format uses sign/magnitude representation for negative numbers.
  • Numbers do not need to be normalized (although in practice they always are).
  • The format is somewhat wasteful, because it explicitly transmits the most-significant bit of the fraction. (Most binary floating-point formats assume that f is is normalized, which means by definition that the exponent b is adjusted and f is shifted left until the most-significant bit of f is one. Most formats then choose to delete the most-significant bit from the encoding. If we were to do that, we would insist that the actual value of f be in the range 2048..4095, and then transmit only f - 2048, saving a bit. However, this complicates the handling of gradual underflow; see next point.)
  • Gradual underflow at the bottom of the range is automatic and simple with this encoding; the more sophisticated schemes need extra logic (and extra testing) in order to provide the same feature.

JavaScript decoder

function sflt162f(rawSflt16)
	{
	// rawSflt16 is the 2-byte number decoded from wherever;
	// it's in range 0..0xFFFF
	// bit 15 is the sign bit
	// bits 14..11 are the exponent
	// bits 10..0 are the the mantissa. Unlike IEEE format,
	// 	the msb is explicit; this means that numbers
	//	might not be normalized, but makes coding for
	//	underflow easier.
	// As with IEEE format, negative zero is possible, so
	// we special-case that in hopes that JavaScript will
	// also cooperate.
	//
	// The result is a number in the open interval (-1.0, 1.0);
	//

	// throw away high bits for repeatability.
	rawSflt16 &= 0xFFFF;

	// special case minus zero:
	if (rawSflt16 == 0x8000)
		return -0.0;

	// extract the sign.
	var sSign = ((rawSflt16 & 0x8000) != 0) ? -1 : 1;

	// extract the exponent
	var exp1 = (rawSflt16 >> 11) & 0xF;

	// extract the "mantissa" (the fractional part)
	var mant1 = (rawSflt16 & 0x7FF) / 2048.0;

	// convert back to a floating point number. We hope
	// that Math.pow(2, k) is handled efficiently by
	// the JS interpreter! If this is time critical code,
	// you can replace by a suitable shift and divide.
	var f_unscaled = sSign * mant1 * Math.pow(2, exp1 - 15);

	return f_unscaled;
	}

uflt16

A uflt16 datum represents an unsigned floating point number in the range [0, 1.0), transmitted as a 16-bit field. The encoded field is interpreted as follows:

bits description
15..12 binary exponent b
11..0 fraction f

The corresponding floating point value is computed by computing f/4096 * 2^(b-15). Note that this format is deliberately not IEEE-compliant; it's intended to be easy to decode by hand and not overwhelmingly sophisticated.

For example, if the transmitted message contains 0xEB, 0xF7, and the transmitted byte order is big endian, the equivalent floating point number is found as follows.

  1. The full 16-bit number is 0xEBF7.
  2. b is therefore 0xE, and b-15 is -1. 2^-1 is 1/2
  3. f is 0xBF7. 0xBF7/4096 is 3063/4096 == 0.74780...
  4. f * 2^(b-15) is therefore 0.74780/2 or 0.37390

Floating point mavens will immediately recognize:

  • There is no sign bit; all numbers are positive.
  • Numbers do not need to be normalized (although in practice they always are).
  • The format is somewhat wasteful, because it explicitly transmits the most-significant bit of the fraction. (Most binary floating-point formats assume that f is is normalized, which means by definition that the exponent b is adjusted and f is shifted left until the most-significant bit of f is one. Most formats then choose to delete the most-significant bit from the encoding. If we were to do that, we would insist that the actual value of f be in the range 4096..8191, and then transmit only f - 4096, saving a bit. However, this complicated the handling of gradual underflow; see next point.)
  • Gradual underflow at the bottom of the range is automatic and simple with this encoding; the more sophisticated schemes need extra logic (and extra testing) in order to provide the same feature.

JavaScript decoder

function uflt162f(rawUflt16)
	{
	// rawUflt16 is the 2-byte number decoded from wherever;
	// it's in range 0..0xFFFF
	// bits 15..12 are the exponent
	// bits 11..0 are the the mantissa. Unlike IEEE format,
	// 	the msb is explicit; this means that numbers
	//	might not be normalized, but makes coding for
	//	underflow easier.
	// As with IEEE format, negative zero is possible, so
	// we special-case that in hopes that JavaScript will
	// also cooperate.
	//
	// The result is a number in the half-open interval [0, 1.0);
	//

	// throw away high bits for repeatability.
	rawUflt16 &= 0xFFFF;

	// extract the exponent
	var exp1 = (rawUflt16 >> 12) & 0xF;

	// extract the "mantissa" (the fractional part)
	var mant1 = (rawUflt16 & 0xFFF) / 4096.0;

	// convert back to a floating point number. We hope
	// that Math.pow(2, k) is handled efficiently by
	// the JS interpreter! If this is time critical code,
	// you can replace by a suitable shift and divide.
	var f_unscaled = mant1 * Math.pow(2, exp1 - 15);

	return f_unscaled;
	}

sflt12

A sflt12 datum represents an signed floating point number in the range [0, 1.0), transmitted as a 12-bit field. The encoded field is interpreted as follows:

bits description
11 sign bit
11..8 binary exponent b
7..0 fraction f

The corresponding floating point value is computed by computing f/128 * 2^(b-15). Note that this format is deliberately not IEEE-compliant; it's intended to be easy to decode by hand and not overwhelmingly sophisticated.

For example, if the transmitted message contains 0x8, 0xD5, the equivalent floating point number is found as follows.

  1. The full 16-bit number is 0x8D5.
  2. The number is negative.
  3. b is 0x1, and b-15 is -14. 2^-14 is 1/16384
  4. f is 0x55. 0x55/128 is 85/128, or 0.66
  5. f * 2^(b-15) is therefore 0.66/16384 or 0.000041 (to two significant digits)
  6. The decoded number is therefore -0.000041.

Floating point mavens will immediately recognize:

  • This format uses sign/magnitude representation for negative numbers.
  • Numbers do not need to be normalized (although in practice they always are).
  • The format is somewhat wasteful, because it explicitly transmits the most-significant bit of the fraction. (Most binary floating-point formats assume that f is is normalized, which means by definition that the exponent b is adjusted and f is shifted left until the most-significant bit of f is one. Most formats then choose to delete the most-significant bit from the encoding. If we were to do that, we would insist that the actual value of f be in the range 128 .. 256, and then transmit only f - 128, saving a bit. However, this complicates the handling of gradual underflow; see next point.)
  • Gradual underflow at the bottom of the range is automatic and simple with this encoding; the more sophisticated schemes need extra logic (and extra testing) in order to provide the same feature.
  • It can be strongly argued that dropping the sign bit would be worth the effort, as this would get us 14% more resolution for a minor amount of work.

JavaScript decoder

function sflt122f(rawSflt12)
	{
	// rawSflt12 is the 2-byte number decoded from wherever;
	// it's in range 0..0xFFF (12 bits). For safety, we mask
	// on entry and discard the high-order bits.
	// bit 11 is the sign bit
	// bits 10..7 are the exponent
	// bits 6..0 are the the mantissa. Unlike IEEE format,
	// 	the msb is explicit; this means that numbers
	//	might not be normalized, but makes coding for
	//	underflow easier.
	// As with IEEE format, negative zero is possible, so
	// we special-case that in hopes that JavaScript will
	// also cooperate.
	//
	// The result is a number in the open interval (-1.0, 1.0);
	//

	// throw away high bits for repeatability.
	rawSflt12 &= 0xFFF;

	// special case minus zero:
	if (rawSflt12 == 0x800)
		return -0.0;

	// extract the sign.
	var sSign = ((rawSflt12 & 0x800) != 0) ? -1 : 1;

	// extract the exponent
	var exp1 = (rawSflt12 >> 7) & 0xF;

	// extract the "mantissa" (the fractional part)
	var mant1 = (rawSflt12 & 0x7F) / 128.0;

	// convert back to a floating point number. We hope
	// that Math.pow(2, k) is handled efficiently by
	// the JS interpreter! If this is time critical code,
	// you can replace by a suitable shift and divide.
	var f_unscaled = sSign * mant1 * Math.pow(2, exp1 - 15);

	return f_unscaled;
	}

uflt12

A uflt12 datum represents an unsigned floating point number in the range [0, 1.0), transmitted as a 16-bit field. The encoded field is interpreted as follows:

bits description
11..8 binary exponent b
7..0 fraction f

The corresponding floating point value is computed by computing f/256 * 2^(b-15). Note that this format is deliberately not IEEE-compliant; it's intended to be easy to decode by hand and not overwhelmingly sophisticated.

For example, if the transmitted message contains 0x1, 0xAB, the equivalent floating point number is found as follows.

  1. The full 16-bit number is 0x1AB.
  2. b is therefore 0x1, and b-15 is -14. 2^-14 is 1/16384
  3. f is 0xAB. 0xAB/256 is 0.67
  4. f * 2^(b-15) is therefore 0.67/16384 or 0.0000408 (to three significant digits)

Floating point mavens will immediately recognize:

  • There is no sign bit; all numbers are positive.
  • Numbers do not need to be normalized (although in practice they always are).
  • The format is somewhat wasteful, because it explicitly transmits the most-significant bit of the fraction. (Most binary floating-point formats assume that f is is normalized, which means by definition that the exponent b is adjusted and f is shifted left until the most-significant bit of f is one. Most formats then choose to delete the most-significant bit from the encoding. If we were to do that, we would insist that the actual value of f be in the range 256 .. 512, and then transmit only f - 256, saving a bit. However, this complicates the handling of gradual underflow; see next point.)
  • Gradual underflow at the bottom of the range is automatic and simple with this encoding; the more sophisticated schemes need extra logic (and extra testing) in order to provide the same feature.

JavaScript decoder

function uflt122f(rawUflt12)
	{
	// rawUflt12 is the 2-byte number decoded from wherever;
	// it's in range 0..0xFFF (12 bits). For safety, we mask
	// on entry and discard the high-order bits.
	// bits 11..8 are the exponent
	// bits 7..0 are the the mantissa. Unlike IEEE format,
	// 	the msb is explicit; this means that numbers
	//	might not be normalized, but makes coding for
	//	underflow easier.
	// As with IEEE format, negative zero is possible, so
	// we special-case that in hopes that JavaScript will
	// also cooperate.
	//
	// The result is a number in the half-open interval [0, 1.0);
	//

	// throw away high bits for repeatability.
	rawUflt12 &= 0xFFF;

	// extract the exponent
	var exp1 = (rawUflt12 >> 8) & 0xF;

	// extract the "mantissa" (the fractional part)
	var mant1 = (rawUflt12 & 0xFF) / 256.0;

	// convert back to a floating point number. We hope
	// that Math.pow(2, k) is handled efficiently by
	// the JS interpreter! If this is time critical code,
	// you can replace by a suitable shift and divide.
	var f_unscaled = sSign * mant1 * Math.pow(2, exp1 - 15);

	return f_unscaled;
	}

Release History

  • Interim bug fixes: added a new API (radio_irq_handler_v2()), which allows the caller to provide the timestamp of the interrupt. This allows for more accurate timing, because the knowledge of interrupt overhead can be moved to a platform-specific layer (#148). Fixed compile issues on ESP32 (#140 and #153). We added ESP32 and 32u4 as targets in CI testing. We switched CI testing to Arduino IDE 1.8.7. Fixed issue #161 selecting the Japan version of as923 using CFG_as923jp (selecting via CFG_as923 and LMIC_COUNTRY_CODE=LMIC_COUNTRY_CODE_JP worked). Fixed #38 -- now any call to hal_init() will put the NSS line in the idle (high/inactive) state. As a side effect, RXTX is initialized, and RESET code changed to set value before transitioning state. Likely no net effect, but certainly more correct.

  • V2.2.2 adds ttn-abp-feather-us915-dht22.ino example, and fixes some documentation typos. It also fixes encoding of the Margin field of the DevStatusAns MAC message (#130). This makes Arduino LMIC work with newtorks implemented with LoraServer.

  • V2.2.1 corrects the value of ARDUINO_LMIC_VERSION (#123), allows ttn-otaa-feather-us915 example to compile for the Feather 32u4 LoRa (#116), and addresses documentation issues (#122, #120).

  • V2.2.0 adds encoding functions and tn-otaa-feather-us915-dht22.ino example. Plus a large number of issues: #59, #60, #63, #64 (listen-before-talk for Japan), #65, #68, #75, #78, #80, #91, #98, #101. Added full Travis CI testing, switched to travis-ci.com as the CI service. Prepared to publish library in the offical Arduino library list.

  • V2.1.5 fixes issue #56 (a documentation bug). Documentation was quickly reviewed and other issues were corrected. The OTAA examples were also updated slightly.

  • V2.1.4 fixes issues #47 and #50 in the radio driver for the SX1276 (both related to handling of output power control bits).

  • V2.1.3 has a fix for issue #43: handling of LinkAdrRequest was incorrect for US915 and AU921; when TTN added ADR support on US and AU, the deficiency was revealed (and caused an ASSERT).

  • V2.1.2 has a fix for issue #39 (adding a prototype for LMIC_DEBUG_PRINTF if needed). Fully upward compatible, so just a patch.

  • V2.1.1 has the same content as V2.1.2, but was accidentally released without updating library.properties.

  • V2.1.0 adds support for the Murata LoRaWAN module.

  • V2.0.2 adds support for the extended bandplans.

Contributions

This library started from the IBM V1.5 open-source code.

  • Thomas Telkamp and Matthijs Kooijman ported V1.5 to Arduino and did a lot of bug fixing.

  • Terry Moore, LeRoy Leslie, Frank Rose, and ChaeHee Won did a lot of work on US support.

  • Terry Moore added the AU921, AS923 and IN866 bandplans, and created the regionalization framework.

  • @tanupoo of the WIDE Project debugged AS923JP and LBT support.

Trademark Acknowledgements

LoRa is a registered trademark of the LoRa Alliance. LoRaWAN is a trademark of the LoRa Alliance.

MCCI and MCCI Catena are registered trademarks of MCCI Corporation.

All other trademarks are the properties of their respective owners.

License

The upstream files from IBM v1.6 are based on the Berkeley license, and the merge which synchronized this repository therefore migrated the core files to the Berkeley license. However, modifications made in the Arduino branch were done under the Eclipse license, so the overall license of this repository is still Eclipse Public License v1.0. The examples which use a more liberal license. Some of the AES code is available under the LGPL. Refer to each individual source file for more details, but bear in mind that until the upstream developers look into this issue, it is safest to assume the Eclipse license applies.