Updated for TTS(CE) = Also Known as TTN V3. The code requires the first server to be 'ttn' protocol.

Overview

Updated

Updated for TTS(CE) = Also Known as TTN V3. The code requires the first server to be 'ttn' protocol.

Introduction

This balenaCloud (previously resin.io) setup is based on the Multi-protocol Packet Forwarder by Jac Kersing.

An alternative guide to use this balenaCloud setup can be found in the official TTN documentation at: https://www.thethingsnetwork.org/docs/gateways/rak831/

Difference between Poly-packet-forwarder and Multi-protocol-packet-forwarder

mp-pkt-fwd uses the new protocolbuffers-over-mqtt-over-tcp protocol for gateways, as defined by TTN and used by the TTN kickstarter gateway. Using this protcol the gateway is authenticated, which means it is registered under a specific user and can thus be trusted. Because it uses TCP, the chance of packet loss is much lower than with the previous protocol that used UDP. Protocolbuffers packs the data in a compact binary mode into packets, using much less space than the plaintext json that was previously used. It should therefore consume less bandwidth.

balenaCloud TTN Gateway Connector for Raspberry Pi

balenaCloud Dockerfile & scripts for The Things Network gateways based on the Raspberry Pi. This updated version uses the gateway connector protocol, not the old packet forwarder. See the TTN documentation on Gateway Registration, you need to create a gateway API key.

Currently any Raspberry Pi with one of the following gateway boards, communicating over SPI, are supported, but not limited to these:

Prerequisites

  1. Build your hardware.
  2. Create a new gateway that uses gateway connector on the TTN Console. Also set the location and altitude of your gateway. Go to API keys and create a new API key with 'link as Gateway to a Gateway Server for traffic exchange, i.e. write uplink and read downlink' rights. Copy the secret and use it for GW_KEY later on.
  3. Create and sign into an account at https://www.balena.io/cloud/, which is the central "device dashboard".

Create a balenaCloud application

  1. On balenaCloud, create an "Application" for managing your TTN gateway devices. I'd suggest that you give it the name "ttngw", select the appropriate device type (i.e. Raspberry Pi 2 or Raspberry Pi 3), and click "Create New Application". You only need to do this once, after which you'll be able to manage one or many gateways of that type.
  2. You'll then be brought to the Device Management dashboard for that Application. Follow the instructions to "Add device" and create a boot SD-card for your Raspberry Pi. (Pro Tip: Use a fast microSD card and a USB 3 adapter if you can, because it can take a while to copy all that data. Either that, or be prepared to be very patient.)
  3. When the (long) process of writing the image to the SD card completes, insert it into your Raspberry Pi, connect it to the network with Ethernet, and power it up.
  4. After several minutes, on the balenaCloud Devices dashboard you'll now see your device - first in a "Configuring" state, then "Idle". Click it to open the Devices control panel.
  5. If you like, enter any new Device Name that you'd like, such as "my-gateway-amsterdam".

Configure the gateway device

Click the "Environment Variables" section at the left side of the screen. This will allow you to configure this and only this device. These variables will be used to pull information about this gateway from TTN, and will be used to create a "global_conf.json" and "local_conf.json" file for this gateway.

For a more complete list of possible environment variables, see CONFIGURATION.

Device environment variables - no GPS

For example, for an IMST iC880A or RAK831 with no GPS, the MINIMUM environment variables that you should configure at this screen should look something like this:

Name Value
GW_TTSCE_CLUSTER The TTS(CE) cluster being used: eu1, nam1, au1
GW_ID The gateway ID from the TTN console
GW_KEY The gateway API KEY secret key you copied earlier
GW_RESET_PIN 22 (optional)

GW_RESET_PIN can be left out if you are using Gonzalo Casas' backplane board, or any other setup using pin 22 as reset pin. This is because pin 22 is the default reset pin used by this balenaCloud setup.

Device environment variables - with GPS

For example a LinkLabs gateway, which has a built-in GPS, you need:

Name Value
GW_TTSCE_CLUSTER The TTS(CE) cluster being used: eu1, nam1, au1
GW_ID The gateway ID from the TTN console
GW_KEY The gateway API KEY secret key you copied earlier
GW_GPS true
GW_RESET_PIN 29

Reset pin values

Depending on the way you connect the concentrator board to the Raspberry Pi, the reset pin of the concentrator might be on a different GPIO pin of the Raspberry Pi. Here follows a table of the most common backplane boards used, and the reset pin number you should use in the GW_RESET_PIN environment variable.

Note that the reset pin you should define is the physical pin number on the Raspberry Pi. To translate between different numbering schemes you can use pinout.xyz.

Backplane Reset pin
Gonzalo Casas backplane
https://github.com/gonzalocasas/ic880a-backplane
https://www.tindie.com/stores/gnz/
22
ch2i
https://github.com/ch2i/iC880A-Raspberry-PI
11
Linklabs Rasberry Pi Hat
https://www.amazon.co.uk/868-MHz-LoRaWAN-RPi-Shield/dp/B01G7G54O2
29
Rising HF Board
http://www.risinghf.com/product/risinghf-iot-dicovery/?lang=en
26
IMST backplane or Lite gateway
https://wireless-solutions.de/products/long-range-radio/lora_lite_gateway.html
29 (untested)
Coredump backplane
https://github.com/dbrgn/ic880a-backplane/
https://shop.coredump.ch/product/ic880a-lorawan-gateway-backplane/
22
RAK backplane
11
Pi Supply IoT LoRa Gateway HAT for Raspberry Pi
https://uk.pi-supply.com/products/iot-lora-gateway-hat-for-raspberry-pi
15

If you get the message ERROR: [main] failed to start the concentrator after balenaCLoud is finished downloading the application, or when restarting the gateway, it most likely means the GW_RESET_PIN you defined is incorrect. Alternatively the problem can be caused by the hardware, typically for the IMST iC880A-SPI board with insufficient voltage, try another power supply or slightly increase the voltage.

Special note for using a Raspberry Pi 3

There is a backward incomatibility between the Raspberry Pi 1 and 2 hardware, and Raspberry Pi 3. For Raspberry Pi 3, it is necessary to make a small additional configuration change.

Click <- to go back to the Device List, and note that on the left there is an option called "Fleet Configuration". Click it.

Add a New config variable as follows:

Application config variables

Name Value
BALENA_HOST_CONFIG_core_freq 250
BALENA_HOST_CONFIG_dtoverlay pi3-miniuart-bt

TRANSFERRING TTN GATEWAY SOFTWARE TO BALENACLOUD SO THAT IT MAY BE DOWNLOADED ON YOUR DEVICES

  1. On your computer, clone this git repo. For example in a terminal on Mac or Linux type:

    git clone https://github.com/kersing/ttn-resin-gateway-rpi-1
    cd ttn-resin-gateway-rpi-1/
  2. Now, type the command that you'll see displayed in the edit control in the upper-right corner of the balenaCloud devices dashboard for your device. This command "connects" your local directory to the balenaCloud GIT service, which uses GIT to "receive" the gateway software from TTN, and it looks something like this:

    git remote add balena [email protected]:youraccount/yourapplication.git
  3. Add your SSH public key to the list at https://dashboard.balena-cloud.com/preferences/sshkeys. You may need to search the internet how to create a SSH key on your operating system, where to find it afterwards, copy the content, and paste the content to the balenaCloud console.

  4. Type the following commands into your terminal to "push" the TTN files up to balenaCloud:

    git add .
    git commit -m "first upload of ttn files to balenaCloud"
    git push -f balena master
  5. What you'll now see happening in terminal is that this "git push" does an incredible amount of work:

  6. It will upload a Dockerfile, a "build script", and a "run script" to balenaCloud

  7. It will start to do a "docker build" using that Dockerfile, running it within a QEMU ARM virtual machine on the balenaCloud service.

  8. In processing this docker build, it will run a "build.sh" script that downloads and builds the packet forwarder executable from source code, for RPi+iC880A-SPI.

  9. When the build is completed, you'll see a unicorn 🦄 ASCII graphic displayed in your terminal.

  10. Now, switch back to your device dashboard, you'll see that your Raspberry Pi is now "updating" by pulling the Docker container from the balenaCloud service. Then, after "updating", you'll see the gateway's log file in the window at the lower right corner. You'll see it initializing, and will also see log output each time a packet is forwarded to TTN. You're done!

Troubleshooting

If you get the error below please check if your ssh public key has been added to you balenaCloud account. In addition verify whether your private key has the correct permissions (i.e. chmod 400 ~/.ssh/id_rsa).

$ git push -f balena master
Connection closed by xxx.xxx.xxx.xxx port 22
fatal: Could not read from remote repository.

Please make sure you have the correct access rights
and the repository exists.
$

Pro Tips

  • At some point if you would like to add a second gateway, third gateway, or a hundred gateways, all you need to do is to add a new device to your existing Application. You needn't upload any new software to balenaCloud, because balenaCloud already knows what software belongs on the gateway. So long as the environment variables are configured correctly for that new device, it'll be up and running immediately after you burn an SD card and boot it.

  • balenaCloud will automatically restart the gateway software any time you change the environment variables. You'll see this in the log. Also, note that balenaCloud restarts the gateway properly after power failures. If the packet forwarder fails because of an error, it will also automatically attempt to restart.

  • If you'd like to update the software across all the gateways in your device fleet, simply do the following:

    git add .
    git commit -m "Updated gateway version"
    git push -f balena master
    
  • For devices without a GPS, the location that is configured on the TTN console is used. This location is only read at startup of the gateway. Therefore, after you set or changed the location, restart the application from the balenaCloud console.

Device statistics

If you want to show nice looking statistics for your gateway(s) there are a couple of additional steps to take. First, copy Dockerfile.template.metering to Dockerfile.template. Next copy start.sh.metering to start.sh. Now use the instructions above to update the balenaCloud image.

Once the new image is deployed, go to the balenaCloud dashboard for your devices and select 'Enable Public device URL' in the drop down menu (the one to the right of the light bulb). That is all that is required to provide metrics. Now you will need to install a metrics collector on a seperate system as outlined in Fleet-wide Machine Metrics Monitoring in 20mins.

(To show packet forwarder graphs you need to add your own graphs to the provided templates)

Credits

You might also like...
Chinese Mandarin tts text-to-speech  中文 (普通话) 语音 合成 , by fastspeech 2 , implemented in pytorch, using waveglow as vocoder,
Chinese Mandarin tts text-to-speech 中文 (普通话) 语音 合成 , by fastspeech 2 , implemented in pytorch, using waveglow as vocoder,

Chinese mandarin text to speech based on Fastspeech2 and Unet This is a modification and adpation of fastspeech2 to mandrin(普通话). Many modifications t

Pytorch Implementation of Google's Parallel Tacotron 2: A Non-Autoregressive Neural TTS Model with Differentiable Duration Modeling
Pytorch Implementation of Google's Parallel Tacotron 2: A Non-Autoregressive Neural TTS Model with Differentiable Duration Modeling

Parallel Tacotron2 Pytorch Implementation of Google's Parallel Tacotron 2: A Non-Autoregressive Neural TTS Model with Differentiable Duration Modeling

Pytorch Implementation of DiffSinger: Diffusion Acoustic Model for Singing Voice Synthesis (TTS Extension)
Pytorch Implementation of DiffSinger: Diffusion Acoustic Model for Singing Voice Synthesis (TTS Extension)

DiffSinger - PyTorch Implementation PyTorch implementation of DiffSinger: Diffusion Acoustic Model for Singing Voice Synthesis (TTS Extension). Status

This is a template for the Non-autoregressive Deep Learning-Based TTS model (in PyTorch).

Non-autoregressive Deep Learning-Based TTS Template This is a template for the Non-autoregressive TTS model. It contains Data Preprocessing Pipeline D

Pytorch implementation of "Grad-TTS: A Diffusion Probabilistic Model for Text-to-Speech"

GradTTS Unofficial Pytorch implementation of "Grad-TTS: A Diffusion Probabilistic Model for Text-to-Speech" (arxiv) About this repo This is an unoffic

PyTorch Implementation of VAENAR-TTS: Variational Auto-Encoder based Non-AutoRegressive Text-to-Speech Synthesis.
PyTorch Implementation of VAENAR-TTS: Variational Auto-Encoder based Non-AutoRegressive Text-to-Speech Synthesis.

VAENAR-TTS - PyTorch Implementation PyTorch Implementation of VAENAR-TTS: Variational Auto-Encoder based Non-AutoRegressive Text-to-Speech Synthesis.

Byte-based multilingual transformer TTS for low-resource/few-shot language adaptation.

One model to speak them all 🌎 Audio Language Text ▷ Chinese 人人生而自由,在尊严和权利上一律平等。 ▷ English All human beings are born free and equal in dignity and rig

Make your AirPlay devices as TTS speakers

Apple AirPlayer Home Assistant integration component, make your AirPlay devices as TTS speakers. Before Use 2021.6.X or earlier Apple Airplayer compon

🗣️ Microsoft Edge TTS for Home Assistant, no need for app_key

Microsoft Edge TTS for Home Assistant This component is based on the TTS service of Microsoft Edge browser, no need to apply for app_key. Install Down

Releases(Alpha)
  • Alpha(Oct 17, 2021)

    A working setup where docker containers communicate through the host networking. This might expose port 1680 which is undesired. A solution should be found by using the docker bridge that will hide port 1680.

    Source code(tar.gz)
    Source code(zip)
Owner
Remko
Wireless (RF) specialist, Teacher Embedded Systems Engineering, IOT and Embedded enthusiast. Scout and Radio amateur.
Remko
Code for WECHSEL: Effective initialization of subword embeddings for cross-lingual transfer of monolingual language models.

WECHSEL Code for WECHSEL: Effective initialization of subword embeddings for cross-lingual transfer of monolingual language models. arXiv: https://arx

Institute of Computational Perception 45 Dec 29, 2022
Show-attend-and-tell - TensorFlow Implementation of "Show, Attend and Tell"

Show, Attend and Tell Update (December 2, 2016) TensorFlow implementation of Show, Attend and Tell: Neural Image Caption Generation with Visual Attent

Yunjey Choi 902 Nov 29, 2022
Learning embeddings for classification, retrieval and ranking.

StarSpace StarSpace is a general-purpose neural model for efficient learning of entity embeddings for solving a wide variety of problems: Learning wor

Facebook Research 3.8k Dec 22, 2022
Camera-caps - Examine the camera capabilities for V4l2 cameras

camera-caps This is a graphical user interface over the v4l2-ctl command line to

Jetsonhacks 25 Dec 26, 2022
Official code for our ICCV paper: "From Continuity to Editability: Inverting GANs with Consecutive Images"

GANInversion_with_ConsecutiveImgs Official code for our ICCV paper: "From Continuity to Editability: Inverting GANs with Consecutive Images" https://a

QingyangXu 38 Dec 07, 2022
TSIT: A Simple and Versatile Framework for Image-to-Image Translation

TSIT: A Simple and Versatile Framework for Image-to-Image Translation This repository provides the official PyTorch implementation for the following p

Liming Jiang 255 Nov 23, 2022
Code for "Finding Regions of Heterogeneity in Decision-Making via Expected Conditional Covariance" at NeurIPS 2021

Finding Regions of Heterogeneity in Decision-Making via Expected Conditional Covariance Justin Lim, Christina X Ji, Michael Oberst, Saul Blecker, Leor

Sontag Lab 3 Feb 03, 2022
TensorFlow code for the neural network presented in the paper: "Structural Language Models of Code" (ICML'2020)

SLM: Structural Language Models of Code This is an official implementation of the model described in: "Structural Language Models of Code" [PDF] To ap

73 Nov 06, 2022
Learning 3D Part Assembly from a Single Image

Learning 3D Part Assembly from a Single Image This repository contains a PyTorch implementation of the paper: Learning 3D Part Assembly from A Single

18 Dec 21, 2022
Real-ESRGAN aims at developing Practical Algorithms for General Image Restoration.

Real-ESRGAN Colab Demo for Real-ESRGAN . Portable Windows executable file. You can find more information here. Real-ESRGAN aims at developing Practica

Xintao 17.2k Jan 02, 2023
Towhee is a flexible machine learning framework currently focused on computing deep learning embeddings over unstructured data.

Towhee is a flexible machine learning framework currently focused on computing deep learning embeddings over unstructured data.

1.7k Jan 08, 2023
A python code to convert Keras pre-trained weights to Pytorch version

Weights_Keras_2_Pytorch 最近想在Pytorch项目里使用一下谷歌的NIMA,但是发现没有预训练好的pytorch权重,于是整理了一下将Keras预训练权重转为Pytorch的代码,目前是支持Keras的Conv2D, Dense, DepthwiseConv2D, Batch

Liu Hengyu 2 Dec 16, 2021
Train SN-GAN with AdaBelief

SNGAN-AdaBelief Train a state-of-the-art spectral normalization GAN with AdaBelief https://github.com/juntang-zhuang/Adabelief-Optimizer Acknowledgeme

Juntang Zhuang 10 Jun 11, 2022
TransFGU: A Top-down Approach to Fine-Grained Unsupervised Semantic Segmentation

TransFGU: A Top-down Approach to Fine-Grained Unsupervised Semantic Segmentation Zhaoyun Yin, Pichao Wang, Fan Wang, Xianzhe Xu, Hanling Zhang, Hao Li

DamoCV 25 Dec 16, 2022
AFL binary instrumentation

E9AFL --- Binary AFL E9AFL inserts American Fuzzy Lop (AFL) instrumentation into x86_64 Linux binaries. This allows binaries to be fuzzed without the

242 Dec 12, 2022
Self-Supervised Pillar Motion Learning for Autonomous Driving (CVPR 2021)

Self-Supervised Pillar Motion Learning for Autonomous Driving Chenxu Luo, Xiaodong Yang, Alan Yuille Self-Supervised Pillar Motion Learning for Autono

QCraft 101 Dec 05, 2022
Apply AnimeGAN-v2 across frames of a video clip

title emoji colorFrom colorTo sdk app_file pinned AnimeGAN-v2 For Videos 🔥 blue red gradio app.py false AnimeGAN-v2 For Videos Apply AnimeGAN-v2 acro

Nathan Raw 36 Oct 18, 2022
GLNet for Memory-Efficient Segmentation of Ultra-High Resolution Images

GLNet for Memory-Efficient Segmentation of Ultra-High Resolution Images Collaborative Global-Local Networks for Memory-Efficient Segmentation of Ultra-

VITA 298 Dec 12, 2022
PyTorch implementation of SwAV (Swapping Assignments between Views)

Unsupervised Learning of Visual Features by Contrasting Cluster Assignments This code provides a PyTorch implementation and pretrained models for SwAV

Meta Research 1.7k Jan 04, 2023
Code implementing "Improving Deep Learning Interpretability by Saliency Guided Training"

Saliency Guided Training Code implementing "Improving Deep Learning Interpretability by Saliency Guided Training" by Aya Abdelsalam Ismail, Hector Cor

8 Sep 22, 2022