LoRa Image and Video Transmission Wireless | ML on EdgeX

Introduction: LoRa Image and Video Transmission Wireless | ML on EdgeX

Hey, what's up, Guys! Akarsh here from CETech.

No Internet No Problem As LoRa is up to the rescue. LoRa is a Technology with the help of which we can easily transmit data to a range of kilometres without any Internet and today we are going to discuss a module named EdgeX from MatchX which works on the amazing LoRa Technology. The best part is that it is ML and AI-based and is very much suitable for the Transmission of Image and Video data over hundreds of kilometres that too without the Internet. So Let's get to the fun part now.

Step 1: Get PCBs for Your Projects Manufactured

You must check out PCBWAY for ordering PCBs online for cheap!

You get 10 good quality PCBs manufactured and shipped to your doorstep for cheap. You will also get a discount on shipping on your first order. Upload your Gerber files onto PCBWAY to get them manufactured with good quality and quick turnaround time. Check out their online Gerber viewer function. With reward points, you can get free stuff from their gift shop.

Step 2: What Is LoRa ?

LoRa (Long Range) is a low-power wide-area network (LPWAN) protocol developed by Semtech. It is based on spread spectrum modulation techniques derived from chirp spread spectrum (CSS) technology.

LoRa uses license-free sub-gigahertz radio frequency bands like 433 MHz, 868 MHz (Europe), 915 MHz (Australia and North America), and 923 MHz (Asia). LoRa enables long-range transmissions (more than 10 km in rural areas) with low power consumption. The technology covers the physical layer, while other technologies and protocols such as LoRaWAN (Long Range Wide Area Network) cover the upper layers.

LoRa devices have geolocation capabilities used for trilaterating positions of devices via timestamps from gateways.LoRa and LoRaWAN permit long-range connectivity for the Internet of things (IoT) devices in different types of industries.

The spread spectrum LoRa modulation is performed by representing each bit of payload information by multiple chirps of information. The rate at which the spread information is sent is referred to as the symbol rate, the ratio between the nominal symbol rate and chirp rate is the spreading factor (SF) and represents the number of symbols sent per bit of information. LoRa can trade-off data rates for sensitivity with a fixed channel bandwidth by selecting the amount of spread used (a selectable radio parameter from 7 to 12). Lower SF means more chirps are sent per second; hence, you can encode more data per second. Higher SF implies fewer chirps per second; hence, there are less data to encode per second. Compared to lower SF, sending the same amount of data with higher SF needs more transmission time, known as airtime. More airtime means that the modem is up and running longer and consuming more energy. The benefit of high SF is that more extended airtime gives the receiver more opportunities to sample the signal power which results in better sensitivity. In addition, LoRa uses forward error correction coding to improve resilience against interference. LoRa's high range is characterized by high wireless link budgets of around 155 dB to 170 dB.

Step 3: Advantages and Disadvantages of LoRa


1) Long Range: LoRa devices can transmit signals over distances from 1km — 10km.

2) Low Power: LoRa end nodes wake up only at a fixed time, which can extend battery life. End node batteries can last for 5-10 years.

3) Security: Data encryption using AES128 between end nodes and network servers/ Data encryption using AES128 at the application level.

4) Network Capability: Single LoRa gateway device is designed to take care of thousands of end devices or nodes and easy to extend network capability by increasing gateways. A LoRaWAN gateway capability is influenced by these factors:

• Tunnels: Different tunnels can receive data from end nodes simultaneously; the greater quantity of tunnels, the more end nodes a gateway can connect to.

• Data size and reporting interval: Large data size and reporting interval will reduce the end nodes that a gateway can connect to.

• ADR (Adaptive Data Rate): The distance between end nodes and gateways is closer, the data rate is higher, which can save the bandwidth of gateways.

5) Low Cost: Work in free frequencies and no upfront licensing cost to use the technology.

6) Easy Deployment: Simple network architecture and easy to deploy by yourself.


1) Not for large data transmission.

2) Not for continuous monitoring.

3) Wake up only at a fixed time, so you can’t communicate with end nodes at any time.

4) The transmission rate is slow and easy to get interference because of using free frequencies.

Step 4: About EdgeX Module

EdgeX is a module that works on LoRaWAN technology due to which it is capable of transferring data to hundreds of kilometres without using any Internet. But the thing that makes EdgeX module special is that it can be used to easily transfer image, video and, audio data as well which is not possible in the conventional LoRa modules as they have limited bandwidth. It has a pre-installed good quality camera similar to the ESP32 camera which is capable of capturing images of its surroundings which can be seen on the LCD attached with the module. It also has a Mic installed to get audio input and perform audio analysis as well.

EdgeX module is AI-powered and uses Neural Network Accelerator to process images and video data, extract relevant data from that which is sufficient for the transfer of data, and its recreation at the receiver end and then send the relevant data to the cloud for transferring it to the receiver. This Neural Network Accelerator is the heart of our EdgeX AI module as all the processing of image and video data such as flattening, pooling, and all other relevant processes are done by this part only.

Being ML and AI-powered this module can easily perform tasks such as object detection which generally requires good processing power to work in real-time. For eg - It can be used to get the registration number of a car from its number plate. To do that the module captures an image of the car's number plate it processes the image on the Neural Network Accelerator, extracts the number, and sends that data only.

Step 5: Structure of EdgeX Module

The Module's basic structure contains the following divisions it is also shown in the Block diagram above:-

1) Input Devices:-

EdgeX AI Module can accept two types of Inputs Audio and Video because it has a pre-installed Camera and a microphone as well.

2) Neural Network Accelerator:-

It is the brain of the EdgeX module as all the processing such as audio analysis, Video Analysis, Image Analysis, and sending the data to the cloud.

3) LCD Display:-

It comes with a 3 inch LCD on which we can see the things that are in front of the camera or the images that are captured by the camera.

4) LoRa:-

In the end, it has the LoRa module which is used to transfer the processed data to the cloud and receive the data if it is at the receiving end.

Step 6: Technical Specifications of EdgeX AI Module

Weight: 0.25 kg

Operating System: FreeRTOS / Bare metal

Protocol: LoRa, (G)FSK, LoRaWAN compatible

CPU: Kendryte K210 400MHz RISC-V

Memory: 8MB RAM, 128MB Flash, SD card extensible

Features: AI acceleration module, LCD and Camera controller, I2S, I2C, UART, SPI, SD-card, secure authentication

Be the First to Share


    • Tinkercad to Fusion 360 Challenge

      Tinkercad to Fusion 360 Challenge
    • Cardboard Speed Challenge

      Cardboard Speed Challenge
    • Chocolate Challenge

      Chocolate Challenge



    11 months ago on Step 6

    hello, great contribution, only where can I find the detailed instructions for rebuilding. or the code. best regards