The MatchX Development Kit is enabling you to test your LoRa prototype and build LoRa sensors quickly and cost-effectively. The Development kit has been used to build LoRa based water and parking sensors in a number of projects around the world that all require long-distance communication and battery longevity. Find the user guide here.
MatchX’s latest and most exciting technological breakthrough yet — EdgeX AI Kit – will enable you to take your LoRa project one step further with an artificial intelligence algorithm designed for your specific data processing needs. EdgeX AI Kitis designed for running and testing artificial intelligence (AI) models on edge devices such as sensor nodes. This enables you to develop new LPWAN use-cases with the efficiency and scalability they require.
EdgeX AI Kit is designed for running and testing artificial intelligence (AI) models on edge devices such as sensor nodes.
LoRaWAN and NB-IoT are low-power wireless solutions with limited bandwidth. Because of this, sensors using these networks can function for up to ten years on a single battery. These devices, however, are incapable of transmitting a live video or audio feed for server-side analysis. This, previously seen as a disadvantage of these networks, is a valuable asset when considering data privacy.
“Privacy is a major concern for citizens and therefore smart city projects around the globe,” said MatchX Co-Founder and CEO Sheen Xin Hu. “By ensuring that no data is stored on the device and only the results of the AI analysis are transmitted to the cloud, we provide companies with a headstart on privacy, especially when building GDPR compliant solutions.”
So, what’s under the hood? A RISC-V chip with built-in neural networks acceleration together with the LoRa radio module enables massive deployment of edge intelligence in LPWAN applications where complex data processing is required, such as videos. EdgeX AI Kit is equipped with a micro-camera bringing you for the first time many new complex-data LPWAN project options. Using edge computing the gigabytes of sensory and special data are analyzed, filtered and compressed before being transmitted on IoT edge Gateways to several systems for further use. This edge processing saves network expenses, storage and operating costs for management IoT traffic.
The EdgeX AI Kit is built around the Kendryte K210. The Kendryte K210 is a system-on-chip (SoC) that integrates machine vision and machine hearing. Using TSMC’s ultra-low-power 28-nm advanced process with dual-core 64-bit processors for better power efficiency, stability, and reliability.
Architecture overview of Kendryte K210
- CPU: RISC-V Dual Core 64bit, with FPU
- Image Recognition: QVGA@60fps/VGA@30fps Audio:
- Advanced encryption standard (AES) hardware accelerator
- One time programmable (OTP) ROM
- Power consumption of typical application scenarios < 1W
- Power consumption of chip < 300mW
- OS: FreeRTOS
- NN Model: TinyYOLOv2 (after pruned)
- DNN Framework: TensorFlow/Keras/Darknet
- Peripherals FPIOA/UART/GPIO/SPI/I²C/I²S/WDT/TIMER/RTC, etc.
Harness AI-powered LPWAN at the Edge
The full-featured MatchX AI Kit includes everything you need to kick off and deploy an IoT project with a processing power of AI module onboard and LoRaWAN’s antenna for secure and uninterrupted data transmission. This unlocks new features for the LPWAN applications that have been yet unimaginable until yesterday: facial recognition, hand gesture detection, key phrase detection, human presence detection, and object counting.
AI System on Module Stack for AI + IoT applications
- Ultra-Low Power (< 1W)
- Low Latency
- LoRaWAN enabled
- Privacy (GDPR Compliant)
The approach that underlies the EdgeX AI Kit is called edge computing. According to Hewlett Packard, “Edge computing is a distributed, open IT architecture that features decentralized processing power, enabling mobile computing and Internet of Things (IoT) technologies. In edge computing, data is processed by the device itself or by a local computer or server, rather than being transmitted to a data center.” Basically it means that the data being processed from cameras, microphones or other sensors never leaves the device.
MatchX Ecosystem. Face Recognition is carried out by the Edge AI Kit and then processed data goes to MatchX cloud through the LoRaWAN
EdgeX AI Kit is ideal to cope with vast amounts of data generated by IoT sensors and connects seamlessly with the MatchX Ecosystem. The MatchX LPWAN gateway takes care of transmitting processed data from AI Kit to MatchX Cloud at low-power usage & long-distance communication, edge computing reduces latency in analytic processing and boosts overall network performance. EdgeX AI Kit has a built-in camera that will reinforce already existing applications of facial recognition thanks to Neural Network Accelerator.
In addition to this, AI Kit is a perfect solution for the Industrial Internet of Things (IIoT) e.g. Grid Edge Control and Analytics, Oil and Gas Remote Monitoring, Predictive Maintenance where real-time response is needed at the machine-level. The edge device does not serve only as a “bridge” for transmitting the data from a source to the cloud center but it also processes, interprets, and acts on the collected data at the edge within milliseconds and then only sends the necessary data to the cloud.
With new abilities come new possibilities
The unique combination of AI module + camera + LoRaWAN+microphone offers new opportunities for the expansion of IoT applications and creates new possibilities in the LPWAN world. For example, a camera installed on a window with an AI module can use deep learning algorithms for detecting free parking spaces in close proximity to your house. The AI module carries out all the computational processes enabling the edge processing of video material. The results are converted into “simple data” that can be sent through the LoRa network energy efficiently over long distances. This is how all the magic happens! The previous deployment of LPWAN was restricted by the amount of data but with AI integration on board, now the possibilities are endless. Forest Fire Detection is another good example of the synergy of these technologies. The AI-driven camera performs with precision accuracy on fire detection while LoRaWAN provides better connectivity and coverage than a cellular network.
MatchX AI Kit not only empowers the LPWAN’s use-cases but it also ensures interoperability and realizes the mixing of a variety of LPWA technologies and provides more efﬁcient and convenient smart services.
For EdgeX AI project consultation please send an email to firstname.lastname@example.org.
The EdgeX AI Dev Kit: https://www.matchx.io/edgexdevkit/
At MatchX, we work actively with partners across the globe to move beyond today’s deployment of IoT. We aim to synchronize the implementation of AI capabilities with existing solutions in IoT space while pursuing energy-saving, scalable, and comprehensive coverage of LPWAN, hence we put forward an AI-driven IoT powered by LoRaWAN. This enables businesses and people to develop new use-cases of the LPWAN technology with the efficiency and scalability they require in their operations since AI brings a new degree of data processing.
MatchX bridges the world of data with Cities, Businesses, and Individuals, making valuable data more accessible for more, making a difference in society and making a difference for the world of tomorrow.