Silicon Labs today announced the BG24 and MG24 families of 2.4GHz wireless SoC chips for Bluetooth and multiprotocol operations, respectively, and a new software toolkit. This new improved hardware and software platform will help bring AI/ML applications and high wireless performance to high-end battery powered devices. The ultra-low-power, off-the-shelf BG24 and MG24 families support multiple wireless protocols and include PSA Level 3 Secure Vault, ideal for diverse smart homes, medical and industrial applications. The SoC and software solution for the Internet of Things (IoT) announced today includes:
- Two new families of 2.4GHz wireless SoCs, Featuring the industry’s first integrated AI/ML accelerators, support for Matter, Zigbee, OpenThread, Bluetooth Low Energy, Bluetooth mesh, proprietary and multi-protocol process, the highest level of industry security certification, ultra-low power capabilities and the largest memory and flash capacity in Silicon Labs wallet.
- New software toolkit Designed to allow developers to quickly build and deploy AI and machine learning algorithms using some of the most popular toolkits such as TensorFlow.
“The BG24 and MG24 represent an impressive suite of industry capabilities including broad wireless multi-protocol support, battery life, machine learning, and security for IoT Edge applications,” said Matt Johnson, CEO of Silicon Labs.
The first integrated AI/machine learning acceleration that improves performance and energy efficiency
IoT product designers see the tremendous potential of artificial intelligence and machine learning to bring greater intelligence to cutting-edge applications such as home security systems, wearable medical monitors, sensors that monitor commercial facilities and industrial equipment, and more. But today, those considering deploying AI or machine learning to the edge face severe performance and power-use penalties that may outweigh the benefits.
The BG24 and MG24 ease these penalties as the first ultra-low-powered devices with built-in AI/ML accelerators. This specialized device is designed to handle complex calculations quickly and efficiently, with internal testing showing up to 4x improvement in performance as well as up to 6x improvement in energy efficiency. Because ML computations occur on the on-premises machine rather than the cloud, network latency is eliminated for faster action and decision making.
The BG24 and MG24 families also have the largest RAM capacity and RAM in the Silicon Labs group. This means that the device can scale to support multi-protocol, trained ML algorithms for large data sets. PSA Level 3 Certified Safe WarehouseTMIt provides the highest level of security certification for IoT devices, the security required in products such as door locks, medical equipment and other sensitive deployments where strengthening the device from external threats is critical.
AI/ML software and material support helps designers create new innovative applications
In addition to supporting TensorFlow on-premises, Silicon Labs has partnered with some of the leading AI and machine learning tool providers, such as SensiML and Edge Impulse, to ensure developers have access to a comprehensive tool chain that simplifies the development of optimized machine learning models for embedded wireless application deployments. Using this new AI/ML toolkit with Simplicity Studio from Silicon Labs and the BG24 and MG24 families of SoCs, developers can build applications that draw information from various connected devices, all of which communicate with each other using Matter to make intelligent machine-learning-based decisions.
For example, in a commercial office building, many lights are controlled by motion detectors that monitor occupancy to determine if the lights should be turned on or off. However, when typing at a desk with limited movement of hands and fingers, workers may be left in the dark when motion sensors alone cannot recognize their presence. By connecting sound sensors to motion detectors through the material application layer, additional audio data, such as typing sound, can be played through a machine learning algorithm to allow the lighting system to make a more informed decision about whether the lights should be on or off.
ML edge computing enables other smart home and industrial applications, including sensor data processing for anomaly detection, predictive maintenance, sound pattern recognition to improve glass breakage detection, simple command word recognition, and vision use cases such as presence detection or people counting with low-resolution cameras .
Alpha software highlights a variety of publishing options
More than 40 companies representing different industries and applications have already started developing and testing this new platform solution in a closed Alpha platform. These companies have been drawn to the BG24 and MG24 platforms by their extremely low power, advanced features, including AI/ML capabilities and Matter support. Global retailers are looking to improve the in-store shopping experience with more accurate asset tracking, real-time price updates, and other uses. Participants from the commercial building management sector are exploring how to make their building systems, including lighting, heating, ventilation and air conditioning, smarter to lower owners’ costs and reduce their environmental footprint. Finally, consumer and smart home solution providers make it easier to connect different devices and expand the way they interact to deliver innovative new features and services to consumers.
Most Capable SoCs in Silicon Lab
The single-die BG24 and MG24 SoCs combine a 78MHz ARM Cortex-M33 processor, high-performance 2.4GHz radio, an industry-leading 20-bit ADC, an improved flash stack (up to 1536KB) and RAM (up to 256 kB) , and an AI/ML hardware accelerator to process machine learning algorithms while unpacking the ARM Cortex-M33, so that applications have more cycles to do other work. Supporting a wide range of 2.4GHz wireless IoT protocols, this chipset combines the highest level of security with the best RF performance/energy efficiency ratio on the market.