Empowers the Creation of Battery-Powered NB-IoT Units, Suitable with All Indian Telecom Carriers

Eoxys is unveiling its XENO+ NB(Slim band)-IoT( Web of Issues) ML(Machine Learnig) SOM (System-On-Module) module, a fully-fledged SOM designed for the environment friendly operation of superior IoT/AIML units in purposes resembling sensible metering, sensible lighting, sensible monitoring, and industrial automation. This outstanding innovation represents an business milestone, as Eoxys’ SOM seamlessly integrates Renesas’s cutting-edge RH1NB200 LTE Cat NB1/NB2 modem with Syntiant’s highly effective NDP120 Neural Choice Processor. This integration gives an thrilling convergence of AI and Deep Studying capabilities with strong mobile connectivity, opening up unprecedented avenues for innovation inside the IoT ecosystem. By harnessing the clever computing prowess and seamless mobile connectivity that this module supplies, prospects can considerably expedite the event of their IoT merchandise, notably within the domains of sensible metering and sensible monitoring.

The SOM module gives a number of key benefits, together with:

• Important Time Financial savings: Clients can scale back {hardware} and software program improvement time by as much as 40%, streamlining their product improvement processes.
• Simplified Software program Improvement: The Software program SDK offered alongside the ML SOM modules aids prospects in chopping down on embedded software program improvement time, making the event course of extra environment friendly.
• Seamless Compatibility: With constant pin mapping throughout all XENO+ ML SOM modules, software program compatibility is ensured amongst all connectivity variant units, simplifying integration efforts.
• Over-the-Air Upgrades: Clients can reap the benefits of the Firmware Over-the-Air (FOTA) characteristic supplied with XENO+ ML SOM to simply replace MCU Software binaries, ML Mannequin binaries, and Modem binaries, protecting their units up-to-date.
• Versatile Neural Processing: The NDP120 is able to operating a number of Deep Neural Networks (DNN) on a wide range of architectures, together with Convolutional Neural Networks (CNNs), Recurrent Neural Networks (RNNs), and totally related networks, supporting as much as 256 layers.
• Future-Prepared Flexibility: Clients can initially deploy IoT use instances and seamlessly improve to ML use instances on the identical machine sooner or later, permitting for scalability and adaptableness.

Clients can streamline product improvement by specializing in sensors and distinctive worth propositions, making certain prime safety and efficiency. The modem module is designed for Indian telecom carriers, prioritizing bands B1, B3, B5, and B8, very best for Indian markets like sensible metering, lighting, and monitoring. Notably, it’s extremely power-efficient.

It options an built-in EAL5+ Safe Factor (SE) to offer unparalleled safety, essential for shielding finish purposes like sensible power and water metering methods. Moreover, it incorporates the Syntiant NDP120 Neural Choice Processor, able to concurrently operating a number of machine studying purposes for audio and time-series sensor information with minimal energy consumption. The NDP120 processor is optimized to effectively deal with numerous deep neural community architectures, together with CNNs, RNNs, and totally related networks with as much as 256 layers, making certain peak efficiency throughout various purposes.

To show the capabilities of NB-IoT ML SOM in sensible embedded units, we’ve created an Audio Classification Machine (ACD) centered across the NB-IoT ML SOM module. This ACD is supplied with a Machine Studying mannequin for figuring out real-time human alert sounds in its atmosphere, together with child cries, gunshots, canine barks, drilling noises, automobile theft alarms, jackhammer sounds, hearth alarms, glass breaks, hammering sounds, males’s screams, and ladies’s screams. The NDP120 processor handles the classification of those sounds, and the outcomes are transmitted to an IoT Dashboard software utilizing LTE NB-IoT connectivity, which effectively makes use of each NIDD and HTTP-based information protocols for information transmission.

Moreover, there’s an AIML Analysis Package (EVK) tailor-made particularly for the NB-IoT SOM module. The AIML EVK Package is designed to counterpoint the IoT product improvement expertise for builders, researchers, product design engineers, and college students by providing hands-on sensor expertise. Customers can experiment with AIML-based IoT purposes utilizing this equipment. Builders have the flexibleness to create embedded C-based purposes for sensor information assortment by means of interfaces like I2C, UART, SPI, ADC, and DAC. They will develop Tiny ML fashions and cargo them onto the AIML EVK Package to run these fashions on audio or time-series sensor information, enabling superior ML classifications for numerous purposes. Furthermore, builders can transmit ML classification outcomes to a cloud server utilizing HTTP, MQTT, or TCP information protocols by means of NB-IoT connectivity. This complete ecosystem empowers builders to discover and implement cutting-edge AIML options in IoT purposes.