InferSens introduces ultra-low-power Edge DL sensor technology for commercial and industrial applications


InferSens introduces ultra-low-power Edge DL sensor technology for commercial and industrial applications. The battery-powered sensor, with built-in Syntiant NDP120 neural decision processor, enables highly accurate on-device processing. InferSens CTO Jonathan Peace presented the first application for Legionella water system monitoring at the TinyML EMEA Innovation Forum 2022 in Cyprus.

CAMBRIDGE-based InferSens today announced the first application of its breakthrough sensor technology, which brings sophisticated deep learning to ultra-low-power sensors and opens up a series of new cutting-edge sensor applications. of technology, rich in data, which were previously difficult or impossible to process.

Deep learning (DL) is the main area of ​​research in artificial intelligence (AI) – in which algorithms learn patterns in data without being explicitly programmed to do so. Historically, these kinds of algorithms for rich sensor data have been processed in the cloud, creating issues with data fidelity, privacy, latency – and most importantly power consumption. Attempts to solve this problem, by moving DL to the edge, have required the use of local power sources which significantly impede large-scale sensor deployment.

InferSens’ sensor technology uses state-of-the-art silicon, state-of-the-art DL models, and innovative mechanical and system engineering to break the power consumption versus performance lock-in and bring to market the first of a new generation of smart sensors for the built environment and beyond.

Silicon is an ultra-efficient next-generation neural accelerator – the Syntiant NDP120 Neural Decision Processor – that uses in-memory computing and native neural network instruction processing for enhanced data optimization, enabling InferSens to achieve a on-device deep learning processing at just 1% of the power required by traditional processors.

The NDP120 can simultaneously run multiple AI algorithms at less than 1mW, among other sensor and voice applications, and is designed to natively run multiple deep neural networks on a variety of architectures, such as CNNs, RNNs and fully connected networks.

Syntiant CEO Kurt Busch said, “We are excited to be working with InferSens to deploy cloud-free sensor solutions with our industry-leading AI processor technology. The NDP120 delivers 25 times the tensor throughput of our first-generation neural network, enabling highly accurate sensor processing with near-zero power consumption.

The first product release of InferSens’ sensor technology – scheduled for Q1 2023 – is a low-cost, battery-powered water temperature and flow sensor for monitoring and detecting the risk of legionella in systems. of water. It can be quickly and easily attached to any pipe without any cutting or plumbing. It is initially aimed at the commercial and public sector property market (7.6 million properties in the UK alone) – where regulations require owners to monitor the risks of Legionella – the cause of a life-threatening form of pneumonia contracted through contaminated water.

InferSens CEO Colin Payne said, “We’ve been developing our high-tech sensor technology since 2017, thinking about the future of on-device deep learning, which is why we were able to achieve pole position. position with this revolutionary technology. Prior to the official product launch, we are engaging with customers for pilots, under commercial agreement, from a variety of industries including universities, hospitals, commercial offices, hotels and other owners and real estate operators.

He continued, “The technology has very significant potential in multiple other practical uses such as multifactor building occupancy, combined air quality and compound sensing for buildings and smart cities, and others. applications where low power, rich data and intelligent sensing requirements converge. We look forward to bringing an exciting product portfolio to market in due course. »

CTO Jonathan Peace will present the technology and the first application at the TinyML EMEA Innovation Forum 2022 in Cyprus on October 11, 2022.

About InferSens

InferSens is a high-tech company focused on transforming sensor technology. Based in Cambridge, UK, InferSens brings sophisticated deep learning models to ultra-low-power sensors, opening up a range of new data-rich sensor applications that were previously difficult or impossible to manage. The sensor technology leverages a new wave of next-generation ultra-efficient neural accelerator silicon, which takes, aggregates and classifies data at a very fast rate, allowing InferSens to perform deep learning processes on the device at nearly 1% of the power required by traditional transformers. Earlier this year, InferSens raised $1 million to accelerate the development of the core capabilities of its high-tech sensor technology from a mix of strategic and technology investors, including the RO Group, the Oxford Innovation Fund and a number of influential serial entrepreneurs.

More information about the company can be found by visiting or by following InferSens on Twitter @InferSens or LinkedIn.

About Syntiant

Founded in 2017 and headquartered in Irvine, CA, Syntiant Corp is a leader in providing end-to-end deep learning solutions for always-on applications by combining purpose-built silicon with a optimized data and training pipeline. Syntiant’s advanced chip solutions merge deep learning with semiconductor design to produce ultra-low-power, high-performance deep neural network processors for cutting-edge artificial intelligence applications in a wide range of cases consumer and industrial use, from headphones to automobiles. The company is backed by several of the world’s leading strategic and financial investors, including Intel Capital, Microsoft’s M12, Applied Ventures, Robert Bosch Venture Capital, Amazon Alexa Fund and Atlantic Bridge Capital.

More information about the company can be found by visiting or by following Syntiant on Twitter @Syntiantcorp or LinkedIn.


Comments are closed.