Emulating the low power and high efficiency of spiking neural networks (SNN) found in brain biology has long been a goal in electronics. By comparison the human brain, which uses SNN processing, works on about 20 watts while planned exascale machines will run (we hope) on around 30 megawatts. Yesterday, Imec unveiled what it is calling the world’s first SNN chip designed for radar signal processing.
While Imec’s novel chip is intended to support electrocardiogram (ECG) and speech processing in power-constrained devices, the company says its generic architecture features a completely new digital hardware design that could easily be reconfigured to process a variety of other sensory inputs like sonar, radar and lidar data.
“A flagship use-case for our new chip includes the creation of a low-latency, low-power anti-collision system for drones. Doing its processing close to the radar sensor, our chip should enable the radar sensing system to distinguish much more quickly – and accurately – between approaching objects. In turn, this will allow drones to nearly instantaneously react to potentially dangerous situations,” said Ilja Ocket, program manager of neuromorphic sensing at Imec.
Imec reports the chip “consumes 100 times less power than traditional implementations while featuring a tenfold reduction in latency – enabling almost instantaneous decision-making. For example, micro-Doppler radar signatures can be classified using only 30 mW of power.”
As described by Ocket, “SNNs operate very similarly to biological neural networks, in which neurons fire electrical pulses sparsely over time, and only when the sensory input changes. As such, energy consumption can significantly be reduced. What’s more, the spiking neurons on our chip can be connected recurrently – turning the SNN into a dynamic system that learns and remembers temporal patterns. The technology we are introducing today is a major leap forward in the development of truly self-learning systems.”
More widely used today are artificial neural networks (ANN) which, according to Imec, consume power than SNNs and whose “underlying architecture and data formatting requires data to undertake a time-consuming journey from the sensor device to the AI inference algorithm before a decision can be made.” Another strength of its novel chip, says Imec, is that contrary to analog SNN implementations, “Imec’s event-driven digital design makes the chip behave exactly and repeatedly as predicted by the neural network simulation tools.”
“This chip meets the industry’s demand for extremely low-power neural networks that truly learn from data and enable personalized AI. For its creation, we rallied experts from various disciplines within imec – from the development of training algorithms and spiking neural network architectures that take neuroscience as a basis, to biomedical and radar signal processing and ultra-low power digital chip design. That is where Imec really makes a difference,” Kathleen Philips, program director of IoT cognitive sensing at Imec, concludes.