Abstract
Research on various neuro-inspired technologies has received much attention. However, while higher-order neural functions such as recognition have been emphasized, the fundamental properties of neural circuits as advanced control systems have not been fully exploited. Here, we applied the functions of central pattern generators, biological neural circuits for motor control, to the control technology of switching circuits for extremely power-saving terminal edge devices. By simply applying a binary waveform with an arbitrary temporal pattern to the transistor gate, low-power and real-time switching control can be achieved. This binary pattern generator consists of a specially designed spiking neuron circuit that generates spikes after a pre-programmed wait time in the six-order range, but consumes negligible power, with an experimental record of 1.2 pW per neuron. This control scheme has been successfully applied to voltage conversion circuits consuming only a few nanowatts, providing an ultra-low power technology for trillions of self-powered edge systems.
Highlights
Research on various neuro-inspired technologies has received much attention
Most of the current hardware research targets the understanding of biological processes or the implementation of machine learning algorithm, and may not fully exploit the tremendous potential of biological neural circuits
Biological neural circuits are excellent control systems that control body movements with low power consumption and low latency. It compromises the trade-off between power consumption and response time by automating and decentralizing individual motor control, rather than fast centralized feedback control of body movements
Summary
Research on various neuro-inspired technologies has received much attention. while higher-order neural functions such as recognition have been emphasized, the fundamental properties of neural circuits as advanced control systems have not been fully exploited. We applied the functions of central pattern generators, biological neural circuits for motor control, to the control technology of switching circuits for extremely power-saving terminal edge devices. By applying a binary waveform with an arbitrary temporal pattern to the transistor gate, low-power and realtime switching control can be achieved This binary pattern generator consists of a specially designed spiking neuron circuit that generates spikes after a pre-programmed wait time in the six-order range, but consumes negligible power, with an experimental record of 1.2 pW per neuron. In parallel with the artificial central pattern generators for actuators and biological systems, the pattern generators for switching circuits have the potential to provide the real-time and energy-saving control scheme in the future IoT devices.
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
