Abstract :
A bump circuit with flexible tuning ability that uses 500 times less power and is smaller than previous circuits has been demonstrated by researchers at the University of Tennessee in the US. The challenges and requirements of the analogue deep-learning system inspired the authors to come up with this radically new design. They implemented the bump circuit by preceding the current correlator with a novel nano-power tunable transconductor to achieve variable width and height. By significantly reducing the power consumption of the bump circuit, this work makes possible the realisation of analogue learning and signal processing systems that achieve better energy efficiency than their digital equivalents, and ultimately fully autonomous systems, which are able to get both information and energy from the environment without external intervention.
Keywords :
CMOS analogue integrated circuits; analogue processing circuits; circuit tuning; low-power electronics; nanoelectronics; CMOS process; flexible tuning ability; machine learning applications; nanopower analogue bump circuit; nanopower tunable transconductor; power consumption; signal processing systems;
