A Neuromorphic Spiking Neural Network Using Time-to-First-Spike Coding Scheme and Analog Computing in Low-Leakage 8T SRAM
C.C. Chen, Yan-Siou Dai, Hao-Chiao Hong
Abstract
This article demonstrates the first functional neuromorphic spiking neural network (SNN) that processes the time-to-first-spike (TTFS) encoded analog spiking signals with the second-order leaky integrate-and-fire (SOLIF) neuron model to achieve superior biological plausibility. An 8-kb SRAM macro is used to implement the synapses of the neurons to enable analog computing in memory (ACIM) operation and produce current-type dendrite signals of the neurons. A novel low-leakage 8T (LL8T) SRAM cell is proposed for implementing the SRAM macro to reduce the read leakage currents on the read bitlines (RBLs) when performing ACIM. Each neuron’s soma is implemented with low-power analog circuits to realize the SOLIF model for processing the dendrite signals and generating the final analog output spikes. No data converters are required in our design by virtue of analog computing’s nature. A test chip implementing the complete output layer of the proposed SNN was fabricated in 90-nm CMOS. The active area is <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$553.4\ttimes118.6$</tex-math> </inline-formula> <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\mu$</tex-math> </inline-formula> m <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$^{2}$</tex-math> </inline-formula> . The measurement results show that our SNN implementation achieves an average inference latency of 196 ns and an inference accuracy of 81.4%. It consumes 242 <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\mu$</tex-math> </inline-formula> W with an energy efficiency of 4.74 pJ/inference/neuron.