SaleNet: A low-power end-to-end CNN accelerator for sustained attention level evaluation using EEG
This work addresses the need for low-power, accurate EEG-based attention monitoring, which is incremental as it applies known compression techniques to a specific domain.
The paper tackled the problem of evaluating sustained attention levels from EEG signals by proposing SaleNet, a CNN accelerator that achieved a state-of-the-art subject-independent classification accuracy of 84.2% with a total compression ratio of 183.11x, implemented on an FPGA with 0.11 W power consumption and 8.19 GOps/W energy efficiency.
This paper proposes SaleNet - an end-to-end convolutional neural network (CNN) for sustained attention level evaluation using prefrontal electroencephalogram (EEG). A bias-driven pruning method is proposed together with group convolution, global average pooling (GAP), near-zero pruning, weight clustering and quantization for the model compression, achieving a total compression ratio of 183.11x. The compressed SaleNet obtains a state-of-the-art subject-independent sustained attention level classification accuracy of 84.2% on the recorded 6-subject EEG database in this work. The SaleNet is implemented on a Artix-7 FPGA with a competitive power consumption of 0.11 W and an energy-efficiency of 8.19 GOps/W.