RTSSNN: Efficient Image Classification for Latency-Critical and Energy-Constrained SNNs Through a Time-Step Reduction Technique
Nada AbuHamra, Baker S. Mohammad, Muhammad Umair Khan, Mahmoud Al‐Qutayri · IEEE Internet of Things Journal · 2025
Spiking Neural Networks (SNNs) are emerging as potent alternatives to Convolutional Neural Networks (CNNs), especially for energy-constrained and latency-sensitive applications, due to their spiked activations and inherent sparsity. Rate-coded SNNs, trained with backpropagation through time (BPTT) and using static data over artificial time-steps, have achieved state-of-the-art results on benchmarks like MNIST. For resource-constrained devices, smaller models help meet memory and power limitations. However, shallow rate-coded SNNs often need numerous time-steps for accurate inference, increasing latency and computational cost. To address this, a technique named RTS (Reduced Time-Step) is proposed to reduce time-steps, optimizing the balance between model size and convergence latency. RTSSNN leverages the periodic dynamics of Leaky-Integrate-and-Fire (LIF) neurons’ membrane potentials when stimulated with constant input by adding a small Fully Connected (FC) layer at the end of the network. At this depth, spikes are sparse and stable, allowing reduced time-steps without losing accuracy. Demonstrated on four-bit quantized SNNs on Raspberry Pi, the method achieves 9x, 4x, and 4.9x operations reduction, 3x, 2x, and 2.2x time-step reduction, as well as 2.3x, 1.5x, and 1.9x runtime reduction during inference on MNIST, FashionMNIST, and GTSRB datasets, respectively, with maintained accuracy. It also shows a 4x, 2x, and 1.9x operations, time-step, and runtime reduction in one-bit quantized SNNs. Additional experiments conducted on the higher complexity CIFAR-10 dataset as well as the dynamic neuromorphic N-MNIST dataset confirmed that RTSSNN is effective primarily on shallow networks with static datasets, where stable activations support periodic membrane behavior in LIF neurons.