Unified Differentiable Architecture Search for Efficient and Low-Latency Spiking Neural Networks in Remote Sensing

Real-time intelligent processing of remote sensing data on resource-constrained edge platforms requires lightweight models with high performance. Conventional lightweight artificial neural networks (ANNs) incur substantial computational costs due to dense multiply-accumulate operations. In contrast, spiking neural networks (SNNs) reduce computation by simplifying calculations to accumulation-based operations driven by sparse spikes. However, existing SNN design methods typically optimize network architecture, quantization, and time steps separately. This fragmented approach often relies on expert heuristics, thereby limiting joint optimization of model efficiency. To address this, we propose a Unified Differentiable Architecture Search (UDAS) framework for SNNs targeting resource-constrained remote sensing applications. It constructs a unified search space integrating critical design parameters for lightweight models, including network topology (operators and channel counts), quantization bit-widths, and time steps. Our framework innovatively treats the time step as a layer-wise optimizable variable through the proposed Temporal Integration strategy, enabling joint spatio-temporal optimization. To accelerate convergence, we introduce the Target Bit-Width Aware Co-Search method, which refines the search space. Experiments on public datasets (UC-Merced, WHU-RS19, AID, and NWPU VHR-10) demonstrate that UDAS-designed SNNs achieve performance competitive with state-of-the-art lightweight ANNs under strict resource constraints. For instance, on WHU-RS19, our SNN achieves 96.94% accuracy (vs. 96.63% for the ANN base-line) while achieving an 8× reduction in storage (1.69MB vs. 13.89MB) and up to 33.23× compression over its full-precision counterpart. Our UDAS framework offers a practical way to fully exploit the efficiency advantages of SNNs in real-time remote sensing applications on resource-constrained devices.