Research map / Architectures and efficiency / Spiking networks
Frequency Matching in Spiking Neural Networks for mmWave Sensing
Zhenyu Liao, Di Yu, Changze Lv, Wentao Tong, Linshan Jiang, Sijie Ji, Xin Du, Hailiang Zhao, Xiaoqing Zheng, Shuiguang Deng
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| replace LeNet ANN with the same LeNet backbone using LIF spiking neurons | LeNet · AOPHand | accuracy (%): 60.86 -> 83.70 | Improved |
| replace LeNet ANN with SpikingLeNet | LeNet · MMActivity | accuracy (%): 59.17 -> 75.00 | Improved |
| replace ANN with SpikingLeNet | mmWave classifiers · energy per sample | theoretical energy: ~3.64x lower than best ANN per dataset | Improved |
| add an explicit front-end low-pass filter to LeNet instead of LIF dynamics (ablation) | LeNet · AOPHand | accuracy | Mixed |
| increase membrane decay factor beta from small to moderate values (ablation) | SpikingLeNet · AOPHand, Pantomime | accuracy | Improved |
| increase beta beyond the optimum (over-low-pass) (ablation) | SpikingLeNet · AOPHand, Pantomime | accuracy | Hurt |
| increase simulation timesteps T (ablation) | SpikingLeNet · AOPHand, Pantomime | accuracy | Mixed |
| use recurrent ANNs (RNN, LSTM, BiLSTM, GRU) instead of CNN ANNs | ANN baselines · mmFiT | accuracy (%): GRU 14.11, LSTM 13.14 vs LeNet 62.36 | Hurt |
| use ViT instead of CNN baselines | ANN baselines · AOPHand | accuracy (%): 21.39 vs LeNet 60.86 | Hurt |
| scale up ANN size (VGG16, ResNet101 vs LeNet-size SNN) | ANN baselines · AOPHand | accuracy (%): VGG9 74.39 vs VGG16 67.92 | No clear effect |
| deploy SpikingLeNet on GPU Jetson instead of LeNet | SpikingLeNet · inference latency | latency: approximately 4x overhead | Hurt |
Closest papers by meaning
- Spiking Neural Networks Need High-Frequency Information (NeurIPS 2025 · similarity 0.82)
- Balanced Resonate-and-Fire Neurons (ICML 2024 · similarity 0.82)
- MITO: A Millimeter-Wave Dataset and Simulator for Non-Line-of-Sight Perception (NeurIPS 2026 · similarity 0.82)
- Are Large Brainwave Foundation Models Capable Yet ? Insights from Fine-Tuning (ICML 2025 · similarity 0.81)
- Traveling Waves Encode The Recent Past and Enhance Sequence Learning (ICLR 2024 · similarity 0.81)
- FSTA-SNN:Frequency-Based Spatial-Temporal Attention Module for Spiking Neural Networks (AAAI 2025 · similarity 0.81)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
