End-to-End Radar and Communication Modulation Recognition with Neuromorphic Computing

2026-06-22

AI Summary

This work addresses the challenge of achieving both high accuracy and energy efficiency in deep learning–based modulation recognition on resource-constrained platforms by proposing EMRFormer, an end-to-end spiking neural network architecture. EMRFormer introduces SpikeFormer—a spiking-driven Transformer—into modulation recognition for the first time, integrating adaptive spiking encoding, integer-based Leaky Integrate-and-Fire neurons, and spiking depthwise separable convolutions to efficiently extract multi-scale temporal features directly from raw IQ signals. The method achieves state-of-the-art accuracy across multiple benchmark datasets, demonstrates robust performance under low signal-to-noise ratios, and reduces theoretical energy consumption by over 90%. When deployed on the KA200 neuromorphic chip, it operates at five times lower power than implementations on an RTX 3090 or Orin NX.

Abstract

Although deep learning-based methods can achieve high accuracy in automatic modulation recognition (AMR) tasks, their high computational cost makes it difficult to strike a balance between accuracy and power consumption, thereby limiting their application on resource-constrained platforms. Neuromorphic architectures that perform spike-driven inference with modest energy budgets have recently been explored for vision and timeseries tasks. Motivated by these works, we propose EMRFormer, a novel end-to-end spiking nerural network (SNN) architecture that applies spike-driven transformer to the constraints of neuromorphic hardware for AMR. The model incorporates an adaptive spike encoder and Integer Leaky Integrate-and-Fire neurons to mitigate the degradation of effective information and enhance SNN representational capacity. By integrating spike-separable Convolution Neural Networks (SSCNN) into Spike-Driven Transformers (SpikeFormer), EMRFormer effectively extracts multi-scale temporal features from the raw IQ waveforms. We validate our approach across various mainstream datasets, the experimental results show that EMRFormer achieves state-of-the-art interms of accuracy, outperforming all the baselines. Furthermore, the model maintains strong performance in low signal-to-noise(SNR) environments and reduces theoretical energy consumption by over 90%. Finally, we evaluate our model on a KA200 neuromorphic chip. The results show that our model achieves up to 5 times reduction in power compared to running on a 3090 GPU or an Orin NX. This work demonstrates a promising pathway for AMR on resource-constrained devices.

Problem

• Automatic Modulation Recognition
• Resource-Constrained Platforms
• Energy Efficiency
• Neuromorphic Computing
• End-to-End Recognition

Innovation

• Spiking Neural Network
• Neuromorphic Computing
• Automatic Modulation Recognition
• Spike-Driven Transformer
• Energy-Efficient AI