{"slug": "end-to-end-radar-and-communication-modulation-recognition-with-neuromorphic", "title": "End-to-End Radar and Communication Modulation Recognition with Neuromorphic Computing", "summary": "Researchers propose EMRFormer, a spiking neural network architecture that applies spike-driven transformers to automatic modulation recognition, achieving state-of-the-art accuracy while reducing theoretical energy consumption by over 90%. The model, validated on neuromorphic hardware, demonstrates a 5x power reduction compared to GPUs, enabling efficient radar and communication signal processing on resource-constrained devices.", "body_md": "arXiv:2606.24075v1 Announce Type: new\nAbstract: 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.", "url": "https://wpnews.pro/news/end-to-end-radar-and-communication-modulation-recognition-with-neuromorphic", "canonical_source": "https://arxiv.org/abs/2606.24075", "published_at": "2026-06-24 04:00:00+00:00", "updated_at": "2026-06-24 04:23:45.009616+00:00", "lang": "en", "topics": ["neural-networks", "machine-learning", "ai-research", "ai-chips", "ai-infrastructure"], "entities": ["EMRFormer", "KA200", "3090 GPU", "Orin NX", "SpikeFormer", "SSCNN", "Integer Leaky Integrate-and-Fire"], "alternates": {"html": "https://wpnews.pro/news/end-to-end-radar-and-communication-modulation-recognition-with-neuromorphic", "markdown": "https://wpnews.pro/news/end-to-end-radar-and-communication-modulation-recognition-with-neuromorphic.md", "text": "https://wpnews.pro/news/end-to-end-radar-and-communication-modulation-recognition-with-neuromorphic.txt", "jsonld": "https://wpnews.pro/news/end-to-end-radar-and-communication-modulation-recognition-with-neuromorphic.jsonld"}}