End-to-End Radar and Communication Modulation Recognition with Neuromorphic Computing
Quick Answer
The EMRFormer is a novel spiking neural network architecture that achieves state-of-the-art accuracy in automatic modulation recognition while reducing energy consumption by over 90%.
Quick Take
Tested on a KA200 neuromorphic chip, it outperforms traditional methods, achieving up to five times lower power usage compared to a 3090 GPU.
Key Points
- EMRFormer integrates spike-driven transformers for efficient automatic modulation recognition.
- Achieves state-of-the-art accuracy across various mainstream datasets.
- Maintains strong performance in low signal-to-noise ratio environments.
- Reduces theoretical energy consumption by over 90%.
- Demonstrates up to five times lower power usage on a KA200 chip compared to a 3090 GPU.
Paper Resources
📖 Reader Mode
~2 min readAbstract: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.
| Subjects: | Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2606.24075 [cs.CV] |
| (or arXiv:2606.24075v1 [cs.CV] for this version) | |
| https://doi.org/10.48550/arXiv.2606.24075 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Xiaohu Li [view email]
[v1]
Tue, 23 Jun 2026 02:39:57 UTC (251 KB)
— Originally published at arxiv.org
Want this in your inbox every morning?
Daily brief at your local 8am — bilingual EN/中文, free.
More from arXiv cs.CV
See more →ProMoE-FL: Prototype-conditioned Mixture of Experts for Multimodal Federated Learning with Missing Modalities
ProMoE-FL introduces a Prototype-conditioned Mixture-of-Experts framework for multimodal federated learning, effectively addressing missing modalities. It outperforms existing methods on four chest X-ray datasets, demonstrating superior feature synthesis capabilities in both homogeneous and heterogeneous settings.