arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2203.04588eess.SPcs.LG
收藏

基于容差离散的FMCW雷达回波无监督域自适应

Unsupervised Domain Adaptation across FMCW Radar Configurations Using Margin Disparity Discrepancy

Fraunhofer Heinrich Hertz Institute(弗劳恩霍夫海因里希·赫兹研究所);Infineon Technologies AG(英飞凌科技股份公司);Technical University of Berlin(柏林工业大学);Technical University of Munich(慕尼黑工业大学)

Rodrigo Hernangomez, Igor Bjelakovic, Lorenzo Servadei, Slawomir Stanczak

摘要:

商业雷达传感越来越重要,机器学习算法是将这种无线电技术推广到监控或医疗保健等领域的关键组成部分之一。然而,雷达数据集仍然是稀缺的,一般还不能实现所有的雷达系统,环境条件或设计参数。因此,部署支持机器学习的雷达应用通常需要一定程度的微调。在这项工作中,我们考虑了在使用调频连续波进行深度学习人类活动分类的背景下,跨雷达配置的无监督域自适应问题。为此,我们专注于理论启发的边缘差异离散技术,这已经被证明是成功的计算机视觉领域。我们的实验将这种技术扩展到雷达数据,对于相同的分类问题,实现了与fewshot监督方法相当的精度。

英文摘要:

Commercial radar sensing is gaining relevance and machine learning algorithms constitute one of the key components that are enabling the spread of this radio technology into areas like surveillance or healthcare. However, radar datasets are still scarce and generalization cannot be yet achieved for all radar systems, environment conditions or design parameters. A certain degree of fine tuning is, therefore, usually required to deploy machine-learning-enabled radar applications. In this work, we consider the problem of unsupervised domain adaptation across radar configurations in the context of deep-learning human activity classification using frequency-modulated continuous-wave. For that, we focus on the theory-inspired technique of Margin Disparity Discrepancy, which has already been proved successful in the area of computer vision. Our experiments extend this technique to radar data, achieving a comparable accuracy to fewshot supervised approaches for the same classification problem.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑