恶劣气候背景下基于CNN的毫米波近程测距技术研究

Research on Millimeter-Wave Short-Range Ranging Technology Based on Convolutional Neural Network under Adverse Weather Conditions

提出了一种在恶劣气候环境下提升毫米波近程测距抗干扰能力的方法。针对传统近程测距中所使用的对称性三角波线性调频多普勒体制对回波相位一致性要求较高,“能量积累”等传统抗干扰方法在强降雨等恶劣气象下无法对杂波进行有效抑制等问题,提出了基于卷积神经网络的信号处理模型,建立了恶劣气候环境下的近程测距智能化抗干扰处理架构,给出了在极低信杂比环境下雨雪杂波的普适性抑制方法。经过抗干扰处理后的测距信号信杂比提升6 dB以上,显著提升了恶劣气候背景下近程测距的精度。

A method was proposed to enhance the anti-interference capability of millimeter-wave short-range ranging technique in harsh weather conditions.In the realm of Doppler systems,the conventional symmetric triangular wave linear frequency modulation(LFM)technique has encountered a plethora of challenges.These challenges encompass the need for high coherence in echo phase,as well as the inadequacy of existing anti-interference methods in adverse weather conditions,such as heavy rain.To overcome these challenges,a signal processing model was proposed based on convolutional neural networks(CNN).An intelligent anti-interference processing architecture was established for short-range ranging in adverse weather conditions,and a universal clutter suppression method was proposed to handle low signal-to-noise ratio clutter caused by rain and snow.The proposed method improves the signal-to-noise ratio of the ranging signal by over 6 dB after anti-interference processing,significantly enhancing the accuracy of short-range ranging under adverse weather conditions.