Two Methods of Target Detection in an Automotive Radar Against the Background of its Own Noise

Abstract

We consider the problem of simultaneous detection of close and distant targets in an automotive millimeter-wave radar. Two approaches for solving the problem are proposed. The first approach is based on the subtraction of more powerful signals in the time domain and the subsequent detection of less powerful signals. The second approach is based on the most plausible estimate of the impulse response of the radar channel. The methods have different computational complexity. Numerical simulation of the proposed methods for processing of signals reflected from targets in terms of the false-alarm and missed-detection probabilities has been carried out. Simulations have shown that both close and distant targets can correctly be detected simultaneously with a low predetermined false-alarm probability.