In automobile collision avoidance radar signals, the insufficiency of stepped FM continuous wave and random binary signals is analyzed. Based on the good characteristics of chaotic signals and DDS technology, the DDS-based chaotic binary signal waveform is designed. The DDS-based chaotic binary signal waveform can be realized in two ways, one is the waveform memory modification method and the other is the limiter method. Simulation analysis shows that compared with other waveforms, the DDS-based chaotic binary signal waveform has the characteristics of good autocorrelation performance, strong anti-interference ability, etc. It is easy to generate a large number of different signal waveforms, which is suitable for automotive collision avoidance radar signal target detection in.