Chaos theory has extensive applications in many fields, and the research on this technique has important theory and practice value, which especially more and more demonstrates its superiority in the weak signal detection. In this paper, application methods of chaos in weak signal detection are instructed. The restraint nonlinear programming method combined optimization algorithm with chaos detection method to detect unknown frequency. Mixed methods about cross-correlation with chaos theory was studied. Properties of chaotic signal were ananlyzed and so on. The methods about detecting weak signal based on chaos theory are novel and completely different signal processing methods compared with the traditional time domain detecting methods.