A new method of weak signal detection based on improved matching pursuit algorithm

In this paper, the theory of sparse decomposition is introduced to weak signal detection, and the improved matching pursuit (MP) algorithm is studied to accomplish anti-interference process of some typical signals, such as a weak sine wave signal submerged in strong noises. The improved matching pursuit algorithm uses dual-parameter Gabor dictionary, and the iterative times can be modified in accordance with the signal to noise ratio (SNR), the genetic algorithm is also used to improve the efficiency of searching time-frequency atoms, thereby achieving high searching efficiency of time-frequency atoms and rapid noise restraining. The results of experiments indicated that the improved algorithm can effectively increase the searching speed by approximately 100 times and reduce the noises above SNR -15.