Research on Identification Algorithm Based on Optimal Orthogonal Matching Pursuit

As an inherent physiological signal of the human body, the Photoplethysmography (PPG) signal contains a large amount of pathological and physiological information of the human body. There are great differences in the PPG signal between different individuals, and it has good confidentiality and Uniqueness. In order to make up for the large amount of calculation of the orthogonal matching pursuit algorithm and the difficulty of finding the optimal atom, this paper proposes an improved orthogonal matching pursuit (OMP) sparse decomposition algorithm to decompose the PPG signals of different individuals. Among them, the particle swarm optimization (PSO) optimization algorithm is used to optimize the search for the best atom to realize the fast sparse decomposition of OMP; then the best atom is used to extract the time-frequency characteristic parameters, and combine the time-domain characteristics of the PPG signal to form a feature vector; finally, the extreme learning machine algorithm is used for identification Experimental results show that compared with existing algorithms, the proposed optimized identity recognition algorithm significantly reduces the amount of calculation and ensures the recognition rate, and provides a feasible solution for fast and accurate human identification.