An Improved Cooperative Navigation Algorithm based on Compressive Sensing Extended Kalman Filter

Abstract

Aiming at the problem of large amount of cooperative navigation transmission and processing data, this paper analyzes the cooperative navigation model of extended Kalman filter and proposes an improved algorithm based on compressive sensing theory. Based on the state vector and the error covariance matrix in extended Kalman filter, the article organizes the process of compressive sensing, include the method of sparse representation, the design of sensing matrix and the process of signal reconstruction. On the basis of the process, relevant models and formulas are put forward derived and verified by simulation experiment between extended Kalman filter and the filter based on compressive sensing. The sensing matrix is used to complete the sparse representation of the signals in the transmit end, and the convex optimization algorithm is utilized to reconstruct the sparse signals in the receive end, which reduce the amount of data transmitted in the cooperative network and decrease the cost of calculation on the each node. The simulation shows that the position errors are close to the result without compressive sensing. The reliability of the algorithm is reflected by the cooperative correction of node positioning accuracy, and the effectiveness of the algorithm is reflected by the dimensionality ratio of the state vector and the error covariance matrix before and after compressed sensing.