A Computationally Efficient Iterative MLE for GPS AOA Estimation

This paper presents a computationally efficient algorithm for GPS signal angle of arrival (AOA) estimation based on an iterative maximum likelihood estimator (iMLE) operating on an antenna array output. The iMLE has global convergence and therefore can estimate the true AOA MLE, regardless of the initial value, and has a simple mathematical form without any derivative or matrix decomposition computation. Simulations are performed for signals with different antenna array parameters, and the performances are compared with that of conventional MUSIC and CAPON methods. The results demonstrate that the iMLE can achieve better noise performance even with a small number of antenna elements. The paper also presents a tracking mode for the iMLE that further improves its performance for signals with low signal-to-noise ratio (SNR) and greatly extends its application into weak GPS signal or multipath environments.