JOINT ANGLE-DELAY ESTIMATION BASED ON SMOOTHED MAXIMUM-LIKELIHOOD ALGORITHM

Abstract

In this paper, a novel maximum likelihood algorithm for joint angle and delay estimation is developed to identify the specular components of channel fading for uniform linear array based on the physical propagation channel model. Frequency domain pre- smoothing is applied to the structured frequency transfer matrix before the estimation procedure in order to utilize substantial observations. Iterative Gauss-Newton method is used to solve the multidimensional optimization problem, and a new compact matrix form is presented. Further simpliflcation of the iteration is derived based on the assumption of independent channel parameters. Both simulations and measurement results are investigated for performance analysis. The simulations reveal that the proposed algorithm leads to higher performance with appropriate complexity. Also, a comparison with other algorithms is carried out to validate the accuracy of algorithm by using the power delay proflle measured in a real environment, and the results show the proposed algorithm performs well.