Least square solver for wireless communication system

In this paper a high performance least square solver is presented which use recursive Cholesky decomposition. Wireless communication systems require solving least square equations in order to obtain taps weights of the FIR filter. It is thus about to develop a fast and efficient algorithm for computation pseudo-inverse matrices. This paper also presents the recursive way to calculate the correlation matrices of receiving signal which is applied to blind channel estimation and for spectrum sensing. The recursive Cholesky algorithm is verified for the Rayleigh channel and the Basis expansion model for known as well as an unknown covariance matrix. The experimental results are very close to analytical results.