Smart antenna beamforming algorithm for mobile communications with high speed moving sources

Low cost sliding window loaded sample matrix inversion (LSMI) smart antenna beamforming algorithm for mobile communication applications is presented. The algorithm consists of the vector inner and outer product operations. With each incoming snapshot it requires to update K vectors of size N, instead of K matrixes of size NtimesN in the counterpart, where K and N is a sliding window and antenna array size, respectively. It is shown that the algorithm can be easy implemented with fixed-point arithmetic, due to simple scaling rule. It is suitable also for highly parallel implementations with FPGAs or DSPs. The choice of K in finite precision environment is discussed.