A low-complexity time delay estimation algorithm based on spectral factorization in OFDM system

Abstract

When the Root-MUSIC algorithm (Root-Multiple Signal Classification) is used to estimate time delay in the Orthogonal Frequency Division Multiplexing (OFDM) system, there exists computational redundancy because of the required roots being conjugating symmetry. Aiming at this problem, a Root-MUSIC algorithm based on spectral factorization—SF-Root-MUSIC algorithm is proposed. Based on the structural characteristics of Laurent polynomials, the proposed algorithm uses spectral factorization to reduce the order for the root polynomial by half thus reducing the computational complexity. The simulation results show that the proposed algorithm has lower computational complexity while maintaining the same precision, and is more suitable for real-time computing.