On the Performance of SAGE Algorithm for ToA Estimation in Dual-Band OFDM Systems

This paper discusses the use of the Space-Alternating Generalized Expectation-maximization (SAGE) algorithm for Time of Arrival (ToA) estimation in Orthogonal Frequency Division Multiplexing (OFDM) systems. This algorithm, which is an extension of the expectation-maximization approach, can exploit the intrinsic properties of a multi-carrier OFDM system to achieve the estimation of a number of parameters of each incoming signal replica, such as its amplitude, ToA, Doppler shift, and angle of arrival. The SAGE method can also be easily tailored to estimate only a desired subset of these parameters. Accordingly, this study focuses on the sole ToA estimation, but extending the SAGE functionalities to dual-band OFDM systems. This extension is achieved by exploiting the availability at the receiver of more than one band from the same transmitter, and combining the derived channel information to increase the bandwidth available for estimation. It will be shown that, in highly correlated scenarios, the performance of the SAGE algorithm for ToA estimation in a dual-band system is comparable to that of a single-band system with equivalent total bandwidth.