Exponential filter-based delay estimation algorithm for active systems in the presence of multi-targets

A multi-targets delay estimation algorithm is presented for band-limited and active systems to estimate delays from noisy received signal that consists of attenuated and delayed replicas of a known reference signal and an additive white Gaussian noise. The normal parameters (delays, amplitudes and number of targets, noise variance) of the received signal are assumed to be unknown. In the particular problem under study, range resolution and output signal-to-noise ratio (SNR) are the most important parameters. However, with the use of the conventional matched filter or inverse filter, one parameter is optimised while the other cannot be guaranteed at a desired level and the receiving system is always fixed by the initial design that is unsuitable for every actual condition. The authors proposed a novel delay estimation filter – exponential filter (EF), with a sensitivity exponent α that controls its output SNR and range resolution. Based on the proposed EF, delay estimation algorithm that is optimal for actual requirements is proposed for multi-targets problem. Moreover, the range resolution can be significantly improved by a technique called CLEAN. Therefore, a CLEAN algorithm-based target separation technique is also applied to the proposed algorithm, which can effectively suppress the interferences on weak targets brought by the stronger targets. Simulation results show the effectiveness of the proposed algorithm.