Propagation delay estimation in asynchronous CDMA overlay

In an asynchronous CDMA overlay system with transmitter filtering, the spreading code sequence is altered to form a filtered sequence with spillover, resulting in intersymbol interference (ISI) in the received signal and further complicating the propagation time delay estimation. To solve this problem, we present a new system model and modify three near-far resistant time delay estimators, i.e., the MMSE estimator, the MUSIC estimator and the large sample maximum likelihood (LSML) estimator, and apply them in this scenario. According to the theoretical analysis and the simulation results presented, it is shown that in an AWGN channel, the modified MUSIC estimator can not work well when the system is heavily loaded since the subspace dimensionality is increased to 3 K instead of 2 K in the unfiltered case, where K is the number of CDMA users. The modified LSML estimator offers satisfactory performance but its computational complexity is high, whereas the modified MMSE estimator can acquire timing delay in a short period without significant additional computation complexity and cost