The long relative propagation delays between the underwater acoustic channels poses a challenge to the detection of the multiple-input multiple-output signals but also gives a chance for a better space-time signal processing scheme. This paper proposes a detection ordering scheme for the layered space-time detection with the successive interference cancellation (SIC) algorithm, where the channel relative delays leading asynchronous arrival of the layered signals are utilized to arrange the detection order that is quite important for a SIC detection. This delay-based ordering is demonstrated as an optimal one for minimizing the detection error probability via the geometrically based model of the SIC detection. The complexity and calculation of the ordering procedure are significantly decreased by means of the delay estimations of the sub-channels. An iterative layered space-time detector combining the delay-base ordered SIC algorithm with the iterative block decision feedback equalizer is employed, where the iterative equalizer is utilized for the cancellation of the multipath interference and the asynchronous arrival interference. Numerical results show that up to 4 dB performance gain obtained by the delay-based ordered SIC detection for a 2 × 2 MIMO system.
