Improved Detection for Single-Carrier Block Transmission Based on Gradient Search

Abstract

The Orthogonal Frequency Division Multiplexing (OFDM) is one of the main technology for modern communications. However, the OFDM system has the problem of high peak to average power ratio (PAPR). With the proper linear precoding for the OFDM system, we can obtain the Single-Carrier block transmission system (SCBT), which does not have the problem of PAPR. Furthermore, the SCBT system has the advantage of multipath diversity. The received SCBT signal can be transformed into frequency domain (FD) for equalization, followed by the time-domain (TD) detection. However, we can further improve the detection of the equalized TD signal. Some methods based on decision feedback have been proposed to attain better detection. In this paper, we apply the gradient search algorithm (GSA) in the equalized TD signal to reduce the error rates. The GSA is based on differential metrics (DM), and it was first proposed for the detection of the multiple antenna system. Now we apply the GSA algorithm and propose several searching procedures that can achieve lower error rates for the SCBT system. The simulation shows that the GSA can efficiently improve the performance of error rates for the detection of SCBT signals. The proposed method also achieves a trade-off between error performance and complexity.