LDPC Coded Iterative Signal Detection Based on QRD-M with CRC Check
Koichi Adachi, Masao Nakagawa · IEEE Vehicular Technology Conference · 2006
Maximum likelihood detection with QR decomposition and M-algorithm (QRD-M) is one of the detection methods in multiple-input multiple-output (MIMO) multiplexing. We recently proposed a Turbo coded iterative modified QRD-M with decision directed channel estimation and showed that the packet error rate (PER) performance close to the ideal channel estimation case can be obtained. In the iterative modified QRD-M, signal ranking is performed based on both received signal-to- interference plus noise power ratio (SINR) and results of cyclic redundancy check (CRC) decoding and then, signal detection is performed by modified M-algorithm based on the results of CRC decoding. In this paper, we apply irregular low density parity check (irregular LDPC) codes to the previously proposed iterative modified QRD-M. The a posteriori probability of each bit (including parity bit) is necessary for decision directed channel estimation and modified M-algorithm in the iterative modified QRD-M. Unlike Turbo decoding, the a posteriori probability of each bit can be obtained in LDPC decoding. Thus, by using LDPC codes, more accurate signal detection is possible than using Turbo codes. To achieve more powerful detection of decoding errors, CRC codes can be jointly used with parity check equation of LDPC codes. It is shown by the computer simulation that the performance of the LDPC coded iterative modified QRD-M is superior to that of the Turbo coded iterative modified QRD-M.