Blind ISI Channel Estimation, Symbol Detection, and Message Recovery Using Clustering

Le résumé fourni par la source

In wireless communication systems, the presence of inter-symbol interference (ISI), resulting from multipath propagation and channel memory, poses a significant challenge to reliable symbol detection and message recovery. A blind estimation of the ISI channel with a memory length-$L$is proposed. In the receiver, the iterative$K$-means clustering algorithm is used for this blind channel estimation process and a cyclic redundancy check (CRC) is employed to ultimately validate the recovered message. The algorithm produces$K^{L}$potential sets of channel taps as a consequence of channel phase uncertainties caused by phase rotation. Subsequently, these are reduced to$K$sets using the squared error comparison between the received and the reconstructed symbols. The transmitted message is subsequently obtained using these$K$candidate channel tap sets through Viterbi demodulation, channel decoding, and verification of CRC. The simulation results show that for the ISI channel with a dominant line-of-sight component, the block error rate of the proposed approach is close to that of a perfectly known channel, yet it nearly aligns with the performance of the conventional MMSE estimator utilizing a few pilot symbols.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.