Reduced-Complexity ML Detection and Capacity-Optimized Training for Spatial Modulation Systems - IEEE Journals & Magazine

As a nonprofit organization, IEEE is the world's largest technical specialty organization dedicated to advancing technology for the benefit of mankind.

© Copyright 2018 IEEE - All rights reserved. By using this website, you agree to these terms of service.

Spatial modulation (SM) is a recently developed low complexity multi-input multi-output scheme that uses antenna index and conventional signal set to communicate information. The Maximum Likelihood (ML) detector of the SM system includes simultaneous detection of the transmit antenna index and transmit symbols and thus it has been shown that the ML search complexity increases linearly with the number of transmit antennas and the transmit antenna size ing. Signal set To avoid this problem, we show that the complexity of the ML search of the SM system can be presented irrespective of the constellation size, as long as the signal set used is a square or rectangular QAM. Furthermore, we derive the limits of the SM system capacity by maximizing the worst-case capacity limit of the SM system operating with incomplete channel state information and estimate the optimal power allocation between the data and the training sequence derive.

For multi-input multiple-output (MIMO) systems, optimum maximum likelihood (ML) detection as antenna number and modulation level requires great complexity. In this paper, we can drastically reduce new algorithms and obtain ML performance in complex case. The minimum mean square error (MMSE) criterion streams candidate unreliable symbols, proposed solutions to reduce search space by excluding data. In order to evaluate the measure of reliability of probability, we have each candidate symbol using normalized likelihood function so that ML detection close to optimum is possible. Also, it supports the effectiveness of the proposed method and becomes a performance analysis. We introduce threshold parameters to balance the trade-off between complexity and performance. In addition, we propose a way to generate a Log Likelihood Ratio (LLR) value of an effective way, in which the value can use the coding scheme