Low Complexity Suboptimal ML Detection for OFDM-IM Systems

TL;DR

This paper introduces a low-complexity suboptimal detection method for OFDM-IM systems that achieves near-ML performance, enabling more practical implementation without significant accuracy loss.

Contribution

A novel suboptimal ML detector for OFDM-IM that improves detection performance over existing methods while maintaining low computational complexity.

Findings

The proposed detector closely matches ML detection performance.

It outperforms the traditional klv detector in high illegal SAP ratios.

The method is suitable for flexible OFDM-IM system implementations.

Abstract

Orthogonal frequency division multiplexing with index modulation (OFDM-IM) is a novel multicarrier scheme, which uses the k out of n subcarriers as active subcarriers to transmit data. For detecting the subcarrier activation pattern (SAP) at the receiver, maximum likelihood (ML) detection cannot be used because of its high computational complexity. Instead, the detector selecting the most likely active k subcarriers is used, which is called a k largest values (klv) detector. However, this method degrades the detection performance especially if the ratio of illegal SAPs to SAPs is high. In this letter, the suboptimal ML detector is proposed, which is a slight modification of the klv detector. However, the proposed detector has a similar detection performance compared to the ML detection, which is suitable for flexible implementation of OFDM-IM systems.