Index Modulation Recognition Based on Projection Residual Analysis

Index modulation recognition (IMR) at secondary user (SU) receiver is a challenging topic for MIMO-OFDM cognitive radio network (MIMO-OFDM-CRN) with index modulation scheme, in order to make SU preferably adapt to the communication environment by adjusting own parameters. For index modulated signals, this paper proposes an effective IMR algorithm based on projection residual analysis (PRA). The proposed algorithm is suitable for various types of modulation such as spatial index modulation (SIM), frequency index modulation (FIM) and space-frequency index modulation (SFIM). Firstly the sparse structure of primary user (PU) signal is detected through removing the joint sparsity of signal matrix. Secondly, according to the detected sparse structure, the problem of whether the signal is index modulation (IM) or unindexed modulation (UIM) is addressed by projection residual analysis with z-test. The hypothesis test judges whether the projection residual power of the received signal is significant compared with that of the UIM case, where the projection residual is obtained through projecting the subcarrier signals in the current index modulation symbol into the subspace of those in the previous symbol. The distribution of the test statistic is derived theoretically under UIM case. Thirdly, combining the detected sparse structure and the results of z-test, the index modulation mode of PU signal is identified. Simulation results verify the performance of the proposed algorithm in terms of bit error rate (BER) and recognition rate, respectively.

Automated summary

This paper proposes an effective index modulation recognition algorithm based on projection residual analysis, which can be applied to various modulation types and identifies the modulation mode of the signal by detecting the sparse structure and conducting a hypothesis test.