Subcarrier selection in OFDM with index modulation via multi-play multi-armed bandits

Abstract

Recently, orthogonal frequency division multiplexing (OFDM) technologies, which utilize an index modulation, have been attracting attention in wireless communication systems such as Internet of Things (IoT) and autonomous vehicle communications requiring low latency and high energy efficiency. The index modulation is a technique of transmitting additional information using index resources of communication systems such as antennas and subcarriers. Since the additional information is transmitted using some index resources of the communication systems, higher energy efficiency can be achieved. Also, since an on/off energy detection is sufficient for the index detection process, it is possible to design a low complexity receiver. There are many studies that utilize the advantages of index modulation technique to the OFDM technologies. In OFDM systems, it is necessary to find an optimal combination of subcarriers since the selected subcarriers for the index modulation have a significant effect on a bit error probability. However, previous studies do not consider the combinatorial effect of the subcarriers since they only select the subcarriers based on the channel gain. In this paper, we study an optimal subcarrier selection scheme that minimizes the bit error probability. We use a Multi-armed bandits approach to devise a novel and efficient algorithm to select optimal combination of subcarriers. The simulation results show that the bit error probability can be reduced by the proposed subcarrier selection scheme.