Modulation Classification for OFDM and MIMO OFDM Systems

Abstract

Blind modulation classification (MC) is an integral part of designing an adaptive or intelligent newlinetransceiver for future wireless communications. Blind MC has several applications in newlinethe adaptive and automated systems of sixth-generation (6G) communications to improve newlinespectral and power efficiency as well as reducing latency. It will become integral to intelligent newlinesoftware-defined radios (SDRs) for future communication. First, we have done the newlineliterature survey on blind MC for orthogonal frequency division multiplexing (OFDM) signals. newlineWe provide various MC techniques for OFDM signals systematically. Our focus has newlinebeen on the most widely used statistical and machine learning (ML) models, emphasize their newlineadvantages as well as their limitations. The statistical-based blind MC includes likelihoodbased newline(LB), maximum a posteriori (MAP), and feature-based methods (FB). The ML-based newlineautomated MC includes k-nearest neighbors (KNN), support vector machine (SVM), decision newlinetrees (DTs), convolutional neural networks (CNNs), recurrent neural networks (RNNs), newlineand long short-term memory (LSTM) based MC methods. This survey can help the reader to newlineunderstand the main characteristics of each technique, their advantages, and disadvantages. newlineWe have also simulated some primary methods, i.e., statistical- and ML-based algorithms, newlineunder various constraints, which allows a fair comparison among different methodologies. newlineThe overall system performance in terms of bit error rate (BER) in the presence of MC is newlinealso provided. We also provide a survey of some practical experiment works carried out newlinethrough National Instrument hardware over an indoor propagation environment. In the end, newlineopen problems and possible directions for blind MC research are briefly discussed. newlineIn the first research problem, automated MC has been addressed as it is becoming a newlinepromising technique for future adaptive wireless transceiver systems. The existing blind newlineMC methods for OFDM fail to achieve the required performance by using statistical-based newlinemethods. Thus, the modu