An Efficient Likelihood Based Automatic Modulation Classification For SISO And MIMO Wireless Communication Systems

Abstract

Automatic modulation classification (AMC) identifies the type of the modulation of the received signal so that it can be demodulated correctly. It is the intermediate step between the signal detection and its demodulation process. AMC is a branch of the non-cooperative wireless communication system, which brings together some features of the traditional cooperative communication theory, which includes channel estimation and tracking, and signal detection.
AMC has wide applications domain in military and civilian scenarios. In the military application, modulation can be considered as an additional variant of encryption, inhibiting the receivers from extracting the information from the signal received without having knowledge of the modulation type. Other than this, automatic modulation classification is also pivotal in identifying the transmission unit for generation of corresponding jamming signals with matching modulation technique.
In today’s modern civilian applications, AMC find its application in intelligent communication systems such as software defined radio and cognitive radio (CR). AMC is nowadays an essential part of the adaptive modulation communication system for improved spectrum efficiency, throughput, link reliability and lowering the bit error rate. In adaptive modulation, the transmitter dynamically varies the modulation parameters or the constellation size of the transmitted waveform according to the channel condition and system specifications.
In this research, we improved the classification accuracy of the existing likelihood-based (LB) modulation classifier for SISO system. The classification process is assisted by the wireless sensor network (WSN) to enhance the classification accuracy. The time complexity of the proposed approach is also presented. The proposed method is shown to have high classification accuracy and improved robustness over the existing methods. But the cost paid for this is high computational complexity.
In the subsequent chapter, we presented the LB modulation classification for MIMO communication systems. The impact of spatial correlation and antenna mutual coupling is also presented. The uniform linear array (ULA) consisting of the dipole antennas is considered for performance analysis.