Modulation classification in adaptive OFDM systems

One possible solution to improve the utilization of the capacity of time-varying radio channels is to make the transceivers adaptive. Application of adaptive modulation in orthogonal frequency division multiplexing (OFDM) systems is of particular interest. The objective of this paper is to investigate the receiver algorithms capable of blindly recognizing the modulation format of the received signal in adaptive OFDM systems of the HiperLAN/2 type. The performance of the studied algorithms is considered by computer simulations in OFDM system with adaptive modulation and channel estimation. The maximum likelihood (ML) and the generalized likelihood ratio test (GLRT) classifiers require only few symbols to provide acceptable performance when good channel estimation and relatively high SNR values are available. The novel classifier requires high SNR and more symbols to achieve acceptable performance, but its computational complexity is still lower than that of the ML and the GLRT classifiers.