Blind Classification of MIMO Wireless Signals

Classification of MIMO wireless signals with no prior information of the number of transmitter antenna elements and channel(s) is of tremendous importance in both military and civilian applications. In this research, we present a novel generalized algorithm for classification of digital modulation combined with the recognition of the number of transmit antennas. Our algorithm is developed to rely on only one receiver equipped with a single antenna and does not require any prior channel knowledge. We have introduced a two- phase classifier system. In the first phase, we use a discrete-wavelet transform on the received complex samples to separate the different modulation classes. Following that, we use the K- nearest neighbor approach coupled with k-means clustering to further classify the signals, utilizing the symmetry and relative distances of the constellation points. The performance of the classifiers at different stages of the algorithm demonstrates a high accuracy in an SNR regime where the packets can be decoded.