Modulation Classification in MIMO Systems

Abstract

Blind modulation classification is an important aspect of today's military systems and is projected to play equal important role in future cognitive radio systems. Modulation classification in MIMO systems present the added difficulty of having to deal with a set of mixed signals as they are received at the receiving antenna array. Whereas, in narrowband SISO systems, the channel was assumed to be a simple complex constant in most cases, the narrowband MIMO channel presents a matrix channel. Without having an estimate of the channel matrix, it does not seem possible to perform modulation classification. Independent Component Analysis (ICA) allows the separation of independent sources in the case the number of receiving antennas is equal to or larger than the number of transmitting antennas. Here we extend previous work in the area of modulation classification in MIMO systems. Though it is applicable for any number of transmitting and receiving antennas, it makes a particular impact on MIMO systems which utilize a large number of transmitting antennas and or complex modulation schemes.