A new CMA-based blind equalization for MIMO systems

We investigate adaptive blind source separation and equalization for linear multiple input multiple output (MIMO) systems. To recover input signals, remove intersymbol interference (ISI) and suppress inter-user interference (IUI) effectively, a new difference-based criterion is proposed, based on which a constant modulus algorithm (CMA) is derived. Computer simulation results illustrate the promising performance of the proposed method. Without channel identification, the proposed method can recover all the system inputs simultaneously and can be adaptive to channel changes without prior knowledge about signals. Furthermore, this algorithm is very robust for any value of signal modulus, which makes it suitable for many practical scenarios.