Multi-class import vector machine for transmit antenna selection in MIMO systems

Antenna selection is a promising solution to reduce the high cost of multiple RF chains in multiple-input multiple-output (MIMO) systems while maintaining the benefits of spatial diversity and multiplexing gain. By modelling the problem of transmit antenna selection as a multi-class classification and/or decision-making task, this Letter proposed a multi-class import vector machine (IVM) based approach to maximise the average received signal-to-noise ratio (SNR). Simulation results prove that IVM outperforms the conventional optimisation driven algorithm and the state-of-the-art learning-based scheme of support vector machine in terms of average received SNR performance with feasible complexity and sparsity.