Online Downlink MIMO Wireless Network Virtualization in Fading Environments

We consider downlink multiple-input multiple-output (MIMO) wireless network virtualization (WNV) in a fading environment, via a base station (BS) precoding design. The BS is owned by an infrastructure provider (InP) and is shared by several service providers (SPs) who are oblivious to each other. The SPs realize their virtual-cell transmissions via MIMO precoding provided by the InP. We aim to minimize the time-averaged expected deviation of the precoding provided by the InP from the SPs' virtualization demands, considering both long-term and short-term transmit power limits at the BS. We propose an online MIMO WNV algorithm to provide a precoding solution through Lyapunov optimization. Our online precoding solution only requires the current channel state information, and it has a semi-closed form with low computational complexity. We provide an upper bound on the performance of the proposed algorithm, showing that it can be arbitrarily close to the optimum over any given time horizon. Simulation results validate the performance of our proposed algorithm under typical urban micro-cell settings.