MISO TX-power minimization under imperfect instantaneous CSIT

Abstract

The signal to interference and noise ratio (SINR)-constrained transmit (TX)-power minimization problem, incorporating perfect and instantaneous channel state information at the transmitter (CSIT), culminates in higher overhead and extensive backhauling, likewise average CSIT suffers from quality fluctuations. Also, methods relying on the statistical SINR constraints require several algebraic steps in converting those constraints to deterministic convex equivalents. This process not only loosens the constraint of the problem, but also significantly subtracts the potential performance. In contrary to above mentioned procedures, we account for imperfect instantaneous CSIT and propose a simple solution based on Perron-Frobenius (PF) theorem and uplink-downlink duality to address the problem in a multi-user multiple input single output (MISO) system under prescribed outage. We derive accurate models for the distribution of true SINR experienced by the receivers subject to imperfect CSIT and conditioned on constraint-matching SINR estimates. Simulation results demonstrate the significant performance gain achieved by the proposed solution when compared with the state of the art.