Energy-efficient transmit beamforming for MISO downlink via sequential convex approximation

We study the problem of energy efficiency maximization (EEmax) with user-specific quality of service constraints in multiple-input single-output broadcast channels with transmit beamforming. For this challenging nonconvex problem, we propose an efficient approximate design based on sequential convex approximation (SCA). By proper reformulations we arrive at a fast converging iterative algorithm where a convex program is solved at each iteration. We further show that the problem in each iteration can also be approximated as a second-order cone program (SOCP), allowing for exploiting computationally efficient state-of-the-art SOCP solvers. Numerical results show that compared to existing two-layer iterative Dinkelbach based algorithms, the SCA-based method can yield a solution within a few iterations and is insensitive to initial points. We also demonstrate that the superior performance to existing scheme is achieved with lower computational complexity.