Joint Time-Frequency-Space-Power Resource Programming for Dense MIMO OFDMA Systems

Focusing on a dense communication that the number of radio-frequency (RF) chains is smaller than the number of users, we formulate a time, frequency, space, and power full-domain resource programming in orthogonal frequency division multiple access (OFDMA) multiple-input multiple-output (MIMO) systems. Firstly, we present a novel routine for handling the mixed integer non-linear programming (MINLP) problem. Explicitly, by exploiting the inner relationship of multi-dimension variables and conjugate theory, we realize joint scheduling of time and frequency resources by one scheduling variable. Different from the existing search-based or mergence-based approaches, we lead the optimization problem to a convex quadratic constraint quadratic programming (QCQP) problem by leveraging the technique of binary variable relaxation and the difference of convex programming procedure. Finally, a low-complexity iterative algorithm is proposed, aided by a dedicatedly designed second order cone problem (SOCP) for providing an initial point. Simulation shows that our algorithm remarkably outperforms the benchmarks while maintaining a fast convergence rate.