Variable-rate Users Research Articles

Power Allocation for Non-Orthogonal Millimeter Wave Systems With Mixed Traffic

We consider the problem of power allocation for the down-link of a 5G cellular system operating in the mm-wave band and serving two sets of users: fix-rate (FR) users (transmitting data at fixed rate), typically seen in device-to-device communications, and variable-rate (VR) users (that can change their transmission data rate), typically requiring high data rate services. The power allocation objective is the maximization of the spectral efficiency of VR users while ensuring that FR users get the required rate. In contrast with the existing literature on power allocation, we exploit the sparsity of the virtual mm-wave channel matrix, obtained by applying fixed discrete-Fourier transform beamformers at both the transmitter and the receiver. Exploiting the channel partial orthogonality, users are first grouped based on the mutual interference and then the power is allocated among and within groups.

Fair Resource Allocation and Admission Control in Wireless Multiuser Amplify-and-Forward Relay Networks

In this paper, we investigate the problem of fair resource allocation based on the Nash bargaining solution (NBS) over wireless amplify-and-forward (AF) relay networks in which variable-rate users and constant-rate users coexist. Via the dual decomposition method, a distributed algorithm, including relay selection, relay power allocation, and rate adaptation, is proposed. When the number of users is large and/or the rate constraints are stringent, the problem becomes infeasible, and admission control is necessary. We formulate the admission control and fair resource allocation problem as a two-stage optimization problem, aiming to first find the maximal number of users the network can support with guaranteed rates and then allocate resources based on NBS for the admitted users. Since the problem is combinatorially hard, we transform it into an equivalent one-stage optimization problem, which can be solved by existing methods but with a higher computational complexity. To further reduce the computational complexity, a suboptimal algorithm is then developed. Simulations verify the convergence and fairness of the distributed algorithm. Simulations also show that the suboptimal algorithm achieves an approximate optimal performance but with a significant running time reduction.