Multi-stream sum rate maximization for MIMO af relay networks

Abstract

We consider a multiple-antenna amplify-and-forward (AF) two-hop interference network with multiple links and multiple relays. Transmit precoders, receive decoders and relay AF matrices are optimized with the purpose to maximize the system sum rate. By pointing out that the existing models all lead to single data stream transmission for each user, we propose a novel multiple stream model. We maximize the Total Signal to Total Interference plus Noise Ratio (TSTINR), with the requirement of orthogonal columns of precoders and total relay transmit power constraint. An efficient algorithm is proposed to solve the corresponding problem. Simulations show that the system sum rate significantly benefits from multiple data streams in medium to high SNR scenarios.Index Terms-MIMO relay network, total signal to total interference plus noise ratio, multiple stream, alternating iteration

Interference Neutralization for Non-Regenerative Two-Way Relaying Networks—Part II: Optimization

“…Clearly, both (27) and (28) can be formulated into standard SDP problems, which are solved using the interior-point algorithm in [5]. Taking into account the guaranteed convergence speed of the Dinkelbach-type algorithms, we conclude that Algorithm 1 provides a polynomial-time solution.…”

“…If the update of is calculated as (26) and in each iteration we solve the following problem (27) 1 The cost function in (20) is quasi-convex. 2 of our problem is bounded between and which is equivalent to it is called Dinkelbach Type-I (DT-1) algorithm in [7] and was used for solving the max-min SINR problem in multi-pair TWR networks with one multi-antenna relay in [12].…”

“…To simplify the optimization problem, we propose to maximize the total SINR. This corresponds to the total signal to total noise and interference power and this term has already been coined in [27] as TSTINR for one-way multi-relay MIMO AF systems. Let us define and .…”

MIMO Beamforming Design towards Maximizing Mutual Information in Wireless Sensor Network

“…The subproblems are shown to be convex trust-region subproblems. The previous approaches inevitably rely on numerical solvers, such as semidefinite relaxation (SDR) [12], or the conventional numerical solution to general trust-region surbproblem [14], to solve these subproblems, and some of them are only applicable to strictly convex subproblems [8]. Exploiting the convexity, we successfully enhance the standard solution to the trustregion subproblem for the convex case by developing near closed-form solutions (possibly up to a simple one-dimension bisection search).…”

“…It is worth noting that there is a coherent MIMO amplifyforward (AF) model developed in the recent relay literature [12]- [14]. The model thereof consists of several sourcedestination pairs communicating via the help of a group of common relays performing amplify-and-forward (AF).…”