On Throughput Maximization in a Multi-hop MIMO Ad Hoc Network

In recent years, there has been a growing research interest in throughput optimization problems in a multi-hop wireless network. MIMO (multiple-input multiple-output), as an advanced physical layer technology, has been employed in multi-hop wireless networks to increase throughput with a given bandwidth or transmit power. It exploits the use of multiple antennas at the transmitter and receiver to increase spectral efficiency by leveraging its spatial multiplexing (SM) and interference cancellation (IC) capabilities. Instead of carrying complex manipulations on matrices, degree-of-freedom(DoF) based MIMO models, which require only simple computations, are widely used in networking research to exploit MIMO's SM and IC capabilities.

In this thesis, we employ a new DoF model, which can ensure feasible solution and achieve a higher DoF region than previous DoF-based models. Based on this model, we study the DoF scheduling for a multi-hop MIMO network. Specifically, we aim to maximize the minimum rate among all sessions in the network. Some researches have been done based on this model to solve throughput optimization problems with the assumption that the route of each session is given priori. Although the fixed routing decreases the size of the problem, it also limits the performance of the network to a great extent.

The goal of this thesis is to employ this new model to solve the throughput maximization problem by jointly considering flow routing, scheduling, and DoF allocation for SM and IC. We formulate it as a mixed integer linear program (MILP), which cannot be solved efficiently by commercial softwares even for moderate sized networks. Thus, we develop an efficient polynomial time algorithm by customizing the sequential fixing framework. Through simulation results, we show that this algorithm can efficiently provide near-optimal solutions for networks with different sizes.