UC Riverside

Multi-hop wireless networks remain an important research frontier of

wireless communications. Multi-hop wireless networks are rapidly

deployable to extend the coverage of the Internet, which can be an

economical alternative to building new base stations. Multi-hop

wireless networks are particularly useful for first responders for

disaster relief, and military operations in battlefields. In this

thesis, we study power scheduling issues for multi-hop wireless

networks. Power scheduling, also known as medium access control

consisting of link scheduling, power control and source beamforming,

fundamentally governs the capacity of multi-hop wireless networks.

In the first part of this thesis, the achievable network throughput

of large-scale multi-hop wireless networks is evaluated under a

power scheduling scheme called opportunistic synchronous array

method (O-SAM). Under O-SAM, a large network is partitioned into

many small subnets, and within each subnet, the link with best

channel gain is scheduled for transmission. We examine the impact of

traffic load, network topology and multiple antennas on the

achievable network throughput. Compared with slotted ALOHA, the

throughput of O-SAM is significantly higher. In addition to O-SAM, a

distributed synchronous array method (D-SAM) is proposed, and its

performance is also evaluated.

In the second part of this thesis, we focus on the power scheduling

problem for multi-input multi-output (MIMO) relay networks. A

generalized water filling (GWF) theorem is established for link rate

maximization with multiple power constraints. The corresponding GWF

algorithm is a fast solution to an important class of convex

optimization problems. The GWF algorithm is a useful building block

for joint source and relay optimization for a multiuser MIMO relay

network. We study the power scheduling problems for both uplink and

down- link cases of the multiuser MIMO relay network. A number of

computational strategies are proposed to maximize the sum rate

subject to power constraints or to minimize the sum power subject to

rate constraints.