Resource management in communication networks via economic models

This thesis investigates resource management for both wireline and wireless communication networks from an economic point of view. A general model based on microeconomics is developed to study bandwidth and buffer allocation in broadband networks and power control in cellular radio systems.; Wireline communication networks carry a wide range of traffic which has different characteristics and quality-of-service (QoS) requirements. To meet these distinct requirements, we propose to decompose each network component into multiple sub-components. Each sub-component has dedicated bandwidth and buffer capacity and also carries only that traffic which has similar QoS requirements. Thus, the interference from other types of traffic, which have totally different characteristics and QoS requirements, will be avoided.; The first problem addressed in this thesis is how to allocate bandwidth and buffer capacity of one ATM component into virtual paths (VP). Each VP supports one class of traffic which has similar characteristics and quality-of-service (QoS) requirements. Two kinds of connection admission control schemes are developed. Also, the pricing schemes for ATM networks are developed. One pricing scheme is based on the amount of allocated bandwidth and buffer capacity. The other pricing scheme is based on the congestion level presented in the network. As the congestion level increases, the pricing scheme will discourage usage of resources. Given a pricing strategy implemented by an ATM network service provider, it is expected that users will prefer to choose the service with the least cost route.; The economic model developed is also applied to wireless networks to study uplink power control for mobile units. A mobile user's preference is defined in terms of a utility function, which depends on transmitter power and signal-to-interference ratio (SIR). The problem of individual optimal power control is formulated in such a way that each user's utility is maximized. It is shown that the utility is a component-wise concave function of the overall transmitter power under certain reasonable assumptions. The social optimal power control problem is also addressed. The objective is to maximize the total utility of the wireless network. We also study special cases by defining utility as a linear function. A novel pricing scheme is developed for wireless networks based on transmitter power. This pricing scheme is intended to aid in the efficient utilization of available transmitter power.; In this thesis, a general economic-based framework has been established to investigate resource management in communication networks. Pricing schemes have also been developed to make efficient usage of network resources. The economic-based framework is applicable to both wireline networks and wireless networks.