This preview shows pages 1–2. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.View Full Document
Unformatted text preview: Project Report: Power Control and Capacity of Spread Spectrum Wireless Networks 1 Introduction In this paper, the authors review two streams of development in power control for spread spectrum wireless networks: the structure of recently proposed power control algorithms, and the character- ization of the achievable information-theoretic capacity region for single-cell systems. In turn, we will review these topics in two settings. 2 Power Control Algorithms A typical cellular system is one where a set of mobile units wish to communicate with a set of base stations. Much like in a traditional network, wireless users must share a fixed amount of resources. Indeed the channel bandwidth and the symbol period being finite, concurrent users lie in a finite dimensional signal space. And as a consequence, users of a multi-cell environment are forced to transmit on non-orthogonal channels. Frequency reuse is only possible through signal attenuation over distance. Thus, any cellular system is ultimately interference limited. Power control has been recognized as an essential requirement in the design of spread spectrum systems, since only by control of transmitted powers can users share radio resources equitably in a multi-cell environment. In their work, the authors adopt the Signal-to-interference ratio (SIR) of estimated symbols as a measure of Quality-of-Service (QoS), with the understanding that the SIR of estimated symbols accurately predicts bit error probabilities. The connection between the two is supported by empir- ical results and can be made rigorous for large systems using the Central Limit Theorem . The goal of power control for spread spectrum systems then becomes minimizing power usage while maintaining acceptable QoS for every user. A distributed power control algorithm is highly de- sirable; the alternative of centralized control involves added infrastructure for data transmission, heavy computations, and delay. In distributed power control algorithms, users are controlled inde- pendently and the control decision concerning the power of a particular user is derived from local measurements only. The implementation of distributed algorithms usually entails iterative power control methods. A successful iterative control policy will guide the evolution of the power level of each user so that eventually all users meet their quality of service requirement and power usage is minimized. Most of the iterative power control algorithms found in the literature can be cast in the framework developed by Yates . Definition 1 Let I : < M < M be an Interference function if for any p the following properties are satisfied: Positivity I ( p ) > Monotonicity If p p , then I ( p ) I ( p ) ....
View Full Document
- Fall '07
- Wireless Networks