Security-Performance Tradeoffs in a Class of Wireless Network Scenarios

Abstract

There has been an extensive and widespread deployment of wireless local area networks (WLANs) for information access. The transmission, being of a broadcast nature, is vulnerable to security threats and hence, the aspect of security provisioning in these networks has assumed an important dimension. The security of the transmitted data over a wireless channel aims at protecting the data from unauthorized access. The objective is achieved by providing advanced security mechanisms. Implementing strong security mechanisms however, affects the throughput performance and increases the complexity of the communication system. In this paper, we investigate the security performance of a WLAN based on IEEE 802.11b/g/n standards on an experimental testbed in congested and uncongested networks in a single and multi-client environment. Experimental results are obtained for a layered security model encompassing nine security protocols in terms of throughput, response time, and encryption overhead. The performance impact of transmission control protocol and user datagram protocol traffic streams on secure wireless networks has also been studied. Through numerical results obtained from the testbed, we have presented quantitative and realistic findings for both security mechanisms as well as network performance. The tradeoff between the strength of the security protocol and the associated performance is analyzed through computer simulation results. The present real time analysis enables the network designers to make intelligent choices about the implementation of security features and the perceived network performance for a given application scenario.

A Modified IEEE 802.11 MAC for Optimizing Broadcasting in Wireless Audio Networks

“…The traic generation parameters are deined in a generic "music audio data traic model" proposed in [23] and summarized in Table 1. This is a stochastic model that emulates the audio data production from a music instrument or singer based on normal distribution around the tempo of 120 bpm [24,25]. OPNET Modeler 17.1 network simulation platform it is also used for this study.…”