Message queueing analysis in wireless networks with mobile station failures and handoffs

Access points play an essential role in fault tolerant architectures for mobile computing environments which engage wireless networks. They are the performance bottleneck in the presence of failures and handoffs of mobile stations. Different message dispatch strategies impose different effects on the message sojourn time in access points. In this paper, we study five dispatch models which are the basic queueing model, the static and the dynamic processor-sharing models, the round-robin model, and the feedback model. We derive the expected message sojourn time in access points under steady state. We observe that the basic model and the static processor-sharing model demonstrate the worst performance. The other three models cut down the sojourn time by dynamically reducing the probability of message blocking which is introduced by failures and handoffs of mobile stations; however, which one is the best dispatch strategy depends on the specific environments. These analysis results can help designers of wireless networks explore better fault tolerant features of mobile systems for their reliability and performance.