Research On Fault Tolerance And Security Routing For Mobile Computing

Mobile computing devices and mobile communication devices give birth torevolutionary changes for today's information society, which brings us from personalcomputing era into ubiquitous computing era. Wireless networks are the best project tointerconnect ubiquitous devices and are becoming explosive growth. Unfortunately,solutions to reliability and security of wireless network are getting more and more difficultcompared with fixed wired networks since the probability of system failures is dramaticallyincreasing while the level of sophistication and knowledge required to carry out an attackhas been decreasing. Avoiding failure occurring and tolerating faults are common ways to deal with theproblem of reliability. Only is fault tolerance a ultimate way since the failure occurring inthe system is inevitably. Traditional fault tolerance technique is not efficient or not fit to thewireless/mobile environment, so it is necessary to develop the new efficient fault tolerancetechnique. Routing is a crucial component of wireless networks' protocols. A significantdifference between wireless networks and fixed wired networks is that routing is vulnerableto additional threats above those for a fixed wired network due to the wirelesscommunication link and the dynamic and the cooperative nature of routing infrastructure.Awfully forms of attacks are getting more various. Research on network security strategiesand especially security routing is a urgent problem faced up us. Through exploration andresearch in this aspect, some fruits are obtained, but there are plenty of problems needed tolucubrate. In the context of computing, checkpointing and recovery scheme is a popular techniqueto tolerate faults. It periodically stores the processor's state into stable storages. After atransient failure occurring, the processor's state before failure can be restored and theprocess continues to run from the breakpoint. Existing recovery schemes for mobileenvironments typically either create global consistent checkpoints that don't guaranteeindependent recovery or can only support mobile-MSS-mobile (m-MSS-m) communicationthat exchanged messages among mobile hosts are forwarded by MSSs. This paper presents arecovery protocol based on message logging for mobile host direct to mobile host(m-m)communication wireless network. The m-m communication benefits a less contention on thewireless network and a lower latency for message transmission than m-MSS-m. A proof ofalgorithm correctness is given. Finally simulation results indicate that the performance ofthe proposed approach is higher than that of traditional approaches in terms of fail-freeoverhead and recovery overhead. The determination of the interval time between consecutive two checkpoints is a vitalproblem for the performance of system that employs checkpointing and recovery scheme.The optimal checkpointing interval depends on many factors such as failure rate,checkpointing overhead, checkpointing latency, reliability level and so on. Based ondifferent system optimal target functions, computation equations of optimal checkpointinginterval are different. Aiming at the shortage of too strict assumption or non-optimal systemperformance of previous computation model, a computation model NSBW of optimalcheckpointing interval based on time segment that regards the system availability as optimalfunction is researched. This NSBW model suits for any fault distribution function and has IIIthe advantage of loose assumption. In the context of communication, routing protocols are pivotal ones for wirelessnetworks. Multicasting is a communication service that allows an application to efficientlytransmit copies of data packet to a set of destination nodes. The problem of finding aminimum cost multicast tree is NP-completeness. MPH algorithm is a famous solution tothis problem. In this paper, a novel solution TPMPH to improve the MPH by generating the...