Replica allocation and data update strategies in mobile ad hoc networks.

摘要

A Mobile Ad hoc Network (MANET) is a collection of wireless mobile nodes that forms a network without the aid of any existing communication infrastructure. Providing reliable and efficient data management in MANET environment is a challenging task. This is due to the fact that this network is characterized by limited resources, subject to frequent network partitions and unpredictable topological changes. Several replica distribution algorithms and caching strategies have been proposed to increase data accessibility but little consideration is given to reliable data transmission and data consistency. This thesis provides an integrated solution for data dissemination, replication and update using a clustering architecture. The clustering architecture is built on top of the Ad hoc On Demand Distance Vector routing algorithm.; A novel data transmission scheme called Undelivered Message Queue (UMQ) is developed to improve data transmission reliability. Two variations of this scheme, Localized Undelivered Message Queue and Distributed Undelivered Message Queue are proposed and evaluated using simulation. The results show that the proposed UMQ algorithms achieve reliable data transmission with reasonable communication overhead. We also proposed a replica distribution algorithm and a hybrid data request mechanism to improve data accessibility in the presence of frequent node disconnections. In order to maintain replica consistency, two update strategies have been proposed. The local update algorithm ensures the data consistency within a cluster while the global update algorithm maintains data consistency in the whole network. We introduce a data lock mechanism to prevent multiple users from updating a data item simultaneously. In our simulation experiments, two types of update strategies have been considered: periodic update and random update. The results show that the proposed schemes have high replica distribution rate, transmission reliability and better data accessibility.