机会网络协作缓存的研究

发布时间：2018-06-17 22:46

本文选题：机会网络 + 协作缓存　；参考：《新疆大学》2014年硕士论文

【摘要】：机会网络是一种不需要源节点和目标节点之间存在完整的通信链路，利用节点移动带来的相遇机会实现数据通信的自组织网络。由于机会网络具有节点流动性较大、数据时延长、资源有限等特点，如何合理和高效地使用节点提供的缓存资源、增加节点间的协作来减少机会网络时延、节省带宽、防止拥塞，，最终实现资源共享是一个关键问题。本文针对机会网络中缓存空间有限的问题，深入研究了人类移动模型中节点运动状态与相对静止状态下的特点，提出了基于人类移动模型的协作缓存策略，该策略利用目标地址匹配的标准来严格选择协作缓存节点，并采用同步缓存数据表来达到局部域内缓存信息共享的目的，以此能够充分利用各个移动节点有限的缓存资源，有效提高了数据缓存命中率。由于会出现缓存空间已满或数据过期的情况，因此本文提出了基于蚁群算法的缓存替换策略，该策略借鉴求解0/1背包问题的蚁群算法模型，再结合蚁群算法中信息素浓度大小，根据数据对象上信息素浓度选取数据对象，从缓存中替换出没有选取的数据对象。通过仿真实验和结果分析，验证了本文提出的协作缓存策略和缓存替换策略有效性，促进了节点间的协作，降低端到端延迟与网络开销。
[Abstract]:Opportunistic network is an ad hoc network which does not need a complete communication link between the source node and the target node and uses the encounter opportunity brought by the node movement to realize the data communication. Due to the characteristics of node mobility, extended data time and limited resources, how to use the cache resources provided by nodes reasonably and efficiently, increase the cooperation between nodes to reduce the delay of opportunity network and save bandwidth, etc. Preventing congestion and ultimately realizing resource sharing is a key problem. In order to solve the problem of limited buffer space in opportunistic networks, the characteristics of node motion state and relative static state in human mobility model are studied in this paper, and a cooperative caching strategy based on human mobility model is proposed. The strategy uses the criterion of target address matching to select the cooperative cache nodes strictly, and uses the synchronous cache data table to share the cache information in the local domain, so as to make full use of the limited cache resources of each mobile node. The data cache hit ratio is improved effectively. Because the buffer space is full or the data is out of date, a cache replacement strategy based on ant colony algorithm is proposed in this paper. The strategy uses ant colony algorithm model for solving 0 / 1 knapsack problem and combines with pheromone concentration in ant colony algorithm. The data object is selected according to the pheromone concentration on the data object, and the unselected data object is replaced from the cache. The effectiveness of the proposed cooperative cache policy and cache replacement strategy is verified by simulation experiments and results analysis. It promotes the cooperation between nodes and reduces the end-to-end delay and network overhead.
【学位授予单位】：新疆大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TP393.06;TP18

【参考文献】