Research On Cooperative Caching Strategy Of Mobile-Learning Resources Under Hierarchical Structure

With the growing popularity of mobile intelligent terminal and the rapid development of mobile Internet, the scale of mobile user has become much bigger than before. Various services and applications on mobile terminal not only directly affects the way people work and live, but also plays a far-reaching impact on people’s learning tools and learning process. Mobile Internet is developed on the basis of traditional Internet and with which can provides users with ubiquitous timely movable personalized and location-related services through mobile applications, but it use mobile network as the network transmission medium, and the connective node is always the mobile terminals. So mobile learning is an application of mobile terminals which can help users to learn via the online or locally downloaded resources.Mobile learning developed on the basis of the digital learning and the two shared the common in the forms of learning resources, just like the forms of text, images, audio, video and animation. But because of the encoding and decoding formats of mobile terminals supported are less than personal computers and the start point that the users required are different, mobile learning is more used for quick searching or access to a certain knowledge point, the users are more likely stay in a shallow level, spend shorter time and require a strong real-time response, plus the limited computation power, storage capacity, short-lived battery of mobile terminals and insufficient bandwidth of mobile network, so in order to ensure a precision and quick search require the resources in a mobile environment must be refined and extract the metadata more carefully, In practice, a large number of studies show that the caching technology is an effective method for ensuring the quality of services on large-scale data and resources access. The caching technology can not only reduce the usage of bandwidth and response time, but also the real-time response can improve the user experience. We can also learn a lot from the caching algorithms of peer-to-peer networks to solve the mobile caching problems because of the innate characteristics the mobile terminals have, such as distribution, self-organized management and high dynamic. So based on the research have done before, we proposed a hierarchical cooperative caching strategy which can make disorder nodes to have a structure index which can make the nodes search the resource object for themselves, the key point of this strategy is to make the nodes’ search focus on the neighbor regions based on location and within the region based on chord protocol. Above all, the access of a resources object’s latency is closely related to the distance of two nodes, in order to get a better placement strategy, we set up a bipartite network model which can help us to choose caching objects and the number of object should be generated through considering the visit time and visit number of the resource object, after that, we then redraw a network topology after the bipartite network topology project on nodes, which reflects the correlation of nodes while accessing the related resources, so we can place the blocks belonged to the selective copy of resource object in the node and his neighbors of the projective network topology. The last, when the request to the top level domain, we can determine the probability of each regions on request-path from multiple aspects such as visit time, the size of resource object, the load of each region etc.Finally, the simulation experiment show that cooperative caching strategy not only can effectively reduce the object’s searching hops and have a shorter distance between the nodes which request the object with the nodes which storage the object, but also can make the object’s placement more balance and can solve the problem of excessive cache.