Trust-Sensitive & Community Based Peer To Peer Resource Management Model And Implementation In Tele-Learning System

Instruction has been an important social activity from time immemorial. With the popularization of digital technology and Internet, Tele-Learning, an emergent way of instruction comes into being subsequently and fully flourishes recently. Along with the development of Tele-Learning and its application in various fields, digital learning resource, which is the foundation and kernel of systematic Tele-Learning, is coming forth abundantly. However, due to absence of scientific and systematic management in the large-scaled Internet environment, effective cooperation is scarce among different digital learning resource providers (servers) and between providers and learners. As a matter of fact, a great amount of substantive digital learning resources are in a disordered state. Consequently, how we can manage all these digital learning resource by dint of modern Information technology to make the best of them is an issue of significant importance in theory and application.After making an intensive study of resource management in Tele-Learning, the author came to the conclusion that the value of learning resources can be assessed form the following five aspects: the amount of information contained, quality, motility, expressiveness and cost. Therefore, the main task of research on Tele-Learning is to increase the amount of information contained in resources, to enhance the quality of resource, to improve its motility and expressiveness, and to reduce its cost. Among the five key elements of resources, the following three would be analyzed with emphasis: the amount of information contained, quality and motility. The details of correlative research work are as follows:With the help of Petri Nets (a classical systematic specification and modeling tool), the author analyzed with formal methods the activities of resources in Tele-Learning, boiled the complicated problems of resource management down to models, and defined the key steps of resource activities in Tele-Learning. Based on the aforementioned work, a TSC (Trust-Sensitive and Community based) Peer-To-Peer Resource Management Model in Tele-Learning is proposed, which is composed of three layers: P2P(Peer-to-Peer) Layer, Trust Layer, and Community Layer. P2P Layer combines digital learning resource providers together in peer-to-peer network infrastructure, which enables any network peer participating in Tele-Learning to be a source provider, and thus enlarges the amount of learning resources. On the basis of P2P Layer, peers and learning resources is evaluated in Trust Layer. The trust management in this layer endows the resources of better qualitywith stronger motility under decentralized mode. Community Layer keeps the movement of resources controlled within a relatively centralized and specific scope compared with the Internet, and enables more efficient accessing, publishing and retrieval of resources, which ulteriorly speeds up the flow of resources. Taking the layers of TSC P2P Resource Management Model as guidance, the author investigated the kernel algorithm of each layer, developed a prototyping system and validated the correctness and effectivity of the model.The main contributions in this dissertation are as follows:(1) A Trust-Sensitive and Community based Peer-To-Peer Resource Management Model is proposed;(2) A DSKTree (Dissymmetrical K-tree) peer-to-peer topology and the cardinal algorithm are proposed. Firstly, DSKTree, as a pure decentralized topology, each peer of it can be an entrance of the network, and it is based on virtual binary tree so that an integrated structure is guaranteed. Secondly, any form of interaction in the network is strictly carried out in an expectable partial extension, without broadcast and multicast used. Thirdly, the system is of great scalability in the following aspects: the amount of network peers, data quantity of the overall network, storage and communication cost. The construction and maintenance of DSKTree, the publishment and retrieval of learning resources algorithms are expatiated in the paper.(3) On the basis of DSK...