面向泛在末梢时变网络的多终端协同机制

Multi-terminals effective synergy plays an important role in supporting future ubiquitous service widely applied in ubiquitous stub time-varying network. This project focus on multi-terminal synergy mechanism with the relationship among terminal, network and service in ubiquitous stub time-varying network. Two-way feedback based on terminal reputation incentive model is proposed to ensure mutual trust between service requesters and providers. In this model, it makes punishment strategy for bad behavior. Meanwhile, it also makes the evaluation basis to distinguish credibility of punishment. Another network topology is changing along with time result from the limitation of terminal resources and dynamic network, which causes the problem of service quality jitter and frequent interruption. To solve it, this project proposes time-varying network based on terminal dynamic aggregation mechanism to support smooth transition among terminals and keep service duration. In addition, terminal selfish and abrupt abnormal in one time window cause service interruption. Time window oriented terminal self-adaptation mechanism is proposed to select optimal terminal for service to keep service sustainability. The research of this project about reputation incentive, dynamic aggregation and self-adaptation which has important theoretical guiding significance and application value for ubiquitous service deploying on multi-terminals in ubiquitous stub time-varying network.

泛在末梢时变网络环境下多终端有效协同对支撑未来泛在业务的广泛应用具有重要作用。本项目以终端、网络、业务及其之间关系为对象，研究面向泛在末梢时变网络的多终端协同机制。针对业务请求者与提供者双方缺乏互信问题，研究基于双向反馈的终端信誉激励模型，既为不良行为制定惩罚策略，也对惩罚的可信程度进行评估并反馈，防止恶意惩罚；针对由于终端资源有限性与动态性引起的网络结构随时间变化造成泛在业务质量抖动与频繁中断的问题，研究面向时变网络的多终端动态聚合机制，实现终端间平滑切换，达到保障业务持续性的目的；针对同一时间窗口内由于终端自私性或突发异常导致业务中断的问题，研究面向时间窗口的多终端自适应维护机制，选择最佳终端维护策略，保证业务质量和持续性。本项目提出的基于双向反馈的信誉激励模型、基于多时间窗口的动态聚合机制和基于可用性的自适应维护机制对末梢环境下泛在业务的执行具有重要的理论指导意义和应用价值。

本项目以终端、网络、业务及其之间关系为对象，围绕着泛在末梢时变网络环境中，从终端信誉激励、资源动态聚合和资源动态维护三方面出发，以“问题需求导向-技术理论创新-应用成效总结”的模式，展开了终端信誉评价模型、边缘网络资源动态分配、终端可信接入与评价、受限资源智能分配和卸载维护等关键技术的研究，并在边缘用户用能管理上进行了关键技术的验证。具体内容包括：结合终端特征，构建了支持反馈的终端信誉评价模型，提出了融合信誉与服务质量的业务动态适配机制，确保末梢服务高质量提供；采用人工智能方法，以保障业务质量为目的，分别构建了面向移动区块链和分布式智能的边缘网络资源动态协同机制，根据资源时变环境下资源动态调整，防止业务服务质量抖动与频繁中断，提升边缘网络服务能力；为了防止边缘终端欺骗行为，提出了融合边缘计算与区块链的分布式终端可信认证机制，用于终端可信认证、安全准入与行为评估，确保终端信誉与服务质量；围绕着边缘资源有限与动态性，构建了边缘受限条件下末梢网络资源智能分配和卸载维护机制，确保资源间平滑迁移，保障业务质量和持续性；同时，将部分关键技术在用户能源管理方面展开技术验证，确保技术的可用性与通用性。.以上部分关键技术以算法、模块、组件等多种形式内嵌入能源领域的系统装置中，分别应用在电力通信网专网管理和终端通信接入网建设中，并以第二和第五完成人身份，获得省部级奖励2项。