信息中心网络（information-centric networking，ICN）作为一种新型未来互联网体系结构应运而生，并广泛应用于物联网领域，其中，缓存技术作为ICN的显著特征，对信息中心物联网的内容传输性能具有重要影响.由于信息中心物联网具有数据频繁更新、用户对数据新鲜度有严格要求等特性，致使传统信息中心网络缓存技术面临挑战.提出一种基于内容流行度和网络拓扑的分布式缓存策略，同时考虑内容新鲜度，各缓存节点通过优先缓存流行度较高且较靠近用户的内容，以最大化缓存效率.为适应物联网内容的频繁更新，提出一种基于灰色预测的内容缓存收益预测方法，便于快速获取新内容的缓存收益值.同时，该策略具有较低的时间和空间开销.仿真实验结果表明：所提方案相比于传统缓存策略，能有效提高缓存效率和命中率，并降低访问延迟，改善用户体验.

As a new future Internet architecture, information-centric networking (ICN) emerges as the time goes and is widely used in the field of Internet of Things. Caching technology is an important feature of ICN, and has an important impact on the content transmission performance in ICN-IoT. Due to the characteristics of data updating frequently and the strict requirements for data freshness in ICN-IoT, the traditional network caching technology of ICN is facing challenges. This study proposes a distributed caching strategy based on content popularity and network topology, considering the freshness of content. To maximize the caching efficiency, each caching node first caches the content with higher popularity and closer to users. In order to adapt to the frequent updating of content in IoT, a content caching reward prediction method based on grey prediction is proposed, which is convenient to obtain the caching reward of new content quickly. Simulation results show that, compared with the traditional caching strategy, the proposed scheme can effectively improve the caching efficiency and hit ratio, reduce the access delay, and improve the user experience.

TP393

国家自然科学基金（61672490，61602436，61601443）；中国科学院国际合作重点项目（241711KYSB20180002）；中国科学院青年创新促进会项目