针对不稳定且传输代价高的网络传输管道环境下的信息传输,提出一种多级元数据按需小费用传输模型及算法.该传输模型通过提取信息体的元数据,将其组织为元数据二叉树,树中任一个非叶子结点的传输代价均小于或等于它的孩子结点的传输代价.基于此模型的传输算法通过按需传输的元数据决定所要传输的信息体,从而获得尽可能少的传输量.该方法比Yang Yubing的异步断点续传方法的传输代价小.实验结果表明:随着论文数的增加,多级元数据模型传输模型所消耗流量明显小于异步断点续传模型.将该模型及算法应用于科技论文在线手持终端服务平台,对提升该平台的论文传输效率有效可行.
Aiming at information transmission in unstable networks with high transfer cost, a model and its algorithm of on-demand small cost transmission on multi-level metadata are presented. This model is to extract metadata from the transferring information and organize them as a metadata binary tree in which the transmission cost of any of a node except leaves is less than or equal to that of its children. The transfer algorithm based on this model is to transfer the information according to the transferred metadata on demand and get the transmission cost which is smaller than that of Yang Yubing's asynchronous resume broken transmission model. The experi- ment results show that the model begins to consume less network traffic than the Yang Yubing's model. The model is applied to the online hand-held terminal service platform of scientific papers and is feasible and effective to enhance the transmission efficiency of the platform.