基于时延和效用的多速率视频的缓存优化

Caching optimization of multi-rate video based on delay and utility

针对无线网络中视频需求急剧增加导致用户接入时延升高系统运行成本增加的问题,提出从最小化蜂窝用户的平均接入时延及最大化缓存系统效用的角度研究视频的缓存优化问题,并结合这两个目标提出时延效用联合最小化函数,通过拉格朗日松弛的原始对偶分解法将原始问题分解为传输和缓存2个子问题,然后分别通过次梯度迭代和贪婪算法解决,得到一种近似最优的系统性能。为了验证方法的有效性,在MATLAB中建立了3种缓存方式的对比实验。最后仿真结果表明,所提出方法能够显著降低蜂窝用户的接入时延,提升系统性能。

The increasing demands of video contents in wireless networks lead to the increase of both the user′s access delay and the system operating costs.Therefore,the paper propose to study the problem of video cache optimization from the point of minimizing the average access delay of the cellular users and maximizing the utility of the caching system,combined with this two objects,the delay and utility joint minimization function is proposed,and original problem is decomposed into two sub problems of delivery and caching by primal dual decomposition method under Lagrangian relaxation,then,the sub problems are solved by iterative and greedy algorithm respectively,and an approximate optimal system performance is obtained.In order to verify the validity of the method,three kinds of comparison experiments were established in MATLAB.Finally,the simulation results show that by using the proposed method,the access delay of the cellular users is greatly reduced and the system performance is significantly improved.