Optimization Research Of Video On Demand Cache System

People are increasingly accustomed to using mobile terminals to watch online videos,but the user experience is limited by battery life.In process of viewing,data transmission is one of the main sources of energy consumption.Therefore,it is of theoretical and practical significance to study efficient data transmission methods to improve data transmission efficiency and save transmission energy consumption,thereby prolonging battery life.Existing research on reducing transmission energy consumption generally assumes that the bandwidth of wireless interface is either unlimited or an unpredictable random variable.Under this assumption,some studies increase the sleep time of the wireless interface of the terminal device by aggregating data stream,or adopt a fast dormancy mechanism to reduce the tail energy consumption.However,some researches have confirmed that in real life,people usually move in fixed routes(such as commute routes),thus the changes in wireless bandwidth along the way can be treated as a known condition(pre-measured and predicts).In this perspective,combined with the known wireless channel bandwidth,there is more space to design an efficient video transmission method,which can minimize transmission energy consumption while satisfying the user's video viewing quality.Efficiently transferring video in combination with known bandwidth faces the following challenges.First,the video download speed is limited by the length of the mobile terminal cache,so downloading data with the best bandwidth is not always the best option.Second,since the video needs to be played continuously after it is started,it is necessary to find some more suitable opportunities outside the good bandwidth in order to prevent running out cache.Therefore,this paper proposes an energy efficiency driven video downloading algorithm,which can achieve a delicate balance between video playback continuity and transmission energy consumption,and achieves efficient data transmission.Specifically,the main contributions of this article are as follows.(1)This paper establishes a bandwidth-known mobile terminal optimized transmission model to minimize video transmission energy consumption.This model discretizes the continuous video time,and characterizes the optimal problem as an integer linear programming problem with discrete time.This problem is constrained by the cache limit to ensure the best playback fluency and minimize the transmission energy consumption.(2)This paper proposes a maximum equivalent energy efficiency based video downloading algorithm,which can reduce the transmission energy consumption while ensuring video playback continuity.The basic idea is to find optimizable segments(according to the Greedy algorithm).For each optimizable segment,it iteratively selects the download job with a maximum equivalent energy efficiency(i.e.,the maximal amount of data that can be transmitted per unit energy consumption time),until the data is completely downloaded.Simulation results demonstrate that the proposed algorithm significantly outperforms the traditional Greedy algorithm and the On-off transmission algorithm with an equivalent energy efficiency decrease of up to 83%and 69%,respectively.At the same time,it achieves the same video playback quality as greedy transmission.(3)This paper proposes a dynamic buffer threshold algorithm for the re-buffering process,it optimizes each re-buffering segment to further reduce the video re-buffering time while the energy consumption optimization is almost unchanged.The basic idea is to ensure the existing playback continuity,keep the starting moment of the next re-buffering process of the adjacent re-buffer unchanged,and try to change the termination time of the previous re-buffering process until the minimum re-buffering time is found without causing extra re-buffering time.The simulation results show that the proposed algorithm can reduce the total re-buffering time by 82%.