Least square based video storing technique on local server

In this paper, the least square based video storing technique is proposed to efficiently store frequently requested video on local server. In order for estimating the relations between the user request and the request time, we utilize the least square. Local server periodically measures the number of user request. Using the number of user request, local server selects the video which represents the last expected time of request using least square for estimating the linear equation. The proposed method is compared with the well-known methods through simulations. The simulation results show that the proposed method performs better than other methods in terms of the block hit rate respectively.