Research On Video Cache Enabling For Heterogeneous Cellular Network

To deal with the explosive growth of mobile data,heterogeneous cellular net-work(HCN)constructed through embedding small cells(e.g.femto-cells and pico-cells)served by low-cost low-power low-coverage small-cell base stations(SBSs)into the existing macro cells served by macro-cell base stations(MBSs)has received much attention.However,limited backhaul links of SBSs can become the bottleneck of the whole network.Caching popular videos at SBSs can efficiently alleviate the backhaul burden,and reduce the transmission delay.Therefore,both enabling caching function-ality in an existing HCN and cell planning with cache deployment for a new HCN are worth studying.This thesis studies the problems mentioned above,and the main contributions are as follows.1?For the problem of enabling caching functionality in an existing HCN,this thesis studies how to jointly optimize the activation of SBSs,the allocation of cache size,and the placement of video contents to maximize the hit ratio of mobile user(MU)requests,while the total cost of activation and storage cannot exceed a given budget.This thesis firstly formulates the problem as a combinatorial optimization problem and proves that the problem is NP-hard even if there's no SBS activation cost.Further,it is proven that there exists no polynomial-time algorithms with an approximation ratio better than 1—1/e for the problem unless P=NP.For the special cases that SBSs have no activation cost and equal activation cost,algorithms based on maximum coverage and partition are proposed,which both have the best approximation ratio of 1—1/e.For the original case where SBSs have different activation cost,this thesis proposes an algorithm that mainly employs a greedy heuristic.And the algorithm achieves an approximation ratio of(1-1/e)/2.Simulation results highlight the importance of allocating a proper cache size to each SBS,and show that the scheme proposed in this thesis can achieve much better performance than activating the SBSs with maximal number of adjacent MUs.2?This thesis studies cache deployment during the cell planning process for a new HCN.Under the budget constraint,several factors including candidate site se-lection for both MBSs and SBSs,bandwidth and power resource allocation,cache allocation on activated SBSs and video content placement should be jointly optimized so that MUs served by the HCN can be maximized.The problem is formulated as a mixed integer programming problem,and divided into several subproblems to deal with.Firstly,bandwidth and power allocation problem is formulated as a convex opti-mization problem and solved.And then an existing algorithm is utilized to deal with the problem of MBS activation and MU association.Afterwards,a greedy heuristic is proposed to cope with the problem of SBS activation?cache size allocation and MU association.Finally,an algorithm that divides the total budget to activate MBSs and SBSs seperately and then exhaustively searches the division is proposed to deal with the original problem.Simulation results show that caching can efficiently enhance the communication capacity of SBSs and enlarge the total number of MUs served by the HCN.