Traffic Offloading In LTE-WiFi Heterogeneous Network

With the explosive growth of mobile devices and mobile applications,the capacity of current cellular network is far from meeting the users' requirements for data traffic.To cope with this problem,wireless heterogeneous network technology is proposed.By deploying base stations with different coverage ability,different access technology and different functions,the capacity of the cellular can be effectively improved.At present,macro base station overlaid with Wi Fi hotspots is widely studied.And the system capacity is improved by resource sharing,including LTE-U(LTE-Unlicensed),LAA-LTE(License Assisted Access-LTE),LWA(LTE+Wi Fi Link Aggregation)and Wi Fi offloading.Compared with the other three ways,Wi Fi offloading is easy to implement and does not need to consider the coexistence of LTE and Wi Fi networks.Hence,Wi Fi offloading provides an efficient and low-cost way to help system improve capacity.Therefore,this paper focuses on traffic offloading in LTE-Wi Fi networks.By analyzing network conditions and user requirements,this paper establishes a Wi Fi offloading optimization model to maximize aggregate user satisfaction,and two constraints are considered to guarantee the rate promotion of offloaded users and less impact on Wi Fi networks.Due to the release of cellular resource after offloading,a RB allocation scheme and a power allocation scheme are designed to reallocate the resource after offloading.In order to solve the problem,the optimization model is transformed into a combinatorial optimization problem.And three search algorithms are adopted to obtain the optimal offloaded user set,including the exhaustive search algorithm,the greedy algorithm and the best response algorithm.In the best response algorithm,the combinatorial optimization problem is defined as a game,and the existence of Nash Equilibrium(NE)is proved according to game theory.Simulation results show that the aggregate user satisfaction can be effectively improved while RB allocation scheme is only used,however,this case cannot ensure the promotion of energy efficiency.While using the combination of RB allocation scheme and power allocation scheme,aggregate user satisfaction and energy efficiency can be improved simultaneously,but the energy efficiency is improved at the cost of aggregate user satisfaction improvement.Therefore,in order to jointly optimize the aggregate user satisfaction and energy efficiency,a multi-objective optimization model is further established based on the above model,and the constraints on Wi Fi offloading and resource allocation are all considered.In resource allocation part,two partical swarms are designed for RB allocation and power allocation respectively.A two-layer framework is proposed to find the global optimal solution.For the outer framework,the best response algorithm is adopted to obtain the optimal offloaded user set.For the inner framework,a multi-objective PSO based resource allocation algorithm is provided to achieve the jointly allocation of RB and power.Simulation results show that the proposed algorithms and the scheme can jointly and effectively improve the aggregate user satisfaction and energy efficiency.