Resource Management For Traffic Offloading Networks Via Dual-Connectivity

The rapid growth of mobile services and mobile terminals makes explosive growth of gobal data traffic,how to effectively cope with such a huge data traffic is an eager issue for network operators.Traffic offloading is regarded as a promising program by the operators and academics to tackle with this issue.Offloading cellular network's traffic to some other complementary networks(such as small-cell network)can effectively alleviate cellular network's pressure,and improve mobile users'(MUs')experience.At the same time,the 3rd Generation Partnership Project(3GPP)proposes a new technology named dual-connectivity,through which users can simultaneously communicate with a macro base station(mBS)and a complementary network's access point(AP),hence dual-connectivity plays an important role in traffic offloading.In this paper,we design the dual connection technology and data streaming technology combined with the design of an efficient traffic offloading scheme based on dual-connectivity to achieve optimal resource management in traffic offloading network.The mian part and contributions of this paper are as follows:1.From the perspective of users,we design a user-orinted traffic offloading scheme to minimize the user data cost for single small cell,in which we realize the jointly optimal traffic scheduling and power allocation.Due to the mutual interference among mobile users(MUs),the proposed problem is strictly nonconvex,which is difficult to solve in general.In spite of this,we perform a series of equivalent transformation of the problem,based on which we design an efficient feasibility checking scheme based on geometric programming.Then we present an algorithm based on polyblock approximation to solve this problem efficiently.Numerical results are presented to validate the performance of our proposed algorithm,it can efficiently save the computational time compared to LINGO.The results also verify the advantages of the traffic offloading scheme based on dual-connectivity to minimize users' data cost,more specifically,it can save more than 75% of the cost compared to no-offloading shceme and it can save more than 65% of the cost compared to fixed-offloading shceme.2.From the perspective of system revenue,we design a system-orinted traffic offloading scheme to maximize system revenue,which is defined as the difference of system revenue and cost.In the scheme,we realize the optimal access selection,traffic scheduling,and power allocation.The proposed problem is a mixed binary nonlinear programming problem,which is very difficult to solve directly.Exploiting the characteristics of the proposed problem,we solve it by two steps.Firstly we consider the transmit power minimization problem for single small cell case and then we continue consider the users' access selection problem for multiple small cell case.The transmit power minimization problem is proved as a convex optimization problem,and we design a bisection based algorithm to solve it efficiently,based on which we design an efficient access selection algorithm based on the idea of simulated annealing to solve the users' access selection problem.Simulation results are provided to show the performance and advantages of our proposed algorithm,it can efficiently enhance the system revenue compared to heuristic access scheme.