Energy-aware Cooperative Traffic Offloading Via Device-to-device Cooperations:Investigation And Realization

With the proliferation of smart wireless devices and mobile internet services in recent years,receiving realtime content(such as video streaming)through mobile users(MUs)has become more and more popular nowadays,which yield a tremendous traffic pressure in cellular networks.Properly offloading the MUs' traffic demands by exploiting the MUs' cooperations has been considered as a promising approach to relieve such a traffic demand.Particularly,the MUs' cooperations have been facilitated by the recently advocated paradigm device-to-device(D2D)communications,which enable two MUs to perform a direct communication instead of traversing through the cellular base station(BS).Thanks to the close proximity among the MUs,the MUs' cooperative content distribution via D2 D communications are expected to bring multifolded benefits,such as relieving network traffic congestions and reducing the energy consumptions.Therefore,using the MUs' D2 D cooperations to realize a resource-efficient contents distribution in cellular networks has attracted lots of interests.In our study,we investigated resource-efficient contents distribution problem in the following ways:Firstly,we investigate the cooperative traffic offloading among MUs which are interested in receiving a common content from a cellular BS.For offloading traffic,the BS first sends the content to some selected MUs which then broadcast the received data to the other MUs,such that each MU can receive the entire content simultaneously.Due to each MU's limited transmit-power and energy budget,the transmission rate of the content should be properly designed,since it strongly influences whether and how long each MU can perform relaying.Therefore,different from most existing MUs cooperative schemes,we focus on a novel joint optimization of the content transmission rate and each MU's relay-duration,with the objective of minimizing the system cost accounting for the energy consumption and the cellular-link usage.To tackle with the technical challenge due to the coupling effect between the content transmission rate and each MU's relay-duration,we exploit the decomposable property of the joint optimization problem,based on which we characterize different possible cases for achieving the optimal solution.We then derive the optimal solution for each case analytically,and further propose an efficient algorithm for finding the globally optimal solution of the original joint optimization problem.Numerical results are provided to validate the proposed algorithm(including its accuracy and computational efficiency)and demonstrate that the optimal MUs' cooperative offloading can significantly reduce the system cost compared to some heuristic schemes.Secondly,exploiting MUs' cooperations for content distribution has been considered a promising approach for offloading the rapidly growing traffic demand in cellular network.This paper proposes a realtime contents distribution framework for energy-efficient cooperation among MUs.Given a group of MUs in close proximity interested in downloading a set of contents from the cellular BS,we propose a cooperative scheme in which,the BS first unicasts the concerned CBs to some of the selected MUs,who then relay(via broadcasting)their received data to the rest of the MUs such that each MU can successfully obtain all its desired contents.Regarding this cooperative scheme,we aim at jointly optimizing each content's transmission rate and each MU's relay-strategy(i.e.,it relays which contents and for how long),such that the total energy consumption of the BS and all MUs can be minimized.Such an optimization problem is challenging to solve,since the contents' transmission rates significantly influence the MUs' relay-strategies due to their limited transmit-powers and energy capacities.We then explore the layered structure of the problem and propose a corresponding layered algorithm(including a top-layer algorithm and a bottom-layer algorithm)to compute the optimal solutions including the contents' transmission rates and the MUs' relay-strategies.Numerical results are provided to validate our proposed algorithm and the advantages of the proposed MUs' cooperative contents distribution.Thirdly,we realize the proposed efficient algorithms on the MATLAB GUI platform.On this platform,after inputing some initialized value,one can easily get the optimal results by the efficient algorithms in a visual way.Moreover,people can also review the optimal results in the animated film.