Research On User Association And Computation Offloading Algorithms For Fog Computing System In Dense Heterogenrous Cellular Network

With the rapid development of information technology,a series of new applications are prevalent gradually.However,the diversified business requirements of new applications and the fast-growing mobile data traffic have brought great challenges to the terminal devices and networks with limited resources in computing and communication,and the cloud computing model far away from the user side and the traditional communication technology are not enough to fully address these challenges.Therefore,the joint deployment of fog computing and dense heterogeneous cellular network can integrate the advantages of the two,and meet user’s service requirements for high-speed,large connection,low delay,low power and other aspects,at the same time,it will add some challenging problems.For example,in view of the limited computing resources of the fog node,and the structural changes of the network environment caused by the sinking of computing services to the user side and many other factors,how to design a reasonable user association strategy and how to effectively divide and unload the computing tasks of the associated users to the collaborative fog nodes.In response to the issues,this thesis mainly studies the following aspects:Inoder to deal with the problem of user association for fog computing in multi small cell scenario,a user association scheme based on multi-attribute decision-making model is proposed for the purpose of improving the quality of experience(Qo E)and system performance,as well as achieveing an effective trade-off between the two.Firstly,six key performance indicators(KPIs)are selected to construct the attribute set of the association scheme and the mathematical models of these attributes are given.At the same time,the combination weights and the evaluation indicator of the association schemes are derivied.Secondly,on the premise of meeting the capacity constraints and the maximum task processing delay constraints of FSBS,all the optional FSBS are taken as the decisionmaking scheme,and the attribute priority is controlled based on the adjustment of attribute weight value,and the joint optimization of multiple KPIs is realized by using the evaluation indicator.Finally,simulation results verify the effectiveness of the proposed scheme.Aiming at solving the computational offloading problem based on the collaborative work of fog nodes in multi small cell scenarios,and in order to take into account the multiple factors of reality collaboration,the contribution degree of collaboration is designed to measure the collaborative value of fog nodes,which integrates the feasibility,fairness and stability of collaboration.Furthermore,a computational unloading scheme based on the contribution degree of collaboration is proposed.Firstly,the contribution degree of collaboration and the contribution ratio coefficient of collaboration are modeled and designed.Secondly,according to the threshold of the remaining computing capacity and the threshold of the contribution degree of collaboration,the algorithm for selecting cooperative fog nodes is given.Thirdly,the objective optimization problem of minimizing the weighted sum of the task execution energy consumption and the user’s payment cost is proposed under the constraint of the maximum tolerable delay,and the optimal unloading decision is obtained by combining the external penalty function method and the direction acceleration method.Finally,simulation results show the performance of the proposed scheme.