| With the continuous development of mobile communication technology,computing-intensive applications such as autonomous driving,augmented reality,and virtual reality have emerged in large numbers,which places a huge burden on smart terminal devices with limited computation capability and battery capacity.In order to solve this problem,the scheme of combining mobile edge computing(MEC),ultra-dense heterogeneous networks(UDHN),and non-orthogonal multiple access(NOMA)technologies came into being.It deploys a large number of small base stations with low power consumption through heterogeneous dense and flexible deployment in the cellular microcell,configures MEC servers with computation,storage,and caching capabilities,and allows multiple terminal devices to share the same resource block through NOMA technology,which can meet the connection and service requirements of more terminal devices and increase system capacity while improving spectrum efficiency.However,the dense deployment of small base stations will lead to much energy consumption and make the interference problem more serious.Therefore,to manage cell interference,reduce energy consumption and make full use of network resources for the NOMA-MEC system in ultra-dense heterogeneous network scenarios,the main research contents of this thesis are as follows.1)A computation offloading strategy based on AGADGM is proposed.For the MEC system in the UDHN scenario,a frequency band division mechanism is introduced to manage the cell interference problem in UDHN.The power domain NOMA technology is introduced in the uplink communication link to improve bandwidth utilization further.At the same time,consider multi-user multi-task.The collaborative offloading strategy and the proportion of computation resources allocation strategy are introduced to balance the network load and fully use computation resources.Based on the above content,a collaborative computation offloading model is constructed to jointly manage resource and task computation offloading to minimize the system’s total energy consumption under user delay constraints.Considering that the formulated optimization problem has the characteristics of a nonlinear mixed-integer,the traditional genetic algorithm(GA)is improved,and the adaptive genetic algorithm with diversity-guided mutation(AGADGM)is introduced to solve the problem.Numerical simulation and analysis show that the proposed algorithm can obtain lower total system energy consumption and better convergence performance compared with other algorithms.2)A computation offloading strategy based on GWA is proposed.For the NOMA-MEC system integrated with the frequency band division mechanism in the UDHN scenario,based on the above collaborative computation offloading model,a security model is introduced from the data security perspective,and a collaborative computation offloading model based on the security mechanism is constructed.It aims to minimize the system energy consumption by jointly managing resource management,task computation offloading,and security management under user delay and security cost constraints.Given the nonlinear mixed-integer form of the objective optimization problem,the genetic whale algorithm(GWA)based on AGADGM and the improved whale optimization algorithm(IWOA)is used to solve the problem.Numerical simulation analysis shows that GWA can obtain lower total system energy consumption and better performance than other algorithms. |
