Research On Computation Offloading Based On Co-frequency In Cognitive Radio NOMA Network

The rapid development of smart devices has inspired the emergence of various computation-intensive and latency-sensitive applications.Due to the limited computing resources and ability of mobile devices,it is a challenging to handle the high-intensity computing workloads with critical latency requirements.Mobile Edge Computing is a key technology to achieve high data rates and high computing capabilities,which brings the computing capabilities of local terminals and remote clouds to the edge of the network.This thesis studies the computation offloading in NOMA-enabled cognitive radio networks.The main contents include the following:1.Due to the limited computing resources of mobile devices in cognitive radio network,the secondary users in the network can suffer long execution latency.this thesis proposes to offload the tasks of secondary users to the edge server through NOMA transmission protocol.To ensure that the interference caused by the secondary users to the primary users is tolerable and the edge server is not overloaded,this thesis optimizes the offloading decision and transmission power of the secondary users,and formulates the computing latency of the secondary network as a mixed integer non-linear programming problem.In order to overcome the computational complexity caused by the non-convexity of the original problem,the original problem is decomposed into two sub-problems of offloading decision and power control,and the results of the two sub-problems are alternately iterated through the block coordinate descent method to obtain the solution of the original problem.Simulation results show that the computation offloading scheme proposed in this thesis can reduce the total computing latency of the secondary network without causing new spectrum resource overhead.2.The uneven distribution of communication and computing resources in the cognitive radio network causes excessive energy consumption when processing computation-intensive workloads.Aiming at this problem,this thesis proposes that in the case of NOMA technology is accepted in uplink,the secondary users can provide computing service through device-to-device communication with primary users in exchange for the opportunity to offload tasks to the edge server.Under the constraints of task execution deadline and maximum transmit power,the offloading decision and transmission power of the primary users and secondary users are optimized to controlinterference between users and the load of the edge server,and formulate energy consumption minimization as a mixed integer nonlinear programming problem.To solve this problem,the binary variables are relaxed to continuous variables,and the relaxed problem is decoupled into two sub-problems.The non-convex sub-problem is approximated as a linear programming problem using continuous convex approximation,and alternated over the results of two sub-problems by the block coordinate descent method.The simulation analysis prove that the proposed scheme can improve the offloading efficiency of the network,thereby reducing the total computing energy consumption of the entire network.