Study On Wireless Channel Modeling Of 5G Indoor Complex Scene Based On Ray Tracing Acceleration Algorithm

With the rapid development of fifth-generation mobile communications,the number of mobile data services in indoor scenarios is increasing,and it is imperative to study the channel characteristics of complex indoor scenarios.The ray tracing algorithm has high computational efficiency and accurate prediction results,and it is widely used in channel modeling.However,the computational efficiency and computational accuracy of the ray tracing algorithm are restricted by various factors such as scene complexity,modeling accuracy and computational depth.Therefore,this thesis adopts the octree partition algorithm and radial basis function neural network model to accelerate the optimization of the ray tracing algorithm,and the computational efficiency and accuracy are effectively improved.In addition,received power,path loss,delay spread,and angle spread in typical indoor environments are analyzed.The research results of this thesis are mainly divided into the following aspects:(1)Aiming at the low computational efficiency of traditional ray tracing algorithms in complex scenes,this thesis uses the octree partition algorithm to accelerate the ray tracing algorithm,and completes the acceleration of the ray tracing algorithm by reducing useless intersection tests.In addition,in a typical laboratory scenario,the performance of the traditional ray tracing algorithm is compared with that of the algorithm after partitioning with an octree structure.The results show that the ray tracing algorithm based on octree partition optimization can reasonably divide the calculation area according to the location and density of obstacles in the scene,reduce the time spent on additional intersection tests,and improve the computational efficiency of the algorithm.(2)In view of the problem that there is a large error between the simulation results and the measured results of the traditional ray tracing algorithm,and the error is affected by many factors,it cannot be quantified by a fixed mathematical formula.Therefore,this thesis combines the machine learning algorithm on the basis of ray tracing,and uses the RBF neural network model to correct the calculation error of the ray tracing algorithm.In addition,in view of the problem that the RBF neural network is easy to fall into the local optimum during the training process,this thesis introduces the adaptive particle swarm optimization to optimize the selection of the parameters of the RBF neural network model,and compares the optimization results with the basic PSO-RBF neural network algorithm.The results show that the adaptive PSO-RBF neural network model can improve the prediction accuracy,reduce the time required for convergence,and effectively improve the performance.(3)Through modeling and simulation in typical L-shaped office scene and large indoor stadium scene,the effectiveness and feasibility of ray tracing acceleration algorithm in indoor channel modeling are verified.Then,the indoor radio wave propagation prediction is carried out in a typical office scenario,and the propagation characteristics of indoor received power and delay spread are analyzed.