Research On Coverage Computing And Optimization Methods In Mobile Communication Network

In the development of mobile communication technology,network coverage quality has always been one of the most important performance indicators for mobile communication networks.Coverage computing and coverage optimization technologies are two main technologies in related research to evaluate and improve coverage in mobile communication networks.The former one evaluates the digital region coverage and constructs the visualized coverage map.The latter one improves the region coverage quality by site planning in the network planning stage or by optimizing the working parameters of the radio frequency antennas installed in base stations in the network operation and maintenance stage.With the technique evolution and the network deployment in mobile communication,especially in scenarios such as large-scale access user number,complex topography and high-density base stations,the techniques involved in the coverage computing and coverage optimization have a large amount of computation and low algorithm convergence speed.Therefore,in view of technical pain points such as the computational complexity and algorithm efficiency for the existing coverage computing and coverage optimization algorithms,this paper designs a series of efficient algorithms related to coverage computing and coverage optimization.The main researches and innovations are listed as follows:(1)In terms of coverage computing,this paper proposes a self-adaption triangulation and area-wise interpolation coverage computing algorithm to construct the visualized coverage map and calculate the digital region coverage ratio efficiently and accurately,which supports multi indicator coverage quality evaluation.First,the method utilizes a triangulation method which is adaptive to the base station location to divide the service region into triangles.Then,the method uses the linear interpolation according to the coverage indicator information of the vertexes in each triangle.Last,the method merges the coverage areas of all triangles and obtains a closed-form solution of the coverage area in service region.The experiment results manifest that the proposed algorithm shows a good performance in constructing visualized coverage map and computing digital coverage ratio,and the proposed algorithm can obtain better accuracy under the same computation complexity compared with the existing coverage computing methods.(2)In terms of the network planning stage,this paper proposes a geographical-induced genetic algorithm to solve the site planning problem towards to the coverage optimization.First,such method divides the service region into several sub-regions and defines the sub-region chromosome.In the mutation operation,the method consciously revises the genes in sub-region chromosome segments where the coverage fitness functions are low to improve the performance of corresponding sub-region or even the overall coverage performance.In the crossover operation,the method exchanges the genes related to the same sub-region of two individuals,which ensures that the site geographical distribution pattern of corresponding sub-region is inherited.The experiment results manifest that the proposed geographical-induced genetic algorithm is valid in the site planning problem,and the proposed algorithm has higher computational efficiency and better near-optimal solution quality compared with the canonical genetic algorithm.(3)In terms of the network operation and maintenance stage,this paper proposes a full gradient descent algorithm and a stochastic gradient descent algorithm to improve the coverage ratio efficiently in service region by adjusting the antenna work parameters such as the azimuths and tilts of antennas.First,such method proposes a novel indicator to measure the network coverage,named as the soft coverage ratio,to approximate the hard version of the original coverage ratio,which use the condition "whether exceed the threshold" as the judgement standard.It changes the binary status,covered and uncovered,representing whether the signal meets the requirement at a certain position,to a continuous measure ranging from 0 to 1,representing how good the signal quality is at this position.Besides,in order to make the objective function differentiable to further support the gradient-driven optimization methods,such method utilizes the derivable versions of the relevant functions to replace the original non-derivative versions.Then,such method maximizes the soft coverage ratio by adjusting the antenna azimuths and tilts based on the derivative of the objective function.The experiment results manifest that the proposed algorithms,especially the statistics gradient descent algorithm,significantly perform well both in their near-optimal solutions and in their computational efficiency compared with the existing meta-heuristic algorithms.This paper also proves that the proposed algorithms have good extensibility and practicality.(4)In terms of the network operation and maintenance stage under the constraint of the network coverage,for the green coverage optimization problem,this paper proposes a gradient descent optimization algorithm to minimize the power consumption with the coverage constraint to acheive green communication.First,the algorithm utilizes a penalty method to convert the power consumption optimization problem which constrained by the coverage condition into a simple form with only lower and upper bound condition.Furthermore,such algorithm transforms the discrete-valued coverage index into a continuous one and uses the sub-gradient to conduct the gradient descent algorithm.The experiment results manifest that the proposed method performs well in green coverage optimization problem.Based on the above research results,this paper carries out systematic researches on coverage computing and coverage optimization technologies in mobile communication network.Many algorithms or key technologies have been proposed for the digital region coverage evaluation,the visual coverage map construction,the site planning toward to the coverage optimization,the base station working parameter optimization under the unconstrained and the constrained conditions.These research results focus on algorithm design with low complexity and adapting to practical mobile communication network scenarios.Some researches have been verified and intergrated in the system of China mobile design institute.The reasearche works of this paper will provide certain technical support for the coverage optimization of the massive network elements in future mobile communication networks and the comprehensive coverage requirements of the space-air-earth-sea integration in 6th generation mobile networks.