Research And System Measurements Of Wireless Signal Propagation Loss Prediction Algorithm In Mobile Communication System

In the process of planning and optimization of the mobile communication system,one of the most important tasks is predicting network coverage, or field strengthdistribution, and then determines whether the planned network meet the requirements.The author participates in major national projects of "The development of TD-LTEnetwork optimization tool" and "The research of LTE coverage enhanced networkingtechnology ". In the project, the author concentrate on the wireless signal propagationloss prediction and coverage prediction in mobile communication system.The author’smain work is consists of theoretical research, measurements and simulation resultanalysis.In this thesis, the indoor and outdoor propagation characteristics on several keybands are analyzed, and then given the main conclusions. The propagation lossprediction method based on the inversion model is introduced, and on this basis, theprediction area division algorithm is proposed to solve a major bottleneck of inversionmodel in engineering applications.The main innovation of this thesis is as follows:1) For the lack of the exsisting propagation model calibration process, we proposeda standard calibration process, and emphasized the importance of noise analysis andfrequency selection. Under the same condition, the propagation loss of some keyfrequency bands are compared, the result is of great reference value for frequencyplanner and network optimization operator.2) The prediction regions must be divided before coverage prediction in inversionmodel, and according to the low efficiency of manual division method, the predictionarea division algorithm is proposed.The algorithm can automatically divide theprediction area, and greatly improved the prediction efficiency, which made importantcontribution to the engineering application of inversion model.The main work of this thesis is as follows: 1) Based on the standard calibration process,700MHz,1.9GHz and2.6GHzoutdoor propagation characteristics are compared;1.7GHz,1.9GHz and2.3GHz indoorpropagation characteristic are compared, and then give the main conclusions.2) The prediction area division algorithm have been proposed, the algorithminclude four processes, which are first sub-area division, second sub-area division,second sub-area border checks and second sub-area boundary correction. Extracted thedata belonging to the same base station from the driving test as a first sub-area, thencalculate the boundary; a first sub-area is divided into multiple second sub-area. anumber of second sub-area can be parallel coverage predicted in order to improvecomputational efficiency.3) Design and implement the coverage prediction system based on inversion model,and details the software overall design and critical code. Measurements and simulationresult shows that the system has higher prediction efficiency, and the prediction error iswhthin the allowed range.In this thesis, the conclusion of indoor and outdoor propagation loss of keyfrequency bands is of important reference value for network planning and optimization.The simulation results of prediction area division algorithm show that this method cangreatly improve the prediction efficiency, saving labor force costs. The design andimplementation of the coverage prediction system is based on the inversion model,which proved the feasible and effective of the algorithms in engineering applicationswith the measured data and performance analysis.