| As wireless communication technology developing into the3G/4G times, wide coverage and high-bandwidth of wireless communication networks is an increasing need for people. In particular, more and more data services occur in the indoor environment, which makes the indoor wireless coverage very important. Because of the complexity of modern architecture and the strong electromagnetic wave shielding effect of the construction material, there comes a big problem in indoor wireless coverage, and it should be resolved as soon as possible.Computer-aided design of wireless network planning platform is an effective way to solve the wireless network planning, and radio propagation model is the core module. Opposed to outdoor environment, the indoor environment has a higher structural complexity and material complexity. This makes the study on indoor radio propagation models more difficult. In the prediction of the indoor environment field strength coverage, how to improve the calculation speed and accuracy of the models is always the focus and difficulty of the study on indoor propagation model.In this paper, in order to find a suitable indoor radio propagation model, the research work about improving the accuracy of the multi-wall model has been done.(1) The typical radio propagation models have been studied. Such as Two-Ray Model, COST231HATA Model, Motley Keenan Model, Walfisch Ikegami Model, and Ray-based models. Especially, the algorithm analysis of multi-wall model has been done, which provides a basis for the optimization of the model.(2) In order to complete the experiments in models study and models performance verification, a simple experiment platform was set up. With the help of this platform we can use different propagation models to simulate the coverage of the field strength of the indoor environment. In addition, this platform provides a visual analysis capability.(3) In this paper an optimizing in accuracy for multi-wall model and an added consideration of indoor reflection ray have been done, which improves the prediction accuracy of the model. The algorithm of the new model has been redesigned. The handling of transmission and reflection are unified by step method. It improves the computational efficiency of the new model.(4) The performance of the new model is verified by experiments. When the new model prediction are compared with non-optimized multi-wall model, which illustrates that the performance of the new model is improved.Through the research work above, optimization of multi-wall model has been done. The new model has higher prediction accuracy in the indoor environment. It adapted better to the simulation prediction of the indoor environment. In the next work, the new model will be used for the development of wireless network planning platform, expecting to get a better efficiency. |