TV Band Radio Environment Mapping in Beijing

Radio Environment Map (REM) is very useful as it provides insight of the spatial characteristics of radio environment. To achieve higher accuracy, a novel REM methodology is proposed in this paper. Noticing the inability of systematic sampling to tackle inaccessible areas, a sampling scheme named simulated annealing assisted electrons repulsion is formulated. In this algorithm, the sampling points and boundary of inaccessible areas are treated as electrons such that the repulsive force between them can ensure the uniform distribution of the sampling points and keep them out of the inaccessible areas at the same time. In addition, simulated annealing is adopted to prevent the sampling location from converging to local optimal. Then, two interpolation methods, namely inverse distance weighting and ordinary kriging, are introduced to estimate the REM using the sampled values. To validate our methodology, a measurement is conducted in Beijing in the TV band. The result shows that our methodology has lower mean absolute error and root mean square error and the superior performance of our methodology is also proved.