On the spatial quantization noise requirements for accurate RF coverage validation and prediction

With the trend of cellular providers shifting to higher frequencies, there is an increasing migration to smaller cells that is further driven by the growing demand for wireless service. This obviously calls for higher resolution radio frequency (RF) validation and prediction. Yet, to the author's knowledge, there has been no study as to what resolution is required for accurate RF modeling and prediction. Many of today's computer prediction tools can provide estimates of RF signal strength at arbitrary spatial resolution. However, the choice of this resolution is often left up to the discretion of the user. Even worse, sometimes the prediction resolution is hard-coded to be the same as that of the terrain data base. Choosing a resolution bin size that is too small is both computationally inefficient and unnecessarily wasteful of valuable memory resources. Choosing a resolution bin size that is too coarse introduces ubiquitous uncertainty about the quality of RF coverage. This paper investigates the spatial quantization noise requirements of RF prediction and RF coverage validation. It is found that the minimum resolution bin size required to mitigate spatial quantization noise effects is about one fortieth of the cell radius.