Identifying coverage holes: Where to densify?

We develop a methodology to identify coverage holes in cellular markets, given demographic and socio-economic information of the population, detailed knowledge of the current macro-cellular infrastructure, and the spectrum capabilities of a cellular operator. Coverage holes are defined as areas where users contending for macro BS resources outnumber those who eventually get serviced with minimum acceptable average rates. We develop a method that begins with filtering out users who connect to the Internet through Wi-Fi or to their data networks through small cell BSs. Since the rate observed by a user is a function of their own SINR as well as that of other connected users, we generate an SINR heat map and use a fair scheduling algorithm that allows us to determine the effective load on a BS as a number of users. By comparing the cell load to the number of users with acceptable rates, we obtain a coverage map that identifies holes that should be targeted by further densification. We observe that more cells will be in major deficit ten years from now, with up to 25% of an operator's cells requiring more than four additional micro cell BSs to fill the gap. Finally, we give a list of extensions of this work.