Improved estimation of TV white spaces in India using terrain data

Cognitive radio offers a novel solution to overcome the problem of spectrum underutilization by providing spectral access to secondary users. The television (TV) bands are of particular interest for secondary usage due to their high penetration power and greater coverage. These licensed bands are occupied only in few regions while in most of the other regions they are unoccupied and are termed as TV white space (TVWS). The estimation of TVWS has mostly been done by using the statistical and empirical propagation models. In this paper, terrain data is incorporated into the estimation of TVWS and shown to improve the quantitative estimation of TVWS. Using the relevant transmitter information and terrain data in the Indian state of Telangana, the efficacy of the proposed approach is demonstrated. The performance of the proposed approach is compared to that of widely used Hata propagation model. It is shown that the accuracy in TV coverage estimation increases on an average by 45% while incorporating terrain data as compared to using only empirical propagation model. As area outside the TV coverage is TVWS, the increased accuracy in the estimation of TV coverage directly translates to improved accuracy in the estimation of TVWS, which in turn translates into more efficient use of spectrum and better interference management.