On the Approximation of Coverage Probability in Broadcast Single-Frequency Networks

Coverage probability is a key performance indicator for both broadcast and cellular networks, but it is calculated and thus defined in different ways between the two. In this paper, the two definitions are compared and the assumptions for which they are equal are derived. Then the conventional broadcast methodology for calculating the coverage probability is improved by replacing the assumption of uncorrelated shadowing with the assumption of inter-site shadowing correlation and by adding new correction terms to the output of the t-LNM V2 method which is used for approximating the coverage probability. Finally, the impacts of these improvements are simulated using an LTE network as an example, and verified against a Monte-Carlo-simulation. Results have shown that the proposed modifications have accurately accounted for inter-site shadowing correlation and yielded smaller approximation error. Moreover, higher coverage probability is achieved if inter-site shadowing correlation is considered.