Cellular network coverage prediction is a cornerstone of mobile network operators and servicernproviders in order to provide good services to users. Without coverage provisioning, it isrnmeaningless to talk about service or Quality of service (QoS) provisioning. Coverage planning isrna complex task for operators during deploying Radio Access Technology (RAT). This is becausernit needs to consider multiple and correlated network parameters as well as environmentalrnconditions that are out of their control. It is impossible to completely avoid the existence ofrncoverage holes in cellular networks during the planning phase. Therefore, coverage predictionrnprocesses are usually required during the operational phase. rnTraditionally, the cellular coverage estimation performed through drive tests, which consist ofrngeographically measuring different network coverage metrics with a motor vehicle equipped withrnmobile radio measurement facilities. The collected coverage measurements through drive test arernaccurate but limited to roads and other regions accessible by motor vehicles. Drive tests cannot bernconducted in the whole region of the network due to many obstacles such as buildings, lakes, andrnvegetation. Therefore, the drive test is quite inefficient means to solve the coverage problems andrncannot offer a complete and reliable picture of the network situation. rnIn this thesis, the performance of two spatial interpolation methods namely, Inverse DistancernWeight (IDW) and Ordinary Kriging (OK) were evaluated to select which method is best forrnUniversal Mobile Telecommunication System (UMTS) network coverage prediction using thernCommon Pilot Channel Received Signal Code Power (CPICH RSCP) collected from drive test.rnThe experimental analysis was performed on a sample data collected from drive test UMTSrnnetwork in Addis Ababa Ethiopia. Two general interpolation methods were employed withrndifferent parameters. The first method is IDW with various powers and number of neighbors andrnthe second method is OK with Gaussian, Spherical and Exponential semivariogram models withrndifferent numbers of neighbors. The performance of estimation those algorithms were evaluatedrnthrough the cross-validation, coefficient of determination (Rrn), Mean absolute error (MAE) andrnRoot Mean Square Error (RMSE). The test results showed the two coverage prediction methods are able to predict coverage.rnHowever, based on the Exponential model of semivariogram with an optimal number of neighborsrnthe OK method estimated with an error of prediction 4.84 RMSE whereas the IDW estimated 5.33rnRMSE with a percentage difference of 17%. This shows that OK is more accurate than IDW. ThernOK method can infer the missing RSCP data and generates a more accurate coverage map than thernIDW algorithm. This could probably be OK was able to take into account the spatial structure ofrndata. Therefore, this thesis proposes the OK method as the optimal interpolation model to build arnradio coverage map for cellular coverage prediction and hole detection purposes.