Title
Coverage Analysis of Dense Heterogeneous Cellular Networks using Dual-Slope Path Loss Model
Abstract
Densifying the cellular networks has been largely considered as a promising solution in meeting the capacity demands of emerging cellular networks that have expeditiously transformed from voice-oriented to data-oriented. Owing to the disparity in transmit power of the base station (BS) tiers, proactive offloading of users is extensively required to ensure a balance in the user load amongst different BS tiers. Meanwhile, it is equally essential to utilize efficient interference management schemes to account for the excessively increasing interference in the densely deployed cellular networks. There is extensive literature available on the performance analysis of dense multi-tier networks that undertake various load-balancing and interference management mechanisms, but it is strictly limited to the usage of the single-slope (SS) path loss model (PLM). Nonetheless, it is also well-known in the literature that the usage of SS-PLM leads to inaccuracies in evaluating the performance of dense networks. Therefore, the primary objective of this research is to use dual-slope (DS)-PLM in analyzing the downlink performance of dense heterogeneous cellular networks (HCNs) and investigate its impact on the utility/pitfall of the commonly employed load balancing and interference management strategies.
In the first part of the thesis, the downlink performance of uniformly distributed two-tier HCNs is analyzed while jointly considering the load balancing and interference management mechanisms. The adopted load balancing strategy is based on the range extension of small BSs (SBSs), whereas a simple interference management scheme based on the static frequency reuse mechanism is considered. Analytical expressions for the tier association and coverage probability are derived for a randomly chosen user using the tools from stochastic geometry and validated through the Monte Carlo simulations. Owing to the better accuracy of DS-PLM in estimating the path loss in dense deployment scenarios, the results obtained precisely demonstrate the benefits and detriments of densifying SBSs, while employing the load balancing and interference management mechanisms. Meanwhile, a thorough comparison of the evaluated network performance using both SS-PLM and DS-PLM is also performed for the tier association and network coverage probabilities. The comparative analysis verifies the limitation of SS-PLM in dense cellular network scenarios. For instance, it is observed that the usage of SS-PLM not only overestimates the user offloading to smaller BS tier, but it also results in the overestimation of network coverage
In a two-tier uniform HCN with BSs having disparity in transmit powers, deployed densities and critical distances, the SBSs located in the immediate vicinity of macro BSs (MBSs) have negligible coverage areas and therefore act as the strong interference sources and cause degradation of the network coverage. This shortcoming of the uniform HCNs is addressed in the second part of the thesis, where deactivation of the SBSs located in the near vicinity of MBSs is proposed. This selective muting of SBSs introduces a partial spatial correlation between the BS locations of both tiers that transforms the network into a non-uniform HCN. This network transformation ensures that the macro associated users within the near proximity of cell-center region of their respective MBSs experience reduced inter-tier interference. In order to further enhance the performance of these non-uniform HCNs, the above-mentioned load balancing and interference abating strategies are jointly adopted as well. Using stochastic geometry, the mathematical expressions for tier association and coverage probability are derived for a randomly chosen user and validated through the Monte Carlo simulations. A comparison of the estimated network performance using both SS-PLM and DS-PLM is also carried out for the above-mentioned performance metrics. The results highlight that the usage of SS-PLM causes underestimation of the benefits associated with SBS densification, load balancing and interference management in non-uniform HCNs
In a two-tier uniform HCN with BSs having disparity in transmit powers, deployed densities and critical distances, the SBSs located in the immediate vicinity of macro BSs (MBSs) have negligible coverage areas and therefore act as the strong interference sources and cause degradation of the network coverage. This shortcoming of the uniform HCNs is addressed in the second part of the thesis, where deactivation of the SBSs located in the near vicinity of MBSs is proposed. This selective muting of SBSs introduces a partial spatial correlation between the BS locations of both tiers that transforms the network into a non-uniform HCN. This network transformation ensures that the macro associated users within the near proximity of cell-center region of their respective MBSs experience reduced inter-tier interference. In order to further enhance the performance of these non-uniform HCNs, the above-mentioned load balancing and interference abating strategies are jointly adopted as well. Using stochastic geometry, the mathematical expressions for tier association and coverage probability are derived for a randomly chosen user and validated through the Monte Carlo simulations. A comparison of the estimated network performance using both SS-PLM and DS-PLM is also carried out for the above-mentioned performance metrics. The results highlight that the usage of SS-PLM causes underestimation of the benefits associated with SBS densification, load balancing and interference management in non-uniform HCNs.