In the evolving landscape of vertical heterogeneous networks, the practice of cell switching particularly for small base stations faces a significant challenge due to the lack of accurate data on the traffic load of sleeping SBSs. This information gap is crucial as it hinders the feasibility and applicability of existing power consumption optimization methods; however, the studies in the literature predominantly assume perfect knowledge about the traffic load of sleeping SBSs. Addressing this critical issue, our study introduces innovative methodologies for estimating the traffic load of sleeping SBSs in a vHetNet including the integration of a high altitude platform as a super macro base station into the terrestrial network. We propose three distinct spatial interpolation-based estimation schemes: clustering-based, distance based, and random neighboring selection. Employing a real data set for empirical validations, we compare the estimation performance of the developed traffic load estimation schemes and assess the impact of estimation errors. Our findings demonstrate that accurate estimation of sleeping SBSs' traffic loads is essential for making network power consumption optimization methods both feasible and applicable in vHetNets.

翻译：在垂直异构网络不断演进的过程中，小区切换（特别是针对小型基站）面临一项重大挑战：缺乏休眠SBS流量负载的准确数据。这一信息缺口至关重要，因为它阻碍了现有功耗优化方法的可行性和适用性；然而，文献中的研究普遍假设对休眠SBS的流量负载具有完美认知。针对这一关键问题，我们的研究引入了创新方法论，用于估计vHetNet中休眠SBS的流量负载，包括将高空平台作为超级宏基站集成到地面网络中。我们提出了三种基于空间插值的估计方案：基于聚类的、基于距离的和随机邻居选择方案。利用真实数据集进行实证验证，我们比较了所开发流量负载估计方案的估计性能，并评估了估计误差的影响。研究结果表明，准确估计休眠SBS的流量负载对于使网络功耗优化方法在vHetNet中兼具可行性和适用性至关重要。