A congestion-aware user–cell association scheme to assist with traffic offloading in a three-tier heterogeneous network

Abstract

In this work, we aim to expand the cell range in congestion areas. To achieve this goal, we integrate UAV network into macro–micro cellular network to enhance coverage. However, the gain of deploying heterogeneous nodes is limit if there are no CRE (cell range expanding)-based method to steer users towards low-power nodes. Therefore, in order to further improve system access capacity, we propose a congestion-aware UA (user–cell association) scheme to offload queued traffic of cellular network into UAV network. To verify the superiority of proposed scheme, we utilize the low-complexity SG (stochastic-geometry) methods. Then in simulation analysis phase, we obtain five primary results: (a). The SG-based methods have lower computation complexity, and the convergence of derived expressions are verified. (b). The SG-based methods do not need the exact users’ location but the users’ density, which have improved the operability and robustness in infrastructure construction of wireless network. (c). The congestion-aware UA scheme can prompt network to adjust association tier flexibly, which is very suitable for traffic unevenly-distributed regions. (d). The optimal offsets of virtual power are estimated as B2 = −4 dB for small-cell and B3 = 3 dB for drone-cell, the resulting improvement in communication capacity is 8%. (e). We suggest the deployment density of UAV sites will no need to exceed λ3 = 41 (pks/km2).