Performance Analysis of Cellular Radio Access Networks Relying on Control- and User-Plane Separation

Abstract

In a cellular radio access network (RAN) relying on control-and user-plane separation, control base stations (CBSs) form the control plane (CP) and take care of control coverage as well as low speed data delivery. In contrast, traffic base stations (TBSs) in the user plane (UP) are used for transmitting high speed data. Based on queuing theory and stochastic geometry, we derive the coverage probability of the CP under non-line-ofsight (NLoS) transmissions, and derive the energy efficiency of the UP under coexisting line-of-sight (LoS) and NLoS propagation modes. Numerical results show that when the traffic load is high, the coverage probability of the CP increases with the increasing CBSs' data rate or the density of CBSs. We also found that the maximum energy efficiency of the UP can be achieved by optimizing the TBS density, while achieving a high area spectral efficiency.

Deep Reinforcement Learning-Based Optimization for End-to-End Network Slicing With Control- and User-Plane Separation

“…Yang et al [12] have shown that the CUPS architecture outperforms the conventional network architecture in terms of the coverage probability and the handover cost in UDNs. Liang et al [13] employed the stochastic geometry and queuing theory to analyze the coverage probability of the CP assuming nonline-of-sight (NLoS) transmissions and the energy efficiency of the UP under the co-existence of both NLoS and line-ofsight (LoS) transmissions.…”

Separation of Control and Data Transmissions in 5G Networks May Not be Beneficial

“…There was no comparison with the energy consumption of legacy system. Similarly, the recent paper [52] calculates and optimizes coverage probability of signaling BSs under non-line-of sight (NLoS) and energy efficiency of data BSs under LoS and NLoS conditions but did not compare the energy efficiency with that of a legacy system.…”

Performance Analysis for Control- and User-Plane Separation Based RAN With Non-Uniformly Distributed Users

“…The utilization of a highfrequency band and shorter transmission distances in the U-plane introduces a more complex wireless environment, leading to the coexistence of line-of-sight (LoS) and non-lineof-sight (NLoS) transmissions. In this scenario, the SE, EE, and coverage have been analyzed in [11]. Additionally, the use of unlicensed bands in the U-plane has been explored in [12], [13].…”

“…Random service requests from users are aggregated at their associated BSs, and channels are allocated for data transfer. Queueing models have been applied to analyze the traffic conditions according to different network models, including M/M/1 [42], M/M/C/C [11], M/G/1 [43] and interacting queues [41], [44]. [45]- [47] adopt the mean-field approximation to settle the "interacting queue" problem.…”

“…However, in the case of a non-uniform and nonhomogeneous spatial network model, introducing this method would amplify the challenges of analysis. Another assumption proposed by [11], [48], [49] posits that the channel allocation time or channel holding time follows an exponential distribution, simplifying performance analysis.…”