Load-Aware Cell Switching in Ultra-Dense Networks: An Artificial Neural Network Approach

Abubakar, Attai Ibrahim; Öztürk, Metin; Rais, Rao Naveed Bin; Hussain, Sajjad; Imran, Muhammad Ali

doi:10.1109/ucet51115.2020.9205365

Cited by 3 publications

(7 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The authors in [14] proposed a reinforcement learning (RL) based cell switching approach to optimize the energy efficiency as well as the CO 2 emission in a HetNet. A cell switching and traffic offloading scheme for energy optimization in ultra-dense network using artificial neural network was proposed in [15]. The authors in [16] developed a scalable RL based cell switching framework using stateaction-reward-state-action (SARSA) algorithm with value function approximation to determine the optimal switching policy that would minimize the energy consumption in an ultra dense network while ensuring that the QoS of the network is maintained.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Revenue Maximization Through Cell Switching and Spectrum Leasing in 5G HetNets

Abubakar

Öztürk

et al. 2022

IEEE Access

Self Cite

View full text Add to dashboard Cite

Even though dynamic cell switching is a prominent approach for energy optimization in heterogeneous wireless communication networks, it results in spectrum under-utilization as the spectrum originally occupied by the base stations that are turned off remain dormant. In order to make the businesses of primary network (PN) operators, who hold the spectrum license, more profitable and sustainable as well as to avoid spectrum under-utilization, this dormant spectrum can be leased to the secondary network (SN) operators who cannot afford to purchase the spectrum license. In this study, first, the cell switching problem, which solely focuses on the amount of energy saved, is translated to a problem of revenue maximization by including the spectrum leasing concept and converting the energy saving to monetary saving from reduced electricity bills. In this regard, two spectrum demand scenarios are considered for the SN operator: delay tolerant (DT), for non-real time applications, and non-delay tolerant (NDT), for real time applications. Then, a cell switching and spectrum leasing framework based on simulated annealing algorithm is developed to maximize the revenue of the PN while respecting the quality-of-service constraints. The simulation results reveal that the DT spectrum demand is more beneficial to both PN and SN operators as it results in greater revenue for the former while the latter is able to access more spectrum to meet higher service demands. This finding suggests that if the application can tolerate delays, then it makes more sense for both PN and SN to adopt the DT scenario. In addition, it is observed that the performance of the proposed framework is very close to that of the optimal solution with a significant reduction in the computation complexity.INDEX TERMS 5G, HetNet, cell switching, energy efficiency, green communications, spectrum leasing, simulated annealing algorithm.

show abstract

Section: Related Workmentioning

confidence: 99%

“…On the other hand, if there is a maximum limit on the amount of traffic that the MBS can handle then we also have to introduce another constraint. For example, let τ i,1 m denote the maximum traffic that MBS can serve in any time slot t. Then, we have the additional constraint in (15).…”

Section: ) Optimization Objectivementioning

confidence: 99%

Revenue Maximization Through Cell Switching and Spectrum Leasing in 5G HetNets

Abubakar

Öztürk

et al. 2022

IEEE Access

Self Cite

View full text Add to dashboard Cite

show abstract

“…In addition, a large memory would also be required to store the learnt Q values. As a result, it is not feasible to use Q-learning for cell switching operation in UDNs [35] In [13], an ANN based cell switching framework for energy optimization in UDN was proposed. The proposed framework is able to determine the optimal switching strategy that would lead to minimum energy consumption without violating the QoS of the network.…”

Section: Related Workmentioning

confidence: 99%

“…Several cell switching frameworks have been developed using machine learning and heuristic approaches. The machine learning techniques proposed for cell switching includes the use of supervised learning (e.g., artificial neural networks (ANN)) [13], unsupervised learning (e.g., k-means) [14], reinforcement learning (e.g., multi-armed bandit, Q-learning) [9], [15] and deep reinforcement learning (e.g., deep and double-deep Qlearning) [16], [17].…”

Section: Introductionmentioning

confidence: 99%

A Lightweight Cell Switching and Traffic Offloading Scheme for Energy Optimization in Ultra-Dense Heterogeneous Networks

Abubakar¹,

Mollel²,

Öztürk³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

One of the major capacity boosters for 5G networks is the deployment of ultra-dense heterogeneous networks (UDHNs). However, this deployment results in tremendous increase in the energy consumption of the network due to the large number of base stations (BSs) involved. In addition to enhanced capacity, 5G networks must also be energy efficient for it to be economically viable and environmentally friendly. Dynamic cell switching is a very common way of reducing the total energy consumption of the network but most of the proposed methods are computationally demanding which makes them unsuitable for application in ultra-dense network deployment with massive number of BSs. To tackle this problem, we propose a lightweight cell switching scheme also known as Threshold-based Hybrid cEll swItching Scheme (THESIS) for energy optimization in UDHNs. The developed approach combines the benefits of clustering and exhaustive search (ES) algorithm to produce a solution whose optimality is close to that of the ES (which is guaranteed to be optimal), but is computationally more efficient than ES and as such can be applied for cell switching in real networks even when their dimension is large. The performance evaluation shows that the THESIS produces a significant reduction in the energy consumption of the UDHN and is able to reduce the complexity of finding a near-optimal solution from exponential to polynomial complexity.

show abstract

“…The authors in [9] proposed a reinforcement learning based cell switching approach to optimize the energy efficiency as well as the CO 2 emission in a HetNet. A cell switching and traffic offloading scheme for energy optimization in ultra-dense network using artificial neural network was proposed in [10]. The authors in [11] developed a scalable reinforcement learning based cell switching framework using state-action-reward-state-action (SARSA) algorithm with value function approximation to determine the optimal switching policy that would minimize the energy consumption in an ultra dense network while ensuring that the QoS of the network is maintained.…”

Section: Related Workmentioning

confidence: 99%

Revenue Maximization through Cell Switching and Spectrum Leasing in 5G HetNets

Abubakar¹,

Öztürk²,

Öztürk³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

One of the ways of achieving improved capacity in mobile cellular networks is via network densification. Even though densification increases the capacity of the network, it also leads to increased energy consumption which can be curbed by dynamically switching off some base stations (BSs) during periods of low traffic. However, dynamic cell switching has the challenge of spectrum under-utilization as the spectrum originally occupied by the BSs that are turned off remains dormant. This dormant spectrum can be leased by the primary network (PN) operators, who hold the license, to the secondary network (SN) operators who cannot afford to purchase the spectrum license. Thus enabling the PN to gain additional revenue from spectrum leasing as well as from electricity cost savings due to reduced energy consumption. Therefore, in this work, we propose a cell switching and spectrum leasing framework based on simulated annealing (SA) algorithm to maximize the revenue of the PN while respecting the quality-of-service constraints. The performance evaluation reveals that the proposed method is very close to optimal exhaustive search method with a significant reduction in the computation complexity.

show abstract

Load-Aware Cell Switching in Ultra-Dense Networks: An Artificial Neural Network Approach

Cited by 3 publications

References 13 publications

Revenue Maximization Through Cell Switching and Spectrum Leasing in 5G HetNets

Revenue Maximization Through Cell Switching and Spectrum Leasing in 5G HetNets

A Lightweight Cell Switching and Traffic Offloading Scheme for Energy Optimization in Ultra-Dense Heterogeneous Networks

Revenue Maximization through Cell Switching and Spectrum Leasing in 5G HetNets

Contact Info

Product

Resources

About