Using Machine Learning to Locate Gateways in the Wireless Backhaul of 5G Ultra-Dense Networks

Chaudhry, Aizaz U.; Raithatha, Mital; Hafez, Roshdy H. M.; Chinneck, John W.

doi:10.1109/isncc49221.2020.9297282

Cited by 2 publications

(2 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The positioning of gateways to increase the capacity of the backhaul network by minimizing the average number of hops (ANH) is addressed by the authors in [ 24 ]. The study has applications in 5G ultra-dense networks but is included in our analysis because of our interest in controlling the number of hops.…”

Section: Related Workmentioning

confidence: 99%

Towards an Efficient Method for Large-Scale Wi-SUN-Enabled AMI Network Planning

Mochinski

Vieira

Biczkowski³

et al. 2022

Sensors

View full text Add to dashboard Cite

In a smart grid communication network, positioning key devices (routers and gateways) is an NP-Hard problem as the number of candidate topologies grows exponentially according to the number of poles and smart meters. The different terrain profiles impose distinct communication losses between a smart meter and a key device position. Additionally, the communication topology must consider the position of previously installed distribution automation devices (DAs) to support the power grid remote operation. We introduce the heuristic method AIDA (AI-driven AMI network planning with DA-based information and a link-specific propagation model) to evaluate the connectivity condition between the meters and key devices. It also uses the link-received power calculated for the edges of a Minimum Spanning Tree to propose a simplified multihop analysis. The AIDA method proposes a balance between complexity and efficiency, eliminating the need for empirical terrain characterization. Using a spanning tree to characterize the connectivity topology between meters and routers, we suggest a heuristic approach capable of alleviating complexity and facilitating scalability. In our research, the interest is in proposing a method for positioning communication devices that presents a good trade-off between network coverage and the number of communication devices. The existing literature explores the theme by presenting different techniques for ideal device placement. Still rare are the references that meticulously explore real large-scale scenarios or the communication feasibility between meters and key devices, considering the detailed topography between the devices. The main contributions of this work include: (1) The presentation of an efficient AMI planning method with a large-scale focus; (2) The use of a propagation model that does not depend on an empirical terrain classification; and (3) The use of a heuristic approach based on a spanning tree, capable of evaluating a smaller number of connections and, even so, proposing a topology that uses fewer router and gateway positions compared to an approach that makes general terrain classification. Experiments in four real large-scale scenarios, totaling over 230,000 smart meters, demonstrate that AIDA can efficiently provide high-quality connectivity demanding a reduced number of devices. Additional experiments comparing AIDA’s detailed terrain-based propagation model to the Erceg-SUI Path Loss model suggest that AIDA can reach the smart meter’s coverage with a fewer router positions.

show abstract

Section: Related Workmentioning

confidence: 99%

Towards an Efficient Method for Large-Scale Wi-SUN-Enabled AMI Network Planning

Mochinski

Vieira

Biczkowski³

et al. 2022

Sensors

View full text Add to dashboard Cite

show abstract

“…for each cluster i do rithm is O(KlN ). In this work, K is the number of clusters, N is the number of APs, and l is the number of iterations of K-means [19]. In our modified K-means algorithm, employment of the weighted distance will not affect the complexity, and K is far lower than N .…”

Section: A Finding Optimal Number Of Vcsmentioning

confidence: 99%

Flexible Virtual Cell Design for Ultradense Networks: A Machine Learning Approach

et al. 2021

View full text Add to dashboard Cite

With the ever-growing demand for even higher throughput, ultradense networks (UDNs) are being deployed for the fifth generation (5G) mobile communications. Although massively distributed radio access points (APs) result in a considerable increase in throughput, they also cause some critical problems. When employing a wireless backhaul, the backhaul capacity becomes a limiting factor, which may result in a high packet loss rate. Furthermore, dense deployment of APs leads to more frequent handoffs for mobile user equipment, which results in heavy measurement and signaling overhead. For a better trade-off between the packet loss rate and the handoff overhead, a machine learning approach for flexible virtual cell (VC) design is proposed that leverages particle swarm optimization (PSO) to quickly find the optimal VC solution. To be responsive to the dynamic traffic demand and backhaul capacity of APs, a new parameter called "weighted distance" is employed in the modified K-means algorithm, which is nested in the PSO procedures for master AP selection and VC boundary determination. Compared with an exhaustive search, optimal VC solutions can be found efficiently through considerably fewer iterations. The proposed method is generic and applicable to disparate UDN application scenarios.INDEX TERMS K-means clustering, mobility management, particle swarm optimization, resource management, ultradense networks, virtual cell.

show abstract

Using Machine Learning to Locate Gateways in the Wireless Backhaul of 5G Ultra-Dense Networks

Cited by 2 publications

References 20 publications

Towards an Efficient Method for Large-Scale Wi-SUN-Enabled AMI Network Planning

Towards an Efficient Method for Large-Scale Wi-SUN-Enabled AMI Network Planning

Flexible Virtual Cell Design for Ultradense Networks: A Machine Learning Approach

Contact Info

Product

Resources

About