DeepSlice: A Deep Learning Approach towards an Efficient and Reliable Network Slicing in 5G Networks

Thantharate, Anurag; Paropkari, Rahul Arun; Walunj, Vijay; Beard, Cory

doi:10.1109/uemcon47517.2019.8993066

Cited by 140 publications

(79 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Thantharate et al [62] trained the 5G model using a deep learning algorithm to train and predict a slice for a device based on the information calculated from previous connections. The dataflow was treated with a high-level API to build and train the data using TensorFlow.…”

Section: F Resources and Policy Managementmentioning

confidence: 99%

“…defined an alternative type of the slice called master slice to adapt the user request until the deep learning model arranges a specific slice for the UE based on the demands after the network failures [62]. On the other hand, other researcher added a new slice type called High-Performance Machine-Type Communications (HMTC), which works with applications in a high data rate, low latency and high availability, as shown in the ATIS report.…”

Section: Traffic Classification For the Future Networkmentioning

confidence: 99%

See 1 more Smart Citation

A Survey of Mobility Management as a Service in Real-Time Inter/Intra Slice Control

et al. 2021

View full text Add to dashboard Cite

In-network softwarization, Network Slicing provides scalability and flexibility through various services such as Quality of Service (QoS) and Quality of Experience (QoE) to cover the network demands. For the QoS, a set of policies must be considered in real-time, accompanied by a group of functions and services to guarantee the end-user needs based on network demand. On the other hand, for the QoE, the service's performance needs to be improved to bring an efficient service to cover the demands of the end-user. The 3G Partnership Project (3GPP) defined the slice as a component of resources used to process a set of packets. These resources need to be flexible, which means the resources can be scaled up or down based on the demand. This survey discusses softwarization and virtualization techniques, considering how to implement the slices for future networks. Specifically, we discuss current advances concerning the functionality and architecture of the 5G network. Therefore, the paper critically evaluates recent research and systems related to mobility management as a service in real-time inter/intra slice control by considering the strengths and limitations of these contributions to identify the research gaps and possible research directions for emerging research and development opportunities. Moreover, we extend our review by considering the slice types and their numbers based on the 3GPP Technical Specification (3GPP TS). The study presented in this paper identifies open issues and research directions that reveal that mobility management at a service level with inter/intra slice management techniques has strong potential in future networks and requires further investigation from the research community to exploit its benefits fully.

show abstract

Section: F Resources and Policy Managementmentioning

confidence: 99%

Section: Traffic Classification For the Future Networkmentioning

confidence: 99%

A Survey of Mobility Management as a Service in Real-Time Inter/Intra Slice Control

et al. 2021

View full text Add to dashboard Cite

show abstract

“…The state of the art on the 5G use cases have been discussed with the QoS (Quality of Service) requirements like: latency, reliability, availability and throughput. In the paper [14], authors have developed deep learning neural network based deepslice model which uses network KPI to train the model to analyze the incoming traffic. The network failure scenario have also been discussed as master slice.…”

Section: Related Workmentioning

confidence: 99%

Towards ultra-latency using deep learning in 5G Network Slicing applying Approximate k-Nearest Neighbor Graph Construction

Gupta

Pannu

Misra

2021

Preprint

View full text Add to dashboard Cite

The 5G Network Slicing with SDN and NFV have expended to support new-verticals such as intelligent transport, industrial automation, remote healthcare. Network slice is intended as parameter configurations and a collection of logical network functions to support particular service requirements. The network slicing resource allocation and prediction in 5G networks is carried out using network Key Performance Indicators (KPIs) from the connection request made by the devices on joining the network. We explore derived features as the network non-KPI parameters using the k-Nearest Neighbor (kNN) graph construction. In this paper, we use kNN graph construction algorithms to augment the dataset with triangle count and cluster coecient properties for ecient and reliable network slice. We used deep learning neural network model to simulate our results with KPIs and KPIs with non-KPI parameters. Our novel approach found that at k=3 and k=4 of the kNN graph construction gives better results and overall accuracy is imroved around 29%.

show abstract

“…With the wide variety of applications and scenarios, it becomes more important to choose the better network parameters and VNF placement to improve the performance of each network slicing without affecting others. ML models fed with network information to choose these parameters and the placement of VNFs in the infrastructure can enhance the network slicing operations, as demonstrated in many works [18]- [20]. Not necessarily involving slicing and NFV placement, a brief review of related works in 5G testbeds is presented next.…”

Section: Related Work: Testbeds and Their Featuresmentioning

confidence: 99%

Testbed for 5G Connected Artificial Intelligence on Virtualized Networks

et al. 2020

View full text Add to dashboard Cite

DeepSlice: A Deep Learning Approach towards an Efficient and Reliable Network Slicing in 5G Networks

Cited by 140 publications

References 12 publications

A Survey of Mobility Management as a Service in Real-Time Inter/Intra Slice Control

A Survey of Mobility Management as a Service in Real-Time Inter/Intra Slice Control

Towards ultra-latency using deep learning in 5G Network Slicing applying Approximate k-Nearest Neighbor Graph Construction

Testbed for 5G Connected Artificial Intelligence on Virtualized Networks

Contact Info

Product

Resources

About