Machine-Learning based Traffic Forecasting for Resource Management in C-RAN

Guerra-Gómez, Rolando; Boqué, Sílvia Ruiz; García-Lozano, Mario; Bonafe, Joan Olmos

doi:10.1109/eucnc48522.2020.9200958

Cited by 9 publications

(2 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…c) Network Resource Management: ML enables networks to become self-configurable, convergent, self-reliant, intelligent, and robust by optimizing resource utilization through innovative resource-sharing and scheduling algorithms. In [57], the underutilization of available network resources due to limited computational capacity in cloud radio access networks (C-RANs) was addressed by integrating C-RAN with SDN and network function virtualization (NFV), resulting in a selfoptimized network. The study compared SVM, time-delay neural network, and LSTM for predicting performanceAnother in [58] emphasized the importance of efficient resource management (RM) in cloud computing, as over-provisioning and provisioning can increase costs for service providers and cause violations of service level agreements.…”

Section: ) ML For Ad In Wireless Communication Networkmentioning

confidence: 99%

Qualitative Survey on Artificial Intelligence Integrated Blockchain Approach for 6G and Beyond

Pathak,

Pandya,

Bhatia

et al. 2023

IEEE Access

View full text Add to dashboard Cite

Utilizing the 0.1 to 10 THz spectrum in the next-generation wireless communication networks holds potential for futuristic applications. However, managing resources to accommodate numerous devices raises privacy and security concerns. Further, technology proliferation entwines devices, infrastructure complexity, and resources. Indeed, the transition from 5G (fifth-generation) to 6G (sixth-generation) signifies a progression towards high-speed data rates, minimal latency, and seamless integration of artificial intelligence, enabling ground-breaking applications and services. However, it complicates network management, privacy, resource allocation, and data processing. Notably, integrating Blockchain Technology (BCT) and Machine Learning (ML) is a promising solution, enhancing security, decentralization, trust in ML decisions, and efficient data sharing. This survey thoroughly reviews the integrated ML and BCT, showcasing their collaborative enhancement of network security, decentralization, trust in ML decisions, immutability, and efficient model sharing. Furthermore, we also delve into various distinctive topics, such as BCT-enabled spectrum refarming, rate splitting multiple access, 6G radar-based communication, reconfigurable intelligent surfaces, visible light communication, and integrated sensing and communication. Moreover, it also explores the integration of ML and BCT in novel 6G communication technologies, including molecular, holographic, and semantic communication. Finally, critical open issues, challenges, solutions, and futuristic scope are identified for forthcoming researchers. INDEX TERMS Terahertz (THz) Communication, Artificial Intelligence (AI), Machine Learning (ML), Blockchain Technology (BCT), and Resource Allocation (RA).

show abstract

Section: ) ML For Ad In Wireless Communication Networkmentioning

confidence: 99%

Qualitative Survey on Artificial Intelligence Integrated Blockchain Approach for 6G and Beyond

Pathak,

Pandya,

Bhatia

et al. 2023

IEEE Access

View full text Add to dashboard Cite

show abstract

“…This approach is based on a spatiotemporal model that seeks a corelation between the incoming traffic and the physical location and time of the UE. Additionally, LSTM, Time-Delay Neural Networks and Support Vector Machine models are used in [9] to forecast the needed resources of a virtualized base station with the RAN parameters as input of the models.…”

Section: Related Work a Slicing Resource Predictionmentioning

confidence: 99%

Design of AI-based Resource Forecasting Methods for Network Slicing

Camargo

Coronado

Gómez

et al. 2022

2022 International Wireless Communications and Mobile Computing (IWCMC)

View full text Add to dashboard Cite

With the forthcoming of 5G networks, the underlying infrastructure needs to support a higher number of heterogeneous services with different QoS needs than ever. For that reason, 5G inherently provides a way to allocate these services over the same infrastructure through the concept of Network Slicing. However, to maximize revenue and reduce operational costs, a method to proactively adapt the resources assigned to each slice becomes imperative. For that reason, this work presents two Machine Learning (ML) models, leveraging Long-Short Term Memory (LSTM) and Random Forest algorithms, to forecast the throughput of each slice and adapt accordingly the amount of resources needed. The models are evaluated using NS-3, which has been integrated with the ML models through a shared memory framework. This enables a closed loop in which the predictions of the models can be used at run time to introduce changes in the network. Consequently, it makes it able to cope with the forecasted requirements, eliminating the need for off-line training and resembling better a real-life scenario. The evaluation performed shows the ability of the models to predict the slices' throughput under various settings and proves that Random Forest provides up to 26% better results than LSTM.

show abstract