Orion

Foukas, Xenofon; Marina, Mahesh K.; Kontovasilis, Kimon

doi:10.1145/3117811.3117831

Cited by 135 publications

(32 citation statements)

References 43 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The degree of complexity is higher in RAN than in CN and TN, due to the difficulties in the segregation of radio resources. Moreover, virtualization in the RAN is still in its infancy, unlike at the CN level [49]. In this respect, in the 3GPP 5G RAN specifications, eight possible RAN virtual functions splitting options are considered, based on the Cloud-RAN (C-RAN) concept [50], where RAN functions are split between Remote Radio Units (RRUs) and Baseband Units (BBUs) hosted in the Next Generation Node B (gNB) and a BBU pool, respectively.…”

Section: ) Ran Slicingmentioning

confidence: 99%

Machine Learning in Network Slicing—A Survey

Phyu¹,

Naboulsi²,

Stanica

2023

IEEE Access

View full text Add to dashboard Cite

5G and beyond networks are expected to support a wide range of services, with highly diverse requirements. Yet, the traditional "one-size-fits-all" network architecture lacks the flexibility to accommodate these services. In this respect, network slicing has been introduced as a promising paradigm for 5G and beyond networks, supporting not only traditional mobile services, but also vertical industries services, with very heterogeneous requirements. Along with its benefits, the practical implementation of network slicing brings a lot of challenges. Thanks to the recent advances in machine learning (ML), some of these challenges have been addressed. In particular, the application of ML approaches is enabling the autonomous management of resources in the network slicing paradigm. Accordingly, this paper presents a comprehensive survey on contributions on ML in network slicing, identifying major categories and subcategories in the literature. Lessons learned are also presented and open research challenges are discussed, together with potential solutions.

show abstract

Section: ) Ran Slicingmentioning

confidence: 99%

Machine Learning in Network Slicing—A Survey

Phyu¹,

Naboulsi²,

Stanica

2023

IEEE Access

View full text Add to dashboard Cite

show abstract

“…However, despite making a step forward in the direction of the virtualization of the RAN, the proposal still lacks a slicing design. The same authors have also proposed Orion [16], which is a RAN slicing design running on the FlexRAN platform that guarantees the functional isolation among slices. Isolation of functions among slices is of paramount importance since it allows a slice-custom FS within a single shared eNB, i.e., different slices sharing the same physical node can be configured with different FSs.…”

Section: Related Workmentioning

confidence: 99%

SlicedRAN: Service-Aware Network Slicing Framework for 5G Radio Access Networks

Ojaghi

Adelantado

Antonopoulos

et al. 2022

IEEE Systems Journal

View full text Add to dashboard Cite

5G mobile networks are envisioned to substantiate new vertical services with diverse performance requirements. Slicing in the Radio Access Network (RAN) promises an efficient solution for these diversified needs of 5G networks, which foresees the separation of the Base Station (BS) functionality between the Central Unit (CU) and the distributed Remote Radio Heads (RRHs). In this paper, we formulate a Mixed Integer Programming (MIP) framework that maximizes the throughput by jointly selecting the optimal Functional Split (FS) and the routing path from a connected User Equipment (UE) to the CU, while satisfying the agreed Service Level Agreements (SLAs) of each service. Furthermore, we propose an effective heuristic, SlicedRAN, which creates isolated RAN slices premised on the service requirements connected through a Fronthaul/Backhaul (FH/BH) network and obtains near-optimal solutions in a short computing time compared to the MIP framework. Our results show that there is a trade-off between the architecture of the FH/BH network and the minimum SLA of each slice, which provides a solution to efficiently design a virtualized network infrastructure. According to the results, the SlicedRAN outperforms existing State-of-the-Art (SoA) up to 112% gain in throughput. Results are shown close to the optimal results, with a loss below 5%.

show abstract

“…Significant research has also been conducted in the field of programming abstractions for SDN. This includes both theoretical/conceptual [50,56,57] works as well as system research works [53][54][55]58,59]. In this section we aim to provide the reader with a brief introduction to the SDWN paradigm, with a particular focus on Control User Plane Separation (CUPS).…”

Section: B Software-defined Wireless Networkmentioning

confidence: 99%