Agora

Ding, Jian; Doost-Mohammady, Rahman; Kalia, Anuj; Lin, Zhouchen

doi:10.1145/3386367.3431296

Cited by 34 publications

(3 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The ACK/NACK is contained in the Physical Downlink Control Channel (PDCCH), located in the first symbols of the DL frame, while the rest symbols transmit the downlink data in the Physical Downlink Shared Channel (PDSCH). This introduces a hard deadline for the completion of the decoding tasks (that are by far the most computing expensive of the full protocol stack [38], [39], [15], [24]). In 5G-NR this deadline is configurable and let it be equal to J − 1 Transmission Time Intervals (TTIs).…”

Section: B Problem: Resource Scheduling In Vranmentioning

confidence: 99%

ATHENA: Machine Learning and Reasoning for Radio Resources Scheduling in vRAN Systems

Apostolakis,

Gramaglia,

Chatzieleftheriou

et al. 2024

IEEE J. Select. Areas Commun.

View full text Add to dashboard Cite

Next-generation mobile networks will rely on their autonomous operation. Virtual Network Functions empowered by Artificial Intelligence (AI) and Machine Learning (ML) can adapt to varying environments that encompass both network conditions and the cloud platform executing them. In this view, it becomes paramount to understand why AI/ML algorithms made a decision, to be able to reason upon those decisions and, eventually, take further decisions related to e.g., network orchestration. In this paper, we present ATHENA, an ML-based radio resource scheduler for virtualized Radio Access Network (RAN) system. Our real-software implementation shows that the proposed MLbased approach can outperform the baseline solution. We discuss how additional re-orchestration actions can be taken by analyzing our scheduling decisions and learning from the past.

show abstract

Section: B Problem: Resource Scheduling In Vranmentioning

confidence: 99%

ATHENA: Machine Learning and Reasoning for Radio Resources Scheduling in vRAN Systems

Apostolakis,

Gramaglia,

Chatzieleftheriou

et al. 2024

IEEE J. Select. Areas Commun.

View full text Add to dashboard Cite

show abstract

“…Next to open interfaces, open commercial-grade software needs to be developed. In [64], a real-time mMIMO base-band (BB) processing software is presented, which is published in open-source. M-Cube [65] is an open hardware and software for a mmWave mMIMO SDR, where the board schematics, real-time phased array controller code, interfaces, and examples are all available.…”

Section: E Open Testbedsmentioning

confidence: 99%

6G Radio Testbeds: Requirements, Trends, and Approaches

Callebaut,

Liu,

Eriksson

et al. 2024

IEEE Microwave

View full text Add to dashboard Cite

The proof of the pudding lies in eating, that is why 6G testbeds are essential in the progress towards the next generation of wireless networks. Theoretical research towards 6G wireless networks proposes advanced technologies to serve new applications and drastically improve the energy performance of the network. Testbeds are indispensable to validate these new technologies under more realistic conditions. This paper clarifies the requirements for 6G radio testbeds, reveals trends, and introduces approaches toward their development.The sixth-generation of mobile networks is expected to be introduced starting 2030. Research and development toward this novel generation of networks started several years ago in different regions of the world, e.g., [2]. A myriad of applications and novel use cases, ranging from augmented reality (AR) and mixed reality (MR) in professional and entertainment contexts to support autonomous systems and connecting a massive number of battery-less devices, are envisioned to be served by these networks [3]. The analysis of these use cases has led to a harmonized view of which functions and capabilities 6G networks should offer. Summarizing, the desired features for 6G can be classified into three main categories:1) Improving diverse wireless network services: delivering higher throughput and network capacity, ultra-reliable operation, imperceptible latency, and connectivity to many devices with an extremely low-energy budget. 2) Providing novel functionalities: precise positioning and sensing [4], and communication with energy-neutral devices requiring wireless-power transfer [5]. 3) Enhancing coverage: delivering more uniform and truly ubiquitous services.

show abstract

“…Moreover, Ope-nAirInterface (OAI) (and solutions based on OAI [59], [60]), despite perhaps being the most advanced open platform that supports 5G-NR and provides software implementations for both RAN and User Equipment (UE), it is still not capable of supporting MU-MIMO processing with a large number of concurrently transmitted information streams. Earlier notable MIMO PHY realizations, including BigStation [61] and Agora [62] explore larger MIMO dimensions, exploiting aggressive pipelining and data parallelization techniques. However, both projects rely on linear processing techniques (MMSE) without accounting for the consequential practical power implications both on the radio side or the processor side.…”

Section: Related Workmentioning

confidence: 99%

Power Efficient and Ultra Dense Open-RAN Vehicular Networks With Non-Linear Processing

Katsaros,

Nikitopoulos

2024

IEEE Access

View full text Add to dashboard Cite

Transitioning to more intelligent, autonomous transportation systems necessitates network infrastructure capable of accommodating both substantial uplink traffic and massive vehicle connectivity. Current approaches addressing these throughput and connectivity requirements rely on the utilization of the multiple input, multiple output (MIMO) technology. However, When traditional linear detection/precoding processing methods are adopted, they require the deployment of an extensive number of co-located, accesspoint antennas to support a comparatively much smaller number of data streams. Such a setup significantly increases the power consumption on the radio side, raising substantial concerns about the operational costs and sustainability of such deployments, particularly in densely deployed scenarios, across extensive road networks. Addressing these concerns, this work proposes an Open Radio Access Network (Open-RAN) deployment that incorporates a Massively Parallelizable, Nonlinear (MPNL) MIMO processing framework and assesses, for the first time, its impact on the power consumption and vehicular connectivity in various Vehicle-to-Infrastructure (V2I) and Network (V2N) scenarios. We show that flexible, Open-RAN physical layer deployments, incorporating MPNL, emerge as a critical power efficiency enabler, especially when flexibly activating/deactivating employed RF elements. Our field-programmable gate array (FPGA) based evaluation of MPNL, reveals that it can lead to significant power savings on the radio side, by eliminating the need for a "massive" number of base station antennas and radio frequency (RF) chains. Additionally, our findings show substantial connectivity gains, exceeding 400%, in terms of concurrently transmitting vehicles compared to traditional processing approaches, without significantly affecting the access point power consumption budgets, thereby catalyzing the evolution towards more intelligent, fully autonomous, and sustainable transportation systems. INDEX TERMSPower Efficiency, C-V2X, Open-RAN, Massive MIMO, Non-linear Processing I. INTRODUCTION T HE standardization of 5G New Radio (5G-NR) based Cellular Vehicle-to-Everything (C-V2X) [1] technology marks a significant milestone in the transformation of our transportation systems. It promises to improve road safety, traffic efficiency, and user convenience by enabling seamless communication between vehicles, infrastructure, pedestrians, and network services, laying the foundation for autonomous and cooperative driving.However, the transition to even more intelligent transportation systems requires that vehicles, infrastructure, and pedestrians continuously transmit substantial volumes of real-time sensor data with high reliability and low latency. For instance, it is projected that a single autonomous vehicle will generate an astounding 1.4 to 40TB of data per hour [2], [3]. When comparing this with the average smartphone data usage in 2021 [4], the projected data output for autonomous vehicles exceeds the average user consumption by over 70...

show abstract

Agora

Cited by 34 publications

References 29 publications

ATHENA: Machine Learning and Reasoning for Radio Resources Scheduling in vRAN Systems

ATHENA: Machine Learning and Reasoning for Radio Resources Scheduling in vRAN Systems

6G Radio Testbeds: Requirements, Trends, and Approaches

Power Efficient and Ultra Dense Open-RAN Vehicular Networks With Non-Linear Processing

Contact Info

Product

Resources

About