Data-driven Predictive Latency for 5G: A Theoretical and Experimental Analysis Using Network Measurements
Authors:
Marco Skocaj,
Francesca Conserva,
Nicol Sarcone Grande,
Andrea Orsi,
Davide Micheli,
Giorgio Ghinamo,
Simone Bizzarri,
Roberto Verdone
Abstract:
The advent of novel 5G services and applications with binding latency requirements and guaranteed Quality of Service (QoS) hastened the need to incorporate autonomous and proactive decision-making in network management procedures. The objective of our study is to provide a thorough analysis of predictive latency within 5G networks by utilizing real-world network data that is accessible to mobile n…
▽ More
The advent of novel 5G services and applications with binding latency requirements and guaranteed Quality of Service (QoS) hastened the need to incorporate autonomous and proactive decision-making in network management procedures. The objective of our study is to provide a thorough analysis of predictive latency within 5G networks by utilizing real-world network data that is accessible to mobile network operators (MNOs). In particular, (i) we present an analytical formulation of the user-plane latency as a Hypoexponential distribution, which is validated by means of a comparative analysis with empirical measurements, and (ii) we conduct experimental results of probabilistic regression, anomaly detection, and predictive forecasting leveraging on emerging domains in Machine Learning (ML), such as Bayesian Learning (BL) and Machine Learning on Graphs (GML). We test our predictive framework using data gathered from scenarios of vehicular mobility, dense-urban traffic, and social gathering events. Our results provide valuable insights into the efficacy of predictive algorithms in practical applications.
△ Less
Submitted 31 August, 2023; v1 submitted 5 July, 2023;
originally announced July 2023.
A New Scheduler for URLLC in 5G NR IIoT Networks with Spatio-Temporal Traffic Correlations
Authors:
Sara Cavallero,
Nicole Sarcone Grande,
Francesco Pase,
Marco Giordani,
Joseph Eichinger,
Roberto Verdone,
Michele Zorzi
Abstract:
This paper explores the issue of enabling Ultra-Reliable Low-Latency Communications (URLLC) in view of the spatio-temporal correlations that characterize real 5th generation (5G) Industrial Internet of Things (IIoT) networks. In this context, we consider a common Standalone Non-Public Network (SNPN) architecture as promoted by the 5G Alliance for Connected Industries and Automation (5G-ACIA), and…
▽ More
This paper explores the issue of enabling Ultra-Reliable Low-Latency Communications (URLLC) in view of the spatio-temporal correlations that characterize real 5th generation (5G) Industrial Internet of Things (IIoT) networks. In this context, we consider a common Standalone Non-Public Network (SNPN) architecture as promoted by the 5G Alliance for Connected Industries and Automation (5G-ACIA), and propose a new variant of the 5G NR semi-persistent scheduler (SPS) to deal with uplink traffic correlations. A benchmark solution with a "smart" scheduler (SSPS) is compared with a more realistic adaptive approach (ASPS) that requires the scheduler to estimate some unknown network parameters. We demonstrate via simulations that the 1-ms latency requirement for URLLC is fulfilled in both solutions, at the expense of some complexity introduced in the management of the traffic. Finally, we provide numerical guidelines to dimension IIoT networks as a function of the use case, the number of machines in the factory, and considering both periodic and aperiodic traffic.
△ Less
Submitted 24 February, 2023;
originally announced February 2023.
