According to the World Health Organization (WHO), 1.3 million people die every year as a result of road traffic accidents caused by human errors. While more severe traffic regulations and a safer road infrastructure design would substantially reduce the number of casualties, Vehicle-to-Everything (V2X) communications will play a crucial role in improving road safety and transportation efficiency. To this end, the Third Generation Partnership Project (3GPP) has introduced its first V2X technology, known as LTE-V2X, in Release 14 specifications. LTE-V2X has been designed to support basic safety-related applications, such as the dissemination of awareness messages in out-of-coverage scenarios, where vehicles directly communicate without the support of the cellular infrastructure. In recent years, the advent of more sophisticated V2X use cases that rely on cooperative perception and maneuvering has prompted the development of the 5G New Radio (NR)-V2X technology. Standardized by 3GPP in Release 16, NR-V2X is characterized by a new physical (PHY) layer design and new Medium Access Control (MAC) features expected to guarantee an improved robustness and flexibility. System-level simulations are instrumental in assessing the large scale performance of LTE-V2X and NR-V2X networks, as real-world tests and measurements are often impracticable due to the limited availability and the high cost of hardware prototypes. This thesis introduces MoReV2X, an open-source ns-3 module for the simulation of LTE-V2X and NR-V2X communications. MoReV2X adheres to 3GPP specifications and features an accurate PHY layer abstraction model based on Block Error Rate (BLER) curves obtained through a detailed link-level analysis. The MoReV2X simulator has been employed to thoroughly investigate the behavior of LTE-V2X and NR-V2X from different perspectives. To begin with, the coexistence of periodic and aperiodic traffic in LTE-V2X has been studied, focusing on the limitations which characterize its distributed resource allocation strategy. Then, the impact of aperiodic traffic dissemination on system performance has been further analyzed in the NR-V2X domain, concentrating on the effectiveness of the new MAC features introduced in Release 16. As simulation results revealed that LTE-V2X and NR-V2X are not able to effectively disseminate aperiodic traffic, a novel Artificial Intelligence (AI)-based strategy to broadcast awareness messages has been put forth. According to it, each vehicle forecasts its message generation times and optimizes the MAC layer configuration. Exploiting AI predictions, the proposed approach outperforms legacy solutions in all respects. Besides simulation studies, this thesis analyzes the results of an extensive measurement campaign aimed at investigating the generation of awareness messages in a real-world context. Field tests have been performed with connected Vulnerable Road Users (VRUs) and cars in the urban, suburban, and highway scenarios. As the findings indicate that many messages from VRUs were generated under non-relevant circumstances, this study proposes an adjustment to reduce their dissemination frequency without missing relevant information about the VRU movements. Last, this thesis presents a novel analytical approach to evaluate the outage probability that Power-Domain (PD) Non-Orthogonal Multiple Access (NOMA) achieves on the uplink. The fundamental limits on system performance are analytically assessed for an arbitrary number of simultaneously transmitting vehicles, and both the case of Rayleigh and lognormal-shadowed Rayleigh fading are examined. The obtained closed-form expressions disclose the potential of PD-NOMA in beyond 5G V2X communications.

Secondo l'OMS, 1.3 milioni di persone muoiono ogni anno in incidenti stradali. Sebbene norme più severe contribuirebbero a diminuire il numero di vittime, le comunicazioni Vehicle-to-Everything (V2X) avranno un ruolo fondamentale nel garantire una maggiore sicurezza stradale. A questo proposito, il Third Generation Partnership Project (3GPP) ha definito la tecnologia LTE-V2X all'interno della Release 14. La tecnologia LTE-V2X è stata progettata per supportare applicazioni di tipo safety, come la trasmissione di semplici awareness messages in scenari di out-of-coverage, dove i veicoli sono in grado di comunicare autonomamente senza il supporto dell'infrastruttura cellulare. Recentemente, l'avvento di applicazioni V2X più sofisticate che prevedono la guida cooperativa ed autonoma, ha reso necessario lo sviluppo della nuova tecnologia 5G New Radio (NR)-V2X. Standardizzata dal 3GPP all'interno della Release 16, la tecnologia NR-V2X è caratterizzata da un nuovo livello fisico (PHY) e da nuove funzionalità di livello MAC pensate per garantire una maggiore robustezza e flessibilità. Le simulazioni hanno un ruolo fondamentale nell'analisi su larga scala delle reti LTE-V2X e NR-V2X. Infatti, l'analisi sperimentale è spesso impossibile a causa dell'elevato costo e della scarsa disponibilità di prototipi hardware. Questa tesi presenta MoReV2X, un modulo ns-3 open-source per la simulazione di comunicazioni LTE-V2X e NR-V2X. Il funzionamento di MoReV2X aderisce alle specifiche 3GPP e, a livello PHY, sfrutta un modello di astrazione basato su curve di Block Error Rate (BLER) ottenute attraverso un'accurata analisi di tipo link-level. Il simulatore MoReV2X è stato utilizzato per analizzare esaustivamente le tecnologie LTE-V2X e NR-V2X. Innanzitutto, è stata studiata la coesistenza di traffico periodico e aperiodico nel contesto LTE-V2X, evidenziando le limitazioni che ne caratterizzano la strategia di accesso. Dopodiché, il simulatore è stato utilizzato per valutare l'impatto del traffico aperiodico sulla tecnologia NR-V2X, ponendo particolare enfasi sulle funzionalità di livello MAC introdotte nella Release 16. Poiché i risultati hanno dimostrato che le tecnologie LTE-V2X e NR-V2X non sono in grado di servire adeguatamente sorgenti di traffico aperiodico, è stata sviluppata una nuova strategia di accesso basata su Intelligenza Artificiale (IA) per la distribuzione di awareness messages. Questa soluzione consiste nel prevedere i tempi di generazione dei messaggi e di sfruttare tale informazione per configurare opportunamente il livello MAC. Sfruttando le previsioni dell'IA, la strategia proposta è in grado di battere le soluzioni esistenti sotto tutti i punti di vista. In aggiunta agli studi di tipo simulativo, questa tesi presenta anche i risultati di una campagna sperimentale destinata ad analizzare la generazione di awareness messages. I test sperimentali sono stati condotti con macchine e Vulnerable Road Users (VRUs) connessi, considerando gli scenari urbano, suburbano e autostradale. Poiché i risultati hanno rivelato un'eccessiva generazione di messaggi da parte dei VRU, è stata proposta una nuova soluzione per ridurre la frequenza di disseminazione di tali messaggi che allo stesso tempo preservasse le informazioni più significative riguardanti i VRU. Infine, questa tesi presenta un nuovo approccio analitico per la caratterizzazione del Power Domain (PD) Non-Orthogonal Multiple Access (NOMA) nella direzione di uplink. Le performance del PD-NOMA sono state esplorate considerando un numero arbitrario di veicoli simultaneamente attivi, analizzando sia il caso di Rayleigh fading che di lognormal-shadowed Raylegih fading. Le espressioni in forma chiusa presentate in questo studio rivelano il potenziale del PD-NOMA nelle comunicazione veicolari 6G.

Connettività Veicolare nei Sistemi 5G e 6G / Luca Lusvarghi , 2023 Mar 08. 35. ciclo, Anno Accademico 2021/2022.

Connettività Veicolare nei Sistemi 5G e 6G

LUSVARGHI, LUCA
2023

Abstract

According to the World Health Organization (WHO), 1.3 million people die every year as a result of road traffic accidents caused by human errors. While more severe traffic regulations and a safer road infrastructure design would substantially reduce the number of casualties, Vehicle-to-Everything (V2X) communications will play a crucial role in improving road safety and transportation efficiency. To this end, the Third Generation Partnership Project (3GPP) has introduced its first V2X technology, known as LTE-V2X, in Release 14 specifications. LTE-V2X has been designed to support basic safety-related applications, such as the dissemination of awareness messages in out-of-coverage scenarios, where vehicles directly communicate without the support of the cellular infrastructure. In recent years, the advent of more sophisticated V2X use cases that rely on cooperative perception and maneuvering has prompted the development of the 5G New Radio (NR)-V2X technology. Standardized by 3GPP in Release 16, NR-V2X is characterized by a new physical (PHY) layer design and new Medium Access Control (MAC) features expected to guarantee an improved robustness and flexibility. System-level simulations are instrumental in assessing the large scale performance of LTE-V2X and NR-V2X networks, as real-world tests and measurements are often impracticable due to the limited availability and the high cost of hardware prototypes. This thesis introduces MoReV2X, an open-source ns-3 module for the simulation of LTE-V2X and NR-V2X communications. MoReV2X adheres to 3GPP specifications and features an accurate PHY layer abstraction model based on Block Error Rate (BLER) curves obtained through a detailed link-level analysis. The MoReV2X simulator has been employed to thoroughly investigate the behavior of LTE-V2X and NR-V2X from different perspectives. To begin with, the coexistence of periodic and aperiodic traffic in LTE-V2X has been studied, focusing on the limitations which characterize its distributed resource allocation strategy. Then, the impact of aperiodic traffic dissemination on system performance has been further analyzed in the NR-V2X domain, concentrating on the effectiveness of the new MAC features introduced in Release 16. As simulation results revealed that LTE-V2X and NR-V2X are not able to effectively disseminate aperiodic traffic, a novel Artificial Intelligence (AI)-based strategy to broadcast awareness messages has been put forth. According to it, each vehicle forecasts its message generation times and optimizes the MAC layer configuration. Exploiting AI predictions, the proposed approach outperforms legacy solutions in all respects. Besides simulation studies, this thesis analyzes the results of an extensive measurement campaign aimed at investigating the generation of awareness messages in a real-world context. Field tests have been performed with connected Vulnerable Road Users (VRUs) and cars in the urban, suburban, and highway scenarios. As the findings indicate that many messages from VRUs were generated under non-relevant circumstances, this study proposes an adjustment to reduce their dissemination frequency without missing relevant information about the VRU movements. Last, this thesis presents a novel analytical approach to evaluate the outage probability that Power-Domain (PD) Non-Orthogonal Multiple Access (NOMA) achieves on the uplink. The fundamental limits on system performance are analytically assessed for an arbitrary number of simultaneously transmitting vehicles, and both the case of Rayleigh and lognormal-shadowed Rayleigh fading are examined. The obtained closed-form expressions disclose the potential of PD-NOMA in beyond 5G V2X communications.
Vehicular Connectivity in 5G and Beyond
8-mar-2023
MERANI, Maria Luisa
File in questo prodotto:
File Dimensione Formato  
PhD_Thesis_Lusvarghi_Final.pdf

embargo fino al 07/03/2026

Descrizione: Tesi definitiva Lusvarghi Luca
Tipologia: Tesi di dottorato
Dimensione 36.78 MB
Formato Adobe PDF
36.78 MB Adobe PDF   Visualizza/Apri   Richiedi una copia
Pubblicazioni consigliate

Licenza Creative Commons
I metadati presenti in IRIS UNIMORE sono rilasciati con licenza Creative Commons CC0 1.0 Universal, mentre i file delle pubblicazioni sono rilasciati con licenza Attribuzione 4.0 Internazionale (CC BY 4.0), salvo diversa indicazione.
In caso di violazione di copyright, contattare Supporto Iris

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11380/1300328
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact