Serviços Determinísticos em Infraestruturas de Cidades Inteligentes

The integration of the proposed techniques on a smart city scale, in such heterogeneous and highly mobile scenarios, has not, to the best of our knowledge, been comprehensively undertaken. This integration will introduce a new set of challenges, typically absent within TSN domains. This unexplored territory holds significant interest, in distinct use cases already explored by us (Lopes23). Thus, the project will address the following objectives: #O1 Heterogeneous Communication Networks with High Mobility: Smart cities are characterized by diverse technologies. Introducing deterministic flows can be complex and, in some instances, not feasible. Our approach involves exploring the utilization of centralized control through SDN, incorporating the Central Network Controller (CNC), and integrating a Centralized User Configuration (CUC). #O2 Time Awareness on Services and Edge: Current edge architectures seldom provide time guarantees. Supporting time-sensitive services will require service differentiation to prioritize such services and preempt non-critical ones. SmartCityDet aims at building a time-aware orchestration framework that is lightweight and integrates fine-tuned execution environments. #O3 Soft-Deterministic Wireless in Vehicular Communications: Wireless technologies already provide some traffic differentiation, e.g., WMM extensions to WiFi. However, this is not enough for time-critical services. Some approaches attempted at controlling the traffic triggering for tight real-time behavior, such as V-FTT. Recent advances, e.g., WiFi 7, are already TSN-capable. SmartCityDet aims at enhancing these mechanisms for dynamic real-time bandwidth reservations in vehicular environments. #O4 Dynamic Flow Classification: In real-time communication, flows are typically described statically. However, the number, heterogeneity and dynamism of flows found at city scale can be rather high. Identifying their timing properties online and classifying them into smaller classes (e.g., PCP, DSCP), mapped onto the networking infrastructure is key to support real-time services. SmartCityDet aims at developing a dynamic approach that classifies flows on-the-fly guaranteeing their time requirements. #O5 Dynamic Flow and Resource Reservation: TSN enhances the data plane with traffic control based on preemption, queue priority, and traffic policy. In a city-wide area, traffic will be rather dynamic, stemming from dynamic flow and resource reservation. These reservations can be requested a priori by the service itself or controlled by the Central Network Controller (CNC), leveraging its global perspective. #O6 End-to-End Transport upon Deterministic Layer: The transport layer ensures end-to-end guarantees. Introducing deterministic layers at the lower layers will impact current transport protocols. The SmartCityDet project will evaluate this impact and compare it with newer protocols that use Network Coding techniques, aiming to balance end-to-end delay with throughput. #O7 Deployment of a Large Deterministic Network: TSNs are usually deployed on small-scale and/or controlled environments with predictable flows. Large networks present additional challenges arising from long propagation delays, limited synchronization, dynamic traffic, and per-class queueing schemes. SmartCityDet aims at leveraging deterministic devices at the city scale to provide reservation and shaping mechanisms suitable to large networks.