A Comprehensive System Architecture using Field Programmable Gate Arrays Technology, Dijkstra's Algorithm, and Edge Computing for Emergency Response in Smart Cities

TL;DR

This paper tackles emergency response challenges in smart cities under infrastructure constraints, particularly in Chad, where cloud-centric solutions falter due to latency and connectivity. It proposes a comprehensive architecture that fuses FPGA-accelerated Dijkstra computations with Edge Computing, distributed within a Surveillance and Alerting System (SAS), Edge Server Network (ESN), Intervention Order Processing System (IOPS), and Intervention Team Guidance System (ITGS). Key contributions include parallel Dijkstra execution on FPGA across a cluster of edge servers, a caching mechanism to reuse computed routes, and decentralized decision-making that remains operational with intermittent connectivity. The work highlights potential improvements in response times and resource management, while acknowledging the theoretical nature and deployment challenges of such a complex, hardware-software system. Overall, the architecture aims to enhance robustness, timeliness, and adaptability of emergency response in resource-constrained smart-city contexts, paving the way for future prototypes and real-world validation.

Abstract

Efficient emergency response systems are crucial for smart cities. But their implementation is highly challenging, particularly in regions like Chad where infrastructural constraints are prevalent. The urgency for optimized response times and resource allocation in emergency scenarios is magnified in these contexts, yet existing solutions often assume robust infrastructure and uninterrupted connectivity, which is not always available. Most of the time, they are based on system architectures pre-designed for other purposes. This paper addresses these critical challenges by proposing a comprehensive system architecture that integrates Field Programmable Gate Arrays (FPGAs), Dijkstra's algorithm, and Edge Computing. The objective is to enhance emergency response through accelerated route planning and resource allocation, addressing gaps left by traditional cloud-based systems. Methodologically, key characteristics of the desired system are identified, then its components are described and their integration is explained; the system leverages FPGA-based computations and a distributed implementation of Dijkstra's algorithm to compute the shortest paths rapidly, while Edge Computing ensures decentralized and resilient processing. A theoretical analysis highlights promising improvements in response times and resource management. The proposed system architecture not only enhances emergency response efficiency but is also adaptable to infrastructural constraints of Chadian-like environments.

Figures (5)

• Figure 1: Overview of the proposed architecture.
• Figure 2: Overview of the edge network.
• Figure 3: The physical architecture of an edge server in the system.
• Figure 4: Splitting adjacency matrix to sub-matrices.
• Figure 5: Parallelism of Dijkstra algorithm using GPE.