Edge-Based Intelligence for Real-Time Monitoring in Low-Connectivity Regions

Lukas Reinhardt

Edge-Based Intelligence for Real-Time Monitoring in Low-Connectivity Regions

Author(s)	Dr. Lukas Reinhardt
Country	United States
Abstract	Reliable real-time monitoring remains difficult in rural, remote, disaster-prone, and infrastructure-limited regions where network availability is intermittent, bandwidth is restricted, and continuous cloud access cannot be guaranteed. Conventional cloud-centric Internet of Things architectures frequently transmit raw sensor observations to centralized platforms before anomaly detection or decision generation. Such architectures can become vulnerable to communication delay, data-transfer overhead, service interruption, and dependence on backhaul availability. This study proposes an Edge-Based Intelligent Monitoring Framework for Low-Connectivity Regions (EIM-LCR) in which lightweight analytical models operate near the point of data generation and continue detecting critical events even when external connectivity is temporarily unavailable. The framework combines local sensor processing, lightweight inference, event prioritization, data compression, store-and-forward synchronization, adaptive communication, local alert generation, and delayed cloud aggregation. A design-science methodology was supported by contemporary edge-computing research and a reproducible scenario-based simulation. The evaluation modeled 8,000 monitoring events across four backhaul conditions: stable connectivity, intermittent connectivity, low connectivity, and severe connectivity loss. Cloud-only monitoring was compared with the proposed edge-intelligent architecture using detection latency, immediate anomaly-alert coverage, and transmitted data volume. Under stable connectivity, cloud-based detection required approximately 1.9 seconds on average, whereas local edge inference required approximately 0.24 seconds. As simulated backhaul availability declined to 10%, cloud means detection latency increased to 273.1 seconds and immediate anomaly-alert coverage fell to 12.5%. In contrast, local edge-detection latency remained approximately 0.24 seconds, while local anomaly-alert coverage remained above 92% across all connectivity conditions. The proposed architecture also reduced modeled backhaul data volume by approximately 93.3% through local filtering and selective synchronization. These results represent controlled simulation outcomes rather than measurements from a deployed rural infrastructure. The findings indicate that low-connectivity monitoring should be designed around local operational continuity rather than continuous cloud dependence. Edge-based intelligence can provide a practical foundation for environmental monitoring, agriculture, community infrastructure, public safety, energy systems, and other applications in which delayed connectivity should not prevent immediate local detection and response.
Keywords	edge intelligence; low-connectivity regions; real-time monitoring; edge computing; Internet of Things; rural connectivity; TinyML; intermittent networks; local inference; resilient monitoring
Field	Engineering
Published In	Volume 1, Issue 1, January-February 2020
Published On	2020-02-05

View / Download PDF File

About AJAMIR Fees & Payment Current Issue Publication Archive	Submit Research Paper Track Submission Status Publication Guidelines Publication Ethics Peer Review & Plagiarism	Join as a Reviewer Editors & Reviewers Reviewer Referral Program Get Reviewer Membership Certi.	Website/Journal Policies Usage Policy Content Policies Privacy Policy

Contact Us		+1-872-302-3851	editor@ajamir.com

American Journal of Advanced Multidisciplinary Innovation and Research

A Widely Indexed Open Access Peer Reviewed Multidisciplinary Bi-monthly Scholarly International Journal

Edge-Based Intelligence for Real-Time Monitoring in Low-Connectivity Regions

Share this