Optimizing Message Delivery Cost in Preexisting, Disruption/Delay Tolerant Messaging Based LoRaWAN Networks for Disaster Communication.

Abstract:
The availability of communication channels and the internet is often taken for granted, but disasters like the 2023 Turkey-Syria earthquake show how quickly that can change. Emergency response solutions are then faced with a dynamic environment making it challenging to deliver messages between peers. The HofBox introduced an emergency communication system built on a LoRaWAN-based mesh network and delay tolerant networking to provide communication services in disaster situations.
In this thesis, I analyze the requirements and challenges for routing algorithms in the HofBox environment. Legal duty cycle limits imposed on the network traffic in LoRaWAN networks are identified as the main factor limiting the amount of messages that can be routed through the network.
A novel routing algorithm focused on improving duty cycle usage while keeping the delivery rate at flooding based approaches is proposed. To optimize the duty cycle usage, the proposed algorithm incorporates multiple strategies to warn about sub-optimal network topologies and to minimize the amount of data transmitted while still providing message delivery rates comparable to flooding-based routing.
To evaluate the algorithm’s performance, the proposed routing algorithm and the current flooding-based approach are implemented in software, to compare the time on air characteristics of both solutions. Finally, current limitations, future optimizations and findings of the evaluation will be discussed.

Betreuer: Franz Kuntke, M.Sc.
Prüfer: Prof. Dr. Dr. Christian Reuter