[Paper Review] CISER: An Amoebiasis inspired Model for Epidemic Message Propagation in DTN

C C Sobin, Snehanshu Saha+3 more|arXiv (Cornell University)|Aug 27, 2016

Opportunistic and Delay-Tolerant Networks13 references3 citations

TL;DR

This paper proposes CISER, a novel epidemic message propagation model for Delay Tolerant Networks (DTNs) inspired by amoebiasis disease dynamics, introducing exposed (E) and carrier (C) states to better model resource-constrained node behavior. The model improves delivery ratio by up to 49%, reduces delivery delay by up to 56%, and lowers overhead compared to the standard SIR model, especially in real-world traces.

ABSTRACT

Delay Tolerant Networks (DTNs) are sparse mobile networks, which experiences frequent disruptions in connectivity among nodes. Usually, DTN follows store-carry-and forward mechanism for message forwarding, in which a node store and carry the message until it finds an appropriate relay node to forward further in the network. So, The efficiency of DTN routing protocol relies on the intelligent selection of a relay node from a set of encountered nodes. Although there are plenty of DTN routing schemes proposed in the literature based on different strategies of relay selection, there are not many mathematical models proposed to study the behavior of message forwarding in DTN. In this paper, we have proposed a novel epidemic model, called as CISER model, for message propagation in DTN, based on Amoebiasis disease propagation in human population. The proposed CISER model is an extension of SIR epidemic model with additional states to represent the resource constrained behavior of nodes in DTN. Experimental results using both synthetic and real-world traces show that the proposed model improves the routing performance metrics, such as delivery ratio, overhead ratio and delivery delay compared to SIR model.

Motivation & Objective

• To address the lack of mathematical models for epidemic message propagation in DTNs that account for node resource constraints.
• To develop a more accurate epidemic model by extending the SIR framework with states reflecting real-world DTN node behavior.
• To evaluate the performance of the proposed model against existing SIR-based routing in terms of delivery ratio, delay, and overhead.
• To analyze the stability and equilibrium dynamics of message propagation under the CISER model.
• To validate the model using both synthetic and real-world mobility traces (Infocom and MIT Reality).

Proposed method

• The CISER model extends the classic SIR epidemic model by adding two new states: Exposed (E) for nodes holding messages but unable to forward due to resource limits, and Carrier (C) for nodes actively forwarding messages.
• The model draws biological inspiration from amoebiasis, where cysts (C) represent persistent infectious carriers and trophozoites (I) represent acute but non-infectious stages.
• Differential equations are used to model the transition rates between S (susceptible), E (exposed), I (infected), C (carrier), and R (recovered) states in the network.
• The model performs stability analysis and equilibrium point evaluation to understand long-term message propagation behavior.
• Simulations are conducted using synthetic DTN traces and real-world mobility data from Infocom and MIT Reality traces to evaluate routing performance.
• Performance metrics include delivery ratio, overhead ratio, and delivery delay, compared against the baseline SIR model.

Experimental results

Research questions

• RQ1 How does incorporating exposed and carrier states improve the modeling of message propagation in resource-constrained DTNs?
• RQ2 What is the impact of the CISER model on delivery ratio, delivery delay, and overhead compared to the SIR model in synthetic and real-world DTN scenarios?
• RQ3 Under what conditions does the CISER model achieve asymptotic stability in message propagation?
• RQ4 How do the endemic equilibrium and dynamic transitions in the CISER model reflect real-world DTN message forwarding behavior?
• RQ5 What are the key parameters that determine the spread and persistence of messages in the network under the CISER framework?

Key findings

• The CISER model achieves a 49% improvement in delivery ratio over the SIR model when evaluated using the Infocom real-world trace.
• In the MIT Reality trace, the CISER model reduces delivery delay by approximately 56% compared to the SIR model.
• The CISER model reduces delivery delay by up to 35% in synthetic trace simulations, demonstrating consistent performance gains.
• Overhead ratio is slightly lower in CISER compared to SIR, particularly in real-world traces where message generation is limited.
• The model demonstrates asymptotic stability in message propagation dynamics, confirming reliable long-term behavior.
• The endemic equilibrium analysis reveals that message propagation stabilizes under specific parameter regimes, indicating predictable network behavior.