Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Delay Analysis of Random Network Coding Enabled ad-hoc M-to-N broadcast network

Zhaohong Lu, Haibo Zeng

TL;DR

This paper addresses delay analysis for dynamic ad-hoc networks using random network coding in an $M$-to-$N$ broadcast setting. It introduces an equivalent-route construction to bound propagation delay and develops a queueing-delay approximation that combines a shortest-path backbone with detour refinements, formalized through $\mathcal{R}_{s_id_j}$ and queuing models. Key contributions include a fixed equivalent route yielding a propagation-delay lower bound, a distinction between same-message and multi-message traffic with a detour-based improvement, and validation via simulations across multiple topologies showing close agreement with observed performance. The results provide practical, provable delay guarantees for multi-source multicast in evolving networks, informing the design of delay-sensitive, broadcast-augmented systems.

Abstract

In this paper, we analyze the delay performance of an ad-hoc dynamic network where random network coding and broadcast are used in combination to distribute the messages. The analysis is comprehensive for that we consider M-to-N broadcast instead of 1-to-N, which allows both different messages and same messages to be transmitted by several sources at the same time. Although the routes between source-destination pairs are subject to change when some nodes have large backlogs, we derive fixed equivalent routes to provide a upper bound of delay. For some special cases, an detour method is also provided to increase the estimation accuracy. Different network topologies are tested in numeric simulation. The results demonstrate the accuracy of our delay performance approximation.

Delay Analysis of Random Network Coding Enabled ad-hoc M-to-N broadcast network

TL;DR

This paper addresses delay analysis for dynamic ad-hoc networks using random network coding in an -to- broadcast setting. It introduces an equivalent-route construction to bound propagation delay and develops a queueing-delay approximation that combines a shortest-path backbone with detour refinements, formalized through and queuing models. Key contributions include a fixed equivalent route yielding a propagation-delay lower bound, a distinction between same-message and multi-message traffic with a detour-based improvement, and validation via simulations across multiple topologies showing close agreement with observed performance. The results provide practical, provable delay guarantees for multi-source multicast in evolving networks, informing the design of delay-sensitive, broadcast-augmented systems.

Abstract

In this paper, we analyze the delay performance of an ad-hoc dynamic network where random network coding and broadcast are used in combination to distribute the messages. The analysis is comprehensive for that we consider M-to-N broadcast instead of 1-to-N, which allows both different messages and same messages to be transmitted by several sources at the same time. Although the routes between source-destination pairs are subject to change when some nodes have large backlogs, we derive fixed equivalent routes to provide a upper bound of delay. For some special cases, an detour method is also provided to increase the estimation accuracy. Different network topologies are tested in numeric simulation. The results demonstrate the accuracy of our delay performance approximation.

Paper Structure

This paper contains 12 sections, 1 theorem, 8 equations, 4 figures, 2 algorithms.

Table of Contents

  1. Introduction
  2. System Model and Preliminaries
  3. System Model
  4. Network coding
  5. Message generation
  6. Broadcast with discards
  7. Approximation Analysis of Queuing Delay
  8. Packets from the same message
  9. Shortest path method
  10. Improvement of shortest path method
  11. Simulation Evaluation
  12. Conclusion

Key Result

Theorem 3.1

Each packet transmitted using Algorithm alg1 has the same propagation delay performance as the Dijkstra shortest path algorithm.

Figures (4)

  • Figure 1: example case
  • Figure 2: Network topology with (a) 30 nodes (b) 100 nodes.
  • Figure 3: Average packet delay versus sum of arrival rate with (a) n = 30 (b) n = 100.
  • Figure 4: Average packet delay of ($s_1$,$d_1$) transmission pair versus sum of arrival rate of the special case

Theorems & Definitions (4)

  • Theorem 3.1
  • proof
  • Definition 1
  • Definition 2