Abstract

In a real network deployment, the diverse sensor applications generate a heterogeneous traffic pattern which may include basic sensing measurements such as temperature readings or high-volume multimedia traffic. In a heterogeneous traffic network, the two standardized objective functions (OFs), i.e., objective function zero (OF0) and the Minimum Rank with Hysteresis Objective Function (MRHOF) for routing protocol for Low-Power and Lossy Networks (RPL) perform poor routing decisions by selecting an already congested parent node and cause more re-transmissions across the network. Therefore, careful consideration is required in designing a new OF for heterogeneous traffic scenarios. In this study, we examine the RPL protocol under a heterogeneous traffic pattern and proposed a new protocol based on queue and workload-based condition (QWL-RPL). The aim of the proposed protocol is to achieve a reliable path with better overall performance. The proposed OF model considers the link workload in addition to mapping the congestion status of the node using the packet queue. We implement the proposed routing model in the Contiki operating system (OS) Cooja environment to compare with the existing technique. The simulation results show that QWL-RPL can improve the performance of a heterogeneous traffic network as compared with both OF0 and MRHOF, specifically in terms of the amount of overhead, packets reception ratio (PRR), average delay, and jitter. Final results indicate that on average, there is a 5%–30% improvement in PRR, 25%–45% reduction in overheads, 12%–30% reduction in average delay, and 20%–40% reduction in jitter.

中文翻译：

---

异构流量网络的低功耗有损网络的路由协议

摘要

在实际的网络部署中，各种传感器应用程序会生成异构流量模式，其中可能包括基本的传感测量值，例如温度读数或大量多媒体流量。在异构流量网络中，用于低功率和有损网络（RPL）的路由协议的两个标准化目标函数（OF），即目标函数零（OF0）和带有滞后的最小秩目标函数（MRHOF），执行的路由较差通过选择一个已经拥塞的父节点来做出决定，并引起更多的网络重传。因此，在为异构流量方案设计新的OF时需要仔细考虑。在这项研究中，我们检查了异构流量模式下的RPL协议，并提出了一种基于队列和基于工作负载的条件（QWL-RPL）的新协议。所提出的协议的目的是获得具有更好的整体性能的可靠路径。提出的OF模型除了使用数据包队列映射节点的拥塞状态外，还考虑了链路工作量。我们在Contiki操作系统（OS）Cooja环境中实现了建议的路由模型，以与现有技术进行比较。仿真结果表明，与OF0和MRHOF相比，QWL-RPL可以提高异构流量网络的性能，特别是在开销，数据包接收率（PRR），平均延迟和抖动方面。最终结果表明，平均而言，PRR改善了5％–30％，开销减少了25％–45％，平均延迟减少了12％–30％，抖动减少了20％–40％。