IoT supports a spectrum of applications, each of which has certain specific requirements. For instance, mission critical applications cannot tolerate delay in data transmission however simple monitoring applications are delay tolerant. The lifetime and performance of IoT sensor networks depend on the metric/constraint (ETX, Energy etc.) selected for routing path, data size and quality of service required. The selection of metric /constraint dependent Objective function (OF) in RPL provides a range of solutions for IoT applications. However, state-of-art solutions mainly focus on single metric/constraint resulting in poor performance of protocol. To understand the protocol behavior for different metrics (single and combined) a complete evaluation of RPL over important performance parameters is needed. Researchers have proposed several routing algorithms which are application specific and do not define a generic parent selection process. We require a structured algorithm for Minimum Rank Hysteresis Objective Function (MRHOF) applicable to majority of IoT applications. In this paper we have proposed a generalized algorithm for MRHOF along with routing path cost evaluation which defines the complete parent selection process. Further, comparative analysis of different RPL OFs has been done to identify suitable OF for enhanced RPL performance. Performance evaluation parameters have been extended to PDR, power consumption, hop count, average ETX, Rt metric and inter packet time, for different network size and link quality. Results are obtained using Cooja simulator of Contiki. RPL with combined metric provide 24% higher PDR, 28% lower power consumption and 39% lower inter-packet time as compared to RPL with single metric.

物联网支持一系列应用，每个应用都有特定的特定要求。例如，关键任务应用程序不能容忍数据传输中的延迟，但是简单的监视应用程序可以容忍延迟。物联网传感器网络的寿命和性能取决于为路由路径，数据大小和所需服务质量选择的指标/约束（ETX，能源等）。RPL中与度量/约束相关的目标函数（OF）的选择为IoT应用提供了一系列解决方案。但是，最新的解决方案主要集中在单一度量标准/约束上，从而导致协议性能较差。为了了解不同度量标准（单个和组合）的协议行为，需要对RPL进行重要性能参数的完整评估。研究人员提出了几种特定于应用的路由算法，它们没有定义通用的父代选择过程。我们需要适用于大多数物联网应用的最小秩滞后目标函数（MRHOF）的结构化算法。在本文中，我们提出了一种用于MRHOF的通用算法以及路由路径成本评估，该算法定义了完整的父代选择过程。此外，已经对不同的RPL OFs进行了比较分析，以确定适合于增强RPL性能的OF。针对不同的网络规模和链路质量，性能评估参数已扩展到PDR，功耗，跳数，平均ETX，Rt度量和数据包间时间。使用Contiki的Cooja模拟器获得结果。结合了指标的RPL可将PDR提高24％，