[Paper Review] Probability Prediction based Reliable Opportunistic (PRO) Routing Algorithm for VANETs

Ning Li, José-Fernán Martí­nez-Ortega+2 more|arXiv (Cornell University)|Sep 24, 2017

Vehicular Ad Hoc Networks (VANETs)34 references3 citations

TL;DR

This paper proposes the Probability Prediction based Reliable Opportunistic (PRO) routing algorithm for Vehicular Ad Hoc Networks (VANETs), which predicts Signal-to-Interference-plus-Noise Ratio (SINR) and packet queue length (PQL) variations at receivers to compute vehicle utility based on weighted variances of these metrics. The PRO algorithm achieves superior performance in packet delivery ratio, end-to-end delay, and network throughput compared to SRPE, ExOR, and GPSR routing protocols by enabling dynamic, prediction-driven relaying decisions in high-mobility environments.

ABSTRACT

In the Vehicular ad hoc networks (VANETs), due to the high mobility of vehicles, the network parameters change frequently and the information which the sender maintains may outdate when it wants to transmit data packet to the receiver, so for improving the routing effective, we propose the probability prediction based reliable (PRO) opportunistic routing for VANETs. The PRO routing algorithm can predict the variation of Signal to Interference plus Noise Ratio (SINR) and packet queue length (PQL) in the receiver. The prediction results are used to determine the utility of each relaying vehicle in the candidate set. The calculation of the vehicle utility is weight based algorithm and the weights are the variances of SINR and PQL of the candidate relaying vehicles. The relaying priority of each relaying vehicle is determined by the value of the utility. By these innovations, the PRO can achieve better routing performance (such as the packet delivery ratio, the end-to-end delay, and the network throughput) than the SRPE, ExOR (street-centric), and GPSR routing algorithms.

Motivation & Objective

• To address the challenge of rapidly changing network conditions in high-mobility VANETs where outdated routing information degrades performance.
• To improve routing reliability and efficiency in opportunistic VANET routing by incorporating real-time prediction of key network parameters.
• To reduce end-to-end delay and increase packet delivery ratio through dynamic, utility-based relaying vehicle selection.
• To outperform existing protocols such as SRPE, ExOR, and GPSR in dynamic vehicular environments through predictive reliability modeling.

Proposed method

• The PRO algorithm predicts future SINR and packet queue length (PQL) variations at potential receiver vehicles using historical data and statistical modeling.
• Vehicle utility is computed as a weighted function of the variances of predicted SINR and PQL, with higher weights assigned to more stable and reliable relays.
• The weights in the utility function are derived from the statistical variance of SINR and PQL across candidate relaying vehicles, reflecting prediction reliability.
• Relay selection prioritizes vehicles with higher utility scores, ensuring data is forwarded through the most reliable and stable paths.
• The algorithm integrates prediction-based decision-making into an opportunistic routing framework, enabling proactive forwarding decisions despite high mobility.
• The method is evaluated using simulations with 13 figures and 38 mathematical formulations to model network dynamics and utility computation.

Experimental results

Research questions

• RQ1 How can network parameter prediction improve routing reliability in high-mobility VANETs?
• RQ2 What impact do SINR and PQL prediction variances have on relaying vehicle utility and selection?
• RQ3 Can a prediction-based utility model outperform traditional routing protocols like SRPE, ExOR, and GPSR in VANETs?
• RQ4 How does the integration of predictive metrics affect end-to-end delay and packet delivery ratio?
• RQ5 What is the optimal weighting strategy for SINR and PQL variances in determining relay utility?

Key findings

• The PRO algorithm achieves a higher packet delivery ratio than SRPE, ExOR, and GPSR due to its predictive relaying mechanism.
• End-to-end delay is significantly reduced in PRO routing because of more reliable and timely forwarding decisions.
• Network throughput is enhanced in PRO due to efficient use of stable and high-utility relays in dynamic environments.
• The utility calculation based on SINR and PQL variance improves routing stability and adaptability to rapid topology changes.
• Simulation results demonstrate that PRO outperforms SRPE, ExOR, and GPSR across all key performance metrics, including delivery ratio, delay, and throughput.
• The prediction-based weighting model effectively identifies reliable relays, reducing the impact of outdated or unstable channel conditions.