TARS: A Novel Mechanism for Truly Autonomous Resource Selection in LTE-V2V Mode 4

Area of Science:

• Vehicular Networks
• Wireless Communication
• Resource Management

Background:

• Effective communication in vehicular networks relies on efficient wireless channel resource scheduling.
• 3GPP Release 14 defines two scheduling types: eNodeB-controlled and Autonomous Resource Selection (ARS).
• ARS is decentralized, suitable for vehicle safety, but prone to resource collisions and lacks adaptability to traffic density.

Purpose of the Study:

• To address resource collisions and adaptability issues in Autonomous Resource Selection (ARS) for vehicular networks.
• To introduce a novel mechanism, Truly Autonomous Resource Selection (TARS), for improved resource allocation.
• To model and analyze the impact of ARS parameters on resource collisions.

Main Methods:

• TARS treats resource allocation as a local detection problem among neighboring vehicles to prevent collisions.
• Discrete Time Markov Chain (DTMC) is used to model counter selection and resource reselection behavior.
• A fair policy for counter and resource reselection is proposed based on DTMC model observations.

Main Results:

• TARS effectively reduces resource collision probability in vehicular networks.
• The proposed TARS mechanism demonstrates a superior packet delivery ratio compared to LTE Mode 4 and other approaches.
• TARS offers improved performance in terms of resource collision and packet delivery.

Conclusions:

• TARS provides a more robust and adaptive solution for resource scheduling in vehicular networks.
• The DTMC modeling offers insights into optimizing ARS parameters for fairness and efficiency.
• TARS represents a significant advancement over existing methods for vehicular communication resource management.