Source-linked AI summary
VANET Routing Protocols: Pros and Cons
Bijan Paul, Md. Ibrahim, Md. Abu Naser Bikas
TL;DR
VANET routing is difficult because rapidly changing topologies and frequent disconnections challenge communication across varying traffic scenarios. This paper reviews unicast routing protocols by summarizing their advantages and disadvantages. It concludes that further evaluation across traffic scenarios and routing-protocol comparisons is needed.
Problem
Rapid topology changes and frequent disconnections make efficient VANET routing difficult, while existing protocols do not meet every traffic scenario.
Method
The paper investigates and summarizes the pros and cons of topology-based and position-based unicast routing protocols for inter-vehicle communication.
Results
The review identifies protocol-specific trade-offs, including FSR's reduced bandwidth and overhead, AODV's updated paths, and geographic routing's scalability and mobility suitability.
Takeaways & Limitations
Further performance evaluation across traffic scenarios and comparisons among routing protocols are needed for future improvement.
Takeaways & Limitations
DSR performs worse under high mobility and cannot repair broken links locally.
Abstract
from arXiv · showhide
VANET (Vehicular Ad-hoc Network) is a new technology which has taken enormous attention in the recent years. Due to rapid topology changing and frequent disconnection makes it difficult to design an efficient routing protocol for routing data among vehicles, called V2V or vehicle to vehicle communication and vehicle to road side infrastructure, called V2I. The existing routing protocols for VANET are not efficient to meet every traffic scenarios. Thus design of an efficient routing protocol has taken significant attention. So, it is very necessary to identify the pros and cons of routing protocols which can be used for further improvement or development of any new routing protocol. This paper presents the pros and cons of VANET routing protocols for inter vehicle communication.
1. INTRODUCTION
VANET is a vehicle-to-vehicle and vehicle-to-roadside wireless network that has attracted broad research interest. The paper summarizes unicast routing protocols' advantages and disadvantages to support understanding and future improvement.
- VANET is an autonomous, self-organizing wireless network supporting vehicle-to-vehicle and vehicle-to-roadside communication.
- VANET nodes can act as servers or clients to exchange and share information.
- Applications include collision warnings, driver assistance, cooperative driving, road-information dissemination, Internet access, parking, and driverless vehicles.
- The paper summarizes unicast routing protocols' pros and cons to improve understanding and support future protocol development.
2. CHARACTERISTICS
VANET routing is challenging because vehicle mobility produces rapidly changing topologies and frequent disconnections. Network density, traffic conditions, obstacles, and sensing capabilities shape the communication environment.
- VANET has unique characteristics that make designing applications challenging compared with MANET.
- Vehicle movement at high speed changes the topology and can make links short-lived; at 20 m/sec with a 160 m radio range, a link lasts 8 sec.
- Frequent disconnections occur during information exchange and are most common in sparse networks.
- Mobility patterns depend on traffic environment, road structure, vehicle speed, and driver behavior.
- Sparse and dense networks differ in obstacles and communication conditions, requiring different routing approaches.
- GPS and onboard sensors can sense node position and movement for communication and routing decisions.
3. ROUTING PROTOCOLS
VANET routing protocols are classified into topology-based and position-based approaches. This classification reflects the importance of handling highly dynamic network topology.
- VANET routing protocols are classified into topology-based and position-based routing protocols.
4. PROS & CONS OF TOPOLOGY BASED ROUTING PROTOCOLS
Topology-based routing uses link information and is divided into proactive table-driven and reactive on-demand approaches. These protocols trade route availability and latency against bandwidth use, flooding, overhead, and performance under mobility or scale.
- Topology-based routing uses network link information and includes proactive table-driven and reactive on-demand protocols.
- Proactive routing avoids route discovery and offers low latency, but unused paths consume substantial bandwidth.
- Fisheye State Routing: FSR reduces bandwidth consumption and routing overhead by exchanging only partial updates with neighboring nodes.
- Fisheye State Routing: FSR performs poorly in small ad hoc networks, provides less information about distant nodes, and incurs increasing storage and processing costs as networks grow.
- Reactive routing reduces traffic by discovering routes only when needed and avoids periodic flooding, but route-finding latency and excessive flooding remain drawbacks.
- AODV: AODV provides updated paths using destination sequence numbers and responds to link failures, but setup, stale entries, reply duplication, and beaconing create overhead.
- Dynamic Source Routing: DSR is beaconless, uses caching, and avoids periodic updates, but high mobility, large route headers, unnecessary flooding, and broken links limit performance.
- Temporally Ordered Routing Protocol: TORA creates a directed acyclic graph when needed, can reduce rebroadcast overhead, and performs well in dense networks, but is not scalable.
5. PROS & CONS OF GEOGRAPHIC ROUTING PROTOCOLS
Geographic routing uses GPS-derived positions rather than routing tables or link-state exchanges. It avoids route discovery and scales to mobile nodes, but depends on positioning services that can fail in tunnels.
- Geographic routing uses each node’s own and neighbors’ geographic positions for routing decisions.It does not maintain routing tables or exchange link-state information.
- It avoids route discovery and route management while offering scalability and suitability for high node mobility.
- GPS-based positioning is required, and GPS devices do not work in tunnels without satellite signals.
5.1 DTN
Delay Tolerant Networking addresses frequent node disconnections through a carry-and-forward strategy. Nodes store packets when direct contact is unavailable and later forward them using neighbor-related metrics.
- DTN uses carry-and-forward to overcome frequent network disconnections.
- When contact is unavailable, a node stores the packet for later forwarding.Forwarding is based on a metric associated with neighboring nodes.
5.2 BEACON
Beacon-based and opportunistic geographic protocols use vehicle or neighbor information for forwarding, trading communication overhead and responsiveness against mobility-related delay or privacy concerns.
- Beacon: A beacon is a short hello message transmitted periodically to expose a node’s presence and position.Neighbor-table entries are removed after beacons from the corresponding node stop arriving for a specified period.
- VADD: VADD uses predictable vehicle mobility in a carry-and-forward approach and reports high delivery ratio against GPSR with buffer, epidemic routing, and DSR.It is intended for multi-hop data delivery.
- GeOpps: GeOpps selects forwarding vehicles using navigation-suggested routes and minimum arrival time toward the destination.Its delivery ratio depends on mobility patterns and road topology rather than high vehicle density.
- GPSR: GPSR selects the neighbor closest to the destination using beacons, switching from greedy to perimeter forwarding when greedy forwarding fails.Forwarding decisions are made dynamically while retaining only one-hop neighbor location.
Cons
The reviewed protocols have limitations involving stale or incomplete position information, flooding and overhead, unrealistic evaluations, environmental dependencies, and trade-offs between delivery, latency, and scalability.
- GPSR: GPSR can use stale neighbor positions under high mobility, while a moving destination’s packet-header information is not updated.
- AGF: AGF detects stale neighbor-table entries and updates moving-destination information, but may fail to find the shortest connected path optimally.
- PBR-DV: PBR-DV lacks comparative evaluation of packet delivery ratio and overhead, while its non-greedy operation requires excessive flooding.
- GRANT: GRANT was evaluated on static traces despite VANET mobility, and beacon overhead and possible packet-delivery inaccuracy were not measured.
- GPCR: GPCR depends on junction nodes, and its junction-detection approach has problems on curved roads.
- GpsrJ+: GpsrJ+ is unsuitable for delay-sensitive applications and was not applied to a realistic city map.
- CAR: CAR may select unnecessary anchors and cannot adjust to different sub-paths when traffic conditions change.
- GyTAR and GSR: GyTAR depends on roadside-unit traffic information and cannot avoid voids, while GSR neglects sparse networks and has higher overhead than GyTAR.
6. CONCLUSION
The paper reviews pros and cons of VANET routing protocols and concludes that broader performance evaluation across traffic scenarios is still required.
- Further evaluation should compare routing protocols across various traffic scenarios.
- Overlay comparisons remain incomplete, including GSR against other position-based routing protocols.
- PBR-DV has not been evaluated against non-overlay routing protocols.