Advertisement

Wireless Personal Communications

, Volume 100, Issue 4, pp 1765–1774 | Cite as

An Efficient Cross Layer Routing Protocol for Safety Message Dissemination in VANETS with Reduced Routing Cost and Delay Using IEEE 802.11p

  • Shafi Shaik
  • D. Venkata Ratnam
  • B. N. Bhandari
Article
  • 94 Downloads

Abstract

For successful message dissemination in urban vehicular ad hoc networks (VANETs) with reduced route cost and delay is challenging task due to high mobility of the vehicles. Existing cross-layer cooperative routing (CLCR) protocol in VANETs utilizes vast transmission power and increases route cost due to unnecessary broadcast of Hello messages in selecting relay nodes for packet routing. In this paper an enhanced CLCR (ECLCR) with IEEE 802.11p has been proposed, by introducing new mechanism: selecting appropriate relay nodes by calculating neighbor set ratio (NSR) using neighborhood knowledge to acquire more robust route for successful dissemination of packet, then adding a parameter to packet header such as mobility, which can store total network NSR for each route request packet to reduce delay. The performance of the newly proposed algorithm has been evaluated with network simulator-2.34 tool and compare with existing protocols under different conditions. Furthermore, an extensive simulation experiments have been carried out and it is observed that the proposed ECLCR using IEEE 802.11p standard outperforms over existing approaches.

Keywords

Cross layer cooperative routing VANETs Relay estimation Average delay Route cost 

1 Introduction

VANET is special case of wireless mobile communication networks, where nodes are assumed as vehicles [1]. VANETS provide communication among vehicles and combines roadside units to vehicles to provide road safety and comfort. It allows vehicles to communicate each other even out of radio coverage area for finding route. The communication among vehicles in VANET depends on the type of routing metric used, as there is no centralized infrastructure and high mobility of vehicles [2]. As per the literature survey, different layered mechanisms have been provided for safety message broadcasting in vehicular networks with proper routing. For example, Ad hoc On Demand Multipath Distance Vector (AOMDV) protocol, is an enhancement to Ad hoc on demand Distance Vector (AODV) routing protocol, can able to identify node-disconnected and link-disjoint routes while identifying paths for successful data transmission due to circumstances of node-disjoint paths are stronger when compared to link-disjoint route, as a result the numbers of node-disjoint paths are less than that of link-disjoint paths [3]. Where the disconnected routes are identified for message dissemination but route cost is high leads to unnecessary broadcast of Control packets.

In Ad Hoc on Demand Multipath Distance Vector with Retransmission counts metric (R-AOMDV), a cross layer design illustrated how routing is implemented with retransmission counts, in turn increased excessive use of control packets [4]. A routing protocol provides reliable Roadside to Vehicle communications in rural areas [5]. In this roadside unit takes care of maintaining proper routes to forward the packets on fly by predicting lifetime of links without any relay identification to reduce the delay and route cost for urban environment. Wu et al. [6] have proposed an efficient AODV with backbone based routing for message dissemination to increase packet delivery ratio. Still this protocol performance is limited during low vehicle density due to frequent link breaks, in turn leads to delay in message dissemination.

Another work in [7] focused on link quality based on total network weight along with link expiration time. However it suffers from route discovery phase which is similar to AODV, where unnecessary flooding of hello packets takes place results into high route cost. A different approach with geographic routing was followed in [8] to improve reliability of the route by combining AODV. In this work new field format called recent positions of destination vehicle added to RREQ packet to reduce the hop count. RREQ will reach the destination, leads to improved delay performance and increased route cost. Ledy et al. [9] have presented an improved AODV (V-AODV), where path discovery is done by identifying delay from node to node through send and receive times of Hello packets. V-AODV makes decision in path discovery with increased route cost due to unnecessary flooding of control packets. Despite the numerous layered protocols [3, 4, 5, 6, 7, 8, 9] designed for successful message dissemination, selects and saves least hops rather stable or robust paths to reduce the delay parameter only.

In [10, 11] authors have proposed new cross layer approach for VANETs by applying mobility model to estimate neighboring node movement to chose longest life time route. However in both the proposed methods, consideration of link quality leads to increased hops, which increases End–End delay. In [12, 13] authors have focused on total network weight along with link expiration time. But it suffers from route discovery phase which is similar to AODV [4, 14] for VANETs. The existing routing schemes from [10, 11, 12, 13] discussed above, in which relay vehicles are identified based on the process of flooding RREQ packets to all the neighbors in the transmission range at times. Thus leads to high overhead due to generation of control packets for path discovery as all the methods come under On-Demand routing protocol, such as AODV.

To address the aforementioned issues, researchers have identified cooperative routing protocols for vehicular networks [15, 16, 17]. Cooperative communication is a new physical layer technique which allows communication among multiple vehicles, targeting at improvising the overall end-to-end throughput. Towards these authors in [18] have proposed Cross-Layer Cooperative Routing (CLCR) mechanism to choose adaptive relay vehicles based on distance parameters in VANETs. Furthermore, CLCR extend the lifetime of routing path by identifying link life time to reduce the frequency of route rediscovery. However CLCR suffers from increased route cost and transmission power in disseminating data packets as route discovery is similar to AODV for VANETs. Furthermore in urban VANETs verbatim adoption of CLCR is still challenging issue where vehicle density and speed are intermittent.

To this end, we intend to propose an Enhanced CLCR (ECLCR) protocol by introducing new mechanism: selecting appropriate relay nodes by calculating Neighbor Set Ratio (NSR) using neighborhood knowledge to acquire more robust route for successful dissemination of packet, then adding a parameter to packet header such as mobility, which can store total network NSR for each Route Request (RREQ) packet to reduce delay. ECLCR (Enhanced CLCR) to fulfill the performance metrics such as improved Packet Delivery ratio with reduced route cost and delay over CLCR.

The paper is structured into different sections as follows. Section II summarizes the proposed work. Section III illustrates performance analysis of proposed algorithm. Finally, section IV concludes the work.

2 Proposed Mechanism

In this, section, we illustrate main idea of our proposed routing protocol named ECLCR (Enhanced CLCR) targeting at successful message dissemination in urban VANETs. Towards this, ECLCR under goes two major aspects: Firstly, During Route discovery reduces the redundant transmission of control packets by a vehicle in a particular interval. This can be done by finding the cooperative intermediate vehicle which has fewest Neighbor Set Ratio (NSR) based on the information available in Vehicle Information Table (VIT) at each vehicle at times Ts seconds. For example at the beginning source vehicle periodically sends RREQ packet to next succeeding vehicles with in its transmission range. Then source vehicle selects only one vehicle as relay with less RREPs and it records and updates the path in VIT at each vehicle. Except at the destination vehicle the above said process will be repeated at each intermediate vehicle for shortest path. Thus the number of hop counts will be reduced to forward the data. This reduces greatly both route cost in terms of reduced control packets and the delay in forwarding data packets to destination by selecting relay vehicles which has least common neighbors.
Secondly proposed system further enhanced by adding mobility parameter RREP packet header, this can store the entire network fewest neighbors for each Route Request (RREQ) packet. Since fewest common neighboring vehicles are calculated for each vehicle it can be included in the RREQ packets header so that vehicles receives RREQ might be able to decide, to forward the packet if the ratio of neighboring vehicles is less else they can drop the RREQ packet for that particular path, there by generating more stable links, this also decreases routing overhead, because unstable paths RREQ packets are dropped and there by decreases the delay, in turn the available channel bandwidth will be improved.
Figure 1 shows the decision on the process route request packet in Enhanced Cross-Layer Cooperative Routing. After calculating the set of common neighboring vehicles at periodic intervals, traversed path information is stored in each VIT.
Fig. 1

Decission to process RREQ in ECLCR

On receiving RREQ, every vehicle first checks the value of common neighboring vehicles for an appropriate vehicle to decide whether to plunge or carry on before proceeding to next process. If data id and data sequence number belongs to VIT, then insert the path traversed along with route link (rt_link). Later broadcast the data packet into neighboring vehicle with less NSR, which greatly reduces route cost. Repeat the same for every neighboring vehicle with less NSR and update the VIT.

3 Performance Analysis

In this paper using NS-2.34 a network simulator, the performance of proposed method is evaluated. For visualization, Network Animator (NAM) is used. We have created different urban traffic scenarios with varying vehicle speeds from 54 to 108 km/h. For every scenario, we employed a fixed topology of 2500 × 2500 m2. In order to get stable results total of 30 independent simulation runs performed and considered the average best result. The remaining simulation metrics are listed in the Table 1. The simulation time set is 300 s. We have simulated the different traffic scenarios with both the usual IEEE 802.11 and IEEE 802.11p [19]. The main intend of simulations is to assess the performance differences of the CLCR and ECLCR routing Protocols both in non cooperative path and cooperative path with proposed algorithms.
Table 1

Urban simulation parameters for NS-2.34 tool

Parameter

Typical value

Dimensions (m2)

2500 × 2500

Channel type

Wireless

Vehicle speed range (km/h)

54/72/90/108

Simulation execution time (s)

300 s

Number of vehicles

100

Traffic type/source

CBR

Propagation model

Two-ray ground

MAC

802.11, 802.11p

Radio transmission range (m)

250 m

Packet size

512 bytes

Protocols employed

CLCR, ECLCR

Bandwidth

2Mbps

Number of connections

10-20-30-40-50

4 Results and Discussion

Figures 2 and 3 represents the packets sent and received at variable speeds from 54 to 108 km/h and it is obsreved that ECLCR outperform over CLCR [18] in both cooperative and non cooperative scenarios interms of reduced transmission of control packets in route identification. Further from the Fig. 2 it is observed that packets sent is decreses gradually in proposed ECLCR as speed increases that means for high mobility scenarios Packets sent is better over CLCR non cooperative and cooperative paths. This is because of NSR mechanism which establishes stable links among vehicles by calculating set of neighboring vehicles at time Ts, if the number of neighboring nodes at time Ts sec are less when compared to other times then path will be established with the node having less NSR. Thus leads to reduced load on the link in case of ECLCR. On the other hand number of packets received in ECLCR is better comaparatively over non cooperative and CLCR-cooperative paths. This is because of establishment of stable links among vehicles using NSR mechanism as described.
Fig. 2

Packets sent for different speeds

Fig. 3

Packets received for various speeds

From Fig. 4 it is observed that in case of non cooperative scheme Routing Cost is gradually increases as speed increases from 54 to 108 km/h. Route cost is reduced greatly as speed increase (i.e. in high mobility scenarios) from 54 to 90 km/h then maintained constant till 108 km/h in case of ECLCR. This is because of NSR mechanism which establishes stable links among vehicles by calculating set of neighboring vehicles at time Ts sec.
Fig. 4

Routing cost for various speeds

Which reduces greatly the use of control packets such as RREQ and RREP packets. It is observed that the ratio of bytes transmitted control packets to that of bytes of transmitted data packets is less for different speeds in proposed method.

Figure 5 Illustrates the comparison between average E2D delay and at variable speeds ranging from 54 to 108 km/h. From the curves it has been observed that delay gradually increases when speed varies from 54 to 90 km/h and then gradually decreases from 90 to 108 km/h, this is because of random selection of scene and trace files.
Fig. 5

End to end delay for various speeds

On whole it is evident that ECLCR protocol performance is much better than CLCR [18] in both cooperative and non cooperative paths in high mobility scenarios i.e. from 90 to 108 km/h. This is because of selection of relay vehicle at times based on number of neighboring vehicles to a particular relay vehicle, which establishes stable links among vehicles and further more adding overall network neighbor vehicle ratio into RREQ packet header, which leads to less delay while transmitting data packets between source to destination. Thus the proposed method is efficient in disseminating emergency messages related to accidents and traffic conditions to nearest vehicles with reduced delay.

From the Fig. 6 it is observed that the percentage of transmitted packets which reach destination is almost above 90% in ECLCR with 802.11p. Whereas CLCR with cooperative and non cooperative schemes the percentage of transmitted packets reach destination is about 80 and 45% respectively. Thus the proposed method is very much useful in successful dissemination of data packets with reduced delay in emergency situations.
Fig. 6

Packet delivery ratio for various speeds

5 Conclusion

In this work, we have presented an enhancement to existing protocols using cross-layered approach to assess most appropriate relay vehicle. The proposed protocol enhanced by adding a parameter such as mobility to the packet header, it can be stored the entire network fewest neighbors for each Route Request (RREQ) packet. With the help of simulation results it is identified that the path elected is stable and accessing speed between the vehicles and route cost has come down greatly. Furthermore, we intend to employ clustering concept into proposed cross layered mechanism based on relative speeds. In turn, control the frequent link failures in meager networks or light vehicle road segment under critical situations. This could satisfy recent advancements in Multimedia applications like live video/audio streaming among many users over wide range for vehicle to vehicle communication.

Notes

Acknowledgements

Authors are very much thankful to both the editor and referee for giving valuable critical and precise comments/suggestions to improve the quality of this manuscript. Dr. D. Venkata Ratnam would like to express his thanks to the Department of Science and Technology, New Delhi, India for funding this research through SR/FST/ESI-130/2013(C) FIST program. The work of Dr. D. Venkata Ratnam is supported through F. 301/2013(SAII)/RA201416GEANP5585. Authors would like to thank Dr. I. A. Pasha, Professor and Head of the Department, Electronics and Communication Engineering, B V Raju Institute of Technology, Medak, Telangana for his consistent support.

References

  1. 1.
    Yousefi, S., Mousavi, M. S., & Fathy, M. (2006). Vehicular ad hoc networks (VANETs): Challenges and perspectives. In Proceedings of the 6th international conference on ITS telecommunications (pp. 761–766). IEEE Press.  https://doi.org/10.1109/itst.2006.289012.
  2. 2.
    Li, F., & Wang, Y. (2007). Routing in vehicular ad hoc networks: A survey. In IEEE vehicular technology magazine (pp. 12–22).Google Scholar
  3. 3.
    Marina, M. K., & Das, S. R. (2001). On-demand multipath distance vector routing in ad hoc networks. In Proceedings of the 9th international conference on network protocols (pp. 14–23). IEEE Press.  https://doi.org/10.1109/icnp.2001.992756.
  4. 4.
    Chen, Y., Xiang, Z., Jian, W., & Jiang, W. (2009). A cross-layer AOMDV routing protocol for V2V communication in urban VANET. In Proceedings of the IEEE international conference on communications (ICC) (pp. 2353–2358).Google Scholar
  5. 5.
    Wan, S., Tang, J., & Wolff, R. S. (2008). Reliable routing for roadside to vehicle communications in rural areas. In Proceedings of the IEEE international conference on communications (ICC) (pp. 3017–3021).Google Scholar
  6. 6.
    Wu, C., Ohzahata, S., Ji, Y., & Kato, T. (2016). How to utilize inter-flow network coding in VANETs: A backbone based approach. IEEE Transactions on Intelligent Transportation Systems, 17(8), 2223–2237.CrossRefGoogle Scholar
  7. 7.
    Yu, X., Guo, H., & Wong, W.-C. (2011). A reliable routing protocol for VANET communications. In The seventh IEEE international wireless communications and mobile computing conference (pp. 1748–1753).Google Scholar
  8. 8.
    Al-Rabayah, M., & Malaney, R. (2012). A new scalable hybrid routing protocol for VANETs. IEEE Transactions on Vehicular Technology, 61(6), 2625–2635.CrossRefGoogle Scholar
  9. 9.
    Ledy, J., Boeglen, H., & Hilt, B. (2009). An enhanced AODV protocol for VANETs with realistic radio propagation model validation. In Intelligent transport systems telecommunications. Google Scholar
  10. 10.
    Abedi, O., Fathy, M., & Taghiloo, J. (2008). Enhancing AODV routing protocol using mobility parameters in VANET. In Computer systems and applications.Google Scholar
  11. 11.
    Pomplum, R., & Datta, A. (2007). A study of long distance traffic using the AODV protocol in a vehicular ad hoc network. In Vehicular technology conference.Google Scholar
  12. 12.
    He, R., Rutagemwa, H., & Shen, X. (2008). Differentiated reliable routing in hybrid vehicular ad-hoc networks. In Proceedings of the IEEE international conference on communications (ICC) (pp. 2353–2358).Google Scholar
  13. 13.
    Abedi, O., Barangi, R., & Abdollahi Azgomi, R. (2009). Improving route stability and overhead on AODV routing protocol and make it usable for VANET. In 29th IEEE international conference on distributed computing systems workshops.Google Scholar
  14. 14.
    Ledy, J., Boeglen, H., Hilt, B., Abouaissa, H., & Vauzelle, R. (2009). An enhanced AODV protocol for VANETs with realistic radio propagation model validation. In 9th International conference on intelligent transport systems telecommunications (ITST) (pp. 398–402). IEEE.Google Scholar
  15. 15.
    Khandani, A. E., Abounadi, J., Modiano, E., & Zheng, L. (2007). Cooperative routing in static wireless networks. IEEE Transactions on Communications, 55, 2185–2192.CrossRefGoogle Scholar
  16. 16.
    Nosratinia, A., Hunter, T. E., & Hedayat, A. (2004). Cooperative communication in wireless networks. IEEE Communications Magazine, 42, 74–80.CrossRefGoogle Scholar
  17. 17.
    Shi, Y., Sharma, S., Hou, Y., & Kompella, S. (2008). Optimal relay assignment for cooperative communications. In Proceedings of the ACM mobihoc (pp. 3–12).Google Scholar
  18. 18.
    Chen, W.-H., Pang, A.-C., Pang, A.-C., & Chiang, C.-T. F. (2010). Cross-layer cooperative routing for vehicular networks. In Proceedings of the IEEE ICS (pp. 67–72).Google Scholar
  19. 19.
    IEEE 802.11p. (2010). Amendement 6: Wireless access in vehicular environments.Google Scholar

Copyright information

© Springer Science+Business Media, LLC, part of Springer Nature 2018

Authors and Affiliations

  1. 1.Department of ECEBVRITNarsapur, MedakIndia
  2. 2.Department of ECEKoneru Lakshmaih Education FoundationGunturIndia
  3. 3.Department of ECEJNTU HyderabadHyderabadIndia

Personalised recommendations