View of Efficient Road Side Framework Placement using VANET for Reducing Network Delays

Efficient Road Side Framework Placement using VANET for Reducing Network Delays

1D.Vijendra Babu, ²Sathish Kumar .L, ³Uma Priyadarsini P S, ⁴Kanagaraj Narayanasamy, ⁵V. Pandimurugan

1Professor, Department of Electronics & Communication Engineering,Aarupadai Veedu Institute ofTechnology, Vinayaka Mission's Research Foundation,Paiyanoor - 603 104, Tamil Nadu, India.

2,5Assistant Professor, School of Computing Science and Engineering,VIT Bhopal University, Kothri Kalan, Madhya Pradesh - 466114, India.

3Assistant Professor(SG), Department of Computer Scienceand Engineering, Saveetha Institute of Medical and Technical Science, Chennai - 600077, Tamil Nadu, India

4Assistant Professor, J.J. College of Arts and Science (affiliated to Bharathidasan University,Trichy), Pudukkottai, Tamilnadu.

Abstract

The Road Side Unit (RSU) is a transmitter, it is facilitate along with route to us for communication between network surface and vehicles. The RSU is one of the components of VANET (vehicular ad hoc network). In this research paper mainly focused on problem of placement of RSU on road side like highway and also avoids the network delay along with efficient network. For this problem the proposed ERSF (Efficient Road Side Framework) avoid the network delays with help of number linear conceptual model along with optimization network delay and under consideration of network. The ERSF framework has been tested that performance using metrics of Generating Traffic Mobility Patterns (GTMP) by VanetMobiSim. The experimental results comparisons has been shows standard distribution and cost effective reduction is 23% and the network delay is 9% respectively and these results are gives clear definition of efficiency of ERSF solutions.

Keywords:GTMP, RSU, ERSF, VanetMobiSim, Network Delays, Road Side Unit.

1. Introduction

Now a day the emerging network technology for Ad-Hoc Network is Vehicular Ad Hoc Network (VANET), that is allows the methods of ITS (Intelligent Transportation System) techniques for making an efficient networking systems for between network surface and vehicles in road infrastructure through Vehicular Ad Hoc Network. The VANET facilitate vehicles interactive with every other network in read unit and get efficient internet on the moving state.

The VANET is a part of Mobile Ad-hoc Networks; these VANET and MANET is self organized, independent and focused for the sharing manner along with self organized authentication [1]. With help of Dedicated Short Range Communication (DSRC) the VANET has gives wireless link for communication for roaming vehicles [2] along with the standard of IEEE 802.11a [3]. In VANET changes sequence is very problem in traffic network, Because of high portability the topology. Besides, long range interaction, the serious issues is inaccessibility of RSU within certain regions which brings about separation and undesirable networklate.

The limit and effectiveness of VANET is unequivocally relies on situation of RSU with the path. One of the very much planned RSU position system can enormously improve the limit along with effectiveness of the system, and simultaneously lessens usage cost. In the RSU designation technique must be contrived with the end goal that it can gather all the traffic information and disperse it progressively over the whole system.

The wide scope for VANET has lot of applications going from diversion to street authentications [4]. Wellbeing applications intensely depend on close to constant transmission of crisis information sharing [5]. For that kinds of uses need for solid activity communication of between vehicle-to-framework and vehicle-to-vehicle.

The principle reason for existing is to stay away from incidents, stay alive of humans and give obstruction less condition to crisis reaction group. On interstates, VANET should be had transmit ready information sharing utilizing RSU is to facilitate speed up the crisis reaction if there should be an occurrence of an episode. A perfect time for helping for medical facility during accident happening time period is called golden time [6], at the point when a mishap occurs, ready messages ought to be engendered to the respect specialists inside portion of the besttime [7].

That is manifestlimiting transmission late is the way to give crisis reaction with in time.The ERSF manages along with fitting of RSUs on road side. Every vehicle able get to RSUs in various approaches; direct communicate conveyance, happens when the vehicle come in to theRSU transmission scope, andhanding-off of multi-bounce, it is happens when the vehicle is exit of theRSU transmission boundary.The proposed ERSFapproach performed defining by whole Number Linear Conceptual Model (NLCM) model.

The goal of thistechnique is to reduce the framework of delay, at the same time as regarding the all out spending plan accessible for the sending. Because of building up the technique has taken both vehicle-to-RSU and vehicle-to-vehicle swapping.Distributions of Vehicle along with versatility designs are produced by utilizing VanetMobiSim[8], simulations. During the ERSF is approvedin NS2 by arrange reproductions acted.

This research paper has been organized as the following phases. Related work and problem statements were placed in phase two. The system models in phase three. In phase four had optimal position placement of RSU. The results of simulations are represented in section five.

The results and discussion and conclusion are placed were phases six and seven respectively.

The vast majority [11, 12, and 14] of the exploration work finished on RSU situation has concentrated on boosting framework execution regarding throughput or by [10, 13 and 15]

and large framework limit. Not many authors have thought about the issue of limiting idleness. The most part cantered [9] around limiting handover inertness and related system.

The analysts [17] considered the issue of how to moderate the effect of multichannel hand-off clash and keep up high data engendering speed and proposed a hand-off conflict calculation.

The displayed and dissected Liu [7] along with Hu [18] the postponement of broadcasting an alarm sharing VANET. The issue of limiting [16] transmission late by ideally putting RSU

was considered. In any case, the proposed arrangement was centered on limiting deferrals in urban conditions.

2. Related Work

The issue of the minimization situation for had tended to by a few research authors. The author Liu et al investigated the absolute literacy of broadcasting cautions in VANET and road[7]. The authors likewise built up a connection between the minimum number for the RSUs along with the separation secured by vehicles or cars. Nonetheless, the authors didn't produce any RSU sending methodology. Another author has Ahmed and Bouk has taken the RSU position issue and contrived a plan that limits handover length from starts RSU to destination RSU. The model ascertains a variable called Inclusion/Administration situation, were is utilized to decide the while time which an endorser station can get administration from the objective RSU. Limiting handover postponement can be gainfulin decreasing system late; however it can't give any bound on the general system inertness.

After that, another one author Cheng has been produced a geometry-based meagre inclusion [10] convention for RSU installation is utilized which taken numerous qualities, for example, the geometry of street systems, traffic examples, and asset confinements. For traditional information, the convention can find the most reasonable inclusion regions. Hereditary and eager has both calculations are utilized to determine the inclusion issue. Be that as it may, that plan doesn't address the issue of limiting network late.

the proposed [11] an RSU produced by Chi et al. situation methodology dependent on the traffic stream. The plan limits the quantity of RSU belongs putting RSUs at highways convergences and spots with most elevated vehicle recurrence. The thought is to group street crossing points and discover capacity RSU placed by utilizing Markov bunching calculation.

Then again, T. J. Wu gave a similar to Chi et al report that demonstrated that the scheme that is not efficient [12]. The authors have proposed the practical methodology of RSU arrangement that plans to amplify the framework limit. The plan gives great throughput however the system delay accomplished by it is not minimum cost.

The author displayed the most extreme inclusion with time edge issues [13] by utilizing a hereditary calculation for taking care of the issue. A total of four certifiable informational indexes were tried and contrasted and an avaricious methodology. Another proposed [14]

Aslam has been delivered two diverse advancement techniques for the sending of a predetermined number of RSUs along with urban territory.A method is called the Inflatable Development Heuristic technique, the remaining is a diagnostic BIP strategy. The BIP uses branch along with headed methodology for finding an ideal scientific arrangement, while the strategy utilizes swell extension relationships get finding an ideal or close to the ideal arrangement. The two strategies were utilized to tackle the enhancement issue of limiting the normal detailing average.

The new method conspires indicated BEH technique performs superior to BIP strategy as far as computational expense and adaptability. The master represented by Patil [15] a novel Voronoi arrange based calculation for a powerful portion of RSU's that shapes a Voronoi

network as far as the measure of deferral brought about by information bouch has been RSU above So also, the calculation for picturing by Jalooli [16] the exhibition of message proliferation in VANET, represented Security-Based Separated RSU Position calculation through which message spread postponement can be limited in urban zones.

The calculation for changing by Rui [17] for proposed an effect of a multichannel condition called Achievement Likelihood-based Transfer Conflict procedure. The property of stochastic to vehicular conveyance and constant area data of vehicles by by SPRCA were investigated.

To investigate the SPRCA performance with stochastic geometry-based approach and determine their properties. No methodology for limiting system delay was proposed by the creators. The nature of administration by Hu, Yun [18] in VANET the defer limits and start to finish excess were considered as the critical measurements.

3. The Proposed ERSF Approach

The ability and effectiveness for VANET is unequivocally based on the position of RSU with rout. A perfect organized framework RSU position method able to perform upgraded the ability and effectiveness of the network in road side unit, during as well as the minimized installation cost also. The ERSF is handling along with installation positioning of RSU on road side unit surface boundary.

Figure 1 Architecture of Efficient Road Side Framework

Aroundable to all things considereda RSU put in a fragment. This ERSF strategy to the effective organization of RSU is as per the following:

 Vehicle versatility designs are created under various situations by utilizing VanetMobiSim to imitate practical vehicular condition.

 A Number Liner Conceptual model is planned that limits the general inactivity of the traffic network and decreases RSU sending cost also.

 The ERSF has been improvement model is unravelled to get delay RSU situation.

 The proposed model is then approved by recreations performed on the streamlined model in NS2 test system.

 Results comparison of experimental is done between the proposed ERSF situation and WKUDS.

The underlying advance of this research paperhas been producing a vehicular portability design by utilizing VanetMobiSim simulation.This is a sensible vehicular [19] development follow generator thatbe utilized for mimicking huge scope transmission traffic networks.In the simulator VanetMobiSim has various situations were produced by picking a road map with 10, 30, 40, 60 and 70 vehicles, separately. The highway comprises of different paths, upstairs roadbe to vehicular development for right to left, and though on the lower street traffic is moving from left to right.

The high way like road surface is divided into five fragments; every segmental separation is group equivalent to the communication scope of an ERSF. Each car are roaming with speed between 65 to 125kilometres by hour. Irregular Outing Originator is utilized for arbitrary age of vehicular models. Subsequent to creating portability design VanetMobiSim gives a yield record that can be utilized as a contribution for simulation NS2. Issue definition is talked about in detail in the following phase.

4. Problem Statement

In this phase, ERSF arrangement issue is planned as a Number LinearConceptualdesign. The goal of this issue is to devise a base inertness costeffective ERSF arrangement plot, for example the model decides what number of ERSFs to convey and where to put them so the general system delays be limited at the base conceivable sending cost. Let take that every vehicle able get to the ERSF in 2 distinct manners on aroad, for example straightinteraction and multibounce transferring. The accompanying choice factors are characterized to encourage Number LinearConceptual model detailing:

𝐴_𝑙^𝑘 = 1 : The vehicles are in fragment l for facilitating by the ERSF in section k. During, suppose𝐴_𝑙^𝑘 = 0there no vehicle in fragment l for facilitating by the ERSF in fragment k.

𝑆_𝑙 : There is not an ERSF in fragment l Sl€{0,1}.

𝐹_𝑙 : The mean lateup on theconnectivity through direct access to an ERSF located in segment l.

𝐹 k_l n − :

The mean late belongs withconnection since a vehicle inside fragment k to an ERSFinsidefragmentkthroughvarious bouncing in towards the back. Suppose it is in k > l.

𝐹 k_l n + :

The mean late belongs withconnection since a vehicle inside fragment k to an ERSFinsidefragmentkthroughvarious bouncing in towards the back. Suppose it is in k < l.

Lth : A integer is representative an positive of capacityof hop count for various level hopsegments.

ERSF _{R :} Consumption minimization cost for one ERSF.

ERSF T : Total budget for ERSF installation.

This problem statement is elaborated as follows:

𝑀𝑖𝑛 𝑆_𝑙𝐹_𝑙

𝑘

𝑙=1

+ 𝐴_𝑙^𝑘

𝑙−1

𝑚 =1

𝐹_𝑙^𝑘𝑚−1 + 𝐴_𝑙^𝑘

𝑘

𝑘=𝑙+1

𝐹_𝑙^𝑘𝑚+1

Where:

Si Є {0, 1}, Є {1, 2, 3...,Y} (I)

𝐴_𝑖^𝑘=0if| k-l | > ERSFth else 𝐴_𝑖^𝑘= Є {0, 1} (II) A^k_i

𝑘

𝑙=0 ≤ 1, k Є {1, 2, 3...,Y} (III)

𝑆𝑖 = 1, 𝑖𝑓 1 − 𝑌_{𝑘 𝑙}^𝑘 = 0,Є {1, 2, 3...,Y} (IV) 𝑆 _𝑙𝐸𝑅𝑆𝐹_𝑘 ≤ 𝐸𝑅𝑆𝐹 _𝑇

𝑘𝑙=1 (V)

The main goal is to limit the whole of deferrals coming about because of both direct associations and multi jump associations. The main term ( ^𝑘_𝑙=1𝑆_𝑙𝐹_𝑙) represents the general postponements because of direct associations as it were. The Equation II and III of the goal speak to the absolute postponements in reverse multi bounce associations and forward multi jump associations, individually. Prototype I demonstrates that at most one ERSFis able sent in a fragment. Prototype IIauthorizes the jump check limit (for example ERSFth) refer to vehicle to vehicle correspondence.

That needs to guarantee palatable execution in multi bounce handingoff associations. As indicated by Prototype III, every multi jump association can be served by all things considered one ERSF at a given moment. Prototype IVdemonstrates that there must be a ERSF sent in section ‘k’ if any were vehicles in some other fragment will be served by the ERSF in that portion. Nonetheless, that is doesn't confine the situation of ERSF in some other section for which the condition isn't valid. Prototype V determines the absolute spending plan for this ERSF organization.

5. MaximizationData Rate Achievement

The maximization data rate achievement is solved by Shannon Capacity Theorem for connection path, here data rate is (R) and bandwidth is (BW):

R=BW log2 (1+SECR) (VI)

Were, the SGR be Signal Error Clearance Ratio (SECR) for the receiver.

Here the ESRF have two situations to regard: strait connection data transfer and multihop data transfer. Thestrait connection data transfer, onevehicle is in connection area circle of a ESRF, and consequently an immediate correspondence connection able be set up between vehicle and ESRF. Because of thePublic Area Method with irrelevant background commotion, the sign force at the collector is following by

𝑇

=

𝐻²

(VII)

Where G_tand H²are data transfer power and the data transfer power receiver, and the H is the strait connection length for both data transmitter and data transferreceiver.

Suppose, In the event that there is a vehicle in the obstruction scope of the beneficiary, the quality of clamour signal at the recipient because of the single impedance source can be given as follows:

𝑇′

=

𝐻′²

(VIII)

Where 𝑇′_𝑡 and 𝑇′_𝑒 are the transmitter intensity of obstruction source and the impedance signal force at the collector separately, and H’ is the most brief separation between the meddling vehicle and the recipient.

In the event that there are V meddling vehicles inside the impedance scope of the collector, the all out obstruction experienced by the recipient is the joined impact of all the impedance data. Most pessimistic scenario, every one of these signs will join usefully at the beneficiary and the complete obstruction able it is shown as follows:

𝑇′

=

.

𝐻′2_𝑖

(IX)

After that the SECR at the destination able to be refer by

SECR=

𝐺′_{𝑡𝑜𝑡𝑎𝑙}

SECR=

𝐻²

𝑥

𝐻′_𝑡𝑖

𝑉𝑘=1

(X)

The maximization data rate achievement able to defined by the combination of equations XI and X:

𝜏 = 𝑋 𝑙𝑜𝑔

1 +

𝐻²

𝑥

𝐺_𝑡𝑖^′

𝑉𝑘=1

(XI)

The fruitful correspondence connects, it was vital that SECR is more prominent than limit esteem (C). Let’s see ß(y) will the likelihood closeness capacity of the vehicle area distribution and Ø be the obstruction scopes of the collector, at that point the likelihood of precisely one vehicle in the impedance go is shows by

𝐺

= ß(y)dx

(XII)

After that mean integer of vehicles inside the Ø is shown by

𝐷

= 𝐷𝐴

(XIII)

Where,D is the complete integer of vehicles the framework. Let Øy,u will the likelihood of a fruitful transmission [20] between the transmitter y and the beneficiary u, at that point Øy,u can be controlled by the accompanying scientific connection.

For a multi jump hand-off, the exchange rate relies on fruitful foundation of the association.

Along these lines, s relies on the likelihood of fruitful broadcast. The normal estimation of s can be controlled by joining equation XII.

6. Mean Value of Dissemination Delay

The mean value achievable late is easy to compute for example it can send any integer of SECRs to such an extent that every dynamic vehicle has an immediate connection with a SECR. Right now, joins are immediate and there is no multi bounce association, and along these lines the all out system spread deferral is the base. It is for V associations it very well may be given by the accompanying condition:

𝐹

=

𝑇

𝐺

(IX)

Be that as it may, this hopeful situation isn't constantly conceivable. The normal engendering deferral of a multibounce interface relies on the normalinteger of transfers and the normal separation between nearby handing off vehicles or motors. Allows first taking the instance of a one hop connect. Right now, proliferation delay relies just on the normal separation between imparting vehicles in nearby portions. Between vehicles dividing on ahighwayable to be which demonstrated by an exponential dispersion with average proliferation separation between two vehicles equivalent to 1/ ¥. In ERSF have expected the transmission scope of a vehicle (D) will equivalent to the fragment distance. From D is a lot littler than the sign engendering speed (S), subsequently the average separation between vehicles x and y in neighboring fragmentation can be overall assuming to fragmentation length 1/¥ = D. The engendering delay can be communicated via the following mathematical formation.

𝑆 =

𝑆

𝑥

(X)

Where the 𝑥_𝑎,𝑏 is represented probability for achieved communication.

The intrigue is to decide the normal number of bounces in a transfer procedure. The Vehicles sequence moving [20] a free homogenous stateof process along withsequencef vehicle force

along the highway. The assuming number of bounces (A(b))has given for circulation able to be explore by the accompanying with mathematical equations.

A b = ^{𝐼−𝐽 −𝐾}_{𝑚 =0}  1 − ^m 1 + A m ∀0 ≤ 𝐼 ≤ 𝐽 ≤ 𝐾 − 1 𝑎𝑛𝑑 𝐴 𝐷 = 0 (XI)

Where D is the communication range of concern vehicle and S is the length between vehicles and communication unit. Along these lines, the total proliferation delay for multi bounce connection can be found by consolidating mathematical equationsX and XI.

7. Simulation Experiments

To validate the results of the proposed ERSFmethodology through simulation process, here take the some highway map like one appeared in figure 1, where the highway comprises of different roads and the elements of path are 1500 metersof length by 200 meters of widths.

The highway is segregated by five parts. The 200 meters length is segregated by equal meters for each part of ERSF.Different re-enactment situations were made along with vehicle conveyance of 10 to 60 vehicles aresequence moving along the interstate with the speed in the scope of 60–120 km for every hour on a read like highway. The cars are moving in two distinct ways: right to left and left to right.Has been utilized the simulation to create mobilitydesign to represents the exact sharing on the road side surface.

The vehicle quantitydoes not consistently for sharing;along with populace congested is not good for equal distributions. The normal transmission late among vehicle and ERSF is taken via NS2 because ofbeing the way that a vehicle able to create both roads and multi bounce associations. The communication scope of everyERSF and vehicle to be350 meteralong with the obstruction run is 600 meters;it was yield about five parts. Every vehicle able to connect near ERSF of arelated for with the closest ERSF for this either towards or backwards ways.

The significant re-enactmentmetrics are appeared in the following table.

Because of every one of the reproduction situation, here performed fifty runs with SECR, equalsharing and model plans, and the normal outcomes are accounted for here. For illustrate

Figure 2, Five ERSF placed on the segmented parts.

Parameter Value

Road Side Area 1500 meters length,200meters width

Number of Cars 10, 20, 30, 40, 50, 60, 70

Speed 60–120 km/h

Count of SCER 1, 2, 3, 4, 5

Direction Right to Left and Left to Right

SCER transmission scope 350 m

SCER interference scope 600 m

vehicle transmission scope 350 m

Vehicle interference scope 600 m

Type Wireless channel

Network interface type Wireless

Time for simulation 510 sec.

Table1. Metrics of simulations

Figure 3, Two ERSF placed with reference of SECR and DSF.

Figure 4, Two ERSF placed with reference of equal sharing.

8. Results and discussion

The proposed ERSF solutions are comparing with CES and UD, along with assess the mean value of networkdelay and validate the average delayfor every installation methodology.

Maximize network limitation is main objective of CSE. The budget minimization is

controlled belongs to cumulative of RSU. In figures four and five are demonstrate for the results of DMP1 refers scheme with DMP and RSU, in the same manner DMP2 is refers with DMP and used two RSU. The figure six and seven also represented the similar figure four and five for the DMP 10 has been taken 10 vehicles measurement.

Figure 5Different quantity of communication of delay of Average

Fig. 6Various quantity communication of average throughput

Fig. 7various quantities ERSF of averagedelay

Fig. 8various quantities ERSF of average throughput

The DMP was fitted in ERSF fragments neighbour surface of hue vehicle quantity. The figure 5 demonstrated about the ERSF has been takenaverage of delay and population of various traffics of CES. It was clearly shown the ERSF has produced efficient performance in the all

kind of aspects using metrics. The throughput average metrics results are represented clearly in figure 6 plot. The proposed ERSF is yielding the 20 % of reduction average and 9% of uniform placement delay and CES also. The delay is the high latency because it should not take the mobility topology was gathering ERSF.

The ERSF has fragments where vehicles remain the maximum, during DMP will in general send proposed framework in portions with most elevated vehicle deluge. In this way, DMP gives connection chances to maximum vehicles per segment time along these lines it can accomplish the most minimal deferral for all aspects.The plot of the picture demonstrates the average of latency for methods of the number for the proposed method segment for situations of start with 5 and end with 70 for each step have increased 5 vehicles it has been represented by Figure 7.

The average latency is high for 1 ERSF along with 5 vehicles or car, for that 1 ERSF for the strait to connection and not many vehicles to set up multi bounce connections, every vehicle needs to hold up a significant measure of at the time it can set up an association and send its information packets are has been sent. The circumstance improves by expanding the quantity of proposed and the quantity of cars, these metrics increment the chance to set up either immediate or multi jump associations.

The relating average of the throughput is demonstrated by figure 8. True to form, the base throughput is acquired for the instance of 1 proposed and 5 car. The presentation improves extensively with an expanding number of ERSF. It is clear, DMP beats Uniform Circulation in everyone has 33% of improvement. In any case, the throughput gave by the existing is on normal 2% all the more for throughput gave by the DMP system. The average from Uniform dispersion doesn't take vehicle versatility designs while conveying proposed. In this manner, ERSF permits vehicles to communicate high information because of a more prominent association lifetime along these lines, it can give efficient throughput.

9. Conclusion

In this research paper has been represented about case of ERSF exploitation methodology to the VANET taken for the highway or road side, which is every vehicle able to get access with ERSF in various paths: strait connection and multi jumping. The two RSU equal sharing models are has been compared namely Uniform Distribution (UD) and another one is Delay Minimization Problem (DMP). The NS2 experimental results are demonstrated efficient performance comparably existing systems. The DMP scheme recommends to the efficient ERSF sending system by carry in shows the complete minimized budget have an extent that the average delay of the system is limited. These able to exceptionally helpful in the opportune delivery of aware if there should be an occurrence of an episode, and furthermore gives upgrades in the total throughput of the system.

References

1. Ranjan Senapati B., Mohan Khilar P. (2020) Optimization of Performance Parameter for Vehicular Ad-hoc NETwork (VANET) Using Swarm Intelligence. In: Rout M., Rout J., Das

H. (eds) Nature Inspired Computing for Data Science. Studies in Computational Intelligence, vol 871. Springer, Cham.

2. Babu Ram & Neelendra Badal (2019) Latency estimation based routing (LEBR) protocol in vehicular adhoc networks (VANETs), Journal of Discrete Mathematical Sciences and Cryptography, 22:8, 1517-1531, DOI: 10.1080/09720529.2019.1700923.

3. Malhi, A.K., Batra, S. & Pannu, H.S. An Efficient Privacy Preserving Authentication Scheme for Vehicular Communications. Wireless Pers Commun 106, 487–503 (2019).

https://doi.org/10.1007/s11277-019-06175-7.

4. Nazib, R. A., & Moh, S. (2020). Routing protocols for unmanned aerial Vehicle-Aided vehicular Ad Hoc networks: A survey. IEEE Access, 8, 77535-77560.

5. A. Arora, A. Mehra and K. K. Mishra, "Vehicle to Vehicle (V2V) VANET based analysis on waiting time and performance in LTE network," 2019 3rd International Conference on Trends in Electronics and Informatics (ICOEI), Tirunelveli, India, 2019, pp. 482-489.

6. S. S. Shah, M. Ali, A. W. Malik, M. A. Khan and S. D. Ravana, "vFog: A Vehicle- Assisted Computing Framework for Delay-Sensitive Applications in Smart Cities," in IEEE Access, vol. 7, pp. 34900-34909, 2019.

7. Maratha, B. P., Sheltami, T. R., & Salah, K. (2017). Performance study of MANET routing protocols in VANET. Arabian Journal of Science and Engineering, 42(8), 3115–

3126.

8. Zhang, Z., Boukerche, A., & Pazzi, R. (2011). A novel multi-hop clustering scheme for vehicular ad-hoc networks. In Proceedings of the 9th ACM international symposium on mobility management and wireless access (pp. 19–26). ACM, Miami.

9. Ahmed, Z., Naz, S., & Ahmed, J. (2020). Minimizing transmission delays in vehicular ad hoc networks by optimized placement of road-side unit. Wireless Networks, 26(4), 2905- 2914.

10.Liu, H., Yang, L., & Zhang, Y. (2015). Improved AODV routing protocol based on restricted broadcasting by communication zones in large-scale VANET. Arabian Journal of Science and Engineering, 40(3), 857–872.

11.Shaik, S., Venkata Ratnam, D., & Bhandari, B. N. (2018). An efficient cross layer routing protocol for safety message dissemination in VANETs with reduced routing cost and delay using IEEE 802.11P. Wireless Personal Communications, 100(4), 1765–1774.

12.Bariah, L., Shehada, D., Salahat, E., Yeun, C.Y.: Recent advances in VANET security: a survey. In: 2015 IEEE 82nd Vehicular Technology Conference (VTC2015-Fall), pp. 1–7, September 2015. https://doi.org/10.1109/VTCFall.2015.7391111.

13.Abdelhamid, S., H. Hassanein, and G. Takahara. 2015. Vehicle as a resource (vaar). IEEE Network 29 (1): 12–17.

14.Bhoi, S.K., D. Puthal, P.M. Khilar, J.J. Rodrigues, S.K. Panda, and L.T. Yang. 2018.

Adaptive routing protocol for urban vehicular networks to support sellers and buyers on wheels. Computer Networks 142: 168–178.

15.Younes, M.B., and A. Boukerche. 2013. Efficient traffic congestion detection protocol for next generation vanets. In 2013 IEEE international conference on communications (ICC), pp.

3764–3768. IEEE.

16.Nayak, J., B. Naik, A. Jena, R.K. Barik, and H. Das. 2018. Nature inspired optimizations in cloud computing: Applications and challenges. In Cloud computing for optimization:

Foundations, applications, and challenges, pp. 1–26. Springer.

17.Kaiwartya, O., and S. Kumar. 2014. Geocasting in vehicular adhoc networks using particle swarm optimization. In Proceedings of the international conference on information systems and design of communication, pp. 62–66. ACM.

18.Chen, Yu-Jia, Kai-Min Liao, and Yung-Fang Chen. "End-to-End Delay Analysis in Aerial-Terrestrial Heterogeneous Networks." IEEE Transactions on Vehicular Technology (2021).

19.J. Tian and F. Meng, "Comparison Survey of Mobility Models in Vehicular Ad-Hoc Network (VANET)," 2020 IEEE 3rd International Conference on Automation, Electronics and Electrical Engineering (AUTEEE), Shenyang, China, 2020, pp. 337-342, doi:

10.1109/AUTEEE50969.2020.9315583.

20.Li, Suoping, Fan Wang, Jaafar Gaber, and Xiaokai Chang. "Throughput and Energy Efficiency of Cooperative ARQ Strategies for VANETs Based on Hybrid Vehicle Communication Mode." IEEE Access 8 (2020): 114287-114304.

21.Boukerche, A., & Sun, P. (2020). Design of Algorithms and Protocols for Underwater Acoustic Wireless Sensor Networks. ACM Computing Surveys (CSUR), 53(6), 1-34.