1 Introduction

Ad-hoc networking is the best solution for cooperative driving between cooperative cars on the road [1, 2]. Such networks, recognized as VANETs, communicate to a rapidly developing research field, being a particularly challenging set of Mobile Ad-hoc Networks (MANET). A VANET network has detailed highlights such as disseminated control and controlled networking administration, the expansive quantity of vehicles, high-speed vehicle, obliged however very inconsistent of network configuration. The transmission signals are hindered by buildings and high mobility leads to frequent partition. On the contrary, as different to MANET, no considerable power requires [2]. There is growing research as well as commercial importance in the development and configuration of VANETs [3]. VANET networks are an unusual example of MANETs (Mobile Ad-Hoc Networks) that are completed up of various vehicles moving through city streets and equipped to communicate with one another without the use of a fixed correspondence foundation.

The proposal for VANET’s simulations is the wide assortment of models of mobility. The most usually model of mobility in the writing is the Random Waypoint (RWM) model [4]. Each vehicle chooses an arbitrary place of destination and random speed, proceeds to that target, stops, and then after that go-ahead to another indiscriminate goal. Random Walk, Random Direction, & Boundless Simulation Area are the additional open-field comparable models [5]. In which classified the mobility models in VANETs based on how they use existing information. In terms of geographical constraints, the mobility models used in VANETs differ slightly from the mobility models used in MANETs.

VANET technology can be used to communicate among vehicles, and with the improvement of intelligent transportation system applications, the various accidents and energy waste from fossil fuels is estimated to decrease. VANETs use short-range wireless communication (WC) within moving vehicles on the roads.

During a Sybil attack, a malicious vehicle delivers different messages to various other vehicles. Each message reveals separate created source individuality in a way that the original originator is not disclosed. The crucial purpose of the malicious vehicle (attacker) is to provide the false impression to different vehicles by sending incorrect messages & to wrongly encourage them to leave the street for the easy or free passage or other benefits of that malicious vehicle (attacker) as shown in (Fig. 1). Numerous messages revealing the diverse source of fictitious identities are sent by the attacker (malicious vehicle) to several other vehicles. Thus the attacker creates a fake notion of the presence of numerous vehicles on the street using these forged messages and an illusion of a traffic jam condition. Several vehicles fall in the trap and deliberately leave the street on a false notion of a traffic jam ahead providing an advantage to the attacker (malicious vehicle) [6, 7].

Fig. 1
figure 1

Sybil Attack [6, 7]

Full size image

2 Related Review

In this section, present the mobility and VANET routing protocol concept related to the present research model. The QoS parameter performance can be increase and decrease of mobility model due to speed, direction variation as well as wireless connectivity with dynamic topologies. [3] Investigate three different mobility models as consider the vehicle movement behaviour and design a estimation method called route probability as well as to measure route setup achievement rate.

[4], investigated the control of a mobility model on the concert of a reactive and proactive vehicular ad-hoc routing protocol. With varying vehicle densities, the packet delivery ratio (PDR), throughput, and average delay are estimated. [5], investigated the effects of Random Walk, reflection, and wrapping, RWP and GPCR performance. The experimental result illustrates that the performance of the protocol such as PDR and delays varies across any various number of vehicles.

[6] Investigated PDR, throughput, and routing overhead in the Random Walk (RW) as well as the Random Waypoint (RWP) mobility model using sensitive routing protocols such as DSR, AODV, and one active routing protocol DSDV. The analysis revealed that the random Walk and RWP mobility models with AODV, DSR and DSDV routing protocols produce the similar performance for the inputs, when paused time is increased, then the performance of the network is also increased. Here demonstrated their movement direction, speed and angle direction for both the mobility models.

[8] Evaluating the Gauss-Markov model of mobility and its effect on network connectivity, hop count the lifetime routes. RWMM has utilized comparator evaluation of simulation results. Comprehensive simulations have been done for the distinctive number of the vehicle, and a different level to the randomness parameters α (0 ≤ α ≤ 1) to Gauss-Markov model. On networks with a little number of vehicles, network connectivity on Gauss-Markov models is basically lower than the Random Waypoint. In networks with a medium and large number of vehicles, network connectivity on both models practically equivalent. Gauss-Markov model of mobility, usual hop-count per lowest hop way impressively large then under Random Waypoint. The minimum hop path on Gauss-Markov mobility model has shorter route lifetime relate with the Random Waypoint.

There is always a trade-off between the delay and the network capacity at various mobility models. From the literature that has been studied, the characteristic of this trade-off is completely impacted by the optimal of the mobility model [9,10,11]. Investigate delay as well as capacity trade-off in MANET. System capacity is shared between a numbers of vehicles, hence as the network size grows bigger each vehicle gets smaller throughput, thereby indicate the static ad hoc networks are not accessible. It is important to systematically consider how much delay can be allowed because of vehicle mobility simultaneous as a result of an increase in the network capability.

The concept of average delay to observe several types of vehicle mobility analyzed in the literature from a general aspect and distinguish and compare them [12,13,14,15,16].

The RWP model of mobility was first designed by Johnson and Maltz1 in 1996 [17]. Not long after it turned into a most regularly utilized model of mobility for scientists for VANET networks. RWP Model of mobility is a model that uses the ideas of an interruption time between two cases of mobility. In begin mobile vehicles pick the random goal for movement. The speed of vehicles would be characterized appropriately and ought to be uniform before its movement. The term delay time alludes to the time when vehicle stop for a determined time in the wake of coming to at the goal. After the lapse of respite time, the vehicle again picks the arbitrary goal to move. In the event that the interruption time is set as '0' at that point, it implies that it is a nonstop model of mobility [18]. The model of mobility is incorporated into Network Simulator (NS-2) [19]. The RPW model is the most straightforward and simplest to utilize [20, 21].

3 Methodology and Algorithms

The methodology and procedure of the proposed algorithm are designed for securing the network. In which Max HopCount is adjusted the seed hashed of the top Hashfield. When a vehicle receives RREQ or RREP HopCount is verified by hashing Max HopCount by Hashfield. The achieved value of resultant is linked with the top Hashfield. However, if the verification is failed, then the packet is discarded. Prior to rebroadcasting an RREQ or forwarding RREP, a vehicle hashes is the field of Hash in the Signature Extension. Hashfunction field points out that is utilized for the computation of hash. The hashfunction is used by a forwarding vehicle, which is creator of the routing packet. The routing packets have selected due to the field that is already signed. If the vehicle does not support the hashfunction, the forwarding routing packet is dropped.

figure e

As per the routing protocol of GPCR-MA [21], a source vehicle needs to transmit the RREQ packet to locate a trail to arrive at the target or destination vehicle. The target vehicles, or some transitional vehicles along the path, can the reply back to that source vehicle. By default, then the source vehicle acknowledges the initial fresh RREP packet delivered to it. However, in routing protocol of GPCR-ARE [22, 23], the target or destination vehicle or any intermediate vehicle while generating the RREP packet also creates another packet of RREP. This is a sort of authentication of the primary packet attached with a sequential number increased by one count.

Hence, there are 2 (two) RREP packets from the destined or destination vehicle or any transitional vehicle that happen to along the route to final destination. One packet carries a normal sequence no. while the other packet bears an incremented value by one count (+ 1) of the previous sequence code. Both packets have the value ‘0' set in the field ‘VERIFIED'. When an intermediate vehicle obtains an RREP packet, it preserves the data regarding the reply-packet. After that, it verifies our attached field ‘VERIFIED' whether it is 0 or 1. If the field value is found as 0, it meant that the packet is yet to be verified or can be an invalid packet. The value as 1 proves the verification and validity of the packet. Hence, it should be forwarded to its destination or next vehicle. The flow chart of the proposed methodology mention in Fig. 2.

Fig. 2
figure 2

Flow chart of GPCR-ARE for Sybil Attack

Full size image

4 Experimental Analysis: Performance Metrics

A VANET topology consider with no. of vehicle for the simulation. The no. of vehicles are deployed with the help of different mobility inside the simulation area. The speed is fixed in all the mobility model but direction of the vehicle are depend on the mobility model structure. The simulation parameter which is used in this research are mention in (Table 1).

Table 1 Scenario Parameters for the current research in NS-2 [24]
Full size table

As per the simulation obtained data during simulation and extract the QoS parameter such as Average end-to-end delay, Average energy consumption, Average network throughput and PDR (packet deliver ratio) using the following formula.

  1. 1.

    Average end-to-end delay

Here, the Eq. (1) represents the formula of Average end-to-end delay, while the total number of successfully delivered packets.

$$Average \;end - to - end\; delay = \frac{{\mathop \sum \nolimits_{i = 1}^{n} \left( {Received \;Packet\; Time - Send \;Packet \;Time} \right) \times 1000\left( {ms} \right)}}{Total \;Number\; of \;Packets\; Delivery \;Successfully}$$
(1)
  1. 2.

    Average Energy Consumption

The consumption of entire energy is the sum total of use up the energy of overall vehicles in the network and spend energy of the vehicle is the summing up of energy used for communication.

  1. 2.

    Average network throughput

Here, the Eq. (2) represents the mathematical formula of throughput where PacketSize represents the size of the ith packet which reaches to the destination; the arrival time of the last packet is designated as PacketArrival while the arrival time of the first packet is marked as PacketStart

$$Throughput = \frac{Packet\;Size}{{\left( {Packet\;Arrival - Packet\;Start} \right)}}$$
(2)
  1. 4.

    Packet Delivery Ratio (PDR)

Mathematical formula to calculate PDR is stated as below.

$${\text{Packet}}\;{\text{delivery}}\;{\text{ratio}} = \frac{{{\text{Received}}\,{\text{packets}}}}{{{\text{Generated}}\;{\text{packets}}}}$$
(3)

5 Simulation Results Analysis

Mobility extends, vehicle speed, and vehicle number are the requirements that should be established for mobility arrangement. In this paper, a road network based on the highway, Gauss Markov, Freeway, Manhattan, RPGM, and RWP mobility models is simulated for the standards 802.11p with the help GPCR-MA and GPCR-ARE protocols.

5.1 Relative Study of Mobility Models Based on GPCR-MA Vanet Protocol on the Simulation

Table 2 displayed the performance value of data transmission delay and energy consumption, which is consumed by data packet transmission based on network state change. According to the table, the maximum delay for Freeway, Gauss-Markov, Manhattan, RPGM, and RWP is 206.37, 203.67, 208.58, 127.14, and 130.39. When the entire mobility model is compared, the RPGM outperforms the other mobility models. Our research did not have a significant impact on the study.

Table 2 Comparison study for mobility models in terms of average delay & energy consumption using GPCR-MA based on simulation time
Full size table

In this Section GPCR-MA Vanet Routing performance presented in Tables 2, 3 with respect graphical representation in Figs. 3, 4. Here, Vanet mobility models examination at 100–500 s of simulation time through GPCR-MA. The Vanet mobility models (Freeway Gauss Markov, Manhattan, RPGM and RWP) performance demonstrates at various simulation time in respect of average delay, consumption of energy during the average network throughput, network, and PDR.

Table 3 Comparative study of performance on the basis of the time of simulation applying protocol for GPCR-MA in respect of average throughput and PDR for models of mobility
Full size table
Fig. 3
figure 3

Comparison of the average delay and energy consumption for mobility model on the basis of the time of simulation applying protocol for GPCR-MA

Full size image
Fig. 4
figure 4

Comparative study for mobility in terms of average throughput & PDR on the basis of the time of simulation applying protocol for GPCR-MA

Full size image

Table 3 displayed the throughput performance value, which indicates the number of packets delivered to the receiver side, and PDR, that indicates network performance as per rapidly changing network size. The maximum average throughput as well as PDR values for Freeway, Gauss Markov, Manhattan, RPGM, and RWP are 71.1, 64.79, 62.79, 72.21, 74.44, and 98.75, 98.76, 98.71, 98.75, 98.75, respectively. When compared to RPGM, these entire mobility models have significantly higher network performance. If the network changes quickly, RPGM has 98 percent network performance.

Figure 3 with their performance table shows the energy consumption. The energy consumption did not show much change over the routing and model of mobility, its value changed due to main 2 reasons (i) density of network traffic & (ii) network running time span. Each vehicle follows from source course to its destination within the network vicinity and the process of transmission for each data packet exhausts the energy unit, hence on that basis, the energy consumption has increased. This consumption can further increase with the increase of the traffic density or the network range. But all these models of mobility are not showing a large amount of difference, this network does not fully depend on the energy, it requires to start the network or network connection can communicate or we can say they manage their connection to other vehicles.

Figure 4 with respect to their performance in Table 3, indicating a better or higher packet delivery ratio when comparing these five mobility models, with RPGM having the highest PDR. Both scenarios of amplified network density & quickly transformed simulation time are compared. Network traffic is directly related to network, which implies enhanced PDR as every one vehicle has a better option in communication and gets more opportunity to locate a vehicle, which can be used to uncover an improved route from the sender to receiver vehicle.

5.2 Relative Study of Mobility Models of GPCR-ARE Vanet Protocol on the Basis of the Simulation Time Variation

Table 4 and Fig. 5, displayed the performance value of data transmission delay and energy consumption, which is consumed by data packet transmission based on network state change. According to the table, the maximum delay for Freeway, Gauss-Markov, Manhattan, RPGM, and RWP When the entire mobility model is compared, the RPGM outperforms the other mobility models. Our research did not have a significant impact on the study.

Table 4 Comparative study of performance on the basis of the time of simulation applying protocol for GPCR-ARE in respect of average delay and energy consumption for models of mobility
Full size table
Fig. 5
figure 5

Comparison of energy consumption and average delay for mobility model on the basis of the time of simulation applying protocol for GPCR-ARE

Full size image

Table 5 and Fig. 6 displayed the throughput performance value, which indicates the number of packets delivered to the receiver side, and PDR, that indicates network performance based on quickly changing network state. The highest values of throughput and PDR are 70.02, 64.79, 65.85, 72.55, 76.30, and 58.65, 98.76, 98.71, 96.75, 95.63, for Freeway, Gauss-Markov, Manhattan, RPGM, and RWP, respectively. As per compare mobility model performance QoS values according to network simulation time. RPGM mobility model network performance for GPCR-ARE routing protocol consistently maintained 96%.

Table 5 Comparative study of performance on the basis of the time of simulation applying protocol for GPCR-ARE in respect of average throughput and PDR for models of mobility
Full size table
Fig. 6
figure 6

Comparison of average throughput and PDR for mobility model on the basis of the time of simulation applying protocol for GPCR-ARE

Full size image

5.3 Relative Study of Mobility Models of GPCR- MA Vanet Protocol on the Basis of the Number of Vehicle Variation

Based on simulation, Table 6 and Fig. 7 presented the QoS performance value of average delay as well as energy consumption. In GPCR-MA, the maximum delay is 683.52, 258.54, 688.42, 129.94, and 130.34 all four-mobility model respectively.

Table 6 Comparative study of performance on the basis of network traffic density applying protocol for GPCR- MA in respect of average delay and energy consumption for models of mobility
Full size table
Fig. 7
figure 7

Comparison of average delay and energy consumption for mobility model on the basis of network traffic density applying protocol for GPCR- MA

Full size image

Table 7 and Fig. 8, displayed the performance value of average throughput and PDR based on rapidly changing network traffic. The highest values of throughput and PDR are 224.97, 185.92, 202.37, 229.45, 178.51 and 98.66, 98.75, 98.66, 98.71, 98.75 for Freeway, Gauss-Markov, Manhattan, RPGM, and RWP, respectively. When we compare all mobility model values based on network traffic. RPGM and RWP mobility models network performance consistently maintained 98% in GPCR-MA routing routing.

Table 7 Comparative study of performance on the basis of network traffic density applying protocol for GPCR-MA in respect of average throughput and PDR for models of mobility
Full size table
Fig. 8
figure 8

Comparison of average throughput and PDR for mobility model on basis of on network traffic density applying protocol for GPCR-MA

Full size image

5.4 Relative Study of Mobility Models of GPCR-ARE Vanet Protocol on the Basis of the Number of Vehicle Variation

Table 8 and Fig. 9, shows the performance value of average delay in packet transmission and energy consumption based on vehicle variation. The maximum value of average delay is 683.52, 334.52, 588.42, 290.04, and 333.20 for Freeway, Gauss-Markov, Manhattan, RPGM and RWP respectively in GPCR-ARE.

Table 8 Comparative study of performance on the basis of network traffic density applying a protocol for GPCR-ARE in term of average delay and energy consumption for models of mobility
Full size table
Fig. 9
figure 9

Comparison of average delay and energy consumption for mobility model on basis of on network traffic density applying protocol for GPCR-ARE

Full size image

Table 9 and Fig. 10, displayed the network throughput and packet delivery ratio performance values based on rapidly changing network traffic. PDR values for Freeway, Gauss-Markov, Manhattan, RPGM, and RWP are 79, 78.76, 48.86, 97.75, and 96.15, respectively. When all mobility model values based on network traffic were compared, it appeared that this combination RPGM > RWP > Freeway > Gauss-Markov > Manhattan in GPCR-ARE Vanet routing.

Table 9 Comparative study of performance on the basis of network traffic density applying a protocol for GPCR-ARE in term of average throughput and PDR for models of mobility
Full size table
Fig. 10
figure 10

Comparison of PDR and average throughput for mobility model on basis of on network traffic density applying protocol for GPCR-ARE

Full size image

6 Conclusion

We offer a proposal for a novel way out or solution to remove the attack effect from the vehicular network using fabricated request message as well as next hop information to mitigate Sybil attacks through a reduction in computational, storage overhead, & routing. Based on the proposed method, we have done practical and observed the above results, which reveal the reduction in delivery delay from end to end and increase in performance of throughput as well as in packet delivery ratio.