Propagation Models For Vehicular Ad Hoc Networks Computer Science Essay

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

Abstract-Lot of work is going on area of VANET. These systems often comprise several tens, hundreds or thousand of nodes, a real-world test is a very costly and time consuming operation. VANET research work is carried out using different network simulators, which allows for fast and cheap evaluation of protocols and applications in a controllable manner. Simulation study uses models in order to make judgment related real-world viability. Model should reflect reality, using propagation model in vehicle to vehicle communication, hence accuracy is an important requirement.

Keywords-VANET;V2V communication;Propagation models;simulation;etc.


INTELLIGENT TRANSPORTATION SYSTEM (ITS) applications are being defined to improve road traffic safety, efficiency and comfort. Many applications rely on communication provided by VANET's. In Vehicle-to-Vehicle (V2V) communication takes place between vehicles which meet by chance. A VANET occurs as soon as two or more vehicles are within communication distance.

No infrastructure is involved; VANET's rely heavily on distributed measures to regulate access to the wireless channel. Protocols for random access, TDMA and flooding are implemented and evaluated in simulators. Reality provides opportunities for two nodes to exchange information which would not have been possible in the simulator due to a simplistic propagation model [3].

Radio propagation model also has a strong impact on the performance of a protocol [4] because the propagation model determines the number of nodes within one collision domain, an important input for the contention and interference.

The mobility often involved in VANET's causes nodes to move in and out of each other's transmission range. Depending on the propagation model a node may share a collision domain with tens or hundreds of other nodes, or with only a handful because the model accounts for buildings [5]. This paper provides the propagation models which can be used in VANET, specifically in simulation.


Network simulator used in VANET often provides a stack of protocols on top of which the protocol or application under test is implemented. A component managing connections between nodes often works in conjunction with the propagation model in order to evaluate which nodes are affected by a transmission. Results could be a node correctly receives a message or receives garbled bits due to a collision.

Mobility model can be used to move the nodes around - general the case in a VANET- either based on measured or generated traffic traces [6], an embedded mobility model [7,8] or a coupling with the traffic simulation tool [9.10].

A simulation have two goals:- 1) Perform a statistical exploration to gain insight in how a system with work in a generic environment, or 2) Perform a site-specific evaluation of a system to gain insight in the operational properties in specific environment.


VANET's are subset of MANET's with several differences between them. Mobility is usually constrained, because the nodes follow roads according to some physical vehicle model. Speed is generally high in VANET's but differ e.g. communication between stopped vehicles or vehicles passing in opposite lanes. Nodes in VANET generally do not have strict weight, size and power consumption limits. VANET nodes can safely be assumed to have access to certain peripherals such as positioning and navigation hardware.

Another important difference is a vehicle may easily travel outside an area covered by a certain legislature. Vehicles from multiple vendors need to be able to cooperate; such standardization is an important accepts which is generally not considered when evaluating a MANET application.

Propagation Environment

The wireless channel is a highly chaotic and unpredictable system [3]. It's a way from transmitter to receiver a signal is being reflected, scattered and obsorded by objects in the propagation environment. As such its magnitude is altered, but due to multiple paths it can also interfere with itself or with signals sent in other frequency ranges.

With context of VANET's comes also a typical radio wave propagation environment. Vehicles generally move on roads, but other scenery can vary from open farmlands to forests to large urban canyons and bridges. VANET propagation environment is the presence of large metal objects which are continuously changing position in the environment, namely the vehicles themselves, such environment is highly dynamic.

Large-Scale effects radio wave propagation are the following three phenomena:-

Refelction: It occurs when a wave encounters a large surface with certain optical properties. In models reflection is often translated to a path loss exponent, such as the 2 in (2) and 4 in Eq. (3)

Diffraction: This phenomenon is explained by Huygen's Principle, which states that every point on a wavefront acts as the seed for a secondary wavefront. This enables waves to propagate around edges or through holes. Which can be modeled with the knife-edge diffraction model [11], which can be used for site-specific modeling of propagation over mountains and large buildings.

Scattering: A radio wave scatters when it encounters an object which is small compared to the wavelength, spreading the waves in all directions. This can account for a received signal which is stronger than would have been predicted by relection and diffraction alone.

Small-Scale effects radio wave propagation is often referred to as fading. At the receiver multiple versions of the original signal arrive; they can be reflected and diffracted and arrive with time and phase difference. These multipath waves interfere with each other, which can cause large fluctuations in signal quality with apparently small changes in time or receiver location. This relative motion causes frequency modulation because each multipath will have a different Doppler Shift; the resulting frequency change is derived as follows:

Here v is the relative velocity, λ the wavelength and θ the angle between the signal path and the direction of move-ment.

Channel Parameters

Mobile channel can be characterized with channel parameters. The reception of multipath components can be seen as a sample which can be expressed by means of statistical quantities. Delay Spread is the standard deviation of the arrival times. Doppler Spread measures the spectr5al broadening caused by relative motion of transmitter and receiver.

Radio Technologies

Many communication technologies are used in VANET's, such as infrared [12] and short range radio. Short range radio technologies used is Wi-Fi, but some research has done in 900MHz band [13] and in the millimeter range (60-78Hz) [14]. VANET research converges to IEEE 802.11p [15], a Wi-Fi used for communication in the vehicular environment part of the Wireless Access in Vehicular Environments (WAVE) standard [16], [17].

IEEE 802.11p builds upon the proven and mature 802.11 standards, providing relatively cheap but powerful and communication devices. It provides low latency access to the medium - nodes do not first have to associate and authenticate with base stations - and is optimised for the ad hoc domain. IEEE 802.11p operates on 7 channels in the 5.8-5.9GHz band (as shown in Fig. 1) and is expected to have a maximum communication range in the order of 1km.

node listens to the Control Channel (CCH) at least a certain amount of time. On the CCH announcements for services can be transmitted, these services can then be provided on the Service Channels (SCH). The WAVE standard does not define if one radio should listen to channels in time slots or if multiple radios can be used to observe several channels simultaneously. The channel access is defined in IEEE 1609.4 [19]. So far, most ITS-related VANET research focuses on applications operating on a single channel as if in isolation.

Signal Parameters

The frequency at which a radio technology operates greatly impacts its propagation properties. Besides its carrier frequency, other metrics are the transmitted power, the bandwidth and the symbol time, these are results of the modulation scheme, a combination of signal and channel parameters can lead to different kinds of fading. This fading is often characterized by a probability distribution and appropriate parametric assumptions [20].

Implementation in Network Simulators

Implementation of propagation model in a simulator usually takes the following steps, illustrated in Fig. 2:

1) For every node n within a relevant distance, perform a calculation of the received signal strength. The received signal strength is calculated using a propagation model.

2) For a transmission instance (e.g. the transmission of message x) all signal strengths from concurrent transmissions other than x received at node n are added as noise.

3) Based on the Signal-to-Interference and Noise Ratio (SINR) and Bit Error Rate (BER) a decision is made whether the message is correctly received or has bit errors. If the SINR is below a certain threshold it is impossible to detect the signal in the received noise, and a collision has occurred. Most propagation models in simulators consider nodes to be stationary for the duration of one transmission.


The propagation environment in the simulator is used to judge the effects of propagation of electro-magnetic waves through the medium, usually this medium is air.

In its most abstract form, this defines success or failure of reception of a message for a certain node. Propagation models can be classified in large scale and fading or small-scale models. From an implementation point of view they can be either deterministic or probabilistic.

Deterministic Models

A deterministic model allows computing the received signal strength, based on actual properties of the environment such as the distance between transmitters T and a receiver R. These models range from simple to very complex where they also account for multipath propagation in the environment modeled exactly as the area of deployment.

1) Free Space model: Which is sometimes also referred to as Friis model, after its inventor [21]? It models a single. unobstructed communication path [20]. The received power depends only on the transmitted power, the antenna gain and the distance between the sender and the receiver, as shown in Fig. 3.a). As a radio wave travels away from an (omni-directional) antenna, the power decreases with the square of the distance.

Where Pt is the transmitted power, Gt and Gr are the gains of the transmitter and receiver antenna gains and λ is the wavelength. α is the path loss exponent and is 2 in Free Space. L is the system loss. Often, Gt, Gr and L are set to 1 From a topology point-of-view, this model regards the nodes as floating in free space.

2) Two-ray Ground model: The two-ray ground model also accounts for a reflection via the ground, given the dielectric properties of the earth in addition to the direct line of sight (LOS). Nodes are positioned on a plane as depicted in Fig. 3.b). This model gives more accurate predictions at longer range than the Free Space model [11] and is given as follows:

Where ht and hr are the heights (in meters) of the transmit and receive antennas respectively. Eq. (3) shows a faster power loss than (2), but does not give good results for short distances because of oscillation caused by the constructive and destructive combination of the two separate paths. Either (2) or (3) is used based on the magnitude of d, the T-R separation.

3) Ray Tracing model: Ray tracing is a technique often used to predict propagation for cellular systems. Modeling the propagation environment plays a critical role in the development, planning and deployment of, for instance, UMTS/IMT2000 cellular systems [22]. Because for these systems not only coverage but also bandwidth is an important issue, careful site planning is in order. Ray tracing models can take into account the exact position, orientation and electrical properties of individual buildings in the environment in which the system is to function. Using the rules for reflection, diffraction and scattering all rays emanating from the source traveling towards a receiver can be modeled, as shown in Fig. 3.c). As a result, a complex impulse response h(t) can be calculated as the sum of all contributions [23]:

The received signal h(t) has N time-delayed impulses (rays), each of which is an attenuated and phase-shifted version of the original transmitted signal. Amplitude An, arrival time Tn and phase are calculated for each ray using Snell's laws, the uniform geometrical theory of diffraction (UTD) and Maxwell's equations. All objects in the environment need to be modeled with characteristics such as permittivity, conductivity and thickness. This method also allows to use antenna radiation patterns. Basically, ray tracing models are computed using 3-D vector mathematics. Evaluating every ray individually for a fixed antenna position is feasible, as it is used in cell planning. In VANET multiple transmitters and multiple receivers are moving in a continuously changing environment and h(t) will need to be recomputed upon a change in the environment. Ray tracing propagation models are not often used in VANET [24].

Probabilistic Models

Probabilistic models allow a more realistic modeling of radio wave propagation [3]. A probabilistic model takes a deterministic model as one as its input parameters in order to get a mean transmission range. For every individual transmission the received power is then drawn from a distribution, as shown in Fig. 4. The result is a more diverse distribution of successful receptions. It can happen with a certain probability that two nodes close to each other cannot communicate, although it can also happen with a certain probability that two nodes beyond the deterministic transmission range can communicate. The distribution of these effects depends on the probabilistic model and its parameters.

1) Log-Normal Shadowing: The Log-Normal Shadowing model uses a normal distribution with variance σ to distribute reception power in the logarithmic domain:

Where Prdet is a deterministic model such as Eq. (2) or (3). As such the received power is given as:

Here α is a path loss exponent like the 2 in Eq. (2) and the 4 in Eq. (3). is a reference path loss measured close to the transmitter. Eq. (6) can be rewritten as:

Which gives a received power by multiplying the deterministic received power with a Power Loss scale factor

in dB:

2) Rayleigh: The Rayleigh propagation model [11] models the situation when there is no LOS, and only multipath components exist. This model incorporates intensive variations in received signal power because multiple paths can either combine constructively or destructively. The amplitude, delay and phase shift of these components greatly depends on the environment.

Like the Log-Normal shadowing model in Eq. (5), the Rayleigh model also depends on a deterministic model to which a certain variation is applied: This can be rewritten to read: where the Power Loss factor is defined by:

3) Longley-Rice: The Longley-Rice model (or Rice model) [3] models the reception powers following the Rayleigh distribution but additionally takes into account the positive effects of a LOS path with a certain scale factor k [25]:

With PL(d) as given in Eq. (11) and F(d) defined as a Ricean PDF with a normal distribution:

With c defined as

4) Nakagami: The Nakagami model is highly generic. Reception power follows a gamma distribution:

The parameter m specifies the intensity of fading effects. Nakagami includes other models, such as:

yet it is probabilistic [26].


VANET is mostly modeled as a cluster of nodes on a at surface in a simulator. This abstracts from obstacles in the environment (such as buildings) which could influence the propagation. This can be accounted for by simply using a path loss exponent α ≠ 2 in the Free Space or Two-ray Ground model, depending on the environment when using a probabilistic model. When using the Nakagami or Rice model, the strength of a LOS component can be set with the m-parameter or the k-factor respectively.

It impacts which nodes are able to communicate and the probability of correct reception. As a result, it can influence the speed at which messages propagate through the network, directly influencing end-to-end delay in a multi-hop scenario. The probability distribution of correct reception also influences the overhead with respect to collisions and medium utilization.