How to make driving much safer with a car to car communication

How to make driving much safer with a car to car communication

Vehicle-to-vehicle communication systems allow cars to share information without driver participation. It creates a wireless network between vehicles that transmits data about the cars’ locations and their speeds. The system continuously analyzes this data and helps prevent auto accidents by warning drivers about dangerous situations on the road in advance. Improving algorithms and implementing new technologies can increase the comfort and safety of drivers.

Quick navigation

Most cars today are equipped with various sensors, including radar that is connected to the cruise control systems or sensors that can detect objects in blind spots. V2V (vehicle to vehicle) systems help drivers be more aware of their surroundings. These features improve car safety without distracting the driver are taking control of the vehicle. It is the car’s sixth sense that notifies drivers about the road conditions ahead, drastically reducing traffic jams and collisions. So, how do we implement features like these in cars to make driving much safer?

Hardware overview

The term V2V has become synonymous with car communication by General Motors, which created it. Vehicle to vehicle communication needs two sets of components to operate properly. One of them is a device for transmitting a trusted and accurate basic safety message (BSM). The other is the set of components that are necessary for receiving and interpreting a BSM sent from another vehicle.

#1 Sending a BSM

Required components:

  • GPS
  • Processing unit
  • Security module

In order to generate a basic safety message, a car needs to know its location, which can be defined with a GPS antenna and receiver. However, position data is worthless without some analysis. That is why a car needs a computer processing unit that can combine its position with other data from onboard sensors such as acceleration, direction, and speed to create a BSM.

Then the car needs to be able to send the data it has generated to another vehicle. This is where the security module is needed. It processes and prepares certificates and security information, which has to be transmitted wirelessly so that the other vehicle can verify that the BSM is valid.

#2 Receiving a BSM

Required components:

  • GPS
  • Processing unit with a specific receiver
  • Security module

BSM transmission methods have to be the same in both the transmitter in one car and the receiver in the other for them to communicate. One way to do this is by using dedicated short-range communications (DSRC). A computer processing unit is also required for proper BSM decoding. Another necessary component is the GPS to verify the distance between the receiving device in one car and a sending device in the other. The receiving device also has to have a security module to process the security data sent from the other vehicle.

V2V system operation components

In order to operate, car to car communication systems require devices in cars and along roadways as well. Roadside equipment (RSE) and vehicle-based devices send security data to the security management system. Due to integrating supplementary vehicle-to-infrastructure (V2I) components in signs and traffic signals, external devices can increase their V2V communication capabilities.

Hardware in a vehicle

Required components:

  • 2 DSRC radios
  • GPS receiver with a processor
  • Memory unit
  • Safety application electronic control unit
  • Driver-vehicle interface
  • Vehicle internal communications network

The basic components of a DSRC device are similar to a common radio that receives and transmits information. A driver-vehicle interface (DVI) contains visual head-ups and blinkers placed in a driver’s field of view. The DVI also has to be able to generate audible noises to warn drivers adequately.

Roadside hardware

Non-vehicle-based equipment is the wireless infrastructure that guarantees communication between a car and V2I components, and between a car and the security credential management system (SCMS).

A vehicle can communicate with 2 types of roadside equipment (RSE):

  • RSEs which broadcast messages supporting vehicle-to-infrastructure interface
  • RSEs which support wireless communication between the SCMS and a car receiving security data and reporting unusual car behavior

It is also possible to use various communication standards:

  • DSRC
  • 3G
  • 4G
  • Wi-Fi​
How to effectively manage fleet and warehouse with custom software solutions

System operation software

The wireless exchange of messages between cars is the basis of vehicle-to-vehicle communication. The idea of these messages is to exchange information about potential danger through a safety application. The main unit of the DSRC device is the software which lets devices define and transmit data about current road and vehicle conditions, analyze the received information, and send warnings if the situation appears to be dangerous.

The minimum requirements for a device are:

  • Simple components for generating and receiving safety messages
  • Ordinary operating system
  • Connection to the DVI
  • Algorithms able to analyze data and issue warnings

Communication between vehicles

Two types of messages represent car to car communication technology: certificate exchange messages and safety messages. The first type of message checks whether a message is sent from a trusted source and the second one supports safety applications. If a standard format of transmitting is used in safety messages, it makes it possible for other vehicles in the network to read these messages. In order to match this condition, each car using the DSRC transmitting method has to send and receive messages in a standard format taking into account such characteristics as range and accuracy. Messages also include information about the time they are sent.

Safety messages contain data such as:

  • Car GPS location
  • Turn rate
  • Speed
  • Vertical and lateral acceleration
  • Predicted path
  • Car type

Communication security

In order to provide message authentication, V2V systems use Public Key Infrastructure (PKI) which contains certificates to let another car know the message is sent from a trusted source. The PKI uses cryptography to send encrypted content in a message.

Asymmetric cryptography is a type of cryptography that is used in V2V security systems. What is asymmetric cryptography? It contains 2 linked keys. One of them encrypts the content and the other one decrypts it. They are called the “public key” and the “private key”, respectively. The first one can be safely distributed but the other has to be kept private by the owner. The way they link prevents deriving one key from the other one. That is why asymmetric cryptography is more secure than symmetric cryptography.

Most of the online transactions use the SSL/TLS protocol (secure socket layers/transport layer security). It uses asymmetric cryptography (AS) for authenticating the server to the client and as an option, the client to the server. The AS also establishes a session key which is also used in symmetric cryptography to encrypt data. For example, the widely known HTTPS also uses SSL and operates as a PKI system.

WAVE V2V system development

Wireless Access in Vehicle Environment (WAVE) is a technology that supports ITS (Intelligent Transportation Systems) taking into account V2I and V2V communication.

Cars periodically exchange information about their position, brake control, and speed. By default, each car uses the highest level of the data range and transmitting power. It makes packet collisions and wireless channel competition even more serious, which creates obstacles to the successful operation of V2V systems. The following dynamic algorithm for controlling the data rate effectively optimizes communication.

Optimizing algorithm

Two performance characteristics are needed to develop a car communication system with an efficient optimization algorithm. These are:

  • Packet delivery rate (PDR)
  • Channel Busy Percentage (CBP)

The PDR is the percentage of the number of received packets from a particular transmitter divided by the total number of packets sent. The CBP is the percentage of time that the wireless channel does not respond (the energy level is higher than the sensing threshold of the carrier), divided by the time that the CBP is being calculated.

The algorithm’s task is to maintain a certain CBP level to guarantee efficient V2V communication performance, despite the number of cars that exist within the communication range.

The algorithm itself includes:

  • EpochMap update
  • Calculating the mean data rate
  • CBP measurement
  • Data rate adaptation
  • Phase control
  • Epoch selection
  • Application jitter selection

The whole process can be described in these two phases:

  • The adaptation of Mean-based data rate
  • Phase control with Epoch
How to organize an efficient development process to build the automotive solution

Phase 1. The adaptation of mean-based data rate

This phase includes 3 actions:

  • EpochMap update
  • Calculating the mean data rate
  • CBP measurement

When the duration of the transmission process decreases by selecting a higher data rate in heavy traffic, the possibility of packet collisions can also be reduced. The increase in the data rate leads to a decrease in the communication coverage. Therefore, it decreases the number of cars waiting for data transmission.

And the opposite is true. The lighter the traffic conditions are, the wider the communication coverage is. The algorithm defines the real-time channel status by using the CBP value, which is measured every 100 msec.

Phase 2. The phase control

This phase consists of 4 actions:

  • Data rate adaptation
  • Phase control
  • Epoch selection
  • Application jitter selection

If two other cars try to transmit a safety message at the same time, packet collisions will be at the same level. Therefore, it is necessary to define the epoch size and the number of basic safety messages each epoch can accept. Each car calculates the utilization value (UV) for the used epoch and the neighboring one. If the calculated UV is lower than the average one, then the car will use this epoch for sending the next BSM. If it exceeds the average value, then the car defines its jump rank.

Vehicular Ad Hoc Networks

VANETs, a subclass of the mobile Ad Hoc networks (MANETs), do not have the fixed infrastructure and provide network operation by relying on other cars. They can use any wireless technology, e.g. short-range radio technologies (Wi-Fi or LTE).

Compared to a cellular system, inter-vehicle communication (IVC) has many advantages, such as wider coverage and lower latency. Bluetooth, Internet access, and car phones all use cellular technologies. Specific VANETs’ properties allow developers to build services aimed to improve convenience and safety in V2V technologies.

Comfort applications

Such applications increase traffic efficiency, passenger comfort, and optimize routes. Examples:

  • Gas station locations
  • Traffic information systems
  • Internet access

Safety applications

These applications are aimed at increasing passenger safety by exchanging safety data via V2V communication. Examples:

  • Lane-changing assistant
  • Sign violation warning
  • Emergency warning system

Recurrent neural networks

Predictions of speed, travel time, and flow are complex nonlinear issues having various factors. Neural network learning is a suitable solution for nonlinear traffic prediction issues. Neural network solutions are less sensitive to incorrect data that makes them independent of a particular prediction location geometry.

For example, it is necessary to predict a car’s behavior in different weather conditions. Regarding traditional neural networks, it is unclear how previous analysis in other conditions influences future ones. RNNs are attempting to solve this issue. They have cycled connections that let the car keep data. Every following parameter is connected to the previous one.

Long short time memory networks

The main advantage of RNNs is the ability to use data about past events to solve future scenarios. However, using data that was received in different scenarios a long time prior, may not apply to RNNs. When we need to predict a car’s behavior after turning the steering wheel left, we do not need any other experience. It is obvious that a car will turn left.

If we need to predict a vehicle’s behavior on an icy, mountain road in heavy traffic conditions and fog, traditional RNNs may not apply. Conversely, LSTM networks can learn long-term dependencies.

These networks are designed to avoid long-term dependency issues. They have a special layer form of a repeated module for processing data based on information received in a cycle threshold. This allows it to predict a car’s behavior in various road conditions and to transmit data to other cars via V2V communications.

Wrapping up

Our company has worked with several partnerships with industry giants, such as Volvo and Renault, to build IT solutions.

Trust our experienced developers to build software to improve the comfort and safety of your drivers. If you have any questions or want to get custom software for your business, feel free to contact us at