AndSus - stock.adobe.com
Vehicles are increasingly being connected to each other and to roadside infrastructure. This growing level of connected car technology can support a wide range of applications and services, as well as safer and more reliable operations of vehicles within the same geographical vicinity.
With every element of the vehicular environment sharing data with each other, this vehicular internet of things, or VIoT, can support real-time situational awareness across all vehicles and supporting infrastructure that is otherwise not available in conventional transportation scenarios.
One example where VIoT is expected to be a game changer is with respect to autonomous vehicle technology, where the decisions made by AV-enabled platforms are entirely dependent on the accuracy and quantity of data regarding its surrounding environment.
Until now, most work on AV systems has focused on onboard sensors, such as lidar, radar and vision systems, which are all limited to what sensors see in front of them. With VIoT, the ability of the AV system to collect information from other nearby vehicles, roadside units and other surrounding infrastructure, such as traffic lights and signs, can profoundly enhance that vehicle's situational awareness, and thus its decisions will be better suited to the environment it is operating in.
To enable this situational awareness, VIoT requires the proper wireless configurations. Learning and decision-making algorithms are vital to enabling the best possible wireless configurations for all transmitters and receivers operating within the VIoT environment. Given the rapidly changing dynamics of vehicular traffic, as well as the challenging propagation conditions experienced by wireless signals, conventional wireless transmitters and receivers will be unable to adapt to the prevailing environmental conditions and yield situations where important information is not reaching intended receivers, thus creating dangerous or even fatal scenarios.
One possible solution to this challenge is to use programmable wireless platforms called software-defined radios (SDRs). Combined with algorithms that can adapt radio operations based on prevailing environmental conditions, SDRs can enhance the reliability of connected car technology, especially in time-critical applications, such as safety and AV operations.
The primary challenge when developing algorithms for SDRs is to create a radio intelligence, referred to as cognitive radio, that is capable of handling a wide range of operational scenarios in a robust and near-optimal manner. When a cognitive radio algorithm faces a scenario it has never encountered before, it should be capable of "learning" the environment and deriving new decisions to produce the best-possible performance outcomes for the connected car communications. As a result, there has been significant work on applying machine learning and AI to cognitive radio, as well as biologically inspired approaches, such as the recently published bumblebee model, in order to advance the current state of the art in cognitive radio technology.
As previously mentioned, SDR technology is an excellent way of introducing a substantial amount of agility and intelligence into a connected car environment. This functional flexibility is necessary given the rapidly changing topology of the vehicle communication network. Using a combination of SDR technology and cognitive radio, it is possible to support a robust and reliable connected car network that can adapt to changing environmental conditions in order to maximize communication performance. As a result, SDR is expected to play a significant role in the future of connected car technology as this communications agility is needed given the challenges any communication system operating in that environment will face.
Dig Deeper on Internet of Things (IoT) Wireless
Have a question for an expert?
Please add a title for your question
Get answers from a TechTarget expert on whatever's puzzling you.