Three Requirements for the Next Generation of Connected Cars

By Christer Boberg

Smart cars have the potential to change the world in a meaningful way.

Connected cars have sparked the imagination for decades. Reduced traffic, fewer accidents and more productive transit times are just a few of the benefits of a world with connected, smart cars.

There's a reason we've been talking about these cars for so long, but none of us is cruising around reading a book while the car drives for us: it's extremely difficult. The amount of data generated is mind-boggling, the connectivity requirements are intense and the stakes are extremely high.

It's one of the biggest technology challenges the industry has ever faced, and one of the biggest management challenges the industry has ever faced. It will require forward thinking and breaking barriers. Specifically, here are three requirements for automated cars to come out of science fiction and become a reality.

New Technology
At their heart, connected cars pose a technology challenge. Next-generation connected car services will generate an enormous amount of data, which will need to be shared between cars and the cloud, pushing exabytes of data per month over the networks.

Yet, the current version of mobile broadband on which the cars must operate was built with consumer services in mind. When this technology was introduced, there was never a thought of its potential in industrial cases, never mind connected and even autonomous vehicles. To bring it up to speed, there are a few important considerations. First is all of that data. Second is that cars are mobile, and they're fast compared to many other IoT appliances. Third, it is often critical either from a safety point of view and/or a business point of view. The technology, then, must be able to deal with many variables, as well as handle different quality and capacity situations in a predictive way.

The technology also has to be future-proof and built to last for decades, rather than merely for years. Think about cellular technology in general, giving superior support for future evolution with backward-compatibility in mind. People switch out cell phones every year or two, so technology can change often without a big effect. When talking about cars, though, they will have to work to support the cars and their many owners for 10 years or more, and the demand for having them connected will grow. That means automakers can't just jump on technology that won't necessarily be around a few years down the road.

Cooperation
When 5G was being developed, the needs and use cases for the automotive industry were taken into account. Telecom companies have been working closely with key automotive actors, and have solicited automotive industry feedback. This type of initial collaboration is a good start, but the next round of technology required to support connected cars—including the Internet of Things (IoT), security, connectivity, the cloud and so forth—will require collaboration on a whole different multi-actor level.

To truly get smart cars off the ground (figuratively), industries as diverse as cybersecurity providers, cloud vendors and automotive software providers, along with automotive OEMs, will need to begin working closely together. Regulators and governments will also need to be part of the discussion. Everyone from the car producers and owners to the network, connectivity and other technology providers, to the ones regulating everything will need to be on the same page and develop requirements that work.

This collaboration has already started in earnest. One example is the Automotive Edge Computing Consortium (AECC), which was launched when automakers began to realize the amount of data needed to make the ecosystem successful. Groups like AECC, 5GAA and others can help lead the way, but the problem requires broad and deep collaboration and understanding across industries.

Standardization
All of the things discussed so far point to the most important requirement: standardization. When creating devices (cars) that operate in a multi-actor environment on a global scale, it's critically important that standard infrastructure be in place. This allows for the development of standardized applications for the cars, as well as a framework to develop against.

Developing a consistent view of the technology to use represents a real challenge. Every industry involved—telecom, automotive, software and so on—has a different focus. To make the connected car dream a reality, companies in the automotive industry—the ones developing the cars themselves, along with the technology providers making this happen—must start moving toward a standard infrastructure. The supporting industries involved will then collaborate to work toward a suitable network architecture, deployment and business model for the automotive sector's specific application need.

Emerging standards like 5G—including improved capacity, network slicing for precise quality of experience support and distributed cloud for improved latency, security and compute utilization—are extremely important. The car industry is focusing on how to use the network for its commercial services, and it always comes back to the huge amount of data and the high bandwidth usage. 5G is promising, but any technology standard used has to work in a multi-network environment that includes 4G and even 3G—the old and the new—and be able to support the many different services the industry requires.

Smart cars have the potential to change the world in a meaningful way. Everything is in play, from eliminating traffic delays to making accidents a thing of the past. It won't come easily, however. These three areas are a start. Once the diverse parties pioneering this technology collaborate to develop the technology and standards required for the leap, the potential for innovation and change is beyond our wildest expectations.

Christer Boberg has worked at Ericsson for more than 20 years. Currently, he is the company's director of service layer technologies, technology strategy, business area technology and emerging business, and is driving technology strategies for the IoT and the cloud.