HARMAN’s Till: 5G is “Game-Changer” for CVs

Greg Hyde talks to HARMAN vice-president of marketing and technology Andrew Till about opportunities and threats within the connected vehicle sector.

There has been much discussion within the last few months about the advantages and pitfalls of connecting one’s car to the IoT and the ‘smart city’ infrastructure that is needed for this to happen. Samsung subsidiary HARMAN is one company producing technology that professes to allow drivers to connect their cars securely. In an exclusive interview with TU-Automotive, Till talked about how he felt the rollout of 5G would bring great opportunities for CVs, but also about how he thinks more can still be done to secure the accompanying infrastructure against potential attacks by hackers.

 

 

Q: Do you think the rollout of 5G will benefit the development of CV tech?

“I think 5G is a fundamental game-changer for the connected car industry, on two levels. One is the ability of 5G and the consciousness that emanates from 5G solutions to enable the commissioning of AI between the edge, the vehicle, and the cloud. We can make many more advances on this tech from where it is today. If you look at 4G, we’ve gone from that to being able to provide the real-time services needed to take real-time decisions in a vehicle. This will have benefits for connecting ADAS systems and, further down the line, autonomous vehicles.”

“The second area that I think is really very powerful, and we will see much more literature on this, is network-slicing and being able to effectively run a VPN network to improve the quality of service capabilities at broadcasting information out to vehicles. I just think a basic system you could have for vehicles on the road receiving network-sliced real-time traffic updates is a very basic first step. What a game-changer.”

“That then becomes one of the challenges for the latest operating systems. CV systems currently lag behind the real-time events on the road when they’re collecting information and sending it to people’s cellphones, and they’re also trying to send data from the telematics grid to the cloud. Many, many connected cars just don’t get those benefits. But I think network-slicing technology will provide both cities and automakers with the ability to deliver the beta services that house the training experience on the data being provided. That is something that, up until now, we just haven’t got close to being able to do.”

Q: Do you think national and civic authorities are doing enough to make our towns’ and cities’ infrastructure compatible with vehicle-to-infrastructure (V2I) tech?

“It’s easy to say that there’s ways that it can be done but I think it’s tempting to rush into it and forgo all the planning that has to go into a smart city environment. I think all the focus is on smart vehicles and connected vehicles and their benefits to the infrastructure. At the same time, you don’t want to rush in too early and have to replace that technology because the technology standards have gone in a different direction. To be honest I think, one of the big challenges any city faces now is: which standards are the automakers going to go into from the point of view of vehicle-to-vehicle (V2V) and V2I communications? Can they make that bet right now with the guarantee that five years down the line, they won’t have to spend $1Bn replacing technology they spend money on now?”

“I think what we’re seeing right now is a lot of focus on getting data architecture and data ontology structured in such a way that data can be shared between departments. A great example of this is traffic accidents. You need to be able to share the data between hospitals, public transport, Highways UK, and today they’re not quite there. There’s a pipeline of data and IT systems, and there’s a project taking place to try and get this ready. You can do that last bit where you push a message to two vehicles on a highway ordering them to move out of the way and allow that ambulance through on the fast lane as quickly as possible. I know there’s so much work taking place from a planning perspective between all those departments, and that’s where I see a lot of activity taking place.”

Q: Threatcare and IBM X-Force Red recently carried out a collaborative study that found seven different vulnerabilities in Battelle’s smart city system, which they said could assist attacks on V2I tech. These included instances of “sensitive functionality” being available without any authentication process, and authentication processes which were in place being easy to bypass. Do you think more needs to be done to make smart city systems secure?

“At a very high level, I would say, no matter what point in time that we are moving forward into the future, that will always be an answer where you say yes. Given the nature of the world we live in today, no matter how safe we make our systems, there will always be people out there that are trying to do something nefarious and malicious. We have to keep focusing on how we progress the systems. So coming back specifically to the Threatcare/IBM X-Force Red study, what it highlighted for smart cities was really the challenge of thinking that through on a system level.”

“The function of security is not about an isolated point in the network, such as attacking the vehicle or attacking one server here and there, it’s to try and provide a level of security where you’re confident you can really provide the protection to citizens, whether they are pedestrians, users of public transport, private vehicle owners. You have to look at it as an end-to-end system and think about security at a system level. That extends from what we do inside the vehicle to make better telematics control units (TCUs) and isolated systems and how we design our rating systems for the next generation of vehicles all the way through to those government facilities and their command-and-control centers for the smart city environment.”

“It also extends to things like smart traffic lights that a car can connect to, and a smart city grid being hackable that way. That’s not what people normally think about when these attacks take place. They tend to think about attacks taking place against data centers and people using this style of attack to explore vulnerabilities and attack that way, attack the physical infrastructure itself. We have to think about this as an end-to-end environment. I do think we are seeing responses to these challenges much more now from vicinities that have chief information officers (CIOs). They are increasingly being recruited from the enterprise world, and they are bringing that enterprising level of system architecture and security awareness to the agenda in the smart city debate, and I think that’s an incredibly positive thing that’s happening.”

“This is really a five-year-old discussion that’s started picking up recently about where this technology can take us and how it enriches life. You’re dealing with technology that in some cases was implemented 14 or 15 years ago. So the really big security processes, that means that they’re in the short term going to keep finding challenges to security, and that’s why they’re getting all their systems updated in terms of the modern security paradigm. With regards to everything, from how they design their software to be secure from the ground up, all the way through to how you might have 400 or 500 different data partners providing weather feeds into the smart city environment, providing information about results from hospitals, etc, etc, how they upload their data in a way that is secure. You can detect in the data flows how an API is being queried, or a payload is being structured, or information is being delivered into the smart city environment. Again, [it means] bringing that philosophy of always verifying, seeing that payloads aren’t being delivered from servers that aren’t what they claim to be.”

“So I think this will undoubtedly be a very important part of people’s lives. We tend to take the view that how these things are identified and talked about, such as the Threatcare/IBM X-Force Red study that identified vulnerabilities, is actually very positive. You should be having ‘white hat’ hackers testing your systems, doing the penetration testing, and then when they do find these issues, I think it’s great that they’re talking about them. These issues get visibility and then people are able to build that into their test plans and their security models moving forward. The more that happens, the more we can be confident that the right level of security is in place to protect everybody.”

 


Leave a comment

Your email address will not be published. Required fields are marked *