Connected Cars Spell Trouble for Cellular, Says Machina

Connected cars might help their drivers to avoid traffic jams but they could make the congestion you find on mobile networks much worse, according to a new piece of research.

A whitepaper published today argues that spikes in network traffic stemming from the use of connected cars could lead to all manner of problems for mobile operators over the next decade, potentially affecting their ability to provide mainstream services.

According to Machina Research, which carried out the study, certain cells could see a 97% increase in data traffic during rush-hour periods, when drivers of connected vehicles are traveling to and from work.

"Potentially it could be a very painful thing if you have knock-on effects on existing services," said Matt Hatton, the founder and CEO of Machina, during an interview with Light Reading.

The problem is more to do with the "localization" of demand than a sharp rise in overall data traffic as M2M technologies are rolled out. Indeed, Machina reckons M2M applications will account for just 4% of overall traffic on cellular networks over the next 10 years. In certain areas and at certain times, however, there could be major congestion resulting from specific types of service.

"If, say, there is a traffic accident, children in connected cars may all start watching videos at the same time, creating a hotspot in that area," says Steve Bowker of analytics company Teoco Corp. , which commissioned the study from Machina.

This "clustering" effect could have ramifications for various parts of the mobile network, according to the research. If twice as many devices are trying to access services as an operator would typically expect, the radio access network may struggle to cope, while the signaling "overheads" associated with M2M devices could lead to core network problems.

Teoco clearly has a vested interest in telling this story because its analytics products are designed to help operators identify spikes in traffic so they can take appropriate measures. Yet both Hatton and Bowker insist the expense of adapting networks is negligible compared with the potential cost of doing nothing, which may reassure operators given the low revenues associated with M2M services.

"Sure, revenues will be low but the costs [of the technology rollout] will be much lower," says Bowker. "You don't want to be fixing problems after they've happened."

Machina spells out the risks in very blunt terms in its whitepaper. "If M2M devices regularly generate spikes in usage in a particular location which cannot be met, there are implications for customer satisfaction, and even the risk of non-compliance with service level agreements."

Want to know more about the Internet of Things? Check out our dedicated IoT content channel here on Light Reading.

Among the technology options for coping with these usage spikes are more dynamic network management and RAN optimization, according to Machina, as well as greater focus on device management and what Hatton describes as "a more considered approach to spectrum re-farming."

Although connected cars are not the only M2M devices that may trigger network problems, they are a particular source of concern because of expectations about growth in this area.

Machina is forecasting there will be 2.3 billion cellular M2M connections worldwide by 2024, with connected cars accounting for about half of the total.

Cellular M2M Connections by Vertical Market
Source: Machina Research
Source: Machina Research

Asked which companies can really make money from the whole M2M and Internet of Things opportunity, Hatton said the professional services providers are currently in the best position.

"The professional services people -- the ones responsible for systems integration and the handholding of customers through the process of developing services -- they come out as the ones generating the greatest revenues," he says.

— Iain Morris, Circle me on Google+ Follow me on TwitterVisit my LinkedIn profile, News Editor, Light Reading

nasimson 5/25/2015 | 11:49:56 PM
Re: Congestion planning At Telenor we are deploying smart meters by thousands. After reading this, this has got me thinking. One of the things I am going to do is to evaluate the network congestion at some of the densely deployed sites.
Kruz 5/25/2015 | 6:02:17 PM
Re: Congestion planning Isn't this localization of demand something that is occuring now and should be already treated? This doesnt seem like a direct impact of M2M to me. The solution would be priorities and a "liquid like" network with both a proactive and reactive CEM approach.
Manageme51384 5/22/2015 | 1:56:09 AM
Issues with M2M The article very aptly points out the issue with M2M. The possible solutions with operators is that they can deploy wifi hotspots throughout the city so as to offload the M2M traffic. Another issue that operators will face apart from traffic times is the no of connect and disconnect request from the machines. M2M devices send data in bursts which creates a problem for the operators. Operators can have dynamic assingment of channels at each cell site so as to increase or decrease the capacity according to traffic
mkanterman 5/21/2015 | 6:13:54 AM
Re: M2M Congestion Isn't this also why carriers are deploying new technologies at the cell site such as massive MIMO and other technologies to be able to better handle dense traffic congestion at the cell site level rather than broader network traffic? Densification can also help simply by making there more cells for the many devices to be spread across.
[email protected] 5/21/2015 | 4:17:18 AM
Congestion planning One would hope that the mobile operators have already identified this potential issue and are already planning their congestion relief strategies accordingly RIGHT NOW. 

And this is why video compression is an important part of that, right?


INteersting nod to how mucg business the systems integrators could get from this... is this about to become a golden age for the global, multi-vertical IT consultancies?
Sign In