When your wireless carrier charges you for the amount of data you used on your cell phone in a given month, how do you know the bill is accurate? It very well might not be, according to a new study.
This question is more important to consumers than ever. Over the past year, the growth in the popularity of smartphones has led the largest U.S. mobile carriers to replace unlimited data plans with ones that place caps on data usage, and charge extra for exceeding those limits.
Working with three colleagues at the University of California, Los Angeles, computer science PhD researcher Chunyi Peng probed the systems of two large U.S. cell-phone networks. She won’t identify them but says that together they account for 50 percent of U.S. mobile subscribers. The researchers used a data-logging app on Android phones to check the data use that the carriers were recording. The carriers were found to usually count data correctly, but they tended to overcount—and hence potentially overcharge—when a person used applications that stream video or audio, and particularly when coverage was weak or unreliable.
The researchers determined that even typical use of a phone could lead the data to be overcounted by 5 to 7 percent, Peng says. That could cost customers money. The two largest U.S. wireless networks, AT&T and Verizon, both charge a user $15 for straying into each new gigabyte of data over the data cap.
The problem stems from the way networks count data use. They count data as it leaves the heart of a company’s network and sets out on the journey to the mobile tower nearest a subscriber. That means data is counted whether a phone receives it or not. If a person on a bus is streaming video but enters a tunnel and loses her connection, for example, video that she never sees will already have been counted toward her plan.
The problem affects video and audio streaming apps in particular because they use protocols that don’t require the receiving device to acknowledge the receipt of every chunk of data, or halt data transmission immediately, as Web browsers or many other apps do. That means a video app will keep sending data for some time, oblivious to the fact that a device can’t receive it.
Using a custom app made to demonstrate the flaw they had uncovered, the UCLA researchers racked up a charge for 450 megabytes of data they never received. “We wanted to explore how bad it could be, and stopped after that,” says Peng. “There’s apparently no limit.”
The researchers also found that data use can be hidden from the two cellular networks they tested. Cellular networks’ data accounting ignores a type of data transfer known as a DNS request, used by Web browsers to translate Web addresses into the numerical address of the server hosting a website. A normal request for data can escape a network’s data accounting when disguised as a DNS request, says Peng. An app developed to exploit that was able to use 200 megabytes of data without the carrier recording any of it, a tactic that mobile carriers would probably like to block.
It should be relatively simple for mobile network operators to tweak how they measure the data transferred to their customers’ devices, says Peng, by adding software to make the phones give some feedback to the central data loggers. She notes that carriers might argue that their current accounting is fair, since they incur the costs of transmitting data whether it successfully reaches a device or not. “From the perspective of a mobile user, I think it’s not fair,” she says, “because I didn’t get to use it.”
The introduction of data caps has been controversial, with some customers and advocacy groups saying that mobile networks are trying to increase profits more than they are genuinely seeking to cut data use as smartphone users overwhelm their networks. A nonpartisan think tank, the New America Foundation, has raised concerns with the U.S. Federal Communications Commission about the lack of transparency in how carriers measure data usage and communicate it to customers. Benjamin Lennett, a technology policy director at the foundation, said that if the FCC is willing to allow usage-based prices, the agency should “ensure that consumers are being charged accurately.”
Peng presented her work at the MobiCom conference on mobile computing research last month in Istanbul, Turkey. The UCLA group is working with the carriers they investigated to explain their findings, and also developing an Android app to help smartphone users track their data use more accurately, to help them dispute erroneous charges.
This new data poisoning tool lets artists fight back against generative AI
The tool, called Nightshade, messes up training data in ways that could cause serious damage to image-generating AI models.
Rogue superintelligence and merging with machines: Inside the mind of OpenAI’s chief scientist
An exclusive conversation with Ilya Sutskever on his fears for the future of AI and why they’ve made him change the focus of his life’s work.
Data analytics reveal real business value
Sophisticated analytics tools mine insights from data, optimizing operational processes across the enterprise.
Driving companywide efficiencies with AI
Advanced AI and ML capabilities revolutionize how administrative and operations tasks are done.
Get the latest updates from
MIT Technology Review
Discover special offers, top stories, upcoming events, and more.