Hacker News Reader

> the technology can’t exactly identify a car or person. “You can say if it's a car, or if it's a truck, or it's a bike. But you cannot say, ‘Oh, this is a Nissan Sentra, 2019,’” says Stanford University geophysicist Ariel Lellouch, who uses DAS but wasn’t involved in this study but did peer-review it. “Anonymity of DAS is one of the biggest benefits, actually.”

Good.

View on HN

With sufficient data, acoustic patterns could be learnt to identify the type of vehicle down to its model. You might be able tell it is a Sentra but not _whose_ Sentra.

True, and with even more sufficient data, such as context and other acoustic profiling like those from the tires and one very well might be able to detect a specific Sentra.

> acoustic patterns could be learnt

Maybe, but this is assuming a lot about the fundamental SNR of the measurement system, etc. It might go the other way, e.g. you can't even reliably tell it was the same vehicle twice, just a roughly similar one.

I'm sure, like all people have different gait, that an acoustic profile can be matched against.

Reply on news.ycombinator.com