My group does research on distributed systems, mobile computing, and operating systems with a focus on privacy and security. Our work is currently supported by several grants from the National Science Foundation and the Intel Science and Technology Center for Secure Computing at UC-Berkeley.

For a better sense of the kinds of problems that interest us, please refer to our papers.

Landon Cox

Landon Cox

Aug 2011

TaintDroid for Android 2.3.4!

Now with Nexus S and JIT support! This is a major update that required a significant investment of time and effort. Many thanks to Peter Gilbert, Seungyeop Han, and Will Enck for a job well done.

To quote
Peter’s announcement to the TaintDroid Google Group:

We are happy to announce an updated source code release of TaintDroid based on Android 2.3.4. Please visit http://www.appanalysis.org/download.html for instructions on how to download and build the code. Major changes include:

  • support for Nexus S in addition to Nexus One
  • support for Dalvik JIT compiler
  • taint propagation for native inline functions
  • SSL socket taint sink
  • and other minor bug-fixes

Please post any questions or problems to this discussion group.
Thanks for your interest in TaintDroid!

YouProve at SenSys

We were really happy when we learned that our paper on YouProve had been accepted to SenSys 2011. YouProve is follow-up work to our HotMobile ’10 position paper on using trusted hardware (i.e., TPMs) to help participatory sensing services verify the authenticity of audio and photos collected from anonymous mobile devices. In other words, if you are building a service that uses photos and other media captured by anonymous mobile phone users, how can you know that the data you receive is authentic? How do you know that it hasn’t been invented or spoofed?

Even if you have secure hardware, there are a number of thorny issues that make these questions hard to answer, including user privacy, developer flexibility, legacy apps, users’ app choice, and devices’ limited bandwidth and battery power. YouProve is not a silver bullet, but we think it is a good start and hope that more people will work on this very important problem. Please drop me an email if you have any feedback.

I would also like to mention that one of the best things about this paper is that it has three undergraduate co-authors: Kyungmin (Jason) Lee, Henry Qin, and DJ Sharkey. We have excellent CS students at Duke, and you should definitely be trying to recruit them to your company or grad school.

In any case, the rest of the SenSys program looks great too, and I hope to see you in Seattle in November.