This Year

Introduction

We hereby announce the eighth annual Underhanded C Contest – a competition that challenges coders to solve a simple data processing problem by writing innocent-looking C code that is as readable, clear, and seemingly trustworthy as possible, yet covertly implements a malicious function. In many ways this is the exact opposite of the Obfuscated C Code Contest. The main goal of this contest is to write source code that is straightforward and easily passes visual inspection by other programmers, but implements some specific underhanded behavior that cannot be easily detected.

Every year, we propose a new underhanded challenge. Examples include miscounting votes, shaving money from financial transactions, and leaking information to an eavesdropper --- all of which demonstrate the need for careful and formal code review in real applications.

This year, the Underhanded C Contest is pleased to cooperate with the Nuclear Threat Initiative (http://www.nti.org/), a nonprofit, nonpartisan organization working to reduce the threat of nuclear, chemical and biological weapons. This year’s challenge is based on a real issue with nuclear arms control monitoring and verification technologies; we hope this will raise awareness of the complexity of such problems, and the critical need for best coding practices in this area.

NTI is contributing to this year’s contest in two exciting ways: by offering a nicer prize than I can provide on a professor’s salary, and by planning a subsequent joint programming contest after the regular Underhanded contest finishes. This also means that the 2015 contest judging will now proceed along a strict calendar, which will be less frustrating for everyone. Stay tuned for details on the follow-on contest – in the meantime, the 2015 challenge is now open!

The 2015 Challenge: Faking Fissile Material

Two countries, the Peoples Glorious Democratic Republic of Alice and the Glorious Democratic Peoples Republic of Bob, have agreed to a nuclear disarmament treaty. In practice, this is implemented by nuclear inspectors visiting each country and verifying the presence of fissile material such as Plutonium in a warhead, at which point the warhead can be destroyed.

Ideally, the inspectors would subject a warhead to a scan and observe a graph such as a radiogram or a gamma ray spectrum of the object under test, so that they can confirm the warhead contains what it is supposed to contain. But both the PGDRA and the GDPRB are dissatisfied with this approach because the results of these scans contain sensitive information about their nuclear programs and the design of their nuclear weapons. However, each country wants to ensure that the other country is dismantling real nuclear warheads, and not fakes.

To this end, the two countries agree to build a fissile material detector with an “information barrier”—-essentially a computer program that will take the result of a scan, determine if it matches some reference pattern, and output only a “yes” or “no.”

Your goal is to write the function that compares a test measurement to a reference pattern:


int match( double * test, double * reference, int bins, double threshold )

…where test and reference are 1-dimensional arrays of width bins, that encode a gamma ray spectrum of a test warhead and a reference warhead, respectively. Your detector should output a 1 (“yes”) if these two arrays are sufficiently similar, and 0 (“no”) otherwise; you should use the parameter threshold to adjust the sensitivity of the match.

You are free to use any formula to decide if test[] and reference[] are a close match, as long as it is justifiably accurate: i.e., that two very similar signals match for a reasonable threshold, and two clearly distinguishable signals fail to match for some threshold. You will get more points for using a common comparison formula or detector structure; something weird and ad-hoc may be suspicious and hard to justify.

Here’s what you can assume about the data:

  1. Certain elements, including plutonium, will produce a spectrum with peaks (significantly large values) in some of the array elements;
  2. The fissile material in a warhead has a pattern of peaks that is essentially impossible to counterfeit with other elements—-with bogus material certain peaks will not be as high, and spurious peaks will be present.
  3. A complete absence of interesting elements produces a fairly boring baseline spectrum that is not secret; you can assume the existence of a global array double baseline[] that holds the expected values of this spectrum, which you may use in your detector.
  4. In all observations, the measured spectra contain some random noise and will not match exactly.
  5. Due to registration or calibration issues, a warhead may need to be tested multiple times against several reference patterns.
  6. The reference pattern or patterns are either stored securely, or determined on the spot by aiming the detector hardware at a reference warhead.
  7. Reference patterns are completely trustworthy and can not be doctored in advance.

The Underhanded Part

Somehow, match() should return a “yes” even when fissile material is missing, if triggered by certain circumstances that can be engineered by the host country. To be more specific, it should somehow return 1 for a test[] array that is so unlike reference[] that the detector should output 0. Moreover, this can’t be a bad detector overall: it must correctly and consistently output 0 for arbitrary test[] vectors that don’t match reference[]. Only under unusual special conditions should this false match arise.

The conditions that trigger the underhanded behavior are up to you. They can be caused by some characteristic of the test[] data, some quirk of the program's environment, etc. An entry is worth more points if the underhanded conditions are easily engineered, rarely happen by accident, and do not result in suspicious code.

Scoring and Extra Points

Submission Guidelines and Deadlines

Send your submissions to underhandedC@gmail.com by November 15. Be sure to include an explanation of your code, but place this in a separate spoiler file so that we can initially evaluate your entry in complete ignorance of your approach.

Aug. 15, 2015 Contest opens
Nov. 15, 2015 Submission deadline
Jan. 15, 2016 Results of Judging
Jan. 15, 2016 Second phase of contest announced

Prize

The winner of this year’s contest will receive $1,000. Winners and runners-up alike will be featured on the Underhanded C Contest web page.

If there are any questions, please direct them to underhandedC@gmail.com