Lectures

The twelve-week argument, one line per week, and the lecture behind each line.

A course is one argument, not twelve topics. Weeks 2 through 8 take one physical channel each and argue it from mechanism to disclosure to what it costs to close; week 9 steps back to the precondition the shared-hardware channels depend on; the last three weeks turn the accumulated list into a rubric, a defensive discipline and a design method.

Below is the argument in twelve lines — what each week changes, rather than what it covers. If a line could be removed without the rest weakening, that week is the wrong week.

  1. Week 1

    A computation's unintended physical effects are not noise around the answer; they are a second output, and this course reads them as one.

  2. Week 2acoustic

    A channel needs no contact and no electronics: a microphone in the room is enough, once a language model supplies what the signal alone cannot.

  3. Week 3power

    A single measurement can be unreadable while ten thousand of the same measurement are decisive: statistics, not sensitivity, makes a channel practical.

  4. Week 4timing

    A channel needs no physical access at all: a clock at the far end of a network is a probe, and duration is the quantity it measures.

  5. Week 5electromagnetic

    Closing a channel is an engineering cost with a shape, and that shape — not the mechanism — decides who ever pays it.

  6. Week 6cache

    Two programs that share nothing but a processor still share its memory hierarchy, and the timing of that hierarchy is enough.

  7. Week 7optical

    A surface nobody designed to be a display can be one, which makes this channel a property of the room rather than of the device.

  8. Week 8thermal

    A channel can outlive the computation that produced it, which defeats every defence assuming a defender present while the leak happens.

  9. Week 9

    Before a shared-hardware channel can be used it has to be found, and locating the sharing is a separate problem with its own separate defence.

  10. Week 10

    "It leaks" is not a finding until it is a number; mutual information makes seven channels argued on their own terms comparable on one.

  11. Week 11

    Detecting a leak and closing one are different skills, and a rewrite that only makes a channel harder to see is not a fix.

  12. Week 12

    No defender closes every channel, so the semester's output is a method for deciding which ones a stated device and adversary make worth closing.

Every week in full

What a machine says without meaning to

The opening lecture — stating the course's thesis and the shape of the twelve weeks that argue it

Week 1 · 22 February 2027

Acoustic emanations

What a keyboard's sound reveals about what was typed on it, and how far the effect generalises

Week 2 · 1 March 2027

Reading a power trace

Differential power analysis, and why a device's current draw is close to a printout of the arithmetic it is doing

Week 3 · 8 March 2027

Timing as a channel

When a comparison's running time depends on how much of the secret it agrees with, and how far that reaches over a network

Week 4 · 15 March 2027

Radiation nobody asked for

Van Eck phreaking, TEMPEST, and what it costs to stop a display's own electronics from broadcasting what it shows

Week 5 · 22 March 2027

What a shared cache gives away

Reading a neighbouring process's memory-access pattern through nothing but the timing of your own cache hits and misses

Week 6 · 29 March 2027

Compromising reflections

Recovering a screen's contents from its reflection in glasses, a teapot, or an eyeball, at a distance and after the fact

Week 7 · 5 April 2027

Heat that outlasts the keystroke

Thermal imaging of a keypad after use, and why this channel's clock keeps running after the exploit's own action stops

Week 8 · 12 April 2027

Establishing that the hardware is shared

Colocation as a precondition every shared-hardware channel depends on, and as a problem a provider can defend on its own terms

Week 9 · 19 April 2027

The leakage taxonomy

Putting one rubric under seven individually-argued channels, and attaching a number to what each of them actually reveals

Week 10 · 26 April 2027

Closing a channel in software

Constant-time programming as a discipline, and what it costs to actually practise it

Week 11 · 3 May 2027

Choosing which channels to defend

The closing lecture — turning a semester of separately-argued channels into one design method with a stated stopping point

Week 12 · 10 May 2027

What the course reads

Each week’s readings are listed on its lecture page. Collected here in the order the field produced them, they make a second argument: a 1973 memo that named the problem before any of these channels had been measured, two decades in which each disclosure was its own isolated result, and then a turn — from roughly 2005 — toward papers whose contribution is a bound or a countermeasure rather than another channel.

  1. 1973Butler W. Lampson. A Note on the Confinement Problem. Communications of the ACM 16(10), 613-615. week 1
  2. 1985Wim van Eck. Electromagnetic Radiation from Video Display Units: An Eavesdropping Risk?. Computers & Security 4(4), 269-286. week 5
  3. 1996Paul C. Kocher. Timing Attacks on Implementations of Diffie-Hellman, RSA, DSS, and Other Systems. CRYPTO '96, LNCS 1109, 104-113. week 1
  4. 1999Jean-Sebastien Coron. Resistance Against Differential Power Analysis for Elliptic Curve Cryptosystems. CHES 1999, LNCS 1717, 292-302. week 11
  5. 1999Paul Kocher, Joshua Jaffe and Benjamin Jun. Differential Power Analysis. CRYPTO '99, LNCS 1666, 388-397. week 3
  6. 1999Suresh Chari, Charanjit S. Jutla, Josyula R. Rao and Pankaj Rohatgi. Towards Sound Approaches to Counteract Power-Analysis Attacks. CRYPTO '99, LNCS 1666, 398-412. week 3
  7. 2001Jean-Jacques Quisquater and David Samyde. ElectroMagnetic Analysis (EMA): Measures and Counter-Measures for Smart Cards. E-smart 2001, LNCS 2140, 200-210. week 5
  8. 2002Joe Loughry and David A. Umphress. Information Leakage from Optical Emanations. ACM TISSEC 5(3), 262-289. week 7
  9. 2002Suresh Chari, Josyula R. Rao and Pankaj Rohatgi. Template Attacks. CHES 2002, LNCS 2523, 13-28. week 10
  10. 2003David Brumley and Dan Boneh. Remote Timing Attacks are Practical. USENIX Security 2003. week 4
  11. 2003Markus G. Kuhn. Compromising Emanations: Eavesdropping Risks of Computer Displays. University of Cambridge Technical Report UCAM-CL-TR-577. week 5
  12. 2004Dmitri Asonov and Rakesh Agrawal. Keyboard Acoustic Emanations. IEEE Symposium on Security and Privacy 2004, 3-11. week 2
  13. 2005Colin Percival. Cache Missing for Fun and Profit. BSDCan 2005. week 6
  14. 2005Daniel J. Bernstein. Cache-timing attacks on AES. Technical report, University of Illinois at Chicago. week 11
  15. 2005Li Zhuang, Feng Zhou and J. D. Tygar. Keyboard Acoustic Emanations Revisited. ACM CCS 2005, 373-382. week 2
  16. 2006Dag Arne Osvik, Adi Shamir and Eran Tromer. Cache Attacks and Countermeasures: The Case of AES. CT-RSA 2006, LNCS 3860, 1-20. week 6
  17. 2007Stefan Mangard, Elisabeth Oswald and Thomas Popp. Power Analysis Attacks: Revealing the Secrets of Smart Cards. Springer, chapters 6-9 on countermeasures. week 3
  18. 2008Michael Backes, Markus Duermuth and Dominique Unruh. Compromising Reflections, or How to Read LCD Monitors Around the Corner. IEEE Symposium on Security and Privacy 2008, 158-169. week 7
  19. 2009Francois-Xavier Standaert, Tal G. Malkin and Moti Yung. A Unified Framework for the Analysis of Side-Channel Key Recovery Attacks. EUROCRYPT 2009, LNCS 5479, 443-461. week 10, week 12
  20. 2009Julien Brouchier, Tom Kean, Carol Marsh and David Naccache. Temperature Attacks. IEEE Security & Privacy 7(2), 79-82. week 8
  21. 2009Michael Backes, Tongbo Chen, Markus Duermuth, Hendrik P. A. Lensch and Martin Welk. Tempest in a Teapot: Compromising Reflections Revisited. IEEE Symposium on Security and Privacy 2009, 315-327. week 7
  22. 2009Scott A. Crosby, Dan S. Wallach and Rudolf H. Riedi. Opportunities and Limits of Remote Timing Attacks. ACM TISSEC 12(3), article 17. week 4
  23. 2009Thomas Ristenpart, Eran Tromer, Hovav Shacham and Stefan Savage. Hey, You, Get Off of My Cloud: Exploring Information Leakage in Third-Party Compute Clouds. ACM CCS 2009, 199-212. week 9
  24. 2011Keaton Mowery, Sarah Meiklejohn and Stefan Savage. Heat of the Moment: Characterizing the Efficacy of Thermal Camera-Based Attacks. USENIX WOOT 2011. week 8
  25. 2014Daniel Genkin, Adi Shamir and Eran Tromer. RSA Key Extraction via Low-Bandwidth Acoustic Cryptanalysis. CRYPTO 2014, LNCS 8616, 444-461. week 2
  26. 2014Yuval Yarom and Katrina Falkner. FLUSH+RELOAD: A High Resolution, Low Noise, L3 Cache Side-Channel Attack. USENIX Security 2014, 719-732. week 6
  27. 2015Ramya Jayaram Masti, Devendra Rai, Aanjhan Ranganathan, Christian Mueller, Lothar Thiele and Srdjan Capkun. Thermal Covert Channels on Multi-core Platforms. USENIX Security 2015, 865-880. week 8
  28. 2015Venkatanathan Varadarajan, Yinqian Zhang, Thomas Ristenpart and Michael Swift. A Placement Vulnerability Study in Multi-Tenant Public Clouds. USENIX Security 2015, 913-928. week 9
  29. 2016Jose Bacelar Almeida, Manuel Barbosa, Gilles Barthe, Francois Dupressoir and Michael Emmi. Verifying Constant-Time Implementations. USENIX Security 2016, 53-70. week 11
  30. 2018Qian Ge, Yuval Yarom, David Cock and Gernot Heiser. A Survey of Microarchitectural Timing Attacks and Countermeasures on Contemporary Hardware. Journal of Cryptographic Engineering 8(1), 1-27. week 10, week 12
  31. 2020Ross Anderson. Security Engineering, third edition, chapter 19: Side Channels. Wiley. week 12