Physically Observable Side Channels Flashcards

1
Q

Basics of an inverter

A

When A == 0, charge capacitor, output becomes 1

When A == 1, discharge capacitor, output becomes 0

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

How do we tell a 0 vs 1 in a power side channel?

A

When A == 0, we need to charge capacitor. Power graph will be logarithmic. When A == 1, we discharge capacitor, will get short burst of energy

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

What information can be obtained from power consumption?

A

How many bits were toggled

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

Difference between power and EM SC

A

EM SCs are a bit more observable than the power side channel, even though they don’t contain any more information

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

How is an EM SC created

A

The changes in current in an inverter create pulses in the EM field

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

What are the two ways IC emit EM

A
  • Conductive emissions - signal is radiated from the pins in the circuit
  • Electric and magnetic near-field emissions - EM field is generated due to current inside the IC
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

how are conductive emissions measured?

A

antennas

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

how are E and M near-field emissions measured?

A

E and M probes

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

Thermal Dissipation Side Channel

A

When the capacitor is charged, the resistor releases heat => not very fine-grained

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

Acoustic Side Channel

A

Mechanical vibrations of capacitor lead to acoustic vibrations

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

What tools can we use to measure power SC?

A

real-time oscilloscope, real-time spectrum analyzer, and software defined radio

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

real-time vs sampling oscilloscope

A

real-time takes samples over one run of execution, sampling averages samples over many runs

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

real-time oscilloscope measures what?

A

A signal with respect to time

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

spectrum analyzer measures what?

A

A signal with respect to frequency

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

Which is better spectrum analyzer or oscilloscope?

A

Spectrum analyzer - has better noise floor, can change frequency bandwidth

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

Software defined radio’s differentiator

A

has many configuration options

17
Q

what can we learn from a power/em SC?

A
  • Repetitive patterns in trace indicate general structure of program (loops, etc)
  • Time - what happens when, program flow
  • Amplitude - what happens at a given moment in time, data flow
18
Q

How will the same operation consume power with different operands?

A

It will have different values

19
Q

Chassis potential side channel?

A

???

20
Q

Approach for simple Power/EM analysis attacks

A
  • Visually inspect traces, looking at best/worst case inputs

- Look for patterns , timing differences, amplitude differences

21
Q

What can patterns in simple power/EM analysis leak?

A
  • Key length (number of rounds or iterations)
  • Memory accesses
  • Implementation details
  • Key (if really bad implementation)
22
Q

What can amplitude in simple power/EM analysis leak?

A
  • Exact operand values (extreme case)
  • Hamming weight or hamming distance
  • Operation being executed
  • RE of implementation details
23
Q

What can timing in simple power/EM analysis leak?

A
  • Data dependent branching

- Execution of additional operations

24
Q

Differential Power/EM Analysis Attack

A

homework assignment! This works because the different groups (LSB = 0 and 1) are split correctly only when the key is correct ==> can actually see the power difference. When key is not correct, differences are averaged out