Featured Project
# Schnorr Identification Protocol Key Fob
Team Members:
- Michael Gamota (mgamota2)
- Vasav Nair (vasavbn2)
- Pedro Ocampo (pocamp3)
# Problem
Current car fobs are susceptible to different types of attacks. Rolling jam attacks are one of such attacks where an attacker jams and stores a valid "unlock" signal for later. Cars with passive keys/cards can be stolen using relay attacks. Since a car can be the most expensive item someone owns, it is unreasonable to allow people to steal them so discreetly by hacking the fob/lock combo.
# Solution
By leveraging public key cryptography, specifically the Schnorr identification protocol, it is possible to create a key fob which is not susceptible to either attack (rolling jam and relay) and also gives no information about the private key of the fob if the signal were to be intercepted.
# Solution Components
# Key Fob
## Subsystem 1
Random number generation - We will use a transistor circuit to generate random numbers. This is required by the Schnorr protocol to ensure security.
## Subsystem 2
Microcontroller - The MCU will run all the computation to calculate the messages. We will likely use an ATtiny MCU so we can use the Arduino IDE for programming. However, some group members have experience with the STM32 family so that is another option.
## Subsystem 3
Power - We plan on using either a 5V battery or 3.3V battery with a boost converter to power the fob.
## Subsystem 4
Wireless Communication - We plan on using the 315 MHz frequency band which is currently used by some car fobs. We will need a transmitter and receiver, since the protocol is interactive.
# Lock
## Subsystem 1
Random number generation - We will use a transistor circuit to generate random numbers. This is required by the Schnorr protocol to ensure security.
## Subsystem 2
Microcontroller - This MCU will also run all the computation to calculate the messages. We will likely use an ATtiny MCU so we can use the Arduino IDE for programming. However, some group members have experience with the STM32 family so that is another option. This MCU will need to have PWM output to control the lock.
## Subsystem 3
Linear Actuator - We plan on using a linear actuator as a deadbolt lock for demonstration purposes.
## Subsystem 4
Wireless Communication - We plan on using the 315 MHz frequency band which is currently used by some car fobs. We will need a transmitter and receiver, since the protocol is interactive.
## Subsystem 5
Power - This subsystem will also likely require 5V, but power sourcing is not an issue since this system would be connected to the car battery. During a demo I would be acceptable to have this plugged into a power supply or a barrel jack connector from an AC-DC converter.
# Criterion For Success
Describe high-level goals that your project needs to achieve to be effective. These goals need to be clearly testable and not subjective.
Our first criteria for success is a reasonably sized fob. There is some concern about the power storage and consumption of the fob.
The next criteria for success is communication between the fob and the lock. This will be the first milestone in our design. We will need to have a message sent from one MCU that is properly received by the other, we can determine this in the debug terminal.
Once we are sure that we can communicate between the fob and the lock, we will implement the Schnorr protocol on the two systems, where the fob will act as the prover and the lock as the verifier. If the Schnorr signature implementation is correct, then we will always be able to unlock the lock using the fob whose public key is associated with full privileges.
Project Videos