Safeguarding Unmanned Aerial Vehicles Against Side Channel Analysis Via Motor Noise Injection

We first investigate the efficacy of side channel anal-ysis (SCA) based attacks to recover the encryption key from an unmanned aerial vehicle (UAV). We then propose a method based on noise injection to safeguard against this type of attack. The proposed noise injection method is unique because it prioritizes low power consumption and uses as a noise source the motors of the UAV system. The effectiveness of the proposed method is demonstrated through experiments on a real quadcopter. It is found that when noise injection is not applied, the encryption key can be recovered through correlation power analysis. In contrast, the same key is not recoverable when the proposed noise injection method is applied. The implemented solution is efficient and does not negatively effect the flight stability or control of the UAV because its execution time overhead is negligible inside the control loop. In addition, it consumes a negligible amount of power. These results demonstrate that mitigation of SCA attacks without expanding the complexity or negatively effecting the flight operations of drones is achievable.