Ultrasonic Jamming for Speech Device Security
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Speech-enabled devices are vulnerable to ultrasonic attacks, where inaudible ultrasonic signals can cause the devices to wake up and perform unintended or malicious tasks by exploiting nonlinearities in microphones and pre-amplifiers, leading to intermodulation distortion that shifts the signal to the normal speech frequency range.
Innovation Solution
Generating and broadcasting ultrasonic jamming signals from a loudspeaker in response to detected voice activity or wake-on-voice key phrases, which interfere with the ultrasonic attack signals without affecting legitimate user commands, using techniques such as white or colored noise, and periodic frequency sweeps.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If ultrasonic jamming signals are broadcast continuously to defend against attacks, then security coverage is improved, but power consumption increases
Solution Approach 1:
The system broadcasts ultrasonic jamming signals periodically rather than continuously. The processor determines broadcast durations based on detected voice activity and wake-on-voice key phrases, limiting the jamming signal transmission to specific time windows when attacks are likely, thus reducing overall power consumption while maintaining security coverage during critical periods
Solution Approach 2:
The system activates jamming signals in advance of detected voice activity or wake-on-voice key phrases. By preparing and broadcasting jamming signals before potential attack vectors are fully executed, the system ensures security coverage is established proactively, allowing the device to respond effectively to ultrasonic attacks while controlling power usage through pre-conditioned activation
2Reliability
If ultrasonic jamming signals are broadcast for extended durations to ensure complete attack prevention, then security coverage is improved, but the duration of action increases
Solution Approach 1:
The system implements periodic broadcasting of ultrasonic jamming signals rather than continuous transmission. The processor calculates appropriate broadcast durations based on the detected voice activity and wake-on-voice key phrase contexts, ensuring security coverage is maintained for the necessary period while avoiding excessive duration that would waste resources
Solution Approach 2:
The system dynamically adjusts the broadcast duration of ultrasonic jamming signals based on real-time detection of voice activity and wake-on-voice key phrases. This dynamic adaptation allows the system to extend security coverage when needed while reducing duration when the threat window closes, optimizing both security coverage and operational efficiency
3Measurement precision
If the device processes all audio signals to ensure no attacks are missed, then detection accuracy is improved, but processing complexity increases
Solution Approach 1:
The system extracts and processes only specific audio signals that contain wake-on-voice key phrases or indicate voice activity. By filtering out irrelevant audio data and focusing processing only on segments with potential security implications, the system maintains high detection accuracy for ultrasonic attacks while significantly reducing overall processing complexity
Solution Approach 2:
The system applies different processing quality levels to different audio segments. High-processing modes are applied locally to segments containing wake-on-voice key phrases or detected voice activity, while lower-processing modes are used for background audio. This localized quality approach ensures detection accuracy where needed without uniformly increasing processing complexity across all audio data
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach effectively defends against ultrasonic attacks with reduced cost and complexity, ensuring that the devices only act on legitimate user commands while conserving power by limiting the jamming signal duration.
Implementation Method 1
Generating and broadcasting ultrasonic jamming signals from a loudspeaker in response to detected voice activity or wake-on-voice key phrases, which interfere with the ultrasonic attack signals
Implementation Method 2
broadcasting the ultrasonic jamming signal over a loudspeaker
Data Source
AI summary
Techniques are provided for defending against an ultrasonic attack on a speech enabled device. A methodology implementing the techniques according to an embodiment includes detecting voice activity in an audio signal received by the device and generating an ultrasonic jamming signal in response to the detection. The jamming signal is broadcast over a loudspeaker for up to the duration of the detected voice activity to defend against the ultrasonic attack. According to another embodiment, the ultrasonic jamming signal is generated in response to detection of a wake-on-voice key phrase in the received audio signal, and the jamming signal is broadcast over the loudspeaker for a time duration selected to be less than or equal to a time window during which spoken commands are accepted by the device following the wake-on-voice key phrase detection. The jamming signal may include white or colored noise, combinations of tones, and/or a periodic sweep frequency.