Deep Learning Approach for Protecting Voice-Controllable Devices From Laser Attacks

The laser-based audio signal injection can be used for attacking voice controllable systems. An attacker can aim an amplitude-modulated light at the microphone's aperture, and the signal injection acts as a remote voice-command attack on voice-controllable systems. Attackers are using vulnerabilities to steal things that are in the form of physical devices or the form of virtual using making orders, withdrawal of money, etc. Therefore, detection of these signals is important because almost every device can be attacked using these amplitude-modulated laser signals. In this project, the authors use deep learning to detect the incoming signals as normal voice commands or laser-based audio signals. Mel frequency cepstral coefficients (MFCC) are derived from the audio signals to classify the input audio signals. If the audio signals are identified as laser signals, the voice command can be disabled, and an alert can be displayed to the victim. The maximum accuracy of the machine learning model was 100%, and in the real world, it's around 95%.

Highly Sensitive Sphere-Tube Coupled Photoacoustic Cell Suitable for Detection of a Variety of Trace Gases: NO2 as an Example

The concentration of trace gases in the atmospheric environment is extremely low, but it has a great impact on the living environment of organisms. Photoacoustic spectroscopy has attracted extensive attention in the field of trace gas detection because of its high sensitivity, good selectivity, and fast response. As the core of a photoacoustic detection setup, the photoacoustic cell has a significant impact on detection performance. To improve detection sensitivity, a sphere-tube coupled photoacoustic cell (STPAC) was developed, which was mainly composed of a diffuse-reflective sphere and an acoustic resonance tube. Modulated light was reflected multiple times in the sphere to increase optical path, and photoacoustic (PA) signals were further amplified by the tube. Based on STPAC, a PA gas detection setup was built with a laser diode (LD) at 450 nm as the light source. The experimental results showed that the minimum detection limit (noise equivalent concentration, NEC) of NO2 was ~0.7 parts per billion (ppb). Compared with the T-type PA cell (TPAC) in which the modulated light passed through the sphere, the signal-to-noise ratio of STPAC was increased by an order of magnitude at the same concentration of the NO2 sample.

TOWARDS AN AVIAN FLICKER VISIBILITY MEASURE

We propose a measure of temporal light quality that could be used to design flicker-free light sources, as perceived by birds. It can be applied to, among others, the poultry industry to reduce the negative impact of the modulated light on chicken health and well-being. The model is built by modifying an existing human flicker visibility model considering properties of bird vision. We discuss the implications of the model responses on the temporal quality of relevant historical light sources.

Theoretical bases of combined nonlinear optical demodulator-amplifer for broadband microwave photonics system

Our demodulator is based on the use of a strong light pump wave along with a signal modulated light wave. The demodulator is linear in terms of the modulation signal, which makes it possible to obtain a demodulated signal without harmonic distortion. The modulator conversion efciency and signal gain increase with increasing microwave baseband.

Uncertainty in the Phase Flicker Measurement for the Liquid Crystal on Silicon Devices

Phase flicker has become an important performance parameter for the liquid crystal on silicon (LCOS) devices. Since the phase response of the LCOS device cannot be measured directly, it is usually derived from the intensity response of the modulated light beam when the LCOS device was placed between a pair of crossed polarisers. However, the relationship between the intensity of the beam and the phase response of the LCOS device is periodic. This would lead to uncertainty in the phase flicker measurement. This paper analyses this measurement uncertainty through both simulation and experiments. It also proposed a strategy to minimise the uncertainty.