Ian Vince McLoughlin

Bio-effects and safety of low-intensity, low-frequency ultrasonic exposure.

Low-frequency (LF) ultrasound (20-100 kHz) has a diverse set of industrial and medical applications. In fact, high power industrial applications of ultrasound mainly occupy this frequency range. This range is also used for various therapeutic medical applications including sonophoresis (ultrasonic transdermal drug delivery), dentistry, eye surgery, body contouring, the breaking of kidney stones and eliminating blood clots. While emerging LF applications such as ultrasonic drug delivery continue to be developed and undergo translation for human use, significant gaps exist in the coverage of safety standards for this frequency range. Accordingly, the need to understand the biological effects of LF ultrasound is becoming more important. This paper presents a broad overview of bio-effects and safety of LF ultrasound as an aid to minimize and control the risk of these effects. Its particular focus is at low intensities where bio-effects are initially observed. To generate a clear perspective of hazards in LF exposure, the mechanisms of bio-effects and the main differences in action at low and high frequencies are investigated and a survey of harmful effects of LF ultrasound at low intensities is presented. Mechanical and thermal indices are widely used in high frequency diagnostic applications as a means of indicating safety of ultrasonic exposure. The direct application of these indices at low frequencies needs careful investigation. In this work, using numerical simulations based on the mathematical and physical rationale behind the indices at high frequencies, it is observed that while thermal index (TI) can be used directly in the LF range, mechanical index (MI) seems to become less reliable at lower frequencies. Accordingly, an improved formulation for the MI is proposed for frequencies below 500 kHz.

Robust sound event classification using deep neural networks

The automatic recognition of sound events by computers is an important aspect of emerging applications such as automated surveillance, machine hearing and auditory scene understanding. Recent advances in machine learning, as well as in computational models of the human auditory system, have contributed to advances in this increasingly popular research field. Robust sound event classification, the ability to recognise sounds under real-world noisy conditions, is an especially challenging task. Classification methods translated from the speech recognition domain, using features such as mel-frequency cepstral coefficients, have been shown to perform reasonably well for the sound event classification task, although spectrogram-based or auditory image analysis techniques reportedly achieve superior performance in noise. This paper outlines a sound event classification framework that compares auditory image front end features with spectrogram image-based front end features, using support vector machine and deep neural network classifiers. Performance is evaluated on a standard robust classification task in different levels of corrupting noise, and with several system enhancements, and shown to compare very well with current state-of-the-art classification techniques.

Fourier Transform-Based Scalable Image Quality Measure

We present a new image quality assessment algorithm based on the phase and magnitude of the 2-D discrete Fourier transform. The basic idea is to compare the phase and magnitude of the reference and distorted images to compute the quality score. However, it is well known that the human visual systems sensitivity to different frequency components is not the same. We accommodate this fact via a simple yet effective strategy of non-uniform binning of the frequency components. This process also leads to reduced space representation of the image thereby enabling the reduced-reference (RR) prospects of the proposed scheme. We employ linear regression to integrate the effects of the changes in phase and magnitude. In this way, the required weights are determined via proper training and hence more convincing and effective. Last, using the fact that phase usually conveys more information than magnitude, we use only the phase for RR quality assessment. This provides the crucial advantage of further reduction in the required amount of reference image information. The proposed method is, therefore, further scalable for RR scenarios. We report extensive experimental results using a total of nine publicly available databases: seven image (with a total of 3832 distorted images with diverse distortions) and two video databases (totally 228 distorted videos). These show that the proposed method is overall better than several of the existing full-reference algorithms and two RR algorithms. Additionally, there is a graceful degradation in prediction performance as the amount of reference image information is reduced thereby confirming its scalability prospects. To enable comparisons and future study, a Matlab implementation of the proposed algorithm is available at http://www.ntu.edu.sg/home/wslin/reduced_phase.rar.

Reconstruction of Normal Sounding Speech for Laryngectomy Patients Through a Modified CELP Codec

Whispered speech can be useful for quiet and private communication, and is the primary means of unaided spoken communication for many people experiencing voice-box deficiencies. Patients who have undergone partial or full laryngectomy are typically unable to speak anything more than hoarse whispers, without the aid of prostheses or specialized speaking techniques. Each of the current prostheses and rehabilitative methods for post-laryngectomized patients (primarily oesophageal speech, tracheo-esophageal puncture, and electrolarynx) have particular disadvantages, prompting new work on nonsurgical, noninvasive alternative solutions. One such solution, described in this paper, combines whisper signal analysis with direct formant insertion and speech modification located outside the vocal tract. This approach allows laryngectomy patients to regain their ability to speak with a more natural voice than alternative methods, by whispering into an external prosthesis, which then, recreates and outputs natural-sounding speech. It relies on the observation that while the pitch-generation mechanism of laryngectomy patients is damaged or unusable, the remaining components of the speech production apparatus may be largely unaffected. This paper presents analysis and reconstruction methods designed for the prosthesis, and demonstrates their ability to obtain natural-sounding speech from the whisper-speech signal using an external analysis-by-synthesis processing framework.

Review: Line spectral pairs

Linear prediction-based coders commonly utilise line spectral pairs (LSPs) to represent linear prediction coefficients for reasons of filter stability and representational efficiency. LSPs have other useful properties such as an ordering related to the spectral properties of the underlying data, which leads to advantages when used for analysing speech and other signals. This paper reviews the LSP representation, conversion and quantization processes, computational issues associated with the implementation of LSP-based methods, and their use in speech analysis and processing. In addition, this paper presents LSP manipulation methods that can be used to alter frequencies within the represented signal in a consistent and relevant way, and considers the use of LSPs for analysis of non-speech information.

Robust sound event recognition using convolutional neural networks

Traditional sound event recognition methods based on informative front end features such as MFCC, with back end sequencing methods such as HMM, tend to perform poorly in the presence of interfering acoustic noise. Since noise corruption may be unavoidable in practical situations, it is important to develop more robust features and classifiers. Recent advances in this field use powerful machine learning techniques with high dimensional input features such as spectrograms or auditory image. These improve robustness largely thanks to the discriminative capabilities of the back end classifiers. We extend this further by proposing novel features derived from spectrogram energy triggering, allied with the powerful classification capabilities of a convolutional neural network (CNN). The proposed method demonstrates excellent performance under noise-corrupted conditions when compared against state-of-the-art approaches on standard evaluation tasks. To the authors knowledge this in the first application of CNN in this field.

Data concealment in audio using a nonlinear frequency distribution of PRBS coded data and frequency-domain LSB insertion

A novel method of transparent data concealment in audio-streams is discussed. The proposed system makes use of sub-band coding, least significant bit coding (LSB) and a pseudo-random bit stream generator (PRBS). A maximum of about 6% of the audio file can be used to hide data transparently with no perceptible distortion. A solution to solve the positive bias problem that is inherent in LSB encoding is also presented.

LSP-based speech modification for intelligibility enhancement

CELP coders commonly use line spectral pairs (LSP) to represent linear prediction parameters, giving stable filters and efficient coding. However, manipulation of LSPs can alter frequencies within the represented signals. This paper describes two computationally efficient LSP-based processing methods designed to enhance the intelligibility of speech degraded by acoustic interference.

Secure Embedded Systems: The Threat of Reverse Engineering

Companies releasing newly designed embedded products typically recoup the cost of development through initial sales, and thus are unlikely to welcome early competition based around rapid reverse engineering of their products. By contrast, competitors able to shorten time-to-market though reverse engineering will gain design cost and market share advantages. Reverse engineering for nefarious purposes appears to be commonplace, and has significant cost impact on industry sales and profitability. In the Embedded Systems MSc programme at Nanyang Technological University, we are aiming to raise awareness of the unique security issues related to the reverse engineering of embedded systems. This effort is largely through devoting 50% of the secure embedded systems course, ES6190 to reverse engineering (the remainder to traditional security concerns). This paper covers the reverse engineering problem scope, and our approach to raising awareness through the secure embedded systems course. A classification of hardware reverse engineering steps and mitigations is also presented for the first time, with an overview of a reverse engineering curriculum. Since the quantity of published literature related to the reverse-engineering of embedded systems lies somewhere between scarce and nonexistent, this paper presents a full overview of the topic before descussing educational aspects related to this.

Effects of Channel Prediction for Transmit Antenna Selection With Maximal-Ratio Combining in Rayleigh Fading

Antenna selection is a low-complexity method for pragmatically exploiting spatial diversity in wireless systems. It has potentially reduced hardware cost compared with space-time or multiple-input-multiple-output (MIMO) coding due to the reduction in the amount of radio-frequency hardware required. Although receive antenna selection is, perhaps, more common, transmit antenna selection (TAS) offers several advantages, particularly for hardware-costly transmit schemes, such as, methods that require linearization. In use, TAS requires at least partial channel knowledge at the transmitter to perform selection. This knowledge usually comes in the form of an index to the best set of antenna/antennas that are fed back from the receiver, which implies a delay between the channel that is sampled (at the receiver) and this knowledge being acted upon (at the transmitter). In this paper, performance degradation due to outdated channel knowledge is analytically determined and related to channel characteristics. A predictive scheme is then developed to mitigate delay-induced degradation. Several factors that are related to TAS system performance under different channel scenarios, both with and without mitigation, are explored. Closed-form expressions for performance metrics such as bit error rate, outage probability, average signal-to-noise ratio (SNR) gain, and higher order moments of the output SNR, are derived and verified by simulations. The impact of prediction is analyzed for different TAS setups and channel prediction scenarios, as well as various system design parameters.