Mark Sterling

Morphological steganalysis of audio signals and the principle of diminishing marginal distortions

Steganographic methods attempt to insert data in multimedia signals in an undetectable fashion. However, these methods often disrupt the underlying signal characteristics, thereby allowing detection under careful steganalysis. Under repeated embedding, disruption of the signal characteristics is the highest for the first embedding and decreases subsequently. That is, the marginal distortions due to repeated embeddings decrease monotonically. We name this general principle as the principle of diminishing marginal distortions (DMD) and illustrate its validity in the audio domain using a morphological distortion metric. The principle of DMD is used to derive a steganalysis tool that detects the presence of hidden messages in uncompressed audio files. Detailed analysis and experimental results are provided for the detection of spread spectrum watermarking and stochastic modulation steganography.

The Development of an Automated Device for Asthma Monitoring for Adolescents: Methodologic Approach and User Acceptability

Background Many adolescents suffer serious asthma related morbidity that can be prevented by adequate self-management of the disease. The accurate symptom monitoring by patients is the most fundamental antecedent to effective asthma management. Nonetheless, the adequacy and effectiveness of current methods of symptom self-monitoring have been challenged due to the individuals’ fallible symptom perception, poor adherence, and inadequate technique. Recognition of these limitations led to the development of an innovative device that can facilitate continuous and accurate monitoring of asthma symptoms with minimal disruption of daily routines, thus increasing acceptability to adolescents. Objective The objectives of this study were to: (1) describe the development of a novel symptom monitoring device for teenagers (teens), and (2) assess their perspectives on the usability and acceptability of the device. Methods Adolescents (13-17 years old) with and without asthma participated in the evolution of an automated device for asthma monitoring (ADAM), which comprised three phases, including development (Phase 1, n=37), validation/user acceptability (Phase 2, n=84), and post hoc validation (Phase 3, n=10). In Phase 1, symptom algorithms were identified based on the acoustic analysis of raw symptom sounds and programmed into a popular mobile system, the iPod. Phase 2 involved a 7 day trial of ADAM in vivo, and the evaluation of user acceptance using an acceptance survey and individual interviews. ADAM was further modified and enhanced in Phase 3. Results Through ADAM, incoming audio data were digitized and processed in two steps involving the extraction of a sequence of descriptive feature vectors, and the processing of these sequences by a hidden Markov model-based Viterbi decoder to differentiate symptom sounds from background noise. The number and times of detected symptoms were stored and displayed in the device. The sensitivity (true positive) of the updated cough algorithm was 70% (21/30), and, on average, 2 coughs per hour were identified as false positive. ADAM also kept track of the their activity level throughout the day using the mobile system’s built in accelerometer function. Overall, the device was well received by participants who perceived it as attractive, convenient, and helpful. The participants recognized the potential benefits of the device in asthma care, and were eager to use it for their asthma management. Conclusions ADAM can potentially automate daily symptom monitoring with minimal intrusiveness and maximal objectivity. The users’ acceptance of the device based on its recognized convenience, user-friendliness, and usefulness in increasing symptom awareness underscores ADAM’s potential to overcome the issues of symptom monitoring including poor adherence, inadequate technique, and poor symptom perception in adolescents. Further refinement of the algorithm is warranted to improve the accuracy of the device. Future study is also needed to assess the efficacy of the device in promoting self-management and asthma outcomes.

An adaptive spread-spectrum data hiding technique for digital audio

In this paper we describe an application of spread spectrum techniques in audio data hiding for watermarking and steganography. The method is self-synchronizing, cover-dependent, and operates in the time domain. We use a special class of frequency-hop signal known as a Welch-Costas array. Welch-Costas arrays have the properties of range and Doppler resolution. This allows us to recover embedded data with a matched filter. We also demonstrate a special case of an adaptive method due to Su and Girod (2002).

Automated Cough Assessment on a Mobile Platform.

The development of an Automated System for Asthma Monitoring (ADAM) is described. This consists of a consumer electronics mobile platform running a custom application. The application acquires an audio signal from an external user-worn microphone connected to the device analog-to-digital converter (microphone input). This signal is processed to determine the presence or absence of cough sounds. Symptom tallies and raw audio waveforms are recorded and made easily accessible for later review by a healthcare provider. The symptom detection algorithm is based upon standard speech recognition and machine learning paradigms and consists of an audio feature extraction step followed by a Hidden Markov Model based Viterbi decoder that has been trained on a large database of audio examples from a variety of subjects. Multiple Hidden Markov Model topologies and orders are studied. Performance of the recognizer is presented in terms of the sensitivity and the rate of false alarm as determined in a cross-validation test.

Evaluating the Validity of an Automated Device for Asthma Monitoring for Adolescents: Correlational Design

Background Symptom monitoring is a cornerstone of asthma self-management. Conventional methods of symptom monitoring have fallen short in producing objective data and eliciting patients’ consistent adherence, particularly in teen patients. We have recently developed an Automated Device for Asthma Monitoring (ADAM) using a consumer mobile device as a platform to facilitate continuous and objective symptom monitoring in adolescents in vivo. Objective The objectives of the study were to evaluate the validity of the device using spirometer data, fractional exhaled nitric oxide (FeNO), existing measures of asthma symptoms/control and health care utilization data, and to examine the sensitivity and specificity of the device in discriminating asthma cases from nonasthma cases. Methods A total of 84 teens (42 teens with a current asthma diagnosis; 42 without asthma) aged between 13 and 17 years participated in the study. All participants used ADAM for 7 consecutive days during which participants with asthma completed an asthma diary two times a day. ADAM recorded the frequency of coughing for 24 hours throughout the 7-day trial. Pearson correlation and multiple regression were used to examine the relationships between ADAM data and asthma control, quality of life, and health care utilization at the time of the 7-day trial and 3 months later. A receiver operating characteristic (ROC) curve analysis was conducted to examine sensitivity and specificity based on the area under the curve (AUC) as an indicator of the device’s capacity to discriminate between asthma versus nonasthma cases. Results ADAM data (cough counts) were negatively associated with forced expiratory volume in first second of expiration (FEV1) (r=–.26, P=.05), forced vital capacity (FVC) (r=–.31, P=.02), and overall asthma control (r=–.41, P=.009) and positively associated with daily activity limitation (r=.46, P=.01), nighttime (r=.40, P=.02) and daytime symptoms (r=.38, P=.02), and health care utilization (r=.61, P<.001). Device data were also a significant predictor of asthma control (β=–.48, P=.003), quality of life (β=–.55, P=.001), and health care utilization (β=.74, P=.004) after 3 months. The ROC curve analysis for the presence of asthma diagnosis had an AUC of 0.71 (95% CI 0.58-0.84), which was significantly different from chance (χ2 1=9.7, P=.002), indicating the device’s discriminating capacity. The optimal cutoff value of the device was 0.56 with a sensitivity of 51.3% and a specificity of 72.7%. Conclusions This study demonstrates validity of ADAM as a symptom-monitoring device in teens with asthma. ADAM data reflect the current status of asthma control and predict asthma morbidity and quality of life for the near future. A monitoring device such as ADAM can increase patients’ awareness of the patterns of cough for early detection of worsening asthma and has the potential for preventing serious and costly future consequences of asthma.

Representation of solo clarinet music by physical modeling synthesis

A physical modeling synthesizer is developed and used to parametrically represent digital audio recordings of clarinet soloists. Empirical data, in the form of acoustic impedance measurements, are incorporated in the model definition. Reed dynamics are also included. Algorithms which calculate control parameters from source recordings are presented and examples of both the impedance measurements and the synthesis output are provided. The synthesis results give further support to the idea that music can, in certain circumstances, be adequately represented by a few low bandwidth control parameters. Additionally, the governing equations of clarinet physical modeling are briefly reviewed.

Non-contact, capacitive biosensor electrodes for electrostatic charge reduction

Non-contact capacitive sensors are receiving much attention as the desire for long-term portable biological sensing devices grows. However, electrostatic discharge (ESD) generation, arising from the sensor electrodes rubbing against the subjects clothing, is a significant challenge. We propose an actively driven shielding electrode with a grid-multilayer structure. We have considered and compared five electrode prototype designs. The signal-to-ESD noise ratio (SENR) and maximum amplifier noise budget were measured for the purposes of comparing the different electrode designs. The results indicate that the grid electrode has relatively moderate SENR and an acceptable noise design budget better than the other proposed prototypes.

Pitch bends and tonguing articulation in clarinet physical modeling synthesis

A physical modeling approach is used to investigate playing effects in woodwind instruments. This builds upon prior work concerning both empirical studies of the acoustics of the clarinet and extensive development of computer simulations of musical instrument systems. Specifically, explicit implementations of two performance gestures for the clarinet are given and demonstrated: tonguing and pitch bending. Physical modeling for the clarinet is briefly reviewed. Following this we show how both tonguing and pitch bending map to changes in the clarinet physical model itself and in the control parameter data. To our knowledge, tonguing in particular is one effect that has not been widely discussed in the literature, at least in the exact form we present. Finally, some possible future research directions are indicated.

Information-Theoretic Approach to A/D Conversion

We propose a new information theoretic approach to understanding analog-to-digital (A/D) conversion principles. An A/D converter is characterized as a communication system that transmits information from an analog source to the digital domain for processing, storage, transmission, and analysis. Accordingly, an A/D conversion system consists of an analog source of information, a means to sample and encode the analog signal for transmission through the comparator, and a means to decode the digital output from the comparator and provide a digital representation of the analog signal. The comparator acts as a channel since it “injects” uncertainty in the form of quantization noise. We provide an upper bound (conversion capacity) on the amount of information that can be transmitted from the analog to digital domain for a specific A/D converter and we describe a necessary and sufficient condition that, if satisfied, allows an A/D converter to achieve capacity. The conversion capacity defines a fundamental upper limit on the performance of an A/D converter in terms of resolution-bandwidth product for a given technology. It is shown how the conversion capacity may be utilized to yield performance limitations in terms of mean square error and SNR. The capacity is, thus, a useful metric for the comparison of ADC designs and may also provide hints to improving existing and conceiving new A/D conversion methods.

Empirical physical modeling for bowed string instruments

The various components needed to define an empirical violin physical model are assembled and their interaction is detailed. We experimentally measure the mechanical admittance at the violin bridge using the impact hammer excitation method and the monopole radiativity in anechoic conditions. These are combined with a digital waveguide model computing the dynamical behavior of the bowed string. Theoretical bridge oscillations and sound radiation are defined by convolution of the bowed string physical model signals with impulse responses determined from experiment.