Applied Acoustics最新文献

Reverberation is one of the most critical obstacles to adopt automatic speech recognition (ASR) in real life environments. Therefore, comprehensive understanding on the effect of reverberation to ASR is required to design robust ASR systems for practical uses. To deepen our understanding on the effect of reverberation to practical ASR, we performed a phonemic analysis on commercial ASR system. The analysis method involves a new metric named mean phoneme coherence (MPC), defined by time–frequency-averaged coherence function between clean and reverberated speech spectrograms of each phoneme. MPC measures the amount of spectral contamination on phonemes under certain reverberation condition thus quantifies not only the amount of reverberation on the phonemes but also contextual influences on the phoneme within sentence spoken in the reverberation condition. MPC was proven to represent the amount of reverberation and intelligibility of speeches under given reverberation condition by comparing MPC with word error rate (WER) in real reverberation conditions. Furthermore, the relationship between phoneme groups’ vulnerability to spectral contamination and ASR performance upon reverberation is analyzed by comparing median of phoneme groups’ MPC distribution with phoneme group word accuracy (PGWA). Analysis has shown that the two quantities show weak correlation, thus reverberation differently affects the intelligibility of phonemes. In addition, a comparative study among phoneme groups has shown that nasals and semivowels show the least robust ASR performances to reverberation while nasals and stops are most vulnerable to cause spectral contamination. The results and discussions present what should be taken into account for ASR robust to reverberation.

When implementing active noise control systems on signal processing hardware, the time delay introduced by electronic components (especially components requiring additional lowpass filters or introducing fixed-sample-size delays) may adversely affect the noise control performance. One common approach to reducing this delay is to use a high sampling rate, but this increases the computation significantly when implementing the ANC filters in real time. In the current work, a polyphase-structure-based filter design method is developed for active noise control systems that can reduce the computation load for real-time filter implementation but do not introduce an additional time delay. Although the computation reduction capability of a polyphase filter structure is well known for multi-rate systems, the traditional use of such multi-rate systems requires additional anti-aliasing and reconstruction filters which introduces an additional time delay. Thus, in delay-sensitive applications, such as active noise control, this method was previously applied only on the filter adaption phase, instead of directly on the real-time filtering process. In this article, a filter decomposition method using the minimum-phase technique is proposed to decompose an ANC filter into two multiplicative causal filters both of which have lowpass frequency response shapes at high frequencies such that the polyphase structure can be applied directly to the two multiplicative causal control filters without introducing additional anti-aliasing and reconstruction filters. Results show that, compared with various traditional low sampling rate implementations, the proposed method can significantly improve the noise control performance. Compared with the non-polyphase high-sampling rate method, the real-time computations that increase with the sampling rate are improved from quadratically to linearly.

In this paper, based on a high- sound-pressure microperforated plate model, a nonlinear sound-absorption model for multi-unit couplings based on coiled-up space structures is proposed. The sound absorption performance and relative impedance of two-unit coupled structures (TUCSs) were studied. The results show that the TUCS sound-absorption performance, which is good at low sound pressures, decreases significantly as the incident sound pressure increases owing to impedance mismatch. Furthermore, the influence of parameters such as aperture size, plate thickness, perforation rate, and equivalent length on the unit’s structural sound-absorption performance was studied. By employing the particle swarm optimization algorithm, we optimized the parameters of an eight-unit coupled structure (EUCS) using the proposed model. The optimization results reveal the nonlinear robust sound-absorption characteristics of the structure, which means the EUCS can maintain a stable and good sound-absorption performance when the incident sound pressure level and frequency are within 125–155 dB and 400–3000 Hz, respectively. Experimental assessments of the EUCS sound-absorption performance within the 300–1900 Hz range confirmed the accuracy of the proposed model and the efficacy of the optimized sound-absorption capabilities of the structure. Consequently, the proposed model and sound-absorption structure demonstrated potential applications in the acoustic lining design of aircraft engines.

This paper proposes a solution to reduce the inter symbol interferences (ISI) in multipath channels with frequency selective fading. The proposed solution is based on a Modified Recursive Non-Quadratic (MRNQ) adaptive algorithm combined with a Decision Feed-forward Equalizer (DFE) structure, referred to as MRNQ-DFE. First, we studied the behavior of the proposed approach by evaluating the impact of some key factors of the convergence rate, relying on the mean squared error (MSE) criterion. Then, we conducted a comprehensive set of experiments to evaluate the performance of the proposed MRNQ-DFE equalizer compared to other robust adaptive equalizers, such as NLMS-DFE and APA-DFE. These experiments rely on a variety of objective criteria, including the constellation diagram, the eye diagram, the Nyquist criterion, and the MSE criterion. The obtained results show the effectiveness of the proposed equalizer to achieve the desired goal with a superior convergence speed compared to the other strong adaptive equalizers, making it an effective solution for multipath channels in digital communication systems.

This paper analyzes the factors that affect the noise reduction performance of feedback control based on remote microphone technique (RMT) and how these factors can be exploited to improve the control performance. First, simulations were conducted to compare the noise reduction performance of employing feedback control directly at the virtual microphone (located at the ear position) and on the physical microphone (located on the headrest), as well as the RMT-based feedback control system. Then the impact of the delay of the virtual secondary path and the coherence between physical and virtual signals on the noise reduction performance of the RMT-based feedback control system was analyzed. It is found that the noise reduction performance can be improved by reducing the delay of the virtual secondary path or increasing the number of physical microphones. Finally, experiments of a dual-channel control system conducted inside an electric car cabin demonstrate that the feedback control strategy based on RMT achieves a binaural noise reduction of 3.4 dBA when employing 4 physical microphones to estimate sound pressure at the ear positions. This approach achieves similar performance but is more cost-effective than a feedforward system using multiple reference sensors.

Detecting weak acoustic signals within a strong background noise environment is of significant importance across various research fields. However, conventional detection sensors for sound source are constrained by the minimal detectable pressure, making them unable to extract sound signals that are contaminated by strong background noise. Although metamaterial devices based on gradient refractive index have been used to realize weak signal detection, the current structures are mainly focused on two-dimensional planes and are not capable of acoustic sensing in three dimensions. In this study, a three-dimensional Crossed T-gradient metamaterial device (CTGM) for acoustically enhanced sensing is proposed by utilizing the strong wave compression effect and the equivalent medium mechanism. The superior amplification signal amplitude capability and frequency selectivity of the CTGM structure is verified by numerical simulations. Compared with conventional gradient metamaterial without a T-shape (GAM) and gradient structure without protrusion (GWPM), CTGM could reduce the operating frequency without any change in volume. It has a stronger ability to control acoustic signals at larger wavelengths. The experimental test results show that CTGM has higher acoustic enhancement capability and frequency selectivity in the detection of acoustic signals with Gaussian pulses. This study demonstrates the potential of the designed acoustic metamaterials for enhancing the subtle fault signals detection in acoustic sensing, providing a pathway to enhance the cost-effectiveness of fault diagnostic techniques.

This paper presents a novel approach for nonlinear active noise control (ANC) systems that addresses the limitations of traditional linear ANC systems, which fail to consider nonlinear factors. A nonlinear ANC system is proposed for non-stationary noise with a nonlinear secondary path using the improved filtered-s least mean square/fourth (IFSLMS/F) algorithm. This system utilizes the real-time complementary ensemble empirical modal decomposition (CEEMD) for noise decomposition, trigonometric functional expansions of the functional link artificial neural network (FLANN) filter, and adaptive weight updating. The IFSLMS/F algorithm uses the fast convergence of the variable step size algorithm to efficiently reduce error perturbation by performing power normalization on the error signal. The performance and robustness of the system have been validated by simulations and experiments.

Background

Subharmonic aided pressure estimation (SHAPE) is an innovative non-invasive technique that leverages ultrasound subharmonic imaging to estimate pressure. This method exploits the “negative correlation between subharmonic amplitude and ambient pressure” observed in experimental settings. Despite extensive experimental validation and some promising results in clinical studies, the underlying mechanism of SHAPE remains incompletely understood. Although some studies have attempted to provide theoretical explanations, definitive conclusions have yet to be reached. In addition, theoretical investigations have mainly focused on the steady oscillation of bubbles under long pulse excitation, which contrasts with the short pulse excitation required for clinical SHAPE applications. An understanding of the SHAPE principle under short pulse excitation is needed.

Methods

The exponential elasticity model (EEM) was used to simulate Sonazoid bubbles, and a probe-to-probe acoustic propagation model was introduced to mimic a practical SHAPE scenario. The simulated acoustic signals in response to three-cycle sinusoidal pulse excitations were analyzed for spectral composition. The relationship between microbubble oscillation patterns and subharmonic characteristics was identified through detailed investigation.

Results

For the excitation pulse of 2.5 MHz frequency and 350 kPa magnitude, bubbles larger than the resonance radius (2.29 μm) exhibited significant subharmonics in the magnitude spectrum, while bubbles smaller than the subharmonic resonance radius (3.85 μm) showed the activity of scattering subharmonic energy and the sensitivity to ambient pressure. The emergence of subharmonics when increasing excitation power was related to the increasing amplification of the bubble self-oscillation and the period-doubling features in the acoustically forced oscillation. The negative correlation between subharmonic amplitude and ambient pressure was attributed to the reduced self-oscillation caused by increasing ambient pressure and hence bubble size reduction. Microbubbles falling between 2 and 3 μm showed the desired subharmonic sensitivity to ambient pressure under the specified excitation conditions.

Conclusion

The transient oscillatory behavior of microbubbles in response to short pulse excitation, characterized by a ringing down self-oscillation after the acoustic forcing effect has ceased, is crucial for understanding the subharmonic emergence and the observed negative correlation between subharmonic amplitude and ambient pressure. The proposed concepts of subharmonic resonance radius, subharmonic-significant bubbles, and subharmonic-active bubbles provide valuable insights into the diverse subharmonic behavior of microbubbles. The theoretical explanation of this negative correlation highlights the importance of using subharmonic-significant-and-active bubbles for S