Stochastic Computing Design and Implementation of a Sound Source Localization System

Schober, Peter; Estiri, Seyedeh Newsha; Aygun, Sercan; Jalilvand, Amir Hossein; Najafi, M. Hassan; TaheriNejad, Nima

doi:10.1109/jetcas.2023.3243604

Cited by 12 publications

(6 citation statements)

References 49 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Previous art was focused on the investigation of immunity of the information processing against the noise present in the input data. That includes: contour edge recognition in images [4][5][6][7], image gamma correction [8], pattern/symbol recognition [9] and speaker/sound localization [10]. While in these works a superior noise immunity of the RPC paradigm has been found, in this study we are up to a completely different notion of immunity, namely the one related to the hardware damage.…”

Section: Introductionmentioning

confidence: 71%

“…where ∆h is the vertical difference between the height of the hoop and the starting point of the ball, ϑ is the launch angle measured from the horizontal plane, D is the horizontal distance between the player and the center of the hoop and v 0 is the launch velocity of the ball. Inserting Equation (9) into Equation (8) yields: (10) and inserting Equation (7) into Equation (10) yields:…”

Section: Physical Modelmentioning

confidence: 99%

See 1 more Smart Citation

Biomimetic Random Pulse Computation or Why Do Humans Play Basketball Better than Robots?

Stipčević

2023

Biomimetics

View full text Add to dashboard Cite

In this work, we compare the basketball scoring performance of two imaginary (simulated) mechanical robots in conditions of erroneous information-processing circuits: Machine, whose moves are controlled by a conventional digital computer and Man, controlled by a random pulse computer composed of biologically-inspired circuits which execute basic arithmetic operations. This is the first comparative study of robustness of the digital and the random pulse computing paradigms, with respect to the error rate of the information-processing circuits (perr), for a mechanical robot. In spite of the fact that Man’s computer consists of only about 100 logic gates while Machine’s requires about 3500 gates, Man achieves a significantly higher scoring probability for perr in the range from 0.01% all the way to 10%, while at lower perr, both converge to the perfect score. Furthermore, Man’s hits make up a smooth Gaussian distribution with a vanishing probability of making large misses even at the highest perr, while Machine is prone to spectacular misses already at perr as low as 1 part-per-million. These findings indicate that the biologically inspired computation requires less hardware for the same task, and ensures higher robustness and better behaving operation than digital computation, which are characteristics of importance for the survivability of living beings.

show abstract

Section: Introductionmentioning

confidence: 71%

Section: Physical Modelmentioning

confidence: 99%

Biomimetic Random Pulse Computation or Why Do Humans Play Basketball Better than Robots?

Stipčević

2023

Biomimetics

View full text Add to dashboard Cite

show abstract

“…Stochastic computing can be leveraged in complex system designs. Schober et al's research [36] would underscore the potential for stochastic computing to enhance the analytical capabilities of BIM systems, particularly in acoustic modeling and simulation, which are crucial for the design of auditoriums, open-plan offices, and other architectural spaces where sound control is essential.…”

Section: Traditional Methodsmentioning

confidence: 99%

Enhancing BIM Integration: A Comparative Analysis of Novel Composite Structure Documentation Methods

Szalai,

Ajtayné Károlyfi

2024

Buildings

View full text Add to dashboard Cite

The proper selection and planning of building materials are crucial tasks in architectural design, as they fundamentally impact the functioning of the structure. In traditional design processes, this information is recorded in text form, typically using word processing software. However, this approach hinders the integration with modern, data-driven design methods and is incompatible with the increasingly popular building information modeling (BIM) processes. To address this, two new methods have been developed: one database-like method in the form of an Excel spreadsheet and the other as a dedicated web application. This article introduces and compares these methods based on pilot projects conducted by university students and an expert. Based on the results of the study conducted among students and expert, the database-like method proves to be the fastest. For students, creating a composite structure took an average of 14–20 min, while for experts, it took an average of 1.2 min. According to the evaluation of participants, the traditional method does not facilitate automatic communication with BIM, while both the database and web solutions promote it. The web-based solution, with its dedicated layout and functionality, offers additional advantages in this regard. The research underscores the importance of structured data in BIM and proposes new methods to streamline composite documentation processes during the design phase.

show abstract

“…Tey have also leveraged microphone arrays to capture spatial information for fault source localization [1,2]. Many researchers have developed various sound source locators, such as Schober et al, who developed a functional sound source locator based on stochastic computing (SC) [3]. However, due to various factors such as noise interference, reverberation, the nature of source signals, and the number of sources, the practical efcacy of source localization systems is infuenced.…”

Section: Introductionmentioning

confidence: 99%

Mechanical Fault Sound Source Localization Estimation in a Multisource Strong Reverberation Environment

Deng,

Liu,

Zhang

et al. 2024

Shock and Vibration

View full text Add to dashboard Cite

Aiming at the sound source localization of mechanical faults in a strong reverberation scenario with multiple sound sources, this paper investigates a mechanical fault source localization method using the U-net deep convolutional neural network. The method utilizes the SRP-PHAT algorithm to calculate the response power spectra of the collected multichannel fault signals. Through the utilization of the U-net neural network, the response power spectra containing spurious peaks are transformed into “clean” estimated source distribution maps. By employing interpolation search, the estimated source distribution maps are processed to obtain location estimations for multiple fault sources. To validate the effectiveness of the proposed method, this paper constructs an experimental dataset using mechanical fault data from electromechanical equipment relays and conducts sound source localization experiments. The experimental results show that the U-net network under 0.2 s/0.5 s/0.7 s reverberation time can effectively eliminate spurious peak interference in the response power spectrum. As the signal-to-noise ratio decreases, it can still distinguish the sound sources with a distance of 0.2 m. In the context of multifault source localization, the method is capable of simultaneously locating the positions of four fault sources, with an average localization error of less than 0.02 m. The method in this paper effectively eliminates spurious peaks in the response power spectra under conditions of multisource strong reverberation. It accurately locates multiple mechanical fault sources, thereby significantly enhancing the efficiency of mechanical fault detection.

show abstract

Stochastic Computing Design and Implementation of a Sound Source Localization System

Cited by 12 publications

References 49 publications

Biomimetic Random Pulse Computation or Why Do Humans Play Basketball Better than Robots?

Biomimetic Random Pulse Computation or Why Do Humans Play Basketball Better than Robots?

Enhancing BIM Integration: A Comparative Analysis of Novel Composite Structure Documentation Methods

Mechanical Fault Sound Source Localization Estimation in a Multisource Strong Reverberation Environment

Contact Info

Product

Resources

About