Environmental noise monitoring using source classification in sensors

With the advent of reliable and continuously operating noise monitoring systems, we are now faced with an unprecedented amount of noise monitor data. In the context of environmental noise monitoring, there is a need to automatically detect, separate, and classify all environmental noise sources. This is a complex task because sources can overlap, vary by location, and have an unbounded number of noise sources that a monitor device may record. In this study, we synthetically generate datasets that contain Gaussian noise and overlaps for several pre-labeled environmental noise monitoring datasets to examine how well deep learning methods (e.g., autoencoders) can separate environmental noise sources. In addition to examining performance, we also focus on understanding which signal features and separation metrics are useful to this problem.

Unattended and continuously running environmental noise monitoring systems can capture an intractable amount of data. The signals captured can include a multitude of sources (e.g., wind noise and anthropogenic noise sources) in addition to the environmental noise sources of interest (e.g., aircraft, vehicles, trains, and military weapons). In this presentation, we explore the use of machine-learning methods to effectively isolate and identify environmental noise sources captured on such a noise monitoring system. Specifically, we consider the use of both unsupervised (e.g., principle components analysis, clustering methods, and deep belief networks) and supervised (e.g., logistic regression, support vector machines, and neural networks) pattern-learning methods to derive the features of interest and classify the signals based on the obtained features. The generalization performance of each method is assessed using a dataset of over 120,000 human classified signals, and the strengths and weaknesses of each approach are discussed.

Reducing environmental noise in urban settings, i.e., unwanted or harmful outdoor sounds produced by human activity, has become an important issue in recent years. Most countries have established regulations that set maximum permitted noise levels. However, enforcing these regulations effectively remains challenging as it requires active monitoring networks and audio analysis performed by trained specialists. The manual evaluation of the audio recordings is laborious, time-consuming, and inefficient since many audios exceeding the noise level threshold do not correspond to a sound event considered by the regulation. To address this challenge, this work proposes a computational pipeline to assist specialists in detecting noise sources in the built environment that do not comply with the Chilean noise regulation. The system incorporates a deep neural model following a pre-trained audio neural network architecture transferred to a dataset compiled from public sources and recordings in Valdivia, Chile. The target dataset follows a customized taxonomy of urban sound events. The system also uses a public API so potential users can post audio files to obtain a prediction matrix reporting the presence of noise sources contributing to environmental noise pollution. Experiments using recordings from two continuous noise monitoring stations showed that the amount of data to be inspected by the specialist is decreased by 97% when the deep-learning tools are used. Therefore, this system efficiently assists trained experts in enforcing noise legislation through machine-assisted environmental noise monitoring.

Auto noise is the main contaminated source of urban noises. Accurate identification of auto main noise sources is the premise of controlling auto noise. Sound intensity measurement is an important means of controlling automobiles noises. The principle of sound intensity measurement was discussed in this paper. T,the device of sound intensity measurement and its operating characteristics was analysed. By using a sound intensity measurement system to measure noises of a certain kind of vehicle, the surface radiation noises distribution in the engine was researched in the engine. Through Aannalysis of zing the noise sources according to the test data result,the research results shows that the exhaust system and intake system are the main noise sources. The position of main noise sources could be determined accurately by the sound intensity measurement method. T,and the frequency characteristics of main noise sources could bewas analysed conveniently, and the rule and relavant characteristics about the engine noise was obtained. Tgot, this method can provide a reasonable basis to correspondingly set down thefor noise control strategies. Auto noise pollution is one of the major noise sources in the world. T, and the main noises of automobiles are engine noises. It is necessary to reduce engine noises in order to reduce car noise. To do this, we must first identify the position and, frequency of its main sound sources at first,and then take some necessary measures must be taken forto address the main noise sourses. With its compact and complicated structure, the internal combustion engine has intense vibrations and a complex working process. Therefore, it is difficult to determine the distribution of the noise source effectively using general acoustic measurement technology. The sound intensity measurement technique is particularly suitable for noise source identification in internal combustion engines. It can effectively measure the size and distribution of the main noise source frequencies. Compared with general acoustic measurement techniques, it does not require a special acoustic environment. It can be done in general internal combustion engine bench test rooms with high efficiency (Pang, 2006).

A noise monitor for transmitting an actual sound (noise) when the noise level becomes higher than a regulation standard. In each noise detecting unit, the level of noise is measured through a microphone and a noise meter and input into a radio set. At the same time, an actual sound as noise is input through a microphone and a microphone amplifier into a PHS. In a noise monitoring unit, a PC communicates with the radio set in each noise detecting unit through a radio set to input the noise level measured by each noise detecting unit. The PC determines whether or not the noise level is higher than a regulation standard. If the noise level is higher than the regulation standard, the PC requests a PHS to call the PHS in the specified noise detecting unit to establish the communication between the PHS and the PHS. At this time, a maintenance worker can hear the actual sound from the PHS. Further, the actual sound is recorded by an MD recorder.

Excessive environmental noise and poor air quality can be adverse to human health, living comfort and the environment itself. Measurement of sound pressure levels and air quality in critical areas including libraries, campus areas, public parks and hospitals thus becomes necessary to monitor and mitigate existing noise levels. In a university environment, student activities will be less disrupted if the locations of the activities are sufficiently away from noise sources. The present study is intended to measure sound levels and air quality around the Engineering Complex, Universiti Teknologi Mara, Shah Alam. The measured data is compared with to acceptable sound pressure levels and air quality index specified by the Department of Environment (DOE), Malaysia. Sound pressure levels are measured using the Castle Sound Level Meter Type 6224 and air quality measurement was done by using the BW Gas Alert MicroClip XT device. Both measurements were conducted at five selected stations around the Engineering Complex for three times each weekday for five weeks. Results obtained indicated that sound levels at some locations and time zones are above the thresholds recommended by the DOE. The air quality is acceptable in most locations except the vicinity of a bus stop. With the growing number of students in the university and other factors like construction and redevelopment of existing roads, a continuously increasing noise situations and air pollution proportional to the traffic flow is inevitable.

Plates known as sound absorbing materials: textiles recovered fabrics plus polyethylene are made layered composite material intended to be used to encapsulate industrial engines, as sources of noise in the production departments. Influence of thickness sound-absorbing material can be supported by the values obtained for the absorption coefficient, but it can be studied as a study case in the anechoic chamber, where the sound pressure level is measured with a sound level meter. Cases captured: a free engine; the engine placed in a metallic cube and motor inserted; metal cube that was doubled separately with one material and then with the composite material proposed. For each situation are eight determinations on the perimeter of a circle with a radius of one meter from the source of the noise. Plot a graph of experimental results that support the idea: greater thickness of the layer of material will cause a better absorption of noise.

Advanced hydrostatic transmissions and hydraulic hybrids show potential in new market segments such as commercial vehicles and passenger cars. Such new applications regard low noise generation as a high priority, thus, demanding new quiet hydrostatic transmission designs. The aim of this paper is to investigate noise sources on a series hybrid transmission through simulation and measurements. A model has been developed to capture the interaction of a pump and motor working in a hydrostatic transmission and to predict overall noise sources. The model describes dynamics of the system by coupling lumped parameter pump and motor models with a one-dimensional unsteady compressible transmission line model including a dynamic model of an accumulator. A semi-anechoic chamber has been designed and constructed for sound intensity measurements that can be used to derive sound power. Sound power measurements were conducted on a series hybrid transmission test bench inside the semi-anechoic chamber in order to study the relationship between sound power and two types of noise sources, fluid and structure borne. The focus of these measurements was to investigate the impact of an accumulator in the high pressure line as well as the influence of varying high pressure line length. Results show a strong influence of changing line length and the addition of an accumulator on pressure ripple, but with little impact on sound power. A high correlation was found between sound power levels and control moment amplitudes on the swash plate. This study demonstrates the usefulness of predicting transmission noise sources, and how this information is beneficial in the design process of a transmission.

Noise pollution can come from noise generated by nature, one of which is noise from river flows. Therefore, the aim of this research was to determine whether the sound intensity level of the river flow is still below the threshold value or whether it has disturbed human hearing. The research method used was to measure the sound intensity level of the Cinangneng-Bogor river flow by using a sound level meter application. Measurement of the sound intensity level was carried out at 5 measurement points, namely in the middle of the river flow, on the river bank, at a distance of 5 m, 10 m and 15 m from the river bank to the land. Each measurement was carried out for 120 seconds. The results showed that the sound intensity level of the Cinangneng-Bogor river flow at several measurement points had different values, namely from the middle of the river to away from the river the values were 96.34 DbA, 92.31 DbA, 78.93 DbA, 76.58 DbA and 76.32 DbA. In conclusion, the further away from the noise source, the lower the noise level. The sound intensity in the middle and on the river bank was above the threshold value.

Environmental noise indicators and acoustic indexes based on fuzzy modelling for urban spaces

Noise assessment and monitoring are essential parts of an acoustician's work since it helps to understand the environment and propose better solutions for noise control and urban noise management. Traditionally, equipment to carry out this task is standardized, and, eventually, expensive for the early career professional. This work develops a high-quality (and cost-effective) prototype for an embedded noise monitoring device based upon a digital I2S MEMS microphone and an Arduino compatible microcontroller, named Teensy. Its small size and low power consumption are also advantages designed for the project. The system captures and processes sound in real-time, computes A and C frequency-weighted equivalent sound levels, along with time-weighted instant levels with a logging interval of 125 ms. Part of the software handles the audio environment, while the biquadratic IIR filters present in the Cortex Microcontroller library are responsible for the frequency- and time-weightings - using floating-point for enhanced precision. The prototype results were compared against a Class 2 Sound Level Meter, rendering very similar results for the tested situations, proving a powerful and reliable tool. Improvements and further testing are also being conducted to refine its functioning and characterization. Ultimately, the prototype achieved promising performance, confirming as a solution for noise monitoring.

Time domain identification and ranking of noise sources in a pneumatic nail gun

In this paper the composition, operation, and performance capabilities of a shipboard noise and vibration monitoring system will be described. The uniqueness of this system is that it consists of readily available off-the-shelf instrumentation designed to measure or monitor both the mechanical and/or acoustic condition of shipboard machinery. For this particular application the system has been tailored for use aboard a new class of Navy ships.

Airport noise has been monitored at Logan International Airport since 1975. Recently, Massport decided to install a new noise monitoring system. The new system will monitor trends in noise impact by time of day, season, and on an annual basis; measure noise levels generated by aircraft taxiing, ground operations, reverse thrust, takeoffs, landings, and overflights; obtain accurate data on aircraft flight tracks and fleet mix to be used in the generation of noise contours; and assess the performance of the Logan Preferential Runway Advisory System in terms of runway utilization and noise exposure. To do this, noise data from some sites will be obtained in 1/3-octave bands as well as in the traditional A-weighting; C-weighting is available; radar tracking information is expected to be part of the system; and weather data will be obtained at about half of the sites. All data are saved in database for study of the correlation of noise with various factors. A complaint database, contour prediction mode (FAA integrated noise model), and mapping software are also part of the system. It is expected that this system will be used to evaluate different noise descriptors against actual complaints, and to suggest improvements to noise prediction models based on geography or meteorological effects.

Airport take-off noise assessment aimed at identify responsible aircraft classes