Traffic noise real-time monitoring and early warning method suitable for residential area

Abstract

The invention discloses a traffic noise real-time monitoring and early warning method suitable for a residential area. The method comprises the following steps: S1, acquiring noise data through a noise acquisition device; analyzing the intensity and frequency information of the noise; s2, carrying out adaptive coding on the obtained noise information, converting the noise information into fractal code streams, and transmitting the fractal code streams to a monitoring center through multipath transmission; s3, restoring the original noise data by using a fractal decoding algorithm, constructing a residential area traffic noise model, and judging whether the current noise exceeds a threshold range specified by a residential area environmental noise standard or not; s4, noise early warning information is sent through wireless communication, and the position of a noise source is determined by using the noise traceability model; and S5, tracking a noise source position and a noise intensity change condition, and performing feedback optimization monitoring and early warning. According to the invention, efficient monitoring, early warning and positioning of residential area traffic noise are realized.

Description

Technical Field

[0001] The present invention relates to the technical field of noise monitoring, and in particular to a real-time monitoring and early warning method for traffic noise in residential areas. Background Art

[0002] With the acceleration of urbanization, traffic volume around residential areas is increasing, and traffic noise is becoming an increasingly prominent problem. Traffic noise not only disrupts residents' normal lives, studies, and rest, but long-term exposure to high noise levels can also negatively impact their health. Real-time monitoring and early warning of traffic noise in residential areas are crucial to promptly identify and address noise pollution issues and protect residents' quality of life and health.

[0003] Among traditional traffic noise monitoring technologies, single-point fixed monitoring is the most common. This method employs noise monitoring equipment installed at specific locations to collect noise data at regular intervals. Its advantages lie in its relatively low equipment cost, ease of installation and maintenance, and ability to provide data support for localized noise monitoring. However, this technology has several significant drawbacks. Single-point monitoring only reflects noise levels near the monitoring point, failing to fully capture the noise distribution across an entire residential area, resulting in monitoring blind spots. Furthermore, the regularly collected data cannot be monitored in real time, hindering the ability to capture sudden noise fluctuations, making it difficult to meet the demand for real-time traffic noise warnings. Furthermore, traditional monitoring methods lack effective means of locating noise sources. When noise levels exceed standards, it is difficult to quickly and accurately determine the source, hindering the implementation of targeted noise reduction measures.

[0004] With technological advancements, existing traffic noise monitoring technologies have improved to some extent. Some technologies utilize wireless sensor networks to implement multi-point distributed monitoring, enabling the acquisition of noise data over a wider area and, to some extent, compensating for the shortcomings of single-point monitoring. Furthermore, some technologies are beginning to incorporate signal processing algorithms to analyze noise data, aiming to achieve more accurate measurements of noise intensity and frequency. However, existing technologies still face numerous challenges. For one thing, wireless sensor networks are susceptible to interference during data transmission, resulting in data loss or errors, which affects monitoring accuracy. Furthermore, existing noise analysis algorithms are not ideal for processing noise data in complex environments. They do not fully exploit the fractal characteristics of traffic noise signals, making them inadequate for more efficient monitoring and early warning. Furthermore, while some methods for tracing noise sources rely on signal time difference of arrival (TDOA) and signal strength, in practice, due to factors such as multipath propagation and signal obstruction, positioning accuracy often falls short of practical requirements.

[0005] When it comes to traffic noise monitoring and early warning in residential areas, existing technologies, whether traditional single-point monitoring or more advanced multi-point distributed monitoring, are ill-suited to the complex and ever-changing urban environment and residents' demands for a high-quality living environment. Existing technologies face numerous limitations when it comes to critical issues such as real-time monitoring of traffic noise, accurate early warning, and precise location of noise sources. Therefore, there is an urgent need for innovative solutions that can overcome these shortcomings, achieve efficient monitoring and early warning of traffic noise in residential areas, and create a quiet and comfortable living environment for residents. Summary of the Invention

[0006] Based on the above, this application discloses a real-time monitoring and early warning method for traffic noise in residential areas, comprising:

[0007] S1. Acquire noise data using a noise collection device. The noise collection device consists of a cavity containing a magnetic fluid and sensing electrodes. When traffic noise causes the cavity to vibrate, the magnetic fluid generates an induced current under the action of a magnetic field. The intensity and frequency of the noise are obtained by analyzing the current changes.

[0008] S2. Adaptively encode the acquired noise information based on the fractal characteristics of the traffic noise signal, convert it into a fractal code stream with self-similarity, and transmit the fractal code stream to the monitoring center through different paths via multipath transmission;

[0009] S3. After receiving the fractal code stream, the monitoring center uses a fractal decoding algorithm to restore it to the original noise data, constructs a residential area traffic noise model, and compares and analyzes the real-time noise data with the model data to determine whether the current noise exceeds the threshold range specified by the residential area environmental noise standard;

[0010] S4. Send noise warning information via wireless communication, use the noise source tracing model, receive multi-path transmission noise data, combine signal arrival time difference and signal strength parameters, and determine the location of the noise source through array signal processing algorithm;

[0011] S5. Track the location of the noise source and the changes in noise intensity, and store the detailed data of the noise event, including the noise intensity change curve, duration, noise source location, warning activation and release time information, to optimize monitoring and early warning.

[0012] Preferably, the noise data in S1 is obtained by a noise collection device, which consists of a cavity with a built-in magnetic fluid and an induction electrode. When traffic noise causes the cavity of the noise collection device to vibrate, the magnetic fluid swings in the magnetic field generated by the bar magnet at the bottom of the cavity, cutting the spiral induction electrode to generate an induced current. By processing the induced current, the frequency distribution of the current is obtained, and combined with a preset current intensity-noise intensity correspondence, the noise intensity and frequency information are obtained.

[0013] Preferably, the induced current is processed to obtain the frequency distribution of the current, and the noise intensity and frequency information are obtained by combining the preset current intensity-noise intensity model. Specifically, the collected induced current signal i(t) is subjected to discrete wavelet transform to obtain the wavelet coefficient matrix W of different frequency bands. j，k , calculate the energy distribution of each frequency band, the formula is: Where j is the wavelet decomposition layer number, k is the time index, α is the frequency band weight coefficient, and j0 is the center layer of the traffic noise characteristic frequency band; the preset current intensity-noise intensity model is: Among them I rm is the effective value of the induced current, P no is the noise sound pressure level, β is the proportional coefficient to the magnetic susceptibility of the magnetic fluid, γ is the nonlinear correction index, and the noise intensity is obtained by inversion. The formula is: And through the energy distribution E f The peak position of the noise determines the main frequency component.

[0014] Preferably, in S2, the noise information obtained is adaptively encoded according to the fractal characteristics of the traffic noise signal and converted into a fractal code stream with self-similarity. Specifically, the noise intensity sequence P(n) and the frequency sequence f(n) of the noise information are combined into a complex noise feature vector Z(n), which is expressed as follows: where f s is the sampling frequency, and fractal encoding is performed through fractal function. The formula is: Where D(n) is the local fractal dimension of the noise signal, D avy is its mean, α s is the fractal enhancement index, S f (k) is the fractal code stream coefficient after encoding, N is the total number of sampling points, and a self-similar fractal code stream containing noise intensity and frequency information is generated.

[0015] Preferably, in said S2, the fractal code stream is transmitted to the monitoring center through different paths by multipath transmission, the fractal code stream is divided into multiple sub-code streams with the same fractal dimension according to self-similarity, and an independent transmission path is allocated to each sub-code stream according to the topological structure of the residential road network and the deployment of wireless communication nodes; the transmission path includes a heterogeneous network consisting of a cellular network, LoRaWAN, Wi-Fi and dedicated short-range communication, and an adaptive redundant transmission strategy of fractal coding is used to calculate the fractal complexity R of each sub-code stream. i Determine the redundancy of each sub-stream and allocate the transmission path. The formula is: Among them D i is the fractal dimension of the ith sub-stream, D max is the maximum fractal dimension of all sub-streams, r max To preset the maximum redundancy, the data is encoded and transmitted along different paths.

[0016] Preferably, after receiving the fractal code stream in S3, the monitoring center uses a fractal decoding algorithm to restore it to the original noise data, specifically: f (k) Perform time-space synchronization, filter out independent code streams, and use the multipath fusion algorithm to decode and restore. The formula is: where ω i is the path weight coefficient, I represents the number of independent code streams, and the original noise feature vector Z(n) is reconstructed by inverse fractal Fourier transform.

[0017] Preferably, the residential area traffic noise model of S3 is based on the fractal code stream of multipath transmission, using a multi-scale fractal recursive neural network, and the noise data transmitted by each path is used as the network input layer according to the self-similar hierarchical relationship of the fractal code stream; the multi-path data is fused by constructing a path feature fusion module, and the formula is: where λ i is the weight coefficient dynamically adjusted according to the path signal-to-noise ratio, Δt i is the time difference between the signal of the i-th path and the reference path, I represents the number of independent code streams, S f，i (k) is the fractal code stream coefficient of the i-th path. The fused data is input into the hidden layer of the recursive neural network, and the network parameters are optimized using the back propagation algorithm to output the noise intensity and frequency prediction distribution.

[0018] Preferably, the S3 compares and analyzes the real-time noise data with the model data to determine whether the current noise exceeds the threshold range specified by the residential area environmental noise standard, specifically: converting the real-time noise data into a vector in the fractal feature space Among them D R is the fractal dimension of real-time noise, β Ris the spectral index, and the model data is mapped to the same feature space to obtain the reference vector Calculate the Mahalanobis distance in the fractal feature space to determine the degree of noise abnormality. The formula is: Where Σ is the covariance matrix of real-time data and model data, combined with the environmental noise standard threshold T std , construct dynamic threshold surface where α r is the environmental correction coefficient, λ is the attenuation factor, and when the energy spectral density E(f) of the real-time noise data satisfies E(f)>T(f) at any frequency f, it is determined that the current noise exceeds the standard threshold range.

[0019] Preferably, the S4 uses a noise source tracing model, combines the noise data of multipath transmission, the signal arrival time difference, and the signal strength parameter, and determines the noise source position through an array signal processing algorithm, specifically: constructing a spatiotemporal covariance matrix R from the noise data of multipath transmission, weighting the spatiotemporal covariance matrix, and obtaining a weighted matrix The signal arrival time difference is used to construct the delay constraint matrix T, and the spatial spectrum function is constructed through the improved multiple signal classification algorithm. The formula is: in is the array manifold vector, θ and They represent the azimuth and elevation angles respectively. By searching the peak position of the spatial spectrum function and combining it with the weighted centroid positioning algorithm, the three-dimensional spatial coordinates of the noise source are determined.

[0020] Preferably, in said S5, the noise intensity change curve of this noise event is stored in a fractal compression format, and the frequency information is stored through feature coding. For the noise source location data, the location information is mapped to the grid nodes by constructing a dynamic topological grid of the geographic information system, and the transition probability matrix between the nodes is recorded. A multipath transmission data check code, fractal decoding parameters, and model calibration factors are generated. The stored noise event data is used as a training sample using a reinforcement learning algorithm to update the monitoring node layout optimization strategy and the multipath transmission path allocation scheme in real time. According to the change trend of the fractal characteristics of the noise data, the fractal coding and decoding parameters are adaptively adjusted.

[0021] Compared with the prior art, the technical solution of this application has the following technical effects:

[0022] This invention utilizes a noise acquisition device consisting of a cavity containing a magnetic fluid and sensing electrodes. The magnetic fluid oscillates in a magnetic field, generating an induced current that cuts through the sensing electrodes. This current is then processed using advanced algorithms such as discrete wavelet transforms to accurately capture noise intensity and frequency information. This not only accurately captures noise signals but also utilizes fractal characteristics for adaptive encoding, converting noise data into a self-similar fractal code stream. This method preserves the inherent characteristics of noise data, providing more accurate data support for building residential traffic noise models. This significantly improves noise monitoring accuracy and makes the acquired noise data more relevant to actual conditions.

[0023] The present invention divides a fractal code stream into multiple sub-code streams based on self-similarity. Based on the residential road network topology and wireless communication node deployment, an independent transmission path is assigned to each sub-code stream, employing a heterogeneous network encompassing multiple communication modes for multipath transmission. This approach improves the reliability and efficiency of data transmission, reducing the risk of data loss or transmission interruption. For noise source tracing, the method utilizes noise data from multipath transmission, combined with signal arrival time difference and signal strength parameters, to determine the location of the noise source using an improved array signal processing algorithm. This algorithm is not limited to simple positioning principles and can more accurately determine the three-dimensional spatial coordinates of the noise source in complex environments. Once excessive noise levels are detected, the noise source can be quickly located, improving the efficiency of noise control.

[0024] The present invention stores detailed data on noise events and uses a reinforcement learning algorithm to use this data as training samples to update the monitoring node layout optimization strategy and multipath transmission path allocation scheme in real time. According to the changing trend of the fractal characteristics of the noise data, the fractal encoding and decoding parameters are adaptively adjusted. Through long-term monitoring data accumulation and analysis, the system can continuously optimize its own performance and adapt to the dynamic changes in the residential environment. The intelligent optimization function of the present invention makes the monitoring and early warning system more flexible and adaptable, and can provide effective noise monitoring services for residential areas in a long-term and stable manner, continuously protecting residents from noise interference and improving their quality of life.

[0025] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application so that it can be implemented in accordance with the contents of the specification, and to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the following is a detailed description of the preferred embodiment of the present application in conjunction with the accompanying drawings.

[0026] Based on the detailed description of the specific embodiments of the present application in conjunction with the accompanying drawings below, those skilled in the art will become more aware of the above and other objects, advantages and features of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following will briefly introduce the drawings required for the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For those of ordinary skill in the art, other drawings can also be obtained based on these drawings without inventive work. In all drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn according to the actual scale.

[0028] Figure 1 This is a flow chart of a real-time monitoring and early warning method for traffic noise in residential areas according to the present invention;

[0029] Figure 2 This is a structural diagram of a real-time monitoring and early warning method for traffic noise in residential areas according to the present invention;

[0030] Figure 3 Schematic diagram of the simulated noise source location;

[0031] Figure 4 This is a comparison diagram of positioning deviations between the present invention and the prior art. DETAILED DESCRIPTION

[0032] In order to make the purpose, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all of the embodiments. In the following description, specific details such as specific configurations and components are provided only to help fully understand the embodiments of the present application. Therefore, it should be clear to those skilled in the art that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the present application. In addition, for clarity and brevity, the description of known functions and structures has been omitted in the embodiments.

[0033] It should be understood that references throughout this specification to "one embodiment" or "this embodiment" mean that a particular feature, structure, or characteristic associated with the embodiment is included in at least one embodiment of the present application. Therefore, the appearance of "one embodiment" or "this embodiment" throughout this specification does not necessarily refer to the same embodiment. Furthermore, these particular features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.

[0034] In addition, the present application may repeat reference numerals and / or letters in different examples. This repetition is for the purpose of simplicity and clarity and does not in itself indicate the relationship between the various embodiments and / or settings discussed.

[0035] The term "and / or" in this article is only a description of the association relationship of associated objects, indicating that three relationships can exist. For example, A and / or B can mean: A exists alone, B exists alone, and A and B exist at the same time. The term " / and" in this article describes another type of association object relationship, indicating that two relationships can exist. For example, A / and B can mean: A exists alone, and A and B exist alone. In addition, the character " / " in this article generally indicates that the previous and subsequent associated objects are in an "or" relationship.

[0036] The term "at least one" in this article is merely a description of the association relationship between associated objects, indicating that three relationships may exist. For example, at least one of A and B can mean: A exists alone, A and B exist at the same time, and B exists alone.

[0037] It should also be noted that, in this document, relational terms such as first and second are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "include," "comprises," or any other variations thereof are intended to cover non-exclusive inclusion.

[0038] Example 1

[0039] This embodiment mainly describes a real-time monitoring and early warning method for traffic noise in residential areas. Figure 1-2 As shown, including:

[0040] S1. Acquire noise data using a noise collection device. The noise collection device consists of a cavity containing a magnetic fluid and sensing electrodes. When traffic noise causes the cavity to vibrate, the magnetic fluid generates an induced current under the action of a magnetic field. By analyzing the current changes, the intensity and frequency of the noise are obtained.

[0041] S2. Adaptively encode the acquired noise information based on the fractal characteristics of the traffic noise signal, convert it into a fractal code stream with self-similarity, and transmit the fractal code stream to the monitoring center through different paths via multipath transmission;

[0042] S3. After receiving the fractal code stream, the monitoring center uses a fractal decoding algorithm to restore it to the original noise data, constructs a residential area traffic noise model, and compares and analyzes the real-time noise data with the model data to determine whether the current noise exceeds the threshold range specified by the residential area environmental noise standard;

[0043] S4. Send noise warning information via wireless communication, use the noise source tracing model, receive multi-path transmission noise data, combine signal arrival time difference and signal strength parameters, and determine the location of the noise source through array signal processing algorithm;

[0044] S5. Track the location of the noise source and the changes in noise intensity, and store the detailed data of the noise event, including the noise intensity change curve, duration, noise source location, warning activation and release time information, to optimize monitoring and early warning.

[0045] Furthermore, the noise collection device in S1 consists of a cavity with a built-in magnetic fluid and an induction electrode. When traffic noise causes the cavity of the noise collection device to vibrate, the built-in magnetic fluid begins to swing in the magnetic field generated by the bar magnet at the bottom of the cavity; this swing causes the magnetic fluid to cut the spiral induction electrode, thereby generating an induced current. The spiral induction electrode is used to increase the effective area and time of the magnetic fluid cutting, improve the efficiency and stability of the induced current generation, and thus more sensitively capture the subtle changes caused by traffic noise; after obtaining the induced current, the noise intensity and frequency information are obtained; the induced current is processed to obtain its frequency The system uses discrete wavelet transform to decompose the induced current signal into different frequency bands, thereby analyzing the characteristics of the current at each frequency, and combining the preset current intensity-noise intensity correspondence to obtain the noise intensity. The system comprehensively considers the characteristics of the magnetic fluid, the magnetic field strength, the material and structural factors of the induction electrode to ensure that the current intensity can be accurately converted into noise intensity. In actual application scenarios, whether it is a residential area next to a busy main road or a relatively quiet residential area near a side road where vehicles occasionally pass by, the noise collection device can accurately obtain noise intensity and frequency information with its ingenious design and scientific processing methods.

[0046] Furthermore, the induced current is processed to obtain the frequency distribution of the current. The noise intensity and frequency information are obtained by combining the preset current intensity-noise intensity model. Specifically, the collected induced current signal i(t) is subjected to discrete wavelet transform to obtain the wavelet coefficient matrix W of different frequency bands. j，k , calculate the energy distribution of each frequency band, the formula is: Where j is the wavelet decomposition layer number, k is the time index, α is the frequency band weight coefficient, and j0 is the center layer of the traffic noise characteristic frequency band; the preset current intensity-noise intensity model is: Among them I rm is the effective value of the induced current, P no is the noise sound pressure level, β is the proportional coefficient to the magnetic susceptibility of the magnetic fluid, γ is the nonlinear correction index, and the noise intensity is obtained by inversion. The formula is: And through the energy distribution E f The peak position of the noise determines the main frequency component.

[0047] Furthermore, in S2, the noise information is adaptively encoded according to the fractal characteristics of the traffic noise signal and converted into a fractal code stream with self-similarity. Specifically, the noise intensity sequence P(n) and frequency sequence f(n) of the noise information are combined into a complex noise feature vector Z(n), which is expressed as follows: where f s is the sampling frequency, and fractal encoding is performed through fractal function. The formula is: Where D(n) is the local fractal dimension of the noise signal, D avy is its mean, α s is the fractal enhancement index, S f (k) is the fractal code stream coefficient after encoding, N is the total number of sampling points, and a self-similar fractal code stream containing noise intensity and frequency information is generated.

[0048] Furthermore, in S2, the fractal code stream is transmitted to the monitoring center through different paths through multipath transmission. The fractal code stream is divided into multiple sub-code streams with the same fractal dimension according to self-similarity. An independent transmission path is assigned to each sub-code stream according to the topological structure of the residential road network and the deployment of wireless communication nodes. The transmission path includes a heterogeneous network consisting of cellular network, LoRaWAN, Wi-Fi and dedicated short-range communication. The adaptive redundant transmission strategy of fractal coding is used to calculate the fractal complexity R of each sub-code stream. i Determine the redundancy of each sub-stream and allocate the transmission path. The formula is: Among them D i is the fractal dimension of the ith sub-stream, D max is the maximum fractal dimension of all sub-streams, r max To preset the maximum redundancy, the data is encoded and transmitted along different paths.

[0049] Furthermore, after receiving the fractal code stream, the monitoring center in S3 uses the fractal decoding algorithm to restore it to the original noise data. Specifically, the received multipath fractal code stream S f (k) Perform time-space synchronization, filter out independent code streams, and use the multipath fusion algorithm to decode and restore. The formula is: where ω i is the path weight coefficient, I represents the number of independent code streams, and the original noise feature vector Z(n) is reconstructed by inverse fractal Fourier transform.

[0050] Furthermore, the residential area traffic noise model of S3 uses a multi-scale fractal recursive neural network based on the fractal code stream of multipath transmission, and uses the noise data transmitted by each path as the network input layer according to the self-similar hierarchical relationship of the fractal code stream. The multi-path data is fused by constructing a path feature fusion module. The formula is: where λi is the weight coefficient dynamically adjusted according to the path signal-to-noise ratio, Δt i is the time difference between the signal of the i-th path and the reference path, I represents the number of independent code streams, S f，i (k) is the fractal code stream coefficient of the i-th path. The fused data is input into the hidden layer of the recursive neural network, and the network parameters are optimized using the back propagation algorithm to output the noise intensity and frequency prediction distribution.

[0051] Furthermore, S3 compares and analyzes the real-time noise data with the model data to determine whether the current noise exceeds the threshold range specified by the residential environmental noise standard. Specifically, the real-time noise data is converted into a vector in the fractal feature space. Among them D R is the fractal dimension of real-time noise, β R is the spectral index, and the model data is mapped to the same feature space to obtain the reference vector Calculate the Mahalanobis distance in the fractal feature space to determine the degree of noise abnormality. The formula is: Where Σ is the covariance matrix of real-time data and model data, combined with the environmental noise standard threshold T std , construct dynamic threshold surface where α r is the environmental correction coefficient, λ is the attenuation factor, and when the energy spectral density E(f) of the real-time noise data satisfies E(f)>T(f) at any frequency f, it is determined that the current noise exceeds the standard threshold range.

[0052] Furthermore, S4 uses the noise source tracing model, combines the noise data of multipath transmission, signal arrival time difference, and signal strength parameters, and determines the location of the noise source through the array signal processing algorithm. Specifically, the multipath transmission noise data is used to construct the spatiotemporal covariance matrix R, and the spatiotemporal covariance matrix is ​​weighted to obtain the weighted matrix The signal arrival time difference is used to construct the delay constraint matrix T, and the spatial spectrum function is constructed through the improved multiple signal classification algorithm. The formula is: in is the array manifold vector, θ and They represent the azimuth and elevation angles respectively. By searching the peak position of the spatial spectrum function and combining it with the weighted centroid positioning algorithm, the three-dimensional spatial coordinates of the noise source are determined.

[0053] Furthermore, S5 stores the noise intensity change curve of this noise event in a fractal compression format, and the frequency information is stored through feature coding. For the noise source location data, the dynamic topological grid of the geographic information system is constructed to map the location information to the grid nodes, and the transition probability matrix between the nodes is recorded. A check code containing multipath transmission data, fractal decoding parameters, and model calibration factors is generated. The reinforcement learning algorithm is used with the stored noise event data as training samples to update the monitoring node layout optimization strategy and the multipath transmission path allocation scheme in real time. According to the changing trend of the fractal characteristics of the noise data, the fractal coding and decoding parameters are adaptively adjusted.

[0054] This embodiment describes in detail the entire process from noise collection to monitoring, early warning, and noise source location. Through noise collection devices and advanced algorithms, noise data can be accurately acquired. Fractal coding and multipath transmission ensure reliable data transmission. The monitoring center uses fractal decoding and model construction to accurately determine whether noise levels exceed standards. The noise source tracing model accurately locates noise sources. Data storage and optimization mechanisms allow the system to continuously evolve, overall improving the accuracy, reliability, and intelligence of residential traffic noise monitoring and early warning.

[0055] Based on Example 1, this embodiment describes in detail the determination of the three-dimensional spatial coordinates of the noise source of the present application, specifically:

[0056] The noise source location is determined by array signal processing algorithm. Specifically, the space-time covariance matrix R is constructed from the noise data of multipath transmission, and the space-time covariance matrix is ​​weighted to obtain the weighted matrix The signal arrival time difference is used to construct the delay constraint matrix T, which reflects the time difference of the noise signal propagating from the source to different receiving paths. In order to balance the reliability of the data of each path, after obtaining the weighted matrix, the improved multiple signal classification algorithm is used to construct the spatial spectrum function in is the array manifold vector, which describes the noise signal from the spatial perspective The response characteristics of the receiving array are obtained by searching for the peak position of the spatial spectrum function to preliminarily determine the azimuth and elevation angles of the noise source. Due to the influence of multipath propagation and environmental noise, a single spectrum peak search will lead to positioning deviation. To improve positioning accuracy, the weighted centroid positioning algorithm is combined to convert the multiple candidate azimuth and elevation angles obtained by the search into three-dimensional spatial coordinates. For each candidate angle position, its corresponding spatial coordinates are calculated, and the value of the spatial spectrum function at that position is used as the weight to construct a weighted centroid model. The formula is: where r i is the three-dimensional coordinate of the i-th candidate position, w ri is the corresponding spatial spectrum function value, N rIn order to calculate the number of candidate locations of the weighted centroid, the peak information of the spatial spectrum function is converted into three-dimensional coordinates. At the same time, a weighting mechanism is used to reduce the impact of false peaks on the positioning results. By integrating fractal noise characteristics and multipath transmission characteristics, this algorithm can achieve high-precision three-dimensional positioning of noise sources in complex urban environments, providing strong technical support for traffic noise control.

[0057] This embodiment describes in detail how to accurately determine the three-dimensional spatial coordinates of noise sources in complex environments. By constructing a space-time covariance matrix, a delay constraint matrix, and weighted processing, combined with an improved multiple signal classification algorithm and a weighted centroid positioning algorithm, it effectively overcomes the influence of multipath propagation and environmental noise, quickly and accurately locates the location of the noise source, provides a key basis for traffic noise control, and helps create a quiet residential environment.

[0058] Based on Example 1, this implementation describes in detail the specific implementation effects and comparative verification of this application, specifically:

[0059] As shown in Figure 3, in the location experiment for residential traffic noise monitoring, a typical residential area with a complex environment was selected. This area contains different types of buildings and a variety of traffic conditions. Three simulated noise sources were set up: point A near the main road, point B next to the road inside the community, and point C in the corner of the community near the commercial area.

[0060] Existing technology uses a triangulation positioning method based on signal strength, relying on multiple sensors distributed in residential areas to receive noise signal strength and using trigonometric function relationships to estimate the location of the noise source. Since the signal is affected by factors such as building obstruction and reflection during propagation, the positioning accuracy is limited. When locating the simulated noise source at point A, the signal is blocked by nearby high-rise buildings, and the actual positioning position deviates from the true position by 6.34 meters. Point B has a positioning deviation of 5.87 meters due to signal reflection interference from the surrounding environment. Point C has a positioning deviation of 7.12 meters due to its proximity to a commercial area with complex signals. The average positioning deviation reaches 6.44 meters, and the positioning error is large and unstable, making it difficult to accurately locate the noise source.

[0061] This application's technical solution combines multipath transmission noise data, signal arrival time difference, and signal strength parameters, and uses an improved array signal processing algorithm for positioning. At point A, a simulated noise source is used, and through comprehensive analysis of multipath transmission information, the positioning error is only 2.15 meters. At point B, using precise calculations such as signal arrival time difference, the positioning error is 2.43 meters. At point C, even in complex environments, the positioning error is 2.78 meters, relying on the multi-parameter processing technology of this application. The average positioning error is 2.45 meters, which is significantly lower than that of existing technologies.

[0062] like Figure 4As shown, it can be clearly seen from the positioning deviation comparison diagram that the point representing the technical solution of the present application is significantly lower than the point representing the prior art, indicating that the technology of the present application is more accurate in locating the noise source, can effectively overcome the interference of complex environments, and provide a more reliable basis for traffic noise control in residential areas, greatly improving the pertinence and efficiency of noise control, and ensuring a quiet and comfortable living environment for residents.

[0063] Through specific implementation comparison, this embodiment verifies that in the experiment of locating traffic noise sources in residential areas, the technology of the present application reduces the average positioning deviation of three simulated noise sources from 6.44 meters to 2.45 meters compared with the existing technology based on signal strength triangulation positioning method. It can accurately locate and effectively overcome the interference of complex environments, provide a reliable basis for noise control, greatly improve the control efficiency, and effectively ensure a quiet and comfortable living environment for residents.

[0064] The above are only preferred embodiments of the present invention, which do not limit the scope of protection of the present invention. For those skilled in the art, the present invention can be modified and varied in various ways. Any changes, modifications, replacements, integrations and parameter changes to these embodiments through conventional substitutions or that can achieve the same functions without departing from the principles and spirit of the present invention fall within the scope of protection of the present invention.

Claims

1. A real-time monitoring and early warning method for traffic noise in residential areas, characterized in that: include: S1. Acquire noise data through a noise collection device; The noise collection device consists of a cavity with a built-in magnetic fluid and an induction electrode. When traffic noise causes the cavity to vibrate, the magnetic fluid generates an induced current under the action of the magnetic field. By analyzing the current changes, the intensity and frequency information of the noise are obtained; S2. Adaptively encode the acquired noise information based on the fractal characteristics of the traffic noise signal, convert it into a fractal code stream with self-similarity, and transmit the fractal code stream to the monitoring center through different paths via multipath transmission; S3. After receiving the fractal code stream, the monitoring center uses a fractal decoding algorithm to restore it to the original noise data, constructs a residential area traffic noise model, and compares and analyzes the real-time noise data with the model data to determine whether the current noise exceeds the threshold range specified by the residential area environmental noise standard; S4. Send noise warning information via wireless communication, use the noise source tracing model, receive multi-path transmission noise data, combine signal arrival time difference and signal strength parameters, and determine the location of the noise source through array signal processing algorithm; S5. Track the location of the noise source and the changes in noise intensity, and store the detailed data of the noise event, including the noise intensity change curve, duration, noise source location, warning activation and release time information, to optimize monitoring and early warning.

2. A method for real-time monitoring and early warning of traffic noise in residential areas according to claim 1, characterized in that: The noise data in S1 is obtained by a noise collection device, which consists of a cavity with a built-in magnetic fluid and an induction electrode. When traffic noise causes the cavity of the noise collection device to vibrate, the magnetic fluid swings in the magnetic field generated by the bar magnet at the bottom of the cavity, cutting the spiral induction electrode to generate an induced current. By processing the induced current, the frequency distribution of the current is obtained, and combined with a preset current intensity-noise intensity correspondence, the noise intensity and frequency information are obtained.

3. A method for real-time monitoring and early warning of traffic noise in residential areas according to claim 2, characterized in that: The induced current is processed to obtain the frequency distribution of the current, and the noise intensity and frequency information are obtained by combining the preset current intensity-noise intensity model. Specifically, the collected induced current signal i(t) is subjected to discrete wavelet transform to obtain the wavelet coefficient matrix W of different frequency bands. j，k , calculate the energy distribution of each frequency band, the formula is: Where j is the wavelet decomposition layer number, k is the time index, α is the frequency band weight coefficient, and j0 is the center layer of the traffic noise characteristic frequency band; The preset current intensity-noise intensity model is: φ), where I rm is the effective value of the induced current, P no is the noise sound pressure level, β is the proportional coefficient to the magnetic susceptibility of the magnetic fluid, γ is the nonlinear correction index, and the noise intensity is obtained by inversion. The formula is: And through the energy distribution E f The peak position of the noise determines the main frequency component.

4. The method for real-time monitoring and early warning of traffic noise in residential areas according to claim 1, characterized in that: In S2, the noise information is adaptively encoded according to the fractal characteristics of the traffic noise signal and converted into a fractal code stream with self-similarity. Specifically, the noise intensity sequence P(n) and the frequency sequence f(n) of the noise information are combined into a complex noise feature vector Z(n), which is expressed as follows: where f s is the sampling frequency, and fractal encoding is performed through fractal function. The formula is: Where D(n) is the local fractal dimension of the noise signal, D avy is its mean, α s is the fractal enhancement index, S f (k) is the fractal code stream coefficient after encoding, N is the total number of sampling points, and a self-similar fractal code stream containing noise intensity and frequency information is generated.

5. The method for real-time monitoring and early warning of traffic noise in residential areas according to claim 1, characterized in that: In S2, the fractal code stream is transmitted to the monitoring center through different paths by multipath transmission. The fractal code stream is divided into multiple sub-code streams with the same fractal dimension according to self-similarity. An independent transmission path is allocated to each sub-code stream according to the topological structure of the residential road network and the deployment of wireless communication nodes. The transmission path includes a heterogeneous network consisting of a cellular network, LoRaWAN, Wi-Fi, and a dedicated short-range communication. The adaptive redundant transmission strategy of fractal coding is used to calculate the fractal complexity R of each sub-code stream. i Determine the redundancy of each sub-stream and allocate the transmission path. The formula is: Among them D i is the fractal dimension of the ith sub-code stream, D max is the maximum fractal dimension among all sub-streams, r max To preset the maximum redundancy, the data is encoded and transmitted along different paths.

6. A method for real-time monitoring and early warning of traffic noise in residential areas according to claim 1 or 4, characterized in that: After receiving the fractal code stream in S3, the monitoring center uses the fractal decoding algorithm to restore it to the original noise data, specifically: f (k) Perform time-space synchronization, filter out independent code streams, and use the multipath fusion algorithm to decode and restore. The formula is: where ω i is the path weight coefficient, I represents the number of independent code streams, and the original noise feature vector Z(n) is reconstructed by inverse fractal Fourier transform.

7. The method for real-time monitoring and early warning of traffic noise in residential areas according to claim 6, characterized in that: The residential area traffic noise model of S3 is based on the fractal code stream of multipath transmission. It uses a multi-scale fractal recursive neural network to take the noise data transmitted by each path as the network input layer according to the self-similar hierarchical relationship of the fractal code stream. The multi-path data is fused by constructing a path feature fusion module. The formula is: where λ i is the weight coefficient dynamically adjusted according to the path signal-to-noise ratio, Δt i is the time difference between the signal of the i-th path and the reference path, I represents the number of independent code streams, S f,i (k) is the fractal code stream coefficient of the i-th path. The fused data is input into the hidden layer of the recursive neural network, and the network parameters are optimized using the back propagation algorithm to output the noise intensity and frequency prediction distribution.

8. The method for real-time monitoring and early warning of traffic noise in residential areas according to claim 1, characterized in that: The S3 compares and analyzes the real-time noise data with the model data to determine whether the current noise exceeds the threshold range specified by the residential area environmental noise standard. Specifically, the real-time noise data is converted into a vector in the fractal feature space. Among them D R is the fractal dimension of real-time noise, β R is the spectral index, and the model data is mapped to the same feature space to obtain the reference vector Calculate the Mahalanobis distance in the fractal feature space to determine the degree of noise abnormality. The formula is: Where Σ is the covariance matrix of real-time data and model data, combined with the environmental noise standard threshold T std , construct dynamic threshold surface where α r is the environmental correction coefficient, λ is the attenuation factor, and when the energy spectral density E(f) of the real-time noise data satisfies E(f)>T(f) at any frequency f, it is determined that the current noise exceeds the standard threshold range.

9. The method for real-time monitoring and early warning of traffic noise in residential areas according to claim 1, characterized in that: The S4 uses the noise source tracing model, combined with the noise data of multipath transmission, signal arrival time difference, and signal strength parameters, to determine the noise source location through the array signal processing algorithm. Specifically, the multipath transmission noise data is used to construct the spatiotemporal covariance matrix R, and the spatiotemporal covariance matrix is ​​weighted to obtain the weighted matrix The signal arrival time difference is used to construct the delay constraint matrix T, and the spatial spectrum function is constructed through the improved multiple signal classification algorithm. The formula is: in is the array manifold vector, θ and They represent the azimuth and elevation angles respectively. By searching the peak position of the spatial spectrum function and combining it with the weighted centroid positioning algorithm, the three-dimensional spatial coordinates of the noise source are determined.

10. The method for real-time monitoring and early warning of traffic noise in residential areas according to claim 1, characterized in that: In the S5, the noise intensity change curve of this noise event is stored in a fractal compression format, and the frequency information is stored through feature coding. For the noise source location data, the location information is mapped to the grid nodes by constructing a dynamic topological grid of the geographic information system, and the transition probability matrix between the nodes is recorded. A multipath transmission data check code, fractal decoding parameters, and model calibration factors are generated. The stored noise event data is used as a training sample using a reinforcement learning algorithm to update the monitoring node layout optimization strategy and the multipath transmission path allocation plan in real time. According to the change trend of the fractal characteristics of the noise data, the fractal coding and decoding parameters are adaptively adjusted.

Images