Audio signal enhancement method in tunnel environment based on directional sound wave

Abstract

This application relates to the field of audio enhancement technology, specifically to a method for enhancing audio signals in a tunnel environment based on directional sound waves. The method includes: real-time acquisition of audio signal data and audio decibel data at preset locations within the tunnel; dividing the entire data acquisition time interval into time periods, obtaining the peak fluctuation coefficient and peak difference coefficient at each location within each time period, and further obtaining the stability coefficient at each location within each time period; obtaining the short-term audio variation at each location within each time period by analyzing the changes in audio signal data over historical time periods; and obtaining the interference coefficient and adjustment urgency at each location within each time period to filter locations requiring broadcast audio signal enhancement. This application aims to improve the propagation quality of broadcast audio, thereby ensuring that drivers and passengers can clearly hear the content of tunnel broadcasts from various locations within the tunnel.

Description

Technical Field

[0001] This application relates to the field of audio enhancement technology, specifically to a method for enhancing audio signals in a tunnel environment based on directional sound waves. Background Technology

[0002] As tunnel engineering expands and the total length of highway tunnels increases, with the longest tunnels reaching significant lengths, public address systems are typically installed within tunnels to ensure traffic safety. These systems provide real-time traffic information, emergency notifications, and vehicle dispatch announcements. However, because tunnels are semi-enclosed environments with complex conditions and varying traffic patterns, broadcast audio is subject to various interferences during propagation. For example, audio signals may reflect multiple times off the tunnel walls, causing time delay expansion (multipath effect). Vehicle noise also interferes with audio signals, leading to echoes, reverberation, and distortion, making it difficult for drivers and passengers to hear the broadcasts clearly. Therefore, current tunnel broadcast systems employ directional sound wave technology and amplification systems to reduce echoes and reverberation and improve the listening experience.

[0003] However, noise levels inside highway tunnels are unstable, and there can be significant differences in noise levels at different locations within the tunnel. Traditional sound reinforcement systems, which only use equalization, may amplify the audio too much in areas with low noise levels, leading to increased reflections and worsened reverberation. Conversely, in areas with high noise levels, the amplification level may be too low, failing to effectively reduce reverberation. Ultimately, even after equalization, the quality of the broadcast audio remains poor, and drivers and passengers still cannot hear the broadcast content clearly. Summary of the Invention

[0004] In view of the above, it is necessary to provide an audio signal enhancement method based on directional sound waves in tunnel environments. Compared with traditional audio signal enhancement methods based on directional sound waves in tunnel environments, this method improves the propagation quality of broadcast audio, thereby ensuring that drivers and passengers can clearly hear the content of tunnel broadcasts from all parts of the tunnel.

[0005] The audio signal enhancement method based on directional sound waves in a tunnel environment proposed in this application adopts the following technical solution:

[0006] One embodiment of this application provides a method for enhancing audio signals in a tunnel environment based on directional sound waves, the method comprising the following steps:

[0007] Real-time acquisition of audio signal data and audio decibel data at preset locations within the tunnel;

[0008] The entire data acquisition time interval is divided into time periods. By comparing the number of peaks in the audio signal data at each location in each time period with the number of peaks in the adjacent time periods, and the peak width of the audio signal data at each location in each time period, the peak fluctuation coefficient at each location in each time period is obtained. In addition, by comparing the peak phase and peak dispersion of the audio signal data at each location in each time period with each previous time period, the peak difference coefficient at each location in each time period is obtained, and the stability coefficient at each location in each time period is obtained.

[0009] By analyzing the changes in audio signal data at each location over historical periods, the short-term audio variation at each location during each period is obtained. The tunnel is divided into left and right sides. By analyzing the spatial distribution of audio signal data at all locations on each side during each period, combined with the differences in short-term audio variation between each location and the location on its side, and the distance from each location to the tunnel exit on its side, the interference coefficient at each location during each period is obtained. Then, by combining the audio decibel data at each location during each period with the stability coefficient, the urgency of adjustment at each location during each period is obtained, in order to screen the locations that need to enhance the broadcast audio signal during each period.

[0010] In one embodiment, the process of obtaining the peak ripple coefficient is as follows:

[0011] Obtain the increase in the number of peaks in the audio signal data at each location within each time period compared to the number of peaks in the previous time period;

[0012] The acquisition times within each time period are numbered sequentially. The peaks and troughs of the audio signal data at each location within each time period are obtained. The difference in the acquisition times between the preceding and following troughs of each peak is taken as the peak width of each peak. The average peak width of all peaks of the audio signal data at each location within each time period is calculated.

[0013] The peak fluctuation coefficient is directly proportional to the increase and inversely proportional to the mean.

[0014] In one embodiment, the process of obtaining the peak difference coefficient is as follows:

[0015] Obtain the average phase difference between any two adjacent peaks of the audio signal data at each location within each time period; calculate the difference between the average value of the audio signal data at each location in each time period and the average value of each previous time period;

[0016] Calculate the dispersion of all peaks of the audio signal data at each location within each time period, and calculate the deviation of the dispersion of the audio signal data at each location between each time period and each previous time period;

[0017] The product of the difference and the deviation value is denoted as the difference product;

[0018] The peak difference coefficient is obtained by multiplying all the differences in the audio signal data at each location in each time period.

[0019] In one embodiment, the peak difference coefficient is the sum of all the difference products corresponding to the audio signal data at each location in each time period.

[0020] In one embodiment, the process of obtaining the stability coefficient is as follows:

[0021] The product of the peak fluctuation coefficient and the peak difference coefficient is denoted as the fluctuation product;

[0022] The stability coefficient is inversely proportional to the fluctuation product.

[0023] In one embodiment, the process of obtaining the short-term audio variation is as follows:

[0024] Calculate the arithmetic mean of the audio signal data at each location during each time period;

[0025] The control periods for each time period are preset, and the short-term change in audio is the mean of the deviation of the audio signal data at each location from the arithmetic mean between any two adjacent control periods in each time period.

[0026] In one embodiment, the process of obtaining the interference coefficient is as follows:

[0027] The tunnel is divided into left and right sides. The arithmetic mean of all locations on any side in each time period is sorted in ascending order according to the distance to the tunnel entrance, and the trend strength of the sorting results is calculated.

[0028] For any given location, the location closest to the tunnel entrance is designated as the tunnel entrance location. The difference in short-term audio variation between each location on any given side and the tunnel entrance location is calculated for each time period.

[0029] Obtain the normalized result of the distance from each position on either side to the tunnel entrance position;

[0030] The interference coefficient is inversely proportional to the difference and the normalization result, and directly proportional to the trend strength.

[0031] In one embodiment, the interference coefficient is calculated by mapping the product of the difference and the normalization result to a constant greater than 0; the interference coefficient is the ratio of the trend intensity to the constant.

[0032] In one embodiment, the process of obtaining the urgency of adjustment is as follows:

[0033] The product of the stability coefficient and the disturbance coefficient is denoted as the disturbance product;

[0034] The interference product is mapped to a positive number, and the adjustment urgency is the normalized value of the ratio of the audio decibel data at each location in each time period to the positive number.

[0035] In one embodiment, the method for selecting locations requiring broadcast audio signal enhancement in each time period is as follows:

[0036] Obtain the segmentation threshold of adjustment urgency at all locations under a preset number of historical time periods;

[0037] If the urgency of adjustment at each location in each time period is less than the segmentation threshold, it is determined that the broadcast audio signal at each location in each time period does not need to be enhanced; otherwise, it needs to be enhanced.

[0038] This application has at least the following beneficial effects:

[0039] This application, by dividing audio signal data into time periods, can segment and process the audio signal data, thereby capturing the dynamic changes in noise and reverberation to more accurately reflect the real-time state of the acoustic environment within the tunnel. By comparing the increase in the number of peaks between adjacent time periods, combined with the peak width, it can capture rapid changes in noise levels and noise stability in real time, quantifying noise fluctuations into specific values ​​to provide a basis for subsequent broadcast audio adjustments. Considering that reverberation in tunnels can cause echoes and distortion in broadcast audio, affecting audio clarity, the application calculates a peak difference coefficient to dynamically reflect reverberation changes, providing a basis for real-time adjustments to broadcast audio and reducing the impact of reverberation on audio clarity. By calculating a stability coefficient, the application comprehensively assesses the impact of noise and reverberation on broadcast audio, providing a more accurate basis for audio clarity assessment.

[0040] Furthermore, to avoid misinterpreting noise changes caused by natural wind speed in the tunnel as noise changes caused by vehicle movement, this study analyzes the propagation and distribution patterns of noise within the tunnel to accurately distinguish the impact of natural wind speed and vehicle movement on noise. By combining this with a stability coefficient, it assesses whether the broadcast audio signal needs to be enhanced at different locations within the tunnel at different times. This targeted enhancement ensures that drivers and passengers can clearly hear the tunnel broadcasts from all locations within the tunnel. This solves the problem of traditional audio signal enhancement methods in tunnel environments, which struggle to adjust to dynamically changing noise levels and reflections, resulting in poor broadcast audio transmission quality. Attached Figure Description

[0041] To more clearly illustrate the technical solutions and advantages in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0042] Figure 1 A flowchart illustrating the steps of the audio signal enhancement method based on directional acoustic waves in a tunnel environment provided in this application;

[0043] Figure 2 A schematic diagram of the acquisition process for adjusting urgency. Detailed Implementation

[0044] In the description of the embodiments in this application, the words "exemplary," "or," and "for example" are used to indicate examples, illustrations, or descriptions. Any embodiment or design scheme described as "exemplary" or "for example" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design schemes. Specifically, the use of the words "exemplary," "or," and "for example" is intended to present the relevant concepts in a specific manner.

[0045] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. It should be understood that, unless otherwise stated, " / " in this application means "or".

[0046] It should also be noted that the terms "first" and "second" in this application are used to distinguish similar objects, rather than to describe a specific order or sequence.

[0047] The following description, in conjunction with the accompanying drawings, details the specific scheme of the audio signal enhancement method based on directional sound waves in a tunnel environment provided in this application.

[0048] This application provides an embodiment of an audio signal enhancement method based on directional sound waves in a tunnel environment. Specifically, the method is described below. Please refer to [link to relevant documentation]. Figure 1 The method includes the following steps:

[0049] Step 1: Collect audio signal data and audio decibel data at preset locations within the tunnel in real time.

[0050] An audio sensor is placed at each pre-set location inside the tunnel to collect audio signal data and audio decibel data at each location in real time.

[0051] In this embodiment, a broadcast amplifier is placed every 200m in the tunnel, a loudspeaker is placed every 50m, and an audio sensor is set between every two broadcast amplifiers. For a tunnel with a total length of 4000m, a total of 17 audio sensors are placed in this embodiment. The audio sensor sampling frequency is 51.2kHz. Among them, 200m, 50m, 17 and 51.2kHz are only one embodiment of this application. Implementers can set specific values ​​according to actual conditions. This application does not impose any special restrictions.

[0052] To avoid the influence of dimensions on subsequent analysis, the collected audio signal data and audio decibel data were normalized respectively.

[0053] In this embodiment, the Min-Max normalization method is used to normalize the audio signal data and the audio decibel data respectively. The Min-Max normalization method is a well-known technology and will not be described in detail in this application.

[0054] Step 2: Divide the entire data acquisition time interval into time periods. By comparing the number of peaks in the audio signal data at each location within each time period with the number of peaks in the adjacent time periods, and the peak width of the audio signal data at each location within each time period, obtain the peak fluctuation coefficient at each location within each time period. Combined with the peak phase and peak dispersion of the audio signal data at each location within each time period with each previous time period, obtain the peak difference coefficient at each location within each time period, and obtain the stability coefficient at each location within each time period.

[0055] The main sources of noise within the tunnel are ventilation noise generated by the continuous operation of the tunnel's ventilation system and various noises produced by vehicles. Regarding ventilation noise, since the ventilation system operates continuously, the tunnel typically experiences a relatively high noise level. As for noise from vehicles entering the tunnel, it exhibits a pattern of high noise in the middle and low noise at both ends. Vehicles begin affecting the tunnel's noise level a short distance before entering, rapidly increasing it before stabilizing. Therefore, when vehicles enter the tunnel, the noise generated at different locations has a significant impact on the broadcast audio throughout the tunnel, necessitating adjustments to the broadcast system to accommodate potential noise variations.

[0056] Specifically, as a vehicle approaches the tunnel, it begins to interfere with the noise level inside the tunnel some distance before the tunnel entrance. New noise peaks gradually appear in the audio signals collected by various audio sensors, and the noise levels received by audio sensors at different locations vary. Specifically, the closer to the noise source, the more peaks, the higher the amplitude, and the more pronounced the differences in peak shape. In particular, if a vehicle is about to pass through the tunnel, more new peaks will appear in the audio signal data collected by the audio sensors, and the peak shapes will differ significantly at different locations.

[0057] In order to analyze the noise situation over a short period of time, the entire data collection time interval is divided into time periods of preset length.

[0058] In this embodiment, the preset length of the time period is 0.05s. The preset length of the time period is preset by the user and can be set by the implementer according to the actual situation. This application does not impose any special restrictions.

[0059] Based on the above analysis, by comparing the number of peaks in the audio signal data at each location within each time period and its adjacent time periods, as well as the peak width of the audio signal data at each location within each time period, the peak fluctuation coefficient at each location in each time period is obtained. The specific process is as follows:

[0060] Taking the t-th time period as an example, the peaks and troughs of the audio signal data at each location in the t-th time period are obtained, and the increase in the number of peaks in the audio signal data at each location in the t-th time period compared to the number of peaks in the (t-1)-th time period is obtained. The acquisition times in each time period are numbered sequentially, and the difference in the acquisition times between the preceding and following troughs of each peak is taken as the peak width of each peak. If the number of peaks in the t-th time period does not increase compared to the (t-1)-th time period, the increase in the number of peaks in the audio signal data at each location in the t-th time period compared to the number of peaks in the (t-2)-th time period is obtained. This process is repeated until the increase in the number of peaks is obtained. If the increase in the number of peaks cannot be obtained in the end, the value of the increase in the number of peaks is set to 1. 1 is merely one embodiment of this application, and the implementer can set its specific value according to the actual situation.

[0061] Calculate the mean peak width of all peaks in the audio signal data at each location during the t-th time period;

[0062] The peak fluctuation coefficient at each location in the t-th time period is directly proportional to the increase and inversely proportional to the mean.

[0063] In this embodiment, the expression for the peak fluctuation coefficient at each location during each time period is as follows:

[0064] In the formula, This represents the peak fluctuation coefficient at the i-th position during the t-th time period; This represents the increase in audio signal data at the i-th position during the t-th time period; This represents the mean value of the audio signal data at the i-th position during the t-th time period.

[0065] In this embodiment, a peak and valley detection algorithm is used to obtain each peak and valley point in the audio signal data. The peak and valley detection algorithm is a well-known technology and will not be described in detail in this application. As other implementation methods, based on the ability to obtain each peak and valley point in the audio signal data, the implementer may use other existing technologies, such as automatic multi-scale peak search algorithm, extreme point detection algorithm, etc. This application does not impose any special restrictions.

[0066] In this embodiment, the difference between the acquisition time numbers is the absolute value of the difference.

[0067] It should be noted that: if the audio signal data at position i in the t-th time period has a larger total number of newly added peaks in a short period and a smaller average peak width, it indicates that the position i is more likely to be affected by noise, resulting in changes in the acoustic environment and a more significant impact on the fluctuations generated by the broadcast.

[0068] In addition to the noise interference from vehicles, the transmission time difference between multiple speakers causes a large phase difference when the sound reaches the ears of drivers and passengers. Furthermore, multiple sound sources exacerbate the sound reflection in the tunnel, ultimately resulting in severe reverberation. Therefore, in addition to considering noise interference, further adjustments need to be made based on the reverberation conditions at different locations within the tunnel before amplifying the broadcast audio.

[0069] Specifically, regarding reverberation within tunnels, staff typically adjust the broadcasting system and loudspeakers beforehand to reduce transmission time differences and echoes between multiple loudspeakers. However, due to variations in traffic conditions and tunnel operations, additional factors such as large vehicles and tunnel equipment can obstruct the sound during loudspeaker operation. This leads to significant differences in the reflection of the broadcast audio signal after it is emitted, resulting in reverberation still present when it reaches the same location. Specifically, this manifests as a decrease in the phase difference and amplitude regularity of the waveform after local superposition in audio signal data collected at different locations and time periods, due to varying reflections of the audio signal on non-tunnel walls.

[0070] Based on the above analysis, by comparing the peak phase and peak dispersion of the audio signal data at each location in each time period with those in previous time periods, the peak difference coefficient at each location in each time period is obtained. The specific process is as follows:

[0071] Obtain the average phase difference between any two adjacent peaks of the audio signal data at each location within each time period; calculate the difference between the average value of the audio signal data at each location in each time period and the average value of each previous time period;

[0072] Calculate the dispersion of all peaks of the audio signal data at each location within each time period, and calculate the deviation of the dispersion of the audio signal data at each location between each time period and each previous time period;

[0073] The product of the difference and the deviation value is denoted as the difference product;

[0074] The sum of all the products of the differences in the audio signal data at each location in each time period is used as the peak difference coefficient at each location in each time period.

[0075] In this embodiment, the expression for the peak difference coefficient at each location during each time period is as follows:

[0076] In the formula, represents the peak difference coefficient at the i-th position in the t-th time period; J represents the total number of time periods before the t-th time period; This represents the difference in the average value of the audio signal data at the i-th position between the t-th time period and the j-th time period preceding it; This represents the deviation value of the dispersion of the audio signal data at position i between time period t and its preceding time period j. This is denoted as the difference product.

[0077] In this embodiment, the difference between the average values ​​is the absolute value of the difference. As another implementation, based on the ability to measure the degree of difference between the average values, the implementer may use other calculation methods, such as ratio, square of difference, etc. This application does not impose any special restrictions.

[0078] In this embodiment, the deviation between the discrete values ​​is the absolute value of the difference. As another implementation, based on the ability to measure the degree of difference between the discrete values, the implementer may use other calculation methods, such as ratio, square of difference, etc. This application does not impose any special restrictions.

[0079] In this embodiment, the dispersion of the peak value is the variance. As other implementation methods, implementers can use other existing technologies, such as standard deviation, coefficient of variation, etc., to measure the degree of unevenness in the distribution of peak values. This application does not impose any special restrictions.

[0080] It should be noted that if the average phase difference of the peaks in the audio signal data at the i-th position during the t-th time period is significantly different from the past, and the peak dispersion of the peaks is also significantly different from the past, then the i-th position during the t-th time period is more likely to be affected by additional factors, resulting in additional reflections of the broadcast audio signal.

[0081] Furthermore, by using the peak fluctuation coefficient and peak difference coefficient at each location during each time period, the stability coefficient at each location during each time period is obtained. The specific process is as follows:

[0082] The product of the peak fluctuation coefficient and the peak difference coefficient at each location in each time period is denoted as the fluctuation product; the stability coefficient at each location in each time period is inversely proportional to the fluctuation product.

[0083] In this embodiment, the expression for the stability coefficient at each location during each time period is as follows:

[0084] In the formula, This represents the stability coefficient at the i-th position during the t-th time period; This represents the peak fluctuation coefficient at the i-th position during the t-th time period; This represents the peak difference coefficient at the i-th position during the t-th time period; This indicates a preset value greater than 0, used to avoid a denominator of 0. The value is preset by a person; in this embodiment... The value is 0.01.

[0085] It should be noted that: the weaker the noise fluctuation interference at the i-th position during the t-th time period and the lower the reverberation change, the smaller the overall acoustic environment change at the i-th position during the t-th time period. When broadcasting based on the current parameters of the loudspeaker, the driver and passengers at the i-th position can clearly hear the broadcast content. Conversely, the greater the overall acoustic environment change at the i-th position during the t-th time period, the more adjustments are needed to the broadcast.

[0086] Step 3: By analyzing the changes in audio signal data at each location over historical periods, obtain the short-term audio variation at each location during each time period; divide the tunnel into left and right sides, and by analyzing the spatial distribution of audio signal data at all locations on each side during each time period, combined with the differences in short-term audio variation between each location and the location on its side, and the distance from each location to the tunnel exit on its side, obtain the interference coefficient at each location during each time period.

[0087] Normally, the wind speed inside a tunnel is determined by the operation of the tunnel's ventilation system. Higher ventilation demand leads to more intensive operation of the ventilation system, resulting in higher wind speeds and stronger noise, and vice versa. However, under normal circumstances, the tunnel's ventilation system operates according to pre-set parameters, meaning the wind speed is relatively constant, thus the noise generated by wind speed changes little. Nevertheless, changes in the wind speed in the external environment also affect the wind speed inside the tunnel. Because external wind directions vary, if the wind is parallel to the tunnel, the wind blowing into the tunnel will affect the overall wind speed inside, leading to a greater overall noise level. However, if the wind is perpendicular to the tunnel, it will only have a certain impact on the tunnel entrances and exits, with less impact on the middle of the tunnel. Therefore, simply calculating using the method in step 2 may misjudge the noise change caused by natural wind speed in the middle of the tunnel as a noise change caused by vehicle movement, thus mistakenly believing that the noise at a certain location changes significantly as a vehicle approaches. Consequently, adjustments are made to the broadcast system, resulting in a still low broadcast clarity. Therefore, further analysis is needed to accurately distinguish the impact of natural wind speed and vehicle movement on noise, thereby more effectively optimizing the broadcast system settings.

[0088] Specifically, wind perpendicular to the tunnel direction has a significant impact on noise at the tunnel entrance. The closer to the tunnel entrance, the more sensitive the external wind speed changes are to the noise, and the more pronounced the resulting noise variations. For example, when the external wind speed is high, the noise level in the audio signal data is higher closer to the tunnel entrance, while the noise level is lower closer to the middle of the tunnel. As the external wind speed increases or decreases, the noise situation at the tunnel entrance changes accordingly with a larger amplitude. In contrast, the noise change is smaller near the middle of the tunnel. Specifically, the noise level inside the tunnel gradually decreases from the tunnel entrance to the middle. When the noise changes, it exhibits a gradual trend overall. The closer to the tunnel entrance, the more drastic the noise change, and the closer to the middle of the tunnel, the weaker the noise change, but the overall gradual change is highly regular.

[0089] Based on the above analysis, the tunnel is divided into left and right sides, and the arithmetic mean of the audio signal data at each location during each time period is calculated. The arithmetic mean of all locations on any side during each time period is then arranged in ascending order of distance to the tunnel entrance, forming an audio variation sequence for that side. The trend intensity of the audio variation sequence is calculated to reflect the spatial variation trend of the noise level on that side. The calculation process for the trend intensity is a well-known technique and will not be elaborated upon in this application.

[0090] Furthermore, by analyzing the changes in audio signal data at each location over historical periods, the short-term changes in audio at each location during each time period are obtained. The specific process is as follows:

[0091] Pre-defined control periods for each time period are used. The mean of the deviation of the audio signal data at each location from the arithmetic mean of any two adjacent control periods in each time period is used as the short-term audio variation at each location in each time period to reflect the dynamic changes in noise. The larger the short-term audio variation, the faster the noise level changes.

[0092] In this embodiment, the reference time period for each time period is the 10 time periods adjacent to each time period. 10 is only one embodiment of this application. The implementer can set it according to the actual situation. This application does not impose any special restrictions. If there is insufficient data, the mean filling method is used to fill the missing data. The mean filling method is a well-known technology and will not be described in detail in this application.

[0093] Furthermore, by analyzing the trend intensity of the audio change sequence at each location during each time period, combined with the difference in short-term audio changes between each location and its corresponding location on the same side, and the distance from each location to the tunnel exit on its corresponding side, the interference coefficient at each location during each time period is obtained. The specific process is as follows:

[0094] For any given location, the location closest to the tunnel entrance is recorded as the tunnel entrance location on that side. The difference in short-term audio variation between each location on that side and the tunnel entrance location is calculated at each time period.

[0095] Obtain the normalized result of the distance from each position on either side to the tunnel entrance position;

[0096] The interference coefficient is inversely proportional to the difference and the normalization result, and directly proportional to the trend strength.

[0097] In this embodiment, the expression for the interference coefficient at each location during each time period is as follows:

[0098] In the formula, This represents the interference coefficient at the i-th position during the t-th time period; This represents the trend strength of the audio change sequence on the side of the i-th position during the t-th time period; This represents the normalized result of the distance between the i-th position and the tunnel entrance position on its side; This represents the difference in short-term audio variation between the i-th position and the tunnel entrance position on the same side during the t-th time period. This indicates a preset value greater than 0, used to avoid a denominator of 0. The value is preset by a person; in this embodiment... The value is 0.01.

[0099] In this embodiment, the Sigmoid function is used to obtain the normalized result of the distance between the i-th position and the tunnel entrance position on its side. The Sigmoid function is a well-known technology and will not be described in detail in this application. The distance between the i-th position and the tunnel entrance position on its side is the Euclidean distance.

[0100] It should be noted that: the stronger the trend of the noise level on the side of the i-th location in the t-th time period, and the closer the distance between the i-th location and the tunnel entrance in the t-th time period, and the smaller the difference in the noise level changes, the more likely the noise fluctuation at the i-th location in the t-th time period is caused by changes in wind speed and direction in the natural environment outside the tunnel.

[0101] Step 4: By combining the interference coefficients at each location during each time period with the audio decibel data at each location during each time period and the stability coefficient, the urgency of adjustment at each location during each time period is obtained, so as to screen the locations that need to be enhanced for the broadcast audio signal during each time period, and then enhance the broadcast audio signal.

[0102] Furthermore, by combining the interference coefficients at each location during each time period with the audio decibel data at each location during each time period, and the stability coefficient, the adjustment urgency at each location during each time period is obtained, expressed as:

[0103] In the formula, This indicates the urgency of adjustment at the i-th position within the t-th time period; This represents the audio decibel data at the i-th position within the t-th time period; This represents the stability coefficient at the i-th position during the t-th time period; Represents the interference coefficient at the i-th position in the t-th time period; norm() represents the normalization function; This indicates a preset value greater than 0, used to avoid a denominator of 0. The value is preset by a person; in this embodiment... The value is 0.01. This is denoted as the interference product.

[0104] It should be noted that: the higher the clarity of the tunnel broadcast at location i in time period t, and the more likely the noise at location i in time period t is to be caused by tunnel wind speed and the lower the decibel level, the better the acoustic environment at location i in time period t, and the clearer the tunnel broadcast will be for drivers and passengers. Conversely, the worse the acoustic environment at location i in time period t, the higher the noise level, the lower the broadcast clarity, and the greater the need to enhance the broadcast audio signal at location i in time period t. A flowchart illustrating the urgency of adjustment is shown below. Figure 2 As shown.

[0105] Obtain the segmentation threshold of the adjustment urgency at all locations under a preset number of historical time periods. If the adjustment urgency at the i-th location under the t-th time period is less than the segmentation threshold, it is determined that the tunnel broadcast clarity at the i-th location under the t-th time period is high, and drivers and passengers can clearly hear the broadcast content. Therefore, the broadcast audio signal at the i-th location under the t-th time period does not need to be enhanced. Otherwise, it is determined that the tunnel broadcast clarity at the i-th location under the t-th time period is low, and drivers and passengers cannot clearly hear the broadcast content. Therefore, the broadcast audio signal at the i-th location under the t-th time period needs to be enhanced.

[0106] In this embodiment, the preset quantity is 500. The preset quantity is preset by a person and the implementer can set it according to the actual situation. This application does not impose any special restrictions.

[0107] In this embodiment, cross-validation is used to obtain the segmentation threshold of adjustment urgency. The process of obtaining the segmentation threshold using cross-validation is a well-known technique and will not be described in detail here. As other implementation methods, based on the ability to obtain the segmentation threshold of adjustment urgency, implementers may use other existing techniques, such as the Otsu threshold segmentation algorithm, iterative threshold segmentation, etc. This application does not impose any special restrictions.

[0108] Once the location requiring broadcast audio signal enhancement is determined through assessment, the tunnel broadcasting system employs a DSP audio processing module to perform frequency equalization and suppression of easily resonant frequencies, as well as delay processing of the broadcast audio signal, to obtain the control signal for the speaker at that location. This control signal is then transmitted via the Internet of Things (IoT). The high-definition AI power amplifier controller within the tunnel receives the control signal and adjusts the playback of the speaker at that location to reduce the impact of noise and echo on broadcast clarity, thereby enhancing the broadcast audio signal in the tunnel environment.

[0109] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. In some alternative implementations, the functions marked in the blocks may occur in a different order than that shown in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. In the descriptions corresponding to the flowcharts and block diagrams in the accompanying drawings, the operations or steps corresponding to different blocks may also occur in a different order than disclosed in the description, and sometimes there is no specific order between different operations or steps. For example, two consecutive operations or steps may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. Each block in a block diagram and / or flowchart, and combinations of blocks in a block diagram and / or flowchart, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0110] It will be apparent to those skilled in the art that this application is not limited to the details of the exemplary embodiments described above, and that this application can be implemented in other specific forms without departing from its essential characteristics. Therefore, the embodiments described above should be considered exemplary and non-limiting in all respects.

Claims

1. A method for enhancing audio signals in a tunnel environment based on directional sound waves, characterized in that, The method includes the following steps: Real-time acquisition of audio signal data and audio decibel data at preset locations within the tunnel; The entire data acquisition time interval is divided into time periods. By comparing the number of peaks in the audio signal data at each location in each time period with the number of peaks in the adjacent time periods, and the peak width of the audio signal data at each location in each time period, the peak fluctuation coefficient at each location in each time period is obtained. In addition, by comparing the peak phase and peak dispersion of the audio signal data at each location in each time period with each previous time period, the peak difference coefficient at each location in each time period is obtained, and the stability coefficient at each location in each time period is obtained. By analyzing the changes in audio signal data at each location over historical periods, the short-term audio variation at each location during each period is obtained. The tunnel is divided into left and right sides. By analyzing the spatial distribution of audio signal data at all locations on each side during each period, combined with the differences in short-term audio variation between each location and the location on its side, and the distance from each location to the tunnel exit on its side, the interference coefficient at each location during each period is obtained. Then, by combining the audio decibel data at each location during each period with the stability coefficient, the urgency of adjustment at each location during each period is obtained, so as to screen the locations that need to enhance the broadcast audio signal during each period. The process of obtaining the stability coefficient is as follows: The product of the peak fluctuation coefficient and the peak difference coefficient is denoted as the fluctuation product; The stability coefficient is inversely proportional to the fluctuation product; The process of obtaining the interference coefficient is as follows: The tunnel is divided into left and right sides. The arithmetic mean of all locations on any side in each time period is sorted in ascending order by distance from the tunnel entrance, and the trend strength of the sorting results is calculated. For any given location, the location closest to the tunnel entrance is designated as the tunnel entrance location. The difference in short-term audio variation between each location on any given side and the tunnel entrance location is calculated for each time period. Obtain the normalized result of the distance from each position on either side to the tunnel entrance position; The interference coefficient is inversely proportional to the difference and the normalization result, and directly proportional to the trend strength.

2. The audio signal enhancement method based on directional sound waves in a tunnel environment as described in claim 1, characterized in that, The process for obtaining the peak fluctuation coefficient is as follows: Obtain the increase in the number of peaks in the audio signal data at each location within each time period compared to the number of peaks in the previous time period; The acquisition times within each time period are numbered sequentially. The peaks and troughs of the audio signal data at each location within each time period are obtained. The difference in the acquisition times between the preceding and following troughs of each peak is taken as the peak width of each peak. The average peak width of all peaks of the audio signal data at each location within each time period is calculated. The peak fluctuation coefficient is directly proportional to the increase and inversely proportional to the mean.

3. The audio signal enhancement method based on directional sound waves in a tunnel environment as described in claim 1, characterized in that, The process for obtaining the peak difference coefficient is as follows: Obtain the average phase difference between any two adjacent peaks of the audio signal data at each location within each time period; calculate the difference between the average value of the audio signal data at each location in each time period and the average value of each previous time period; Calculate the dispersion of all peaks of the audio signal data at each location within each time period, and calculate the deviation of the dispersion of the audio signal data at each location between each time period and each previous time period; The product of the difference and the deviation value is denoted as the difference product; The peak difference coefficient is obtained by multiplying all the differences in the audio signal data at each location in each time period.

4. The audio signal enhancement method based on directional sound waves in a tunnel environment as described in claim 3, characterized in that, The peak difference coefficient is the sum of all the differences in the audio signal data at each location during each time period.

5. The audio signal enhancement method based on directional sound waves in a tunnel environment as described in claim 1, characterized in that, The process for obtaining the short-term audio changes is as follows: Calculate the arithmetic mean of the audio signal data at each location during each time period; The control periods for each time period are preset, and the short-term change in audio is the mean of the deviation of the audio signal data at each location from the arithmetic mean between any two adjacent control periods in each time period.

6. The method for enhancing audio signals in a tunnel environment based on directional sound waves as described in claim 1, characterized in that, The interference coefficient is calculated as follows: the product of the difference and the normalization result is mapped to a constant greater than 0; the interference coefficient is the ratio of the trend intensity to the constant.

7. The method for enhancing audio signals in a tunnel environment based on directional sound waves as described in claim 1, characterized in that, The process for obtaining the urgency of the adjustment is as follows: The product of the stability coefficient and the disturbance coefficient is denoted as the disturbance product; The interference product is mapped to a positive number, and the adjustment urgency is the normalized value of the ratio of the audio decibel data at each location in each time period to the positive number.

8. The audio signal enhancement method based on directional sound waves in a tunnel environment as described in claim 1, characterized in that, The method for selecting locations that require broadcast audio signal enhancement in each time period is as follows: Obtain the segmentation threshold of adjustment urgency at all locations under a preset number of historical time periods; If the urgency of adjustment at each location in each time period is less than the segmentation threshold, it is determined that the broadcast audio signal at each location in each time period does not need to be enhanced; otherwise, it needs to be enhanced.

Images