Communication interconnection method and system applied to vehicle-mounted navigator

Abstract

The invention relates to the technical field of vehicle-mounted communication, in particular to a communication interconnection method and system applied to a vehicle-mounted navigator. The method comprises the following steps: collecting a rolling noise signal of a tire through a chassis microphone of a vehicle-mounted navigator, recording the rolling noise signal as a noise frequency spectrum reference value when approaching a tunnel is detected, and identifying a tunnel entrance moment; calculating the difference between the tire noise signal after entering the tunnel and the noise spectrum reference value based on the tunnel entrance moment, and extracting a newly added formant frequency value and formant sharpness value; and calculating the cross section area of the tunnel according to the formant frequency value, judging the traffic flow density in the tunnel according to the formant sharpness value, and combining the cross section area of the tunnel and the traffic flow density in the tunnel to form tunnel environment characteristic parameters. According to the method, the communication signal intensity distribution of each position in the tunnel can be predicted in advance by analyzing the resonance characteristics of the tire noise.

Description

Technical Field

[0001] This invention relates to the field of vehicle communication technology, and in particular to a communication interconnection method and system for vehicle navigation devices. Background Technology

[0002] A car navigation system is an intelligent electronic device integrated into a vehicle. It primarily uses satellite positioning systems (such as GPS and BeiDou) to obtain the vehicle's real-time location and combines this with built-in or network-connected electronic map data to provide drivers with accurate route planning, real-time traffic information, and voice navigation guidance, thereby achieving efficient and safe travel management. It typically includes core components such as a positioning module, map database, processor, display screen, and human-machine interface. It can automatically calculate the optimal route based on the driver's set destination and dynamically adjust it along the way to avoid congestion, accidents, or restricted road sections.

[0003] In existing technologies, vehicles rely on mobile networks for real-time navigation, traffic updates, and information interconnection while in motion. However, when vehicles enter tunnels, underpasses, or areas with signal blockage, mobile communication signals are easily affected by factors such as tunnel structure, vehicle density, and multipath fading, leading to a significant decrease in signal strength and even communication blind spots. This results in problems such as navigation service interruptions, untimely traffic information updates, and delays in emergency information. This not only affects the driving experience but may also pose safety hazards. Summary of the Invention

[0004] Therefore, it is necessary for the present invention to provide a communication interconnection method and system for vehicle navigation devices to solve at least one of the above-mentioned technical problems.

[0005] To achieve the above objectives, a communication interconnection method for in-vehicle navigation devices includes the following steps: Step S1: Collect the rolling noise signal of the tires through the chassis microphone of the vehicle navigation system, record the rolling noise signal as the noise spectrum reference value when the vehicle approaches the tunnel, and identify the tunnel entrance time; Step S2: Calculate the difference between the tire noise signal after entering the tunnel and the noise spectrum reference value based on the tunnel entrance time, and extract the newly added resonant peak frequency value and resonant peak sharpness value; Step S3: Calculate the tunnel cross-sectional area based on the resonance peak frequency value, determine the traffic density inside the tunnel based on the resonance peak sharpness value, and combine the tunnel cross-sectional area and the traffic density inside the tunnel to form tunnel environmental characteristic parameters; Step S4: Query the preset communication attenuation mapping table according to the tunnel environment characteristic parameters, calculate the predicted signal strength at each location in the tunnel, and generate a communication quality prediction curve; Step S5: Adjust the data transmission strategy of the vehicle navigation system based on the communication quality prediction curve to achieve communication interconnection under the tunnel.

[0006] The present invention also provides a communication interconnection system for in-vehicle navigation devices, used to execute the communication interconnection method for in-vehicle navigation devices as described above, wherein the communication interconnection system for in-vehicle navigation devices includes: The noise reference acquisition module is used to collect the rolling noise signal of the tires through the chassis microphone of the vehicle navigation system, record the rolling noise signal as the noise spectrum reference value when the vehicle approaches the tunnel, and identify the tunnel entrance time. The formant extraction module is used to calculate the difference between the tire noise signal after entering the tunnel and the noise spectrum reference value based on the tunnel entrance time, and to extract the newly added formant frequency value and formant sharpness value. The tunnel environment analysis module is used to calculate the tunnel cross-sectional area based on the resonant peak frequency value, determine the traffic density in the tunnel based on the resonant peak sharpness value, and combine the tunnel cross-sectional area and the traffic density in the tunnel to form tunnel environmental characteristic parameters. The communication quality prediction module is used to query a preset communication attenuation mapping table based on tunnel environment characteristic parameters, calculate the predicted signal strength at each location in the tunnel, and generate a communication quality prediction curve. The transmission strategy adjustment module is used to adjust the data transmission strategy of the vehicle navigation system based on the communication quality prediction curve, so as to achieve communication interconnection under the tunnel.

[0007] This invention achieves automatic detection and real-time analysis of tunnel environmental characteristics by collecting vehicle tire rolling noise and combining it with an onboard camera to identify tunnel entrances. Through spectral processing and resonance peak identification of tire noise within the tunnel, the cross-sectional area of ​​the tunnel and traffic density can be accurately calculated, thereby enabling precise modeling of signal propagation characteristics. By utilizing environmental attenuation parameters and the inverse square law to predict the signal strength of each node, and combining multipath fading correction to generate continuous communication quality prediction curves, the system can identify weak signal areas and communication blind spots in advance. Before a vehicle enters a weak signal area, local caching can pre-download navigation instructions, traffic information, and map data for the next 3 minutes, ensuring continuous navigation service even under communication constraints. Simultaneously, by adjusting the communication module's operating state, unnecessary data transmission is reduced in weak signal areas, and the communication module is completely shut down when entering a blind spot, with a timer pre-restoring the connection, achieving energy-efficient and stable data management. This method also continuously corrects the prediction curve through a real-time verification mechanism, ensuring the accuracy and reliability of signal strength prediction. Overall, this method enables in-vehicle navigation systems to maintain highly reliable communication and continuous navigation in complex environments such as tunnels or underground passages, dynamically adapt to different tunnel structures and traffic conditions, reduce the risk of communication interruption, improve driving safety and user experience, and optimize data transmission and system energy consumption management, providing stable and predictable technical support for the interconnection of intelligent vehicles in closed environments. Attached Figure Description

[0008] Other features, objects, and advantages of the invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings: Figure 1 This is a flowchart illustrating the steps of a communication interconnection method for vehicle navigation devices according to the present invention. Figure 2 This is a schematic diagram of a communication interconnection system applied to an in-vehicle navigation device according to the present invention; Figure 3 This is a time-domain waveform diagram of an embodiment of the present invention; Figure 4 This is a noise spectrum diagram of one embodiment of the present invention; Figure 5 This is a differential spectrum diagram according to an embodiment of the present invention; Figure 6 This is a resonance peak marking diagram of one embodiment of the present invention.

[0009] Figure 7 This is a schematic diagram of the hardware layout of a vehicle navigation communication interconnection system according to an embodiment of the present invention. Detailed Implementation

[0010] The technical method of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0011] Furthermore, the accompanying drawings are merely illustrative of the invention and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and therefore repeated descriptions of them will be omitted. Some block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor methods and / or microcontroller methods.

[0012] It should be understood that although the terms "first," "second," etc., may be used herein to describe various units, these units should not be limited by these terms. These terms are used merely to distinguish one unit from another. For example, without departing from the scope of the exemplary embodiments, a first unit may be referred to as a second unit, and similarly, a second unit may be referred to as a first unit. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0013] To achieve the above objectives, please refer to Figures 1 to 7 This invention provides a communication interconnection method for in-vehicle navigation devices, the method comprising the following steps: Step S1: Collect the rolling noise signal of the tires through the chassis microphone of the vehicle navigation system, record the rolling noise signal as the noise spectrum reference value when the vehicle approaches the tunnel, and identify the tunnel entrance time; Step S2: Calculate the difference between the tire noise signal after entering the tunnel and the noise spectrum reference value based on the tunnel entrance time, and extract the newly added resonant peak frequency value and resonant peak sharpness value; Step S3: Calculate the tunnel cross-sectional area based on the resonance peak frequency value, determine the traffic density inside the tunnel based on the resonance peak sharpness value, and combine the tunnel cross-sectional area and the traffic density inside the tunnel to form tunnel environmental characteristic parameters; Step S4: Query the preset communication attenuation mapping table according to the tunnel environment characteristic parameters, calculate the predicted signal strength at each location in the tunnel, and generate a communication quality prediction curve; Step S5: Adjust the data transmission strategy of the vehicle navigation system based on the communication quality prediction curve to achieve communication interconnection under the tunnel.

[0014] Furthermore, step S1 includes the following steps: Step S11: Simultaneously collect the rolling noise of the left front wheel and the right front wheel using two microphones at the front and rear of the chassis, with a sampling frequency of 8000Hz, to obtain the rolling noise signal; In one embodiment, two high-sensitivity microphones, one pointing towards the ground and the other pointing towards the ground, are installed at the front and rear of the vehicle chassis, corresponding to the positions of the left and right front wheels, respectively. When the vehicle travels at 60 km / h on an open road, the two microphones simultaneously collect the rolling noise generated by the friction between the tires and the ground. The sampling frequency is set to 8000 Hz to ensure the capture of effective frequency band information from 0-4000 Hz, covering common structural resonances and tunnel reflection noise. The resulting two parallel time-domain rolling noise signals are then used for subsequent spectral analysis.

[0015] For example, when a vehicle is traveling on a highway and approaches a tunnel, the sampling signals of the left front wheel and the right front wheel may differ due to slight unevenness in the road surface. Simultaneous sampling can reduce the impact of abnormal noise from a single wheel position.

[0016] Step S12: Perform frame segmentation on the rolling noise signal, with adjacent frames overlapping by 50%, and perform a fast Fourier transform on each frame to obtain noise spectrum data. In one embodiment, the acquired rolling noise signal is divided into multiple time-domain frames, each with a length of 1024 sampling points. Adjacent frames have a 50% overlap to avoid information loss. A Fast Fourier Transform (FFT) is then performed on each frame to obtain the frequency domain energy distribution, i.e., the noise spectrum data, for each frame.

[0017] For example, if the signal amplitude in the current frame is significantly higher than other frequency points around 500Hz and 720Hz, it indicates the presence of resonant components or structural reflections from the environment during that time period. By switching frames one by one, the changes in noise characteristics of a vehicle as it approaches a tunnel can be continuously tracked.

[0018] Step S13: Within a range of 50-100 meters from the tunnel entrance, calculate the average value of twenty consecutive frames of noise spectrum data as the noise spectrum benchmark value for the open road; In one embodiment, 20 consecutive frames of noise spectrum data are extracted within a range of 50-100 meters from the tunnel entrance, and the amplitude of each frequency point is averaged to obtain a stable background spectrum. This average spectrum serves as a reference value for the noise spectrum under open road conditions. This effectively eliminates sporadic noise interference and provides a reference for subsequent change detection after entering the tunnel.

[0019] For example, if the energy in the 300-800Hz range remains relatively stable for 20 consecutive frames of spectrum within this range, the energy curve in this range is recorded as the "open road reference". If a significant new peak appears in this range after entering the tunnel, it can be determined as resonance caused by the tunnel structure effect.

[0020] Step S14: Use the vehicle-mounted camera to identify tunnel entrance signs or sudden changes in light to determine the vehicle's tunnel entrance time.

[0021] In one embodiment, a camera installed on the vehicle navigation system captures images of the road ahead, and an image recognition algorithm is used to detect tunnel entrance signs, tunnel outlines, or sudden changes in light. When the system detects that a vehicle is about to enter the tunnel, it automatically records this moment as the tunnel entrance time. This time point is linked to the noise reference spectrum obtained in step S13 to provide a reference for subsequent differential calculations.

[0022] For example, when a clear tunnel entrance outline appears in front of the vehicle and the brightness decreases by more than 40% within 0.3 seconds, the system will immediately determine that the vehicle has entered the tunnel and record that moment as the entry time.

[0023] It should be noted that the entry time can be achieved through camera recognition, or it can be confirmed by means of significant changes in GPS signal attenuation characteristics to improve robustness.

[0024] Furthermore, step S2 includes the following steps: Step S21: Calculate the tunnel spectral data of the tire noise signal after entering the tunnel based on the tunnel entrance time, and perform point-by-point differential operation with the noise spectrum reference value to obtain the differential spectrum; In one embodiment, based on the determined tunnel entrance time, starting from that time, an FFT is calculated for each frame of the time-domain rolling noise signal (referencing the frame parameters in step S12, such as a sampling frequency of 8000Hz, 1024-point FFT, and 50% overlap) to obtain the spectrum of each frame within the tunnel. Then, the spectrum of each frame is subtracted point-by-point from the open road noise spectrum reference value obtained in step S13 at each frequency point to obtain the differential spectrum of each frame, which reflects the additional spectral energy after entering the tunnel. To reduce the impact of instantaneous noise, a moving average or median filter can be applied to the differential spectrum for several adjacent frames (e.g., five consecutive frames) to generate a smoothed differential spectrum for subsequent formant detection.

[0025] For example, if a frame after entry has a linear amplitude of 0.8 at 400Hz and a reference amplitude of 0.2, then the difference at that frequency point is 0.6. The significant positive difference indicates that the frequency point may be affected by the tunnel structure and resonate.

[0026] It should be noted that the difference operation can be performed in the linear amplitude domain or the dB domain—the dB domain is convenient for threshold comparison, but attention must be paid to the handling of zero or near-zero amplitude values; in addition, a window (such as the Hann window) can be pre-windowed before the difference operation to reduce spectral leakage.

[0027] Step S22: Within the differential spectrum range of 200Hz-1000Hz, with a window width of 30Hz, the movement is in 10Hz increments. At each window position, the average amplitude of all frequency points within the window is calculated as the local average. When the amplitude of the center of the window and its two adjacent frequency points both exceed 1.5 times the local average and the amplitude of the center point is the highest, the center frequency point is marked as a candidate resonance peak.

[0028] In one embodiment, a search interval of 200Hz–1000Hz is defined within the obtained differential spectrum, and a sliding window with a width of 30Hz and a step size of 10Hz is used for scanning. When mapping the physical frequency to FFT frequency points, it is necessary to convert according to the FFT resolution and take an approximate integer (for example, for a sampling of 8000Hz and a 1024-point FFT, the frequency resolution is approximately 7.8125Hz, 30Hz ≈ 3.84 frequency points, usually 4 frequency points are taken, 10Hz ≈ 1.28 frequency points, and the step size can be 1 or 2 frequency points with an approximate 10Hz movement). At each window position, the average amplitude of all frequency points within the window is calculated as a local average value; if the amplitude of the center point of the window and its two adjacent left and right frequency points both exceed 1.5 times the local average value and the amplitude of the center point is the maximum value within the window, then the center frequency point is marked as a candidate resonant peak.

[0029] For example, under a 1024-point FFT, if the local average amplitude of a certain window is 0.45, and the amplitudes of the center point and the two points to its left and right are 0.95, 0.9, and 0.88 respectively (all greater than 1.5 × 0.45 = 0.675), and the center has the maximum amplitude of 0.95, then the center is identified as a candidate formant.

[0030] In another embodiment, the threshold can be set between 1.4 and 1.6, or an equivalent dB threshold (e.g., >3 dB gain) can be used in the dB domain; alternatively, interpolation (such as quadratic / parabolic interpolation) can be used to refine the sub-frequency points of the peak to obtain more accurate frequency values. It should be noted that rounding the frequency points corresponding to the window will introduce slight deviations. In actual implementation, it is recommended to use frequency interpolation or zero-padding to improve frequency resolution. At the same time, DC removal and spectral smoothing should be performed before peak determination to reduce false alarms.

[0031] Furthermore, step S2 also includes the following steps: Step S23: Expand the search to both sides for each candidate resonance peak until the amplitude drops to 70% of the peak value, and record this frequency range as the 3dB bandwidth. Divide the peak frequency by the 3dB bandwidth to obtain the sharpness value of the resonance peak. In one embodiment, for each candidate resonant, parabolic interpolation is used to refine the sub-frequency points near the peak to obtain a more accurate peak frequency and peak amplitude. Then, a frequency-by-frequency search is performed from the peak point to the left and right until the amplitude drops to 70% of the peak (approximately −3dB) or reaches a preset search boundary. The accurate frequencies at the two points where the amplitude drops to 70% are recorded (the accurate frequencies can be obtained between two adjacent FFT points using linear interpolation). The difference between the two frequencies is the 3dB bandwidth of the peak. The sharpness value (quality factor Q) of the resonant is obtained by dividing the peak frequency by the 3dB bandwidth.

[0032] For example, if a 1024-point FFT is used and interpolation determines that the peak is at 500.4Hz with a peak amplitude of 1.0, the left side drops to 0.7 at 470.2Hz, and the right side drops to 0.7 at 532.8Hz, then the 3dB bandwidth is approximately 62.6Hz, and the sharpness is approximately 500.4 / 62.6≈8.0.

[0033] It should be noted that when the bandwidth is close to the frequency resolution or a very narrow value appears (which may result in abnormally high sharpness), the resolution should be increased by the minimum bandwidth threshold or zero padding and an upper limit clipping should be added to avoid division by zero or abnormal Q values.

[0034] Step S24: Track the frequency drift of the candidate resonance peak within five consecutive frames. When the drift is less than 20Hz, it is confirmed as a stable resonance peak. Take the median of the frequencies of the five frames as the resonance peak frequency value. In one embodiment, for each candidate resonant peak, the frequency peak closest to the current peak is searched in chronological order in the subsequent five consecutive frames of differential spectrum (matching can be done within a matching window of ± a certain number of frequency points, which can be set according to vehicle speed and FFT resolution, for example, ±30Hz). The peak frequency values ​​corresponding to the five frames are recorded and their frequency drift is calculated (defined as the difference between the maximum and minimum peak frequencies of the five frames). When the drift is less than 20Hz, the resonant peak is confirmed as a stable resonant peak, and the median of the peak frequencies of the five frames is used as the final frequency value of the peak to reduce the influence of outliers.

[0035] For example, if the matching peak frequencies of five consecutive frames are [502Hz, 499Hz, 505Hz, 501Hz, 500Hz], then the drift is 6Hz < 20Hz, and the median of 501Hz is taken as the final value.

[0036] It should be noted that if a matching peak is not found in a certain frame (e.g., it is submerged by noise), interpolation or neighboring frames can be used to fill in the gap. However, if the number of missing frames exceeds the threshold (e.g., ≥2 frames), it is considered discontinuous and will not be included in the stability determination. The matching strategy should prioritize the peak with the closest frequency and the amplitude within a reasonable range to prevent the wrong peak from being selected.

[0037] Step S25: Calculate the standard deviation of the sharpness value of the stable resonance peak within five frames. When the standard deviation is less than 10% of the sharpness mean, take the sharpness mean as the final resonance peak sharpness value.

[0038] In one embodiment, for a stable formant, the sharpness value of the peak is calculated for each of five consecutive frames, and the mean and standard deviation of these five sharpness values ​​are obtained. If the standard deviation is less than 10% of the mean sharpness (i.e., the variation is small), the mean is taken as the final sharpness value of the formant. Otherwise, the peak is determined to be unstable in sharpness, and the observation frame count can be extended (e.g., expanded to seven frames) or the peak can be discarded and the next candidate peak can be selected.

[0039] For example, if the sharpness values ​​of the five frames are [9.8, 10.1, 9.9, 10.2, 9.7], the mean is approximately 9.94, and the standard deviation is approximately 0.19, which is about 1.9% of the mean (less than 10%). In this case, 9.94 is adopted as the final sharpness value.

[0040] It should be noted that, to prevent the influence of extreme values, the median or truncated mean can be used instead of the simple mean; in addition, when calculating the sharpness standard deviation, the reliability of the bandwidth measurement for each frame should be ensured (non-zero, not dominated by amplitude noise), otherwise abnormal frames should be filtered out before calculation.

[0041] See Figure 3 As shown, the time-domain waveform diagram illustrates the variation of vehicle tire rolling noise signals over time. The horizontal axis represents time in seconds, and the vertical axis represents signal amplitude, ranging from -1 to 1. In practice, rolling noise signals from the left and right front wheels are simultaneously collected using two microphones at the front and rear of the chassis, and the sampled signals are plotted as waveforms in chronological order. For example, when the sampling frequency is set to 8000Hz, the rolling noise waveform within 0 to 1 second clearly shows the periodic variation characteristics of the signal, providing basic data for subsequent spectrum analysis. It should be noted that this waveform diagram can be used to determine the overall amplitude variation trend and periodic characteristics of the signal, providing a reference for formant extraction.

[0042] See Figure 4 As shown, the noise spectrum diagram displays the amplitude distribution of the rolling noise signal in the frequency domain. The horizontal axis represents frequency in Hz, and the vertical axis represents amplitude in dB. The spectrum data obtained by performing a Fast Fourier Transform on the rolling noise signal provides a clear picture of the signal strength distribution at different frequencies. For example, the peak frequencies of 500Hz and 750Hz are marked in the diagram, corresponding to potential tunnel resonance characteristic frequencies. This spectrum diagram can be used for preliminary identification and differential analysis of candidate resonance peaks, providing data for tunnel structure identification and traffic density assessment. It should be noted that lower amplitude frequencies in the spectrum diagram do not affect the identification of the main resonance peak; the focus should be on frequency regions with prominent peaks.

[0043] See Figure 5As shown, the differential spectrogram displays the difference between the rolling noise signal after a vehicle enters the tunnel and the reference noise spectrum of the open road. The horizontal axis represents frequency in Hz, and the vertical axis represents the amplitude difference in dB. By calculating the average of the spectrum over several consecutive frames and performing point-by-point differential operations with the reference spectrum, the influence of the tunnel environment on rolling noise can be highlighted. For example, the prominent peaks at 500Hz and 750Hz indicate the resonance enhancement effect of the tunnel on these frequencies, thus providing a basis for extracting stable resonance peaks. It should be noted that the differential spectrogram can effectively suppress the interference of open road noise, making the tunnel characteristics more prominent.

[0044] See Figure 6 As shown, the resonance peak marker diagram is used to mark identified candidate resonance peaks and their sharpness information based on the differential spectrum. The horizontal axis represents frequency in Hz, and the vertical axis represents amplitude in dB. By calculating the 3dB bandwidth and analyzing the sharpness values ​​of the candidate resonance peaks, the main resonance peak and its stability are determined. For example, the diagram marks resonance peaks with f1=500Hz and Q=8.5 and f2=750Hz and Q=6.2, representing frequency and sharpness values ​​respectively, which are used to further calculate the tunnel cross-sectional area and traffic density. It should be noted that this diagram not only shows the peak positions but also provides a quantitative basis for selecting the main peak in multi-peak cases.

[0045] Of particular importance, step S2 also includes: When there are multiple stable resonance peaks, calculate the quality factor of each resonance peak. The quality factor is the product of the resonance peak frequency value and its sharpness value. In one embodiment, when multiple stable resonance peaks are detected, a quality factor needs to be calculated for each resonance peak. This quality factor is obtained by multiplying the frequency value of the resonance peak by its sharpness value. For example, if a resonance peak has a frequency of 480 Hz and a sharpness of 9.5, its quality factor is 4560; if another resonance peak has a frequency of 520 Hz and a sharpness of 8.0, its quality factor is 4160.

[0046] Sort all stable resonance peaks in descending order of quality factor, compare the top two resonance peaks, and select the first one as the main resonance peak when the frequency of the first peak is lower than that of the second peak and the sharpness value is higher than that of the second peak. In one embodiment, all stable resonance peaks are sorted from highest to lowest quality factor, and the top two resonance peaks are compared. If the frequency of the first-ranked resonance peak is lower than that of the second-ranked peak, but its sharpness value is higher, then the first peak is selected as the dominant resonance peak. For example, when the frequency of the first resonance peak is 480 Hz and its sharpness is 9.5, while the frequency of the second peak is 520 Hz and its sharpness is 8.0, the first peak will be selected as the dominant resonance peak because it has a lower frequency and higher sharpness.

[0047] When the frequency of the first peak is higher than that of the second peak, but the sharpness value is also more than 1.5 times higher than that of the second peak, the first peak is selected as the main resonance peak. In one embodiment, if the frequency of the first resonant peak is higher than that of the second after sorting, but its sharpness value is at least 50% higher than that of the second, that is, its sharpness value is more than 1.5 times that of the second, then the first resonant peak is still selected as the main resonant peak. For example, if the frequency of the first resonant peak is 650 Hz and its sharpness is 12.0, while the frequency of the second resonant peak is 600 Hz and its sharpness is 7.0, then although the frequency of the first resonant peak is higher, its sharpness value is much higher than that of the second resonant peak by 1.5 times, so the first resonant peak is still selected.

[0048] Otherwise, calculate the weighted score of the two formants, with the reciprocal of the frequency accounting for 0.6 and the sharpness value accounting for 0.4, and select the one with the highest weighted score as the main formant; In one embodiment, when neither the condition of "lower first frequency and higher sharpness" nor the condition of "higher first frequency but sharpness at least 1.5 times higher" is met, a weighted calculation is needed to select the main resonant peak. Specifically, a weighted score is calculated for each of the two candidate peaks, with the reciprocal of the frequency accounting for 60% and the sharpness value accounting for 40%. The scores are then compared, and the resonant peak with the highest score is selected as the main resonant peak. For example, when the first frequency is 820Hz and the sharpness is 10.0, and the second frequency is 790Hz and the sharpness is 7.5, the weighted score of the first frequency is higher after score conversion, therefore the first frequency is selected.

[0049] It should be noted that the weighted scores here are normalized or scaled to avoid direct superposition issues caused by the different dimensions of frequency and sharpness. If the scores are very close, for example, the difference is less than 1%, the peak with higher sharpness or a larger quality factor can be selected as a supplementary rule.

[0050] The resonance peak frequency and resonance peak sharpness values ​​of the main resonance peak are used as the final resonance peak frequency and resonance peak sharpness values.

[0051] In another embodiment, a simplified strategy can be adopted to achieve fast real-time judgment. For example, the inverse of the frequency component can be multiplied by a fixed scaling constant to make its magnitude close to that of sharpness, and then the frequency and sharpness components can be directly weighted and summed to obtain a score for comparison. This method requires less computation and is more suitable for online processing, but a reasonable scaling constant needs to be determined experimentally during actual system operation.

[0052] It should be noted that during the calculation process, protection against abnormal situations should also be considered. For example, if an extremely small bandwidth causes abnormal amplification of sharpness, a sharpness cap can be set or abnormal frames can be removed to avoid incorrect selection. Finally, the confirmed frequency and sharpness values ​​of the main formant will be used as the final formant frequency and sharpness output for the entire process.

[0053] Furthermore, step S3, calculating the tunnel cross-sectional area based on the resonant peak frequency value, includes: The shape of the tunnel cross section is determined by the resonance peak frequency value. When the resonance peak frequency value is between 400-600Hz, it is determined to be a circular tunnel, and the correction factor is kept at 1000; when the frequency is between 600-800Hz, it is determined to be a rectangular tunnel, and the correction factor is adjusted to 850; when the frequency exceeds 800Hz, it is determined to be a horseshoe-shaped tunnel, and the correction factor is adjusted to 750. In one embodiment, the specific method for determining the tunnel cross-sectional shape by the resonance peak frequency value is as follows: when the detected resonance peak frequency is between 400 and 600 Hz, it is determined to be a circular tunnel, and the correction factor is set to 1000; when the frequency is between 600 and 800 Hz, it is determined to be a rectangular tunnel, and the correction factor is set to 850; when the frequency exceeds 800 Hz, it is determined to be a horseshoe-shaped tunnel, and the correction factor is set to 750.

[0054] For example, if the final identified main resonant frequency is 500Hz, then the tunnel is determined to be circular according to the rules and a correction factor of 1000 is applied.

[0055] It should be noted that these frequency bands and correction coefficients are empirical divisions and calibration values ​​for common tunnel shapes in this method, and can be adjusted appropriately based on the field calibration results.

[0056] The equivalent diameter of the tunnel is calculated based on the resonant peak frequency value, where the equivalent diameter of the tunnel = 340 / (2 × resonant peak frequency value) × correction factor; In one embodiment, the specific method for calculating the equivalent diameter of the tunnel based on the resonant frequency value is as follows: first, calculate the reference diameter using the frequency according to the given formula, and then multiply it by the corresponding correction factor to obtain the equivalent diameter; the formula used in this method is to divide 340 by (2 × resonant frequency) and then multiply it by the above correction factor to obtain the equivalent diameter (output in millimeters according to this implementation).

[0057] For example, if the resonant frequency is 500Hz, first calculate 340 / (2×500)=0.34, then multiply by the correction factor 1000 to get the equivalent diameter of 340 (i.e. 340mm).

[0058] It should be noted that 340 is an empirical value of the sound speed order constant, and the equivalent diameter obtained after substituting the correction factor is in millimeters. The actual unit calibration should be consistent with that of the sensor and the system.

[0059] The cross-sectional area of ​​a tunnel is calculated based on its equivalent diameter. For circular tunnels, the area formula for a circle is used; for rectangular tunnels, the width-to-height ratio is 1.5; and for horseshoe-shaped tunnels, the equivalent diameter is 0.85 times.

[0060] In one embodiment, the specific method for calculating the cross-sectional area of ​​a tunnel based on its equivalent diameter is as follows: For a circular tunnel, the equivalent diameter is directly used as the diameter of the circle, and the cross-sectional area is calculated according to the formula for the area of ​​a circle; for a rectangular tunnel, the width and height are derived by taking a width-to-height ratio of 1.5 (for example, let the width equal the equivalent diameter, and therefore the height equal the equivalent diameter divided by 1.5), and then the area of ​​the rectangle is obtained by multiplying the width by the height; for a horseshoe-shaped tunnel, the equivalent diameter is first multiplied by 0.85 to obtain a corrected value for the equivalent diameter of the horseshoe shape, and then the cross-sectional area is calculated using the formula for approximating a circle based on this corrected diameter as an estimated value.

[0061] For example, if the equivalent diameter is 340mm (in the case of a circle) and the radius is 170mm, then the area of ​​the circle is approximately (about If the equivalent diameter is 206.43 mm (in the case of a rectangle), we can let the width = 206.43 mm and the height = 206.43 / 1.5 ≈ 137.62 mm, then the area of ​​the rectangle ≈ (about If the equivalent diameter is 141.67 mm (horseshoe case), we first get the corrected diameter ≈ 0.85 × 141.67 ≈ 120.42 mm, then calculate the area approximately as a circle ≈ (about ).

[0062] It should be noted that the treatment of rectangles and horseshoes is an engineering estimation method: the width and height of rectangles are directly derived from the aspect ratio, and the horseshoes are approximated as circles after being corrected by 0.85. These are simplified approximations used to quickly estimate the cross-sectional area in real-time scenes. If necessary, more refined shape models or on-site calibration data can be used to replace the above approximations.

[0063] Furthermore, in step S3, determining the traffic density within the tunnel based on the resonance peak sharpness value includes: The temporal variation characteristics of the formant sharpness value are analyzed. The first-order difference value of the sharpness value is calculated within ten consecutive frames. When the first-order difference value is continuously positive, it indicates that the traffic flow is increasing. When it is continuously negative, it indicates that the traffic flow is decreasing. When the first-order difference value fluctuates between ±0.1, it indicates that the traffic flow is stable, thus obtaining the dynamic trend of the traffic flow. In one embodiment, based on the time variation characteristics of the formant sharpness value, the system calculates the first-order difference (i.e., the sharpness of each frame minus the sharpness of the previous frame) for the sharpness values ​​of 10 consecutive frames, obtaining 9 first-order difference values, and uses the arithmetic mean of these difference values ​​as a traffic flow dynamic trend indicator: when the average value is greater than 0.1, it is determined that "traffic flow is increasing"; when the average value is less than -0.1, it is determined that "traffic flow is decreasing"; when the average value fluctuates between -0.1 and 0.1, it is determined that "traffic flow is stable".

[0064] For example, if the sharpness of 10 frames is [9.5, 9.6, 9.7, 9.8, 9.9, 10.0, 10.0, 10.1, 10.1, 10.2], then the average difference is about 0.07 (close to but less than 0.1), which is considered to indicate stable traffic flow; if the average value is 0.15, then it is considered to indicate increased traffic flow.

[0065] It should be noted that, in order to improve robustness, obvious abnormal frames should be removed first during implementation, and the sharpness can be smoothed for a short time (e.g., by the median of 3 frames or by moving average) to avoid incorrect trend judgment caused by abrupt changes in a single frame.

[0066] Traffic density is determined based on the absolute value of the resonance peak sharpness and the dynamic trend of traffic flow. If the sharpness value is less than the preset first threshold and the trend is stable, it is determined to be sparse traffic flow. If the sharpness value is greater than the preset first threshold and less than the preset second threshold and the trend is rising, it is determined to be medium traffic flow. Otherwise, it is determined to be dense traffic flow, thus obtaining the traffic density inside the tunnel.

[0067] In one embodiment, traffic density is determined based on the absolute value of the formant sharpness combined with the traffic flow dynamic trend obtained above: when the sharpness value is less than a preset first threshold and the trend is determined to be stable, it is determined to be sparse traffic flow; when the sharpness value is greater than the first threshold and less than the second threshold and the trend is determined to be rising, it is determined to be moderate traffic flow; in other cases (including abnormal combinations such as sharpness higher than the second threshold, sharpness between the thresholds but the trend is not rising, or sharpness lower than the first threshold but the trend is rising / falling), it is determined to be dense traffic flow.

[0068] For example, the first threshold can be set to 6 and the second threshold can be set to 12: if the detected sharpness is 4.8 and the trend is stable, it is judged as sparse; if the sharpness is 8.5 and the trend is increasing, it is judged as medium; if the sharpness is 13 or the sharpness is 9 but the trend is decreasing, it is judged as dense.

[0069] It should be noted that the threshold should be determined through on-site calibration and can be adaptively adjusted according to tunnel type or vehicle speed; to avoid frequent jumps between "medium / dense", a time-series smoothing or lag strategy can be added (for example, requiring the same judgment to be made 2 or 3 times in a row before outputting the final density).

[0070] In another embodiment, the trend can be determined using the "majority rule" instead of the mean method: if at least 8 of the 9 differences are positive, it is judged as "increasing"; if at least 8 are negative, it is judged as "decreasing"; otherwise, if the vast majority of differences fall within the ±0.1 range, it is judged as "stable". This method is more tolerant of abnormal extreme values, but is slightly less sensitive to small and continuous changes.

[0071] It should be noted that, regardless of the implementation method used, protective measures should be taken to handle frame drops, short-term noise, and abnormal sharpness values ​​to ensure the stability of the judgment.

[0072] Furthermore, step S4 includes the following steps: Step S41: Calculate the basic signal attenuation intensity based on the tunnel cross-sectional area using tunnel environmental characteristic parameters. The tunnel cross-sectional area is inversely proportional to the signal propagation loss, thus obtaining the basic attenuation parameters. In one embodiment, the basic signal attenuation intensity is calculated based on the obtained tunnel cross-sectional area parameters. An empirical model "inversely proportional to the cross-sectional area" is used to represent the damping effect of the tunnel on signal propagation. That is, the basic attenuation parameter is set as an empirical constant divided by the tunnel cross-sectional area, and the result is expressed in dB / m (or a system-standard distance attenuation unit). This empirical constant is obtained through indoor and outdoor field measurements or simulation calibration, and different constants can be used for different frequency bands to reflect frequency dependence.

[0073] For example, if the calibration constant is taken as 1 (unit: dB / m) and the tunnel cross-sectional area is... The basic attenuation parameter is approximately 0.1 dB / m, indicating that the signal attenuates by about 0.1 dB for every 1 m it advances.

[0074] It should be noted that this inverse proportional model is a rapid estimation model. In actual engineering, additional terms such as the sound absorption coefficient of the wall or the curvature of the tunnel can be added to improve the accuracy.

[0075] Step S42: Based on the tunnel environmental characteristic parameters, correct the environmental impact of the basic attenuation parameters by adjusting the traffic flow density inside the tunnel, and obtain the corrected environmental attenuation parameters; In one embodiment, an environmental correction is applied to the basic attenuation parameter based on the determined traffic density within the tunnel, using a multiplicative correction factor to amplify or reduce the basic attenuation. The correction factor can be a discrete level mapping, for example, a factor of 0.9 for sparse traffic flow, a factor of 1.0 for medium traffic flow, and a factor of 1.2 for dense traffic flow; alternatively, a continuous mapping function can be used to map the sharpness value or normalized traffic flow index to a factor in the range of 0.8–1.5.

[0076] For example, if the base attenuation parameter is 0.1 dB / m and the traffic flow is determined to be dense, and the correction factor is 1.2, then the corrected environmental attenuation parameter is 0.12 dB / m.

[0077] It should be noted that the correction factor should be calibrated using measured data, and secondary corrections can be made considering more dimensions (such as the proportion of heavy vehicles and vehicle speed distribution); to avoid abrupt changes, time smoothing or low-pass filtering can be applied to the correction factor.

[0078] Step S43: Starting from the current position of the vehicle, set up signal strength prediction nodes at equal intervals along the tunnel travel direction at preset lengths, where the preset length is determined based on the total length of the tunnel.

[0079] In one embodiment, signal strength prediction nodes are arranged at preset intervals along the tunnel travel direction, starting from the vehicle's current position. The preset interval is determined by the total tunnel length and the desired number of nodes, aiming to balance prediction accuracy with computational / storage overhead. A simple engineering strategy is to first set a target number range (e.g., 20–80 nodes), and then divide the total tunnel length by the target number of nodes to obtain the node spacing; a smaller spacing (e.g., 5–20m) can be selected for short tunnels, and a larger spacing (e.g., 20–50m) can be selected for medium-length tunnels.

[0080] For example, if the total length of the tunnel is 800m and the number of target nodes is 40, then the node spacing is 20m. The system will set 40 prediction nodes sequentially from the current position with a step size of 20m until the end of the tunnel.

[0081] It should be noted that since GPS fails inside the tunnel, node positioning should be combined with vehicle speedometer, wheel mileage or inertial measurement unit for cumulative mileage positioning, and the node spacing should be matched with or finer than the subsequent actual sampling / verification interval (e.g., sampling once every 80–120m in subsequent steps) for interpolation correction.

[0082] Furthermore, step S4 also includes the following steps: Step S44: The signal strength of the current location is detected by the vehicle navigation system as the reference signal strength. The signal strength loss value of each node is calculated based on the inverse square law using the environmental attenuation parameter and the signal strength prediction node. The signal strength loss value is the product of the environmental attenuation parameter and the distance of the current signal strength prediction node from the tunnel entrance. The predicted signal strength of each node is calculated by combining the reference signal strength at the tunnel entrance. In one embodiment, the actual signal strength at the current location of the vehicle is detected by the vehicle navigation system. This signal strength is used as the reference signal strength. The signal strength loss value of each prediction node is calculated by combining the calculated environmental attenuation parameter with the cumulative distance of each signal strength prediction node from the tunnel entrance according to the inverse square law. That is, the node signal strength loss value is equal to the environmental attenuation parameter multiplied by the distance of the node from the tunnel entrance. Then, the predicted signal strength of the node is obtained by subtracting the loss value from the reference signal strength at the tunnel entrance.

[0083] For example, if the signal strength at the tunnel entrance is -70dBm, the environmental attenuation parameter is 0.12dB / m, and the node is 100m away from the tunnel entrance, then the predicted signal strength is -70 - 0.12 × 100 = -82dBm.

[0084] It should be noted that this method assumes that the signal mainly propagates along the tunnel axis and ignores instantaneous large reflection interference, making it suitable for quickly estimating continuous signal attenuation.

[0085] Step S45: Add a Rayleigh distribution random correction within the range of ±2 to ±4dB to the signal strength loss value of each node. The correction range is determined according to the traffic flow density. It is ±2dB for sparse traffic flow, ±3dB for medium traffic flow, and ±4dB for dense traffic flow to obtain multipath fading compensation signal strength data. In one embodiment, a Rayleigh-distributed random correction is added to the signal strength of each node to simulate multipath fading and random environmental disturbances within the tunnel. The correction magnitude is determined by traffic density: ±2dB for sparse traffic, ±3dB for medium traffic, and ±4dB for dense traffic.

[0086] For example, if the predicted signal strength of a node is -82dBm and the traffic density is moderate, then a value within the range of 3dB that conforms to the Rayleigh distribution is randomly added or subtracted from -82dBm to obtain the signal strength after multipath fading compensation.

[0087] Step S46: Arrange the multipath fading compensation signal strength data in the order of spatial location within the tunnel, and smoothly connect adjacent nodes to generate a continuous communication quality prediction curve with the travel distance within the tunnel as the abscissa and the signal strength as the ordinate.

[0088] In one embodiment, the signal strength data of each node after multipath fading correction are arranged in ascending order according to the spatial position of the tunnel travel direction, and the adjacent nodes are smoothly connected, for example, by using linear interpolation or cubic spline interpolation methods, so that the signal strength change curve is continuous and without abrupt changes, and a continuous communication quality prediction curve is generated with the cumulative travel distance in the tunnel as the abscissa and the signal strength as the ordinate.

[0089] For example, the corrected signal strengths of the 40 predicted nodes are used to generate smooth curves through cubic spline interpolation, which are then used for display and subsequent transmission strategy adjustments.

[0090] Furthermore, step S46 includes the following steps: Step S461: Arrange the multipath fading compensation signal strength data in ascending order according to the spatial coordinates of each prediction node in the tunnel, perform cubic spline interpolation smoothing on adjacent nodes, and generate an initial communication quality prediction curve with the cumulative driving distance in the tunnel as the abscissa and the signal strength value as the ordinate. In one embodiment, the signal strength data of each predicted node after multipath fading compensation are arranged in ascending order according to the spatial coordinates of the node in the tunnel. Then, the signal strength between adjacent nodes is smoothed by cubic spline interpolation to generate a continuous initial communication quality prediction curve with the cumulative travel distance in the tunnel as the abscissa and the signal strength value as the ordinate.

[0091] For example, the signal strength of 40 predicted nodes is used to generate a smooth curve through cubic spline interpolation, making the curve smooth and continuous, reflecting the signal attenuation trend in the tunnel.

[0092] Step S462: Establish a real-time verification mechanism for the prediction curve. When the vehicle enters the tunnel, the actual signal strength data is collected every 80-120 meters. The collected actual signal strength is compared with the predicted value at the corresponding position in the prediction curve to generate the prediction deviation. In one embodiment, a real-time verification mechanism for the prediction curve is established. When the vehicle enters the tunnel, the actual signal strength is collected every 80-120 meters through the vehicle navigation system or signal receiving module. The collected real signal strength is compared with the predicted value at the corresponding distance position in the prediction curve, and the deviation between the two is calculated to form the prediction deviation.

[0093] For example, if the predicted signal for a node is -80dBm, while the measured signal is -88dBm, then the deviation is 8dB.

[0094] Step S463: When the prediction deviation of two consecutive collection points both exceed 6dB, calculate the average value and trend of the current cumulative deviation, and correct the initial communication quality prediction curve of the tunnel section that the vehicle has not yet passed according to the deviation direction and magnitude to obtain the corrected communication quality prediction curve. In one embodiment, when the prediction deviation of two consecutive acquisition points both exceed 6dB, the average value and trend of the current cumulative deviation are calculated, and the initial communication quality prediction curve of the tunnel section that the vehicle has not yet passed is corrected according to the direction and magnitude of the deviation, so that the corrected curve is closer to the actual signal change.

[0095] For example, if the average deviation is -7dB and the deviation is trending downwards, the prediction signal for subsequent nodes will be downgraded by 7dB overall, and the slope of the curve will be adjusted appropriately to match the trend.

[0096] Step S464: Mark the weak signal range with signal strength below -85dBm and the no signal range with signal strength below -100dBm on the corrected communication quality prediction curve to form the communication quality prediction curve.

[0097] In one embodiment, on the modified communication quality prediction curve, the interval with signal strength below -85dBm is marked as a weak signal area, and the interval with signal strength below -100dBm is marked as a no-signal area, thus forming a communication quality prediction curve that can be used for vehicle navigation and communication strategy planning.

[0098] Of particular importance, step S5 includes the following steps: Identify the normal communication range with signal strength above -85dBm, the weak signal range between -85dBm and -100dBm, and the communication blind zone below -100dBm in the communication quality prediction curve, and calculate the time required for the vehicle to pass through each range at the current speed. In one embodiment, the communication quality prediction curve is analyzed to identify the normal communication range with signal strength above -85dBm, the weak signal range from -85dBm to -100dBm, and the communication blind zone below -100dBm, and the time required to pass through each range is calculated based on the vehicle's current speed.

[0099] In the last normal communication interval before entering a weak signal area or communication blind spot, the navigation instructions, turning prompts and traffic information needed for the next 3 minutes are downloaded in advance and stored in the local cache. The download priority is executed in the order of navigation instructions, traffic information, and map updates. In one embodiment, during the last normal communication interval before entering a weak signal zone or communication blind spot, the system pre-downloads and stores the navigation instructions, turn prompts, and traffic information needed for the next 3 minutes into a local cache, based on the estimated time required to pass through the weak signal zone or blind spot. The download priority is executed in the order of navigation instructions first, followed by traffic information, and then map updates, to ensure that critical navigation information is prepared first.

[0100] It should be noted that the downloaded content should be dynamically adjusted according to the length of the weak signal area or blind spot and the vehicle speed to avoid insufficient cache or excessive downloading.

[0101] When a vehicle enters a weak signal area, all non-emergency data upload requests are suspended, only the minimum heartbeat connection is maintained, and the local cache navigation mode is switched to continue providing navigation services according to the pre-downloaded navigation instructions. In one embodiment, when a vehicle enters a weak signal area, all non-emergency data upload requests are suspended, only the minimum heartbeat connection is maintained, and the system switches to local cache navigation mode to continue providing navigation services according to pre-downloaded navigation instructions.

[0102] It should be noted that the heartbeat connection is only used to maintain the necessary communication channel so that data can be quickly synchronized when the signal is restored.

[0103] Within the communication blind zone, data transmission and reception are completely disabled. A timer is set based on the blind zone length estimated by the communication quality prediction curve. The communication module is automatically woken up 20 seconds before the vehicle is expected to leave the blind zone to prepare for reconnection, thus achieving communication interconnection under the tunnel.

[0104] In one embodiment, data transmission and reception are completely disabled within the communication blind zone. A timer is set based on the blind zone length estimated by the communication quality prediction curve. The communication module is automatically woken up 20 seconds before the vehicle is expected to leave the blind zone to prepare for reconnection, thereby achieving communication interconnection within the tunnel.

[0105] It should be noted that the blind spot length estimation and timer settings should be dynamically adjusted according to the vehicle speed and the actual communication curve to ensure that the communication module recovers in time before the vehicle leaves the blind spot.

[0106] See Figure 2 The present invention also provides a vehicle navigation communication interconnection system 100 for executing the vehicle navigation communication interconnection method described above, wherein the vehicle navigation communication interconnection system includes: The noise reference acquisition module 101 is used to acquire the rolling noise signal of the tires through the chassis microphone of the vehicle navigation system, record the rolling noise signal as the noise spectrum reference value when the vehicle approaches the tunnel, and identify the tunnel entrance time. The formant extraction module 102 is used to calculate the difference between the tire noise signal after entering the tunnel and the noise spectrum reference value based on the tunnel entrance time, and to extract the newly added formant frequency value and formant sharpness value. The tunnel environment analysis module 103 is used to calculate the tunnel cross-sectional area based on the resonant peak frequency value, determine the traffic density in the tunnel based on the resonant peak sharpness value, and combine the tunnel cross-sectional area and the traffic density in the tunnel to form tunnel environmental characteristic parameters. The communication quality prediction module 104 is used to query a preset communication attenuation mapping table based on tunnel environment characteristic parameters, calculate the predicted signal strength at each location in the tunnel, and generate a communication quality prediction curve. The transmission strategy adjustment module 105 is used to adjust the data transmission strategy of the vehicle navigation system based on the communication quality prediction curve to achieve communication interconnection under the tunnel.

[0107] See Figure 7The hardware layout of the vehicle navigation communication interconnection system of the present invention is shown in the figure. The upper top view of the vehicle shows a length of 4.5 meters and a width of 1.8 meters. Microphone 1 and microphone 2 are installed at the front and rear wheels respectively, spaced 2.7 meters apart. The vehicle navigation system is located on the front center console, and a camera is installed at the front of the vehicle to identify tunnel entrances. The lower side view of the vehicle shows both microphones installed on the chassis at a height of 0.3 meters above the ground, with a sampling angle of a 30-degree fan-shaped range, pointing towards the ground to capture tire rolling noise. The 2.7-meter distance between the front and rear microphones ensures simultaneous acquisition of noise signals from the left and right front wheels, providing a data basis for spectrum analysis. (See reference...) Figure 7 The layout design, microphone installation height and acquisition angle have been optimized to ensure the best signal-to-noise ratio and sound coverage, and the entire hardware system forms a complete in-vehicle communication and interconnection solution.

[0108] Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of the equivalents of the application are intended to be included within the invention.

[0109] The above description is merely a specific embodiment of the present invention, enabling those skilled in the art to understand or implement the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features of the invention herein.

Claims

1. A method for communication interconnection applied to a car navigation, characterized in that, The method comprises the following steps: Step S1: Collect the rolling noise signal of the tire through the chassis microphone of the vehicle-mounted navigator, record the rolling noise signal as the noise spectrum reference value when detecting the approach of the tunnel, and identify the tunnel entrance time; Step S2: Calculate the difference between the tire noise signal after entering the tunnel and the noise spectrum reference value based on the tunnel entrance time, extract the newly added resonance peak frequency value and resonance peak sharpness value; Step S3: Calculate the tunnel cross-sectional area according to the resonance peak frequency value, judge the traffic density in the tunnel according to the resonance peak sharpness value, and combine the tunnel cross-sectional area and the traffic density in the tunnel to form the tunnel environment characteristic parameter; Step S4: Query the preset communication attenuation mapping table according to the tunnel environment characteristic parameter, calculate the predicted signal strength of each position in the tunnel, and generate a communication quality prediction curve; Step S5: Adjust the data transmission strategy of the vehicle-mounted navigator based on the communication quality prediction curve to realize communication interconnection under the tunnel.

2. The method for communication interconnection applied to the car navigation according to claim 1, characterized in that, Step S1 comprises the following steps: Step S11: Collect the rolling noise of the left front wheel and the right front wheel through the front and rear microphones of the chassis simultaneously, with a sampling frequency of 8000 Hz, to obtain the rolling noise signal; Step S12: Frame processing is performed on the rolling noise signal, with a 50% overlap between adjacent frames, and fast Fourier transform is performed on each frame to obtain noise spectrum data; Step S13: Within a range of 50-100 meters from the tunnel entrance, the average value of twenty continuous frames of noise spectrum data is calculated as the noise spectrum reference value of the open road; Step S14: Determine the tunnel entrance time of the vehicle by identifying the tunnel entrance sign or light mutation through the vehicle-mounted camera.

3. The method for communication interconnection applied to the car navigation according to claim 2, characterized in that, Step S2 comprises the following steps: Step S21: Calculate the in-tunnel spectrum data of the tire noise signal after entering the tunnel based on the tunnel entrance time, and perform point-by-point difference operation on the noise spectrum reference value to obtain the difference spectrum; Step S22: Within the range of 200Hz-1000Hz of the difference spectrum, the window width is 30Hz, and the step is 10Hz, the average amplitude of all frequency points in each window position is calculated as the local average value, when the amplitudes of the window center and its adjacent two frequency points are all more than 1.5 times of the local average value and the center point amplitude is the highest, the center frequency point is marked as a candidate resonance peak.

4. The method for communication interconnection applied to the car navigation according to claim 3, characterized in that, Step S2 further comprises the following steps: Step S23: Expand the search to both sides of each candidate resonance peak until the amplitude decreases to 70% of the peak value, and record the frequency range as the 3dB bandwidth, and the sharpness value of the resonance peak is obtained by dividing the peak frequency by the 3dB bandwidth; Step S24: Track the frequency drift of the candidate resonance peak within five continuous frames, and confirm it as a stable resonance peak when the drift is less than 20Hz, and take the median of the five frame frequencies as the resonance peak frequency value; Step S25: Calculate the standard deviation of the sharpness value of the stable resonance peak within five frames, and take the sharpness mean value as the final resonance peak sharpness value when the standard deviation is less than 10% of the sharpness mean value.

5. The method for communication interconnection applied to the car navigation according to claim 4, characterized in that, In step S3, the tunnel cross-sectional area is calculated according to the resonance peak frequency value, which comprises: The tunnel cross-section shape is determined by the resonance peak frequency value, when the resonance peak frequency value is 400-600Hz, it is determined that the tunnel is circular, and the correction coefficient is kept as 1000; when the frequency is 600-800Hz, it is determined that the tunnel is rectangular, and the correction coefficient is adjusted to 850; when the frequency is more than 800Hz, it is determined that the tunnel is horseshoe-shaped, and the correction coefficient is adjusted to 750; The tunnel equivalent diameter is calculated according to the resonance peak frequency value, wherein the tunnel equivalent diameter = 340 / (2*resonance peak frequency value)*correction coefficient; The tunnel cross-sectional area is calculated according to the tunnel equivalent diameter and the tunnel equivalent diameter, wherein the circular tunnel uses the circular area formula, the rectangular tunnel is calculated according to the width-height ratio of 1.5, and the horseshoe-shaped tunnel is calculated according to 0.85 times the equivalent diameter.

6. The method for communication interconnection applied to the car navigation according to claim 5, characterized in that, The tunnel flow density in step S3 is determined according to the resonance peak sharpness value, which includes: The time variation characteristics of the resonance peak sharpness value are analyzed, the first-order difference value of the sharpness value is calculated within ten consecutive frames, when the first-order difference value is continuously positive, it indicates that the vehicle flow is increasing, when the first-order difference value is continuously negative, it indicates that the vehicle flow is decreasing, and when the first-order difference value fluctuates between positive and negative 0.1, it indicates that the vehicle flow is stable, and the dynamic trend of the vehicle flow is obtained; The vehicle flow density is determined according to the absolute value of the resonance peak sharpness value and the dynamic trend of the vehicle flow, when the sharpness value is less than a preset first threshold value and the trend is stable, it is determined that the vehicle flow is sparse, when the sharpness value is greater than the preset first threshold value and less than a preset second threshold value and the trend is rising, it is determined that the vehicle flow is moderate, otherwise, it is determined that the vehicle flow is dense, and the vehicle flow density in the tunnel is obtained.

7. The method for communication interconnection applied to the car navigation according to claim 6, characterized in that, Step S4 includes the following steps: Step S41: calculating the basic signal attenuation intensity based on the tunnel cross-sectional area of the tunnel environment characteristic parameter, the tunnel cross-sectional area is inversely proportional to the signal propagation loss, and the basic attenuation parameter is obtained; Step S42: correcting the environmental influence of the basic attenuation parameter according to the vehicle flow density in the tunnel environment characteristic parameter, and obtaining the corrected environmental attenuation parameter; Step S43: setting signal intensity prediction nodes at equal intervals according to a preset length along the driving direction of the tunnel with the current position of the vehicle as the starting point.

8. The method for communication interconnection applied to the car navigation according to claim 7, characterized in that, Step S4 further includes the following steps: Step S44: detecting the signal intensity of the current position as the reference signal intensity through the vehicle-mounted navigator, and calculating the signal intensity loss value of each node based on the inverse square law using the environmental attenuation parameter and the signal intensity prediction node; Step S45: adding Rayleigh distribution random correction amount in the range of ±2 to ±4dB to the signal intensity loss value of each node, and the correction amplitude is determined according to the vehicle flow density, which is ±2dB for sparse vehicle flow, ±3dB for moderate vehicle flow, and ±4dB for dense vehicle flow, to obtain the multipath fading compensation signal intensity data; Step S46: arranging the multipath fading compensation signal intensity data in the order of spatial position in the tunnel, and smoothly connecting adjacent nodes to generate a continuous communication quality prediction curve with driving distance as the horizontal coordinate and signal intensity as the vertical coordinate.

9. The method for communication interconnection applied to the car navigation according to claim 8, characterized in that, Step S46 includes the following steps: Step S461: The multipath fading compensation signal strength data is arranged in ascending order according to the spatial position coordinates of each predicted node in the tunnel, and a three-spline interpolation smoothing processing is performed on adjacent nodes to generate an initial communication quality prediction curve with the cumulative driving distance in the tunnel as the horizontal coordinate and the signal strength value as the vertical coordinate; Step S462: A real-time verification mechanism of the prediction curve is established, and actual signal strength data is collected every 80-120 meters after the vehicle enters the tunnel, and the deviation is compared between the collected real signal strength and the predicted value at the corresponding position of the prediction curve to generate a prediction deviation; Step S463: When the prediction deviations of two consecutive collection points both exceed 6 dB, the average value and the change trend of the current cumulative deviation are calculated, and the initial communication quality prediction curve of the tunnel section that has not been passed by the vehicle is modified according to the deviation direction and amplitude to obtain a modified communication quality prediction curve; Step S464: The weak signal interval with a signal strength lower than -85 dBm and the no signal interval with a signal strength lower than -100 dBm are marked on the modified communication quality prediction curve to form a communication quality prediction curve.

10. A communication interworking system applied to a car navigation, characterized in that, The application of the communication interconnection method applied to the vehicle-mounted navigator communication interconnection method as claimed in claim 1, the application of the communication interconnection system applied to the vehicle-mounted navigator communication interconnection system comprises: A noise reference collection module is used to collect the rolling noise signal of the tire through the chassis microphone of the vehicle-mounted navigator, record the rolling noise signal as a noise spectrum reference value when approaching the tunnel is detected, and identify the tunnel entrance time; A resonance peak extraction module is used to calculate the difference between the tire noise signal after entering the tunnel and the noise spectrum reference value based on the tunnel entrance time, extract the newly added resonance peak frequency value and resonance peak sharpness value; A tunnel environment analysis module is used to calculate the tunnel cross-sectional area according to the resonance peak frequency value, judge the traffic density in the tunnel according to the resonance peak sharpness value, and combine the tunnel cross-sectional area and the traffic density in the tunnel to form a tunnel environment characteristic parameter; A communication quality prediction module is used to query a preset communication attenuation mapping table according to the tunnel environment characteristic parameter, calculate the predicted signal strength at each position in the tunnel, and generate a communication quality prediction curve; A transmission strategy adjustment module is used to adjust the data transmission strategy of the vehicle-mounted navigator based on the communication quality prediction curve to realize the communication interconnection under the tunnel.

Images