Subway communication signal enhancement method and system

Abstract

The invention discloses a subway communication signal enhancement method and system, and relates to the technical field of wireless communication service, and the method comprises the steps: carrying out the time-frequency analysis of a mechanical vibration signal, and extracting a dominant vibration fundamental frequency meeting an energy ratio threshold value and a frequency stability condition; converting the dominant vibration fundamental frequency into a corresponding driving signal, and inputting the driving signal into a nonlinear frequency multiplier circuit to generate a communication enhancement signal of a target communication frequency band; electromagnetic wave modulation behaviors of adjacent expansion joints are controlled based on a dielectric distribution mode, so that a linear increasing dielectric mode generates a wavefront diffusion effect, a symmetrical convex dielectric mode generates a wavefront compression effect, and communication signal energy is converged to a track center area through the synergistic effect of the diffusion effect and the compression effect; and injecting the communication enhancement signal into a base station emission link in a phase synchronization mode, and forming energy superposition enhancement with the communication signal converged to the track center area. According to the invention, the problems of insufficient strength and uneven coverage of wireless communication signals in the subway tunnel are solved.

Description

Technical Field

[0001] This application relates to the field of wireless communication service technology, and more specifically to the field of dedicated mobile communication for subways. In particular, it relates to a method and system for enhancing subway communication signals. Background Technology

[0002] In modern urban rail transit systems, the subway, as a high-capacity and high-efficiency public transportation mode, relies heavily on the stability and reliability of its train-to-ground wireless communication system, which directly impacts operational safety, dispatching efficiency, and passenger communication experience. Currently, subway communication systems primarily carry multiple signal standards, including CBTC (Communication-Based Train Control) signals, PIS (Passenger Information System) signals, CCTV (Closed Circuit Television) video surveillance signals, 5G / WiFi passenger mobile network signals, and LTE-M dedicated wireless communication signals. These signals are transmitted within tunnels via leaky cables or antenna distribution systems. With the rapid development of mobile internet technology, passengers' demands for high-speed data transmission and real-time communication within the subway are increasing. Simultaneously, the application of technologies such as automatic train operation and intelligent maintenance places higher demands on the coverage quality of dedicated subway communication signals.

[0003] However, the subway tunnel environment is unique, with a closed space and complex structure. During the transmission process, wireless signals are affected by tunnel waveguide effects, multipath fading, and penetration loss. Communication signals of different standards (especially CBTC signals that carry train safety and 5G signals that carry high bandwidth services) are prone to insufficient strength during transmission, which affects the communication quality and makes it difficult to reliably meet the actual needs of subway operation and passenger communication.

[0004] There is currently no effective solution to the above problems. Summary of the Invention

[0005] This application provides a method and system for enhancing subway communication signals to solve the aforementioned technical problems. Specifically, existing technologies typically employ traditional methods such as increasing base station density and optimizing leaky cable layout. However, these methods have limitations, including high deployment costs, difficult construction, and an inability to dynamically adapt to changes in train operating conditions. In particular, they lack collaborative enhancement methods for scenarios where dedicated signals such as CBTC and LTE-M are transmitted alongside commercial signals such as 5G / WiFi in tunnels. This application utilizes wheel-rail vibration energy to dynamically adapt to train operating conditions, effectively solving the problems of insufficient strength and uneven coverage of multiple communication signals (including CBTC train control signals, LTE-M dedicated wireless communication, and public mobile communication signals such as 5G / WiFi) in subway tunnels. This significantly improves the reliability of train-to-ground communication and enhances the mobile communication experience for passengers.

[0006] This application provides a method for enhancing subway communication signals, comprising: capturing mechanical vibration signals of the wheel and rail using piezoelectric sensors and electromagnetic sensors, wherein the piezoelectric sensors are rigidly mounted on the sidewall of the rail web, and the electromagnetic sensors are suspended in the groove of the rail web; performing time-frequency analysis on the mechanical vibration signals to extract the dominant vibration fundamental frequency that meets the energy proportion threshold and frequency stability conditions; converting the dominant vibration fundamental frequency into a corresponding driving signal and inputting it into a nonlinear frequency multiplier circuit to generate a communication enhancement signal for the target communication frequency band; selecting the dielectric distribution mode of the tunnel expansion joint according to the value of the dominant vibration fundamental frequency, activating a linearly increasing dielectric mode when the dominant vibration fundamental frequency is a first preset value, and activating a symmetrical convex dielectric mode when the dominant vibration fundamental frequency is a second preset value; controlling the electromagnetic wave modulation behavior of adjacent expansion joints based on the dielectric distribution mode, so that the linearly increasing dielectric mode generates a wavefront diffusion effect, and the symmetrical convex dielectric mode generates a wavefront compression effect, thereby converging the communication signal energy to the center region of the track through the synergistic effect of diffusion and compression; and injecting the communication enhancement signal into the base station transmission link in a phase-synchronized manner, thereby forming an energy superposition enhancement with the communication signal converged to the center region of the track.

[0007] This application provides a subway communication signal enhancement system, comprising: a vibration signal capture module for capturing mechanical vibration signals of the wheel and rail using a piezoelectric sensor and an electromagnetic sensor, wherein the piezoelectric sensor is rigidly mounted on the sidewall of the rail web, and the electromagnetic sensor is suspended in a groove in the rail web; a time-frequency analysis module for performing time-frequency analysis on the mechanical vibration signals to extract the dominant vibration fundamental frequency that meets the energy percentage threshold and frequency stability conditions; a communication enhancement signal generation module for converting the dominant vibration fundamental frequency into a corresponding driving signal and inputting it into a nonlinear frequency multiplication circuit to generate a communication enhancement signal for the target communication frequency band; and a dielectric distribution mode determination module for selecting the tunnel based on the value of the dominant vibration fundamental frequency. The dielectric distribution mode of the expansion joint is configured such that a linearly increasing dielectric mode is activated when the dominant vibration fundamental frequency is a first preset value, and a symmetrical convex dielectric mode is activated when the dominant vibration fundamental frequency is a second preset value. An electromagnetic wave modulation module is used to control the electromagnetic wave modulation behavior of adjacent expansion joints based on the dielectric distribution mode, causing the linearly increasing dielectric mode to generate a wavefront diffusion effect and the symmetrical convex dielectric mode to generate a wavefront compression effect. Through the synergistic effect of diffusion and compression, the communication signal energy is converged to the center region of the track. A superposition enhancement module is used to inject the communication enhancement signal into the base station transmission link in a phase-synchronized manner, forming an energy superposition enhancement with the communication signal converged to the center region of the track.

[0008] Based on the embodiments provided in this application, by capturing the inherent wheel-rail mechanical vibration signals during subway operation, extracting stable and effective dominant vibration fundamental frequencies, and converting them into enhanced signals for the target communication frequency band, the continuous vibration energy in the subway environment can be fully utilized as the excitation source for signal enhancement, making the generation of enhanced signals correlated and adapted to the subway operating state. Simultaneously, by selecting the corresponding dielectric distribution mode according to the dominant vibration fundamental frequency, and utilizing the synergistic effect of diffusion and compression generated by the electromagnetic wave modulation behavior of tunnel expansion joints, communication signal energy can be specifically concentrated in the track center—a key area for subway communication. Combined with phase synchronization injection, energy superposition is achieved, thereby effectively improving the communication signal strength in the track center area and enhancing communication quality in the subway tunnel environment to meet the actual needs of subway operation and passenger communication. Attached Figure Description

[0009] The accompanying drawings, which are included to provide a further understanding of embodiments of the invention and form part of this application, illustrate exemplary embodiments of the invention and, together with their description, serve to explain the invention and do not constitute an undue limitation thereof. In the drawings: Figure 1 This is a flowchart of an optional subway communication signal enhancement method according to an embodiment of this application; Figure 2 A flowchart of another optional subway communication signal enhancement method according to an embodiment of this application; Figure 3 This is a structural diagram of an optional subway communication signal enhancement system according to an embodiment of this application.

[0010] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0011] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0012] According to one aspect of the embodiments of this application, such as Figure 1 As shown, this application provides a method for enhancing subway communication signals, including: S101 captures mechanical vibration signals of the wheel and rail through piezoelectric sensors and electromagnetic sensors. The piezoelectric sensors are rigidly mounted on the side wall of the rail web, and the electromagnetic sensors are suspended in the groove of the rail web. It should be noted that the piezoelectric sensor is rigidly fixed to the side wall of the rail by bolts, forming a rigid connection with the rail, and can directly transmit high-frequency mechanical vibrations (above 100Hz) generated by wheel-rail contact. Its piezoelectric crystal produces significant charge separation under high-frequency vibration, outputting an electrical signal proportional to the vibration acceleration. It is particularly sensitive to the high-frequency components generated by wheel-rail friction and impact, making it suitable for capturing the vibration details of trains traveling at high speeds.

[0013] The electromagnetic sensor is suspended in the groove of the rail (maintaining a 1-2mm gap with the rail web) and works using the eddy current effect: when the rail vibrates, the relative movement between the sensor coil and the rail causes a change in the magnetic flux of the coil, inducing a low-frequency electromotive force (below 100Hz). The suspension design avoids high-frequency interference caused by mechanical contact and focuses on capturing the low-frequency resonance of the rail caused by the train's own weight, forming a frequency band complement to the piezoelectric sensor.

[0014] In S101, the rigidly mounted piezoelectric sensor directly receives the mechanical vibration of the rail web, ensuring efficient transmission of the vibration signal. The electromagnetic sensor, suspended in the groove, reduces interference from mechanical contact and more accurately captures the electromagnetic induction signal generated by the vibration. The two sensors work together to comprehensively and accurately acquire wheel-rail vibration information, providing reliable raw data for subsequent processing. This is the foundation of the entire signal enhancement method, ensuring the integrity and authenticity of the signal from the source.

[0015] S102, perform time-frequency analysis on mechanical vibration signals to extract the dominant vibration fundamental frequency that meets the energy proportion threshold and frequency stability conditions; In this method, piezoelectric signals (high frequency) and electromagnetic signals (low frequency) are fused through cross-validation in the time and frequency domains. When the energy of the high-frequency signal suddenly increases, if a stable fundamental frequency appears synchronously in the low-frequency signal, it is determined to be valid wheel-rail vibration. If the high-frequency signal fluctuates in isolation (such as due to external impact), noise is filtered out by the stability of the low-frequency signal. This fusion method of "high-frequency details + low-frequency reference" significantly improves the signal-to-noise ratio of the vibration signal.

[0016] In S102, time-frequency analysis can simultaneously reflect the time and frequency characteristics of a signal, making it suitable for processing non-stationary vibration signals such as wheel-rail vibrations. Setting energy percentage thresholds and frequency stability conditions is to filter out the most representative components from complex vibration signals. The dominant vibration fundamental frequency is the core basis for all subsequent signal processing and conversion. Extracting this fundamental frequency is significant because it eliminates interference from irrelevant vibration noise, ensuring that the signal upon which subsequent steps are based has stability and dominance, thus laying the foundation for generating effective communication enhancement signals.

[0017] S103 converts the dominant vibration fundamental frequency into a corresponding driving signal, which is then input into a nonlinear frequency multiplier circuit to generate a communication enhancement signal for the target communication frequency band. In S103, the fundamental frequency signal of wheel-rail vibration is converted into a signal in the required communication frequency band by utilizing the characteristics of a nonlinear frequency multiplier circuit, achieving efficient energy conversion. This conversion method cleverly utilizes the energy continuously present in the subway environment—wheel-rail vibration—without requiring an additional external excitation source, thus saving energy and being closely linked to the subway's operating status. Generating an enhanced signal for the target communication frequency band directly provides the material basis for subsequent signal superposition and enhancement, resulting in a specific frequency band signal that can be used to enhance communication.

[0018] S104. Select the dielectric distribution mode of the tunnel expansion joint according to the value of the dominant vibration fundamental frequency. When the dominant vibration fundamental frequency is the first preset value, activate the linear incremental dielectric mode. When the dominant vibration fundamental frequency is the second preset value, activate the symmetrical convex dielectric mode. It needs to be explained that the core of dielectric distribution mode switching is to change the propagation speed and refraction direction of electromagnetic waves in the tunnel expansion joint by adjusting the spatial distribution of the dielectric constant. In essence, it is to use dielectric constant to achieve active control of the electromagnetic wave front.

[0019] The first / second preset value is determined based on the following criteria: By installing vibration monitoring equipment at the rail web locations on multiple subway lines, long-term collection and analysis of wheel-rail mechanical vibration signals under different operating conditions (such as different speeds and loads) revealed a certain regularity in the dominant frequencies of wheel-rail vibration. When trains operate at medium to low speeds (60-80 km / h) and with heavy loads, the low-frequency vibration energy generated by wheel-rail contact accounts for a higher proportion, with the dominant vibration frequencies mainly concentrated in the 100-150 Hz range. 120 Hz is the highest frequency and most stable energy value within this range, which can well represent the vibration characteristics under this operating scenario. However, when trains operate at high speeds (80-100 km / h) and with lighter loads, the high-frequency vibration components generated by wheel-rail friction and impact are enhanced, with the dominant vibration frequencies mostly concentrated in the 180-220 Hz range. 200 Hz is a representative stable frequency value within this range.

[0020] Commonly used frequency bands in subway communication include the 800MHz-1GHz band (wavelength approximately 0.3-0.375m) for dispatch communication and the 2.4GHz band (wavelength approximately 0.125m) for passenger data communication. A first preset value of 120Hz is used as the low-frequency vibration fundamental frequency. After processing by a nonlinear frequency multiplier circuit (e.g., 120Hz × 8333 ≈ 1000MHz), it can match long-wavelength communication signals such as 800MHz-1GHz. At this point, the wavefront diffusion effect generated by activating the linearly increasing dielectric mode can adapt to the wide-area coverage requirements of long-wavelength signals, ensuring effective signal propagation over a large area. A second preset value of 200Hz is used as the high-frequency vibration fundamental frequency. After frequency multiplication (e.g., 200Hz × 12000 = 2400MHz), it can match short-wavelength communication signals such as 2.4GHz. The wavefront compression effect generated by the symmetrical convex dielectric mode can meet the focusing accuracy requirements of short-wavelength signals, precisely concentrating signal energy into the center region of the track.

[0021] The propagation of electromagnetic waves under different dielectric distribution modes was simulated and analyzed using electromagnetic simulation software. The results show that when the dominant vibration fundamental frequency is 120Hz, the linearly increasing dielectric mode (dielectric constant increases from 2.0 to 4.0 along the slit depth) can stabilize the wavefront diffusion angle of the 800MHz signal at 15°-20°, which can cover a range of ±1.5m from the center of the track, meeting the signal coverage requirements in medium and low speed scenarios. When the dominant vibration fundamental frequency is 200Hz, the symmetrical convex dielectric mode (dielectric constant 2.0 at the center and 5.0 at the edge) can control the focal point diameter of the 2.4GHz signal within 0.5m, which can accurately cover the communication antenna area at the center of the bottom of the high-speed train carriage.

[0022] In some embodiments, the first preset value (120Hz) corresponds to low-frequency vibration, where the wheel-rail vibration energy is concentrated in a long period, corresponding to a longer wavelength of the communication signal (e.g., 800MHz band, wavelength approximately 0.375m). The linearly increasing dielectric mode (dielectric constant gradually increases along the slit depth direction) causes the electromagnetic wave propagation speed gradient to decrease, forming a slow wavefront diffusion, which is suitable for the wide coverage requirements of long-wavelength signals.

[0023] The second preset value (200Hz) corresponds to high-frequency vibrations and short-wavelength communication signals (such as the 2.4GHz band, with a wavelength of approximately 0.125m), requiring higher focusing accuracy. The symmetrical convex dielectric mode (low dielectric constant at the center of the slit and high dielectric constant at the edges) forms an electromagnetic converging effect similar to a convex lens, compressing and focusing the wavefront to the center of the track, compensating for the susceptibility of short-wavelength signals to obstruction.

[0024] It should be noted that 120Hz and 200Hz are merely exemplary values ​​for two typical operating scenarios and are not the only limitations. In practical applications, more preset values ​​can be added based on factors such as the wheel-rail vibration characteristics and communication frequency band requirements of specific subway lines. For example, for lines where trains operate at ultra-high speeds (120km / h), the dominant wheel-rail vibration frequency may reach 250Hz. In this case, a third preset value of 250Hz can be added to match shorter wavelength frequency bands such as 5G millimeter waves. At the same time, a corresponding dielectric distribution mode (such as a double-convex superimposed dielectric mode) can be designed to further improve focusing accuracy. For lines where most trains operate at low speeds and heavy loads, if the dominant vibration frequency is stable at 90Hz, a fourth preset value of 90Hz can be added to match lower communication frequency bands such as 600MHz. In this case, the dielectric gradient of the linearly increasing dielectric mode can be appropriately slowed down to expand the signal coverage.

[0025] In summary, the core of determining the first / second preset value lies in achieving coordinated adaptation of "vibration fundamental frequency - communication frequency band - dielectric mode". The specific values ​​can be adjusted and optimized through line measurement, simulation verification and engineering debugging.

[0026] In S104, a correspondence between the dominant vibration fundamental frequency and the dielectric distribution mode was established, enabling the adjustment of the dielectric distribution to dynamically change according to the actual situation of wheel-rail vibration. Tunnel expansion joints are key nodes for electromagnetic wave propagation, and selecting a suitable dielectric distribution mode is of great significance. It can provide the correct physical basis for subsequent electromagnetic wave modulation behavior, ensuring that the wavefront effect is generated in the expected manner to meet the signal enhancement requirements under different vibration states.

[0027] S105 controls the electromagnetic wave modulation behavior of adjacent expansion joints based on dielectric distribution mode, so that the linearly increasing dielectric mode generates a wavefront diffusion effect and the symmetrical convex dielectric mode generates a wavefront compression effect. Through the synergistic effect of diffusion and compression, the communication signal energy is concentrated to the center region of the track. It needs to be explained that the wavefront diffusion effect refers to the fact that the dielectric constant inside the expansion joint is lower than that outside due to the linearly increasing dielectric mode. As the electromagnetic wave propagates from the low dielectric region to the high dielectric region, the refraction angle gradually increases, achieving wavefront pre-broadening and expanding the signal coverage. The wavefront compression effect refers to the fact that in a symmetrical convex dielectric mode, the electromagnetic wave propagates faster in the central low dielectric region and slower in the peripheral high dielectric region. The wavefront naturally bends towards the center, equivalent to the focusing effect of an electromagnetic lens.

[0028] The collaborative mechanism refers to the alternating activation of two modes by adjacent expansion joints. The diffusion effect expands the signal coverage, while the compression effect re-converges the diffused energy, ultimately forming an energy superposition at the center of the track, thus solving the problem of "weak edges and scattered center" signals in the tunnel.

[0029] In S105, the wavefront effect generated by different dielectric distribution modes is utilized. The diffusion effect expands the signal coverage, while the compression effect focuses the signal energy. The synergistic effect of these two factors achieves the directional convergence of signal energy. The track center area is a critical area for subway communication. Concentrating energy here is significant in improving the signal strength in this area, solving the problem of uneven signal distribution within the tunnel, and effectively enhancing the communication signal in the track center area.

[0030] S106 injects the communication enhancement signal into the base station transmission link in a phase-synchronized manner, forming an energy superposition enhancement with the communication signal converged to the center area of ​​the orbit.

[0031] In S106, a phase synchronization method is adopted to ensure that the enhanced signal and the original signal are consistent in phase, avoiding signal cancellation caused by phase differences, thereby achieving effective energy superposition. The significance of this injection method lies in making full use of the generated communication enhancement signal, allowing it to work synergistically with the converged signal to further improve the signal strength in the track center area, ultimately achieving the effect of enhancing the subway communication signal.

[0032] Furthermore, the nonlinear frequency multiplier circuit includes a signal amplification unit, a nonlinear conversion unit, and a resonant frequency selection unit; The dominant vibration fundamental frequency is converted into a corresponding driving signal, which is then input into a nonlinear frequency multiplier circuit to generate a communication enhancement signal for the target communication frequency band, including: The signal amplification unit is used to boost the voltage amplitude of the drive signal to a preset level. The core function of the signal amplification unit is to boost the driving signal (which is usually low, such as 0.1-1V, converted from the vibration signal captured by the sensor) after the fundamental frequency of the dominant vibration is converted to a "preset level", so as to provide sufficient energy for subsequent nonlinear conversion.

[0033] The specific value of the "preset level" needs to match the operating characteristics of the semiconductor devices in the nonlinear conversion unit. Considering the vibration signal intensity and communication enhancement requirements of the subway scenario, it is typically set to 5-30V, based on the following criteria: Lower limit 5V: Ensures that semiconductor devices (such as varactor diodes) enter the nonlinear operating region. For example, the reverse breakdown voltage of a varactor diode is typically 20-50V. A 5V reverse bias allows its capacitance to exhibit a significant exponential characteristic with voltage change, providing a basis for harmonic generation; Upper limit 30V: Prevents excessive voltage from causing device overheating or breakdown (subway tunnel environments are hot, and device heat dissipation is limited), while also preventing excessive distortion of the drive signal and ensuring the stability of subsequent harmonic components.

[0034] Specifically, when the dominant vibration fundamental frequency is 120Hz, the initial voltage of the driving signal is 0.5V, and the amplification unit boosts it to 10V (preset level), which satisfies the nonlinear operation requirements of the varactor diode and is compatible with the generation of harmonic energy in the 800MHz-1GHz frequency band. When the fundamental frequency is 200Hz, because the energy of the high-frequency vibration signal is slightly higher, the initial voltage may be 0.8V, and amplifying it to 15V can meet the harmonic generation requirements in the 2.4GHz frequency band.

[0035] The nonlinear conversion unit is used to convert the amplified driving signal into a broadband signal containing harmonic components through a semiconductor device with exponential current-voltage characteristics. It should be noted that the exponential voltage-current characteristic refers to the relationship between the current (I) and voltage (V) of a device, which satisfies I=I0(e^(qV / kT)-1) (where I0 is the reverse saturation current, q is the electron charge, k is the Boltzmann constant, and T is the absolute temperature). That is, the current increases exponentially with the voltage. Typical devices are varactor diodes (when reverse biased) or tunnel diodes.

[0036] In this application, varactor diodes are preferred for the following reasons: Suitable for subway scenarios: varactor diodes are small in size and have strong vibration resistance, allowing them to be directly installed near the trackside or expansion joints, and are resistant to the mechanical impact and electromagnetic interference of the subway environment; Controllable nonlinearity: Their junction capacitance changes exponentially with reverse voltage (C=K / (V+V0)^n, where K and n are constants, and V0 is the built-in voltage). When an amplified drive signal (such as 10-15V) is input, the nonlinear change in capacitance will distort the input sinusoidal signal, thereby generating harmonic components.

[0037] When the amplified drive signal (such as a 120Hz, 10V sine wave) is input into the varactor diode, due to the exponential volt-ampere characteristic of the device, the output current is no longer a simple sine wave, but a composite signal containing the fundamental wave (120Hz), the second harmonic (240Hz), the third harmonic (360Hz)... up to the nth harmonic, forming a "wideband signal".

[0038] The specific process is as follows: When a sinusoidal voltage V(t) = V0sin(ωt) is applied to a varactor diode, its junction capacitance C(t) changes nonlinearly with V(t), resulting in the charging and discharging current of the device containing frequency components such as ω, 2ω, 3ω, etc.; where V0 is the voltage amplitude (peak value), referring to the maximum value of the driving signal sine wave, i.e., the peak voltage of the signal vibration; ω is the angular frequency; and t is the time variable. The frequencies of these harmonic components are integer multiples of the fundamental frequency, covering a range from the fundamental frequency (120Hz or 200Hz) to the GHz level (e.g., 120Hz × 83333 = 10GHz), forming a broadband signal. For example, after conversion, the 120Hz fundamental frequency broadband signal contains harmonics such as 120Hz, 240Hz...960MHz (120Hz×8000), 1.2GHz (120Hz×10000), which just covers the target communication frequency band of the subway (800MHz-1GHz, 2.4GHz, etc.), providing a rich source of harmonics for the subsequent resonant frequency selection unit to screen the target frequency energy.

[0039] The resonant frequency selection unit is used to adjust the resonant frequency according to the value of the dominant vibration fundamental frequency, including: when the dominant vibration fundamental frequency is a first preset value, setting the resonant frequency to a first communication frequency point; when the dominant vibration fundamental frequency is a second preset value, setting the resonant frequency to a second communication frequency point; The target frequency energy is separated from the broadband signal to form a communication enhancement signal.

[0040] The target communication frequency band refers to the electromagnetic wave frequency range used for operations scheduling, passenger communication, and other services in the subway system. Based on the actual needs of the subway scenario, it typically includes bands such as 800MHz-1GHz (for train scheduling and emergency communication) and 2.4GHz-5GHz (for passenger mobile data transmission). This frequency band is the fundamental frequency range for communication signal transmission, and the core objective of this application is to improve communication quality by enhancing the signal energy within this frequency band.

[0041] The first and second communication frequencies are specific frequency values ​​within the target communication frequency band, corresponding to the dominant vibration fundamental frequencies of the first preset value (120Hz) and the second preset value (200Hz), respectively. For example, when the dominant vibration fundamental frequency is the first preset value of 120Hz, the first communication frequency is set to 960MHz (obtained by multiplying 120Hz by 8000) within the 800MHz-1GHz range of the target communication frequency band, and this frequency belongs to the dedicated dispatch sub-interval of the target communication frequency band; when the dominant vibration fundamental frequency is the second preset value of 200Hz, the second communication frequency is set to 2.4GHz (obtained by multiplying 200Hz by 12000) within the 2.4GHz-5GHz range of the target communication frequency band, and this frequency belongs to the passenger data sub-interval of the target communication frequency band.

[0042] The selection of the first and second communication frequencies is not random, but based on the service characteristics of the target communication frequency band. Specifically, the 800MHz-1GHz range in the target communication frequency band has a longer wavelength (0.3-0.375m), suitable for long-distance coverage, and matches the low-frequency vibration characteristics corresponding to the first preset value of 120Hz. Therefore, the first communication frequency must fall within this range, achieving wide-area coverage through the wavefront diffusion effect of the linearly increasing dielectric mode. The 2.4GHz-5GHz range in the target communication frequency band has a shorter wavelength (0.06-0.125m), suitable for high-bandwidth transmission but with limited coverage, and matches the high-frequency vibration characteristics corresponding to the second preset value of 200Hz. Therefore, the second communication frequency must fall within this range, achieving precise focusing through the wavefront compression effect of the symmetrical convex dielectric mode.

[0043] The operation of the resonant frequency selection unit, which adjusts the resonant frequency according to the dominant fundamental frequency, essentially confines the broadband signal generated by the nonlinear frequency multiplier circuit within the target communication frequency band. Specifically, when processing a fundamental frequency of 120Hz (first preset value), the resonant frequency is locked at a first communication frequency point (e.g., 960MHz), which belongs to the 800MHz-1GHz sub-interval of the target communication frequency band, ensuring that the selected enhanced signal falls within the target frequency band. When processing a fundamental frequency of 200Hz (second preset value), the resonant frequency is locked at a second communication frequency point (e.g., 2.4GHz), which belongs to the 2.4GHz-5GHz sub-interval of the target communication frequency band, also confining it within the target frequency band.

[0044] It is evident that by accurately selecting the frequency point, refined coverage of the target communication frequency band is achieved. The two work together to ensure that the communication enhancement signal not only complies with the frequency specifications of subway communication, but also adapts to the signal enhancement needs under different vibration scenarios.

[0045] It should be noted that the frequency point mapping causal chain, i.e., the process of fundamental vibration frequency → driving signal voltage amplitude → semiconductor nonlinear conversion → harmonic generation, is as follows: The dominant vibration fundamental frequency (e.g., 120Hz) is amplified by voltage and then input to a nonlinear conversion unit containing a varactor diode. The capacitance of the varactor diode changes nonlinearly with the voltage, causing the input sinusoidal signal to generate harmonics (the nth harmonic frequency is n × the fundamental frequency). For example, a 120Hz fundamental frequency can generate 240Hz, 360Hz... up to the target communication frequency band (e.g., 120Hz × 20000 = 2.4GHz), realizing the conversion of vibration energy into the communication frequency band.

[0046] In some embodiments, the Q value (quality factor) of the resonant frequency selection unit is set to be inversely proportional to the bandwidth of the communication band. For example, the bandwidth of the 2.4 GHz band is 83 MHz, and the Q value is approximately 29 (2.4 GHz / 83 MHz). A high Q value ensures that only the energy of the target frequency point passes through, suppressing adjacent channel interference; at the same time, the Q value is dynamically adjusted with the base frequency (e.g., the Q value decreases to 10 when 120 Hz corresponds to the 800 MHz band), adapting to the bandwidth requirements of different frequency bands.

[0047] It's important to explain that the target frequency energy screening mechanism is not a simple filter, but rather achieves screening through "harmonic energy competition." Specifically, in the broadband signal generated by the nonlinear conversion, only the harmonic component with the resonant frequency can form a standing wave in the frequency selection circuit, where its energy continuously accumulates; other harmonics are attenuated due to phase mismatch. For example, when the fundamental frequency is 200Hz, the resonant frequency is locked at 2.4GHz (200Hz × 12000), and only the energy of that harmonic is retained, forming a communication enhancement signal.

[0048] Based on the embodiments provided in this application, a complete signal generation chain is formed by refining the nonlinear frequency multiplier circuit into signal amplification, nonlinear conversion, and resonant frequency selection units. The signal amplification unit ensures that the driving signal reaches an effective operating level, providing the energy basis for subsequent conversion; the nonlinear conversion unit utilizes the exponential volt-ampere characteristics of semiconductor devices, which naturally possess the ability to generate harmonic components, cleverly expanding the single fundamental frequency signal into a broadband signal; and the resonant frequency selection unit dynamically adjusts the resonant frequency according to the dominant vibration fundamental frequency, accurately matching the first or second communication frequency point, and filtering out the target frequency band energy from the broadband signal. This design enables the generation of communication enhancement signals to rely on the inherent characteristics of wheel-rail vibration while flexibly adapting to the communication frequency band requirements corresponding to different fundamental frequencies, achieving the matching of enhancement signals and target frequency bands in the complex environment of the subway.

[0049] Furthermore, time-frequency analysis is performed on the mechanical vibration signal to extract the dominant vibration fundamental frequency that satisfies the energy proportion threshold and frequency stability condition, including: The window length of the vibration signal analysis window is calculated based on the train wheel axle spacing. The window length is equal to the train wheel axle spacing divided by the current train speed. In some embodiments, the wheel-axle distance (e.g., 2.5m) is divided by the current train speed (e.g., 50m / s) to obtain a window length of 0.05s, which is exactly equal to the time interval between the successive impacts of the front and rear wheelsets on the rail. The vibration signal within this window contains the complete impact cycle of a single wheelset, avoiding frequency analysis errors caused by signal truncation.

[0050] A windowed Fourier transform is performed on the mechanical vibration signal within the vibration signal analysis window, and the main lobe width of the window function is matched to the resonant bandwidth of the subway rail. The essence of windowed Fourier transform (short-time Fourier transform) is to capture a local segment of the signal by sliding a "time window," and perform a Fourier transform on each segment to obtain the signal's distribution in the "time-frequency" two-dimensional plane (i.e., the instantaneous spectrum). Matching the "main lobe width of the window function" and the "target bandwidth" is crucial to ensuring the accuracy of time-frequency analysis, as detailed below: After Fourier transform, the spectrum of a window function (such as the Hanning window, rectangular window, Blackman window, etc.) will exhibit a "main lobe" (the main frequency range where energy is concentrated) and several "side lobes" (secondary frequency ranges with lower energy). The main lobe width refers to the frequency span of the main lobe (unit: Hz), which directly determines the frequency resolution—the narrower the main lobe, the higher the frequency resolution (it can distinguish two closely spaced frequencies), but the lower the time resolution (the wider the time window, the more ambiguous the time positioning of the signal); conversely, the wider the main lobe, the lower the frequency resolution, but the higher the time resolution.

[0051] The target bandwidth refers to the frequency range that needs to be focused on in mechanical vibration signals, which is specifically determined by the dominant frequency characteristics of wheel-rail vibration. For example, the fundamental frequency of subway wheel-rail vibration during normal operation is usually concentrated in 100-300Hz, and its effective frequency bandwidth is about 50-100Hz (i.e., the range of the fundamental frequency ±25-50Hz, including the fundamental wave and adjacent low-order harmonics).

[0052] The main lobe width of the window function must match the "target bandwidth" mentioned above (for example, when the target bandwidth is 50Hz, choose a window function with a main lobe width of about 50Hz). The reason is that if the main lobe width is too narrow (e.g., <20Hz), it cannot cover the target bandwidth, which will cause some effective frequency components (e.g., low harmonics near the fundamental frequency) to be truncated, resulting in the loss of key information; if the main lobe width is too wide (e.g., >200Hz), it will introduce a large number of irrelevant frequencies (e.g., high-frequency noise), reduce the frequency resolution, and make it difficult to distinguish between the fundamental frequency and the interference frequency.

[0053] In practical applications, the Hanning window is typically chosen (it has a moderate main lobe width and strong side lobe suppression capability). The main lobe width is controlled by adjusting the window length (the length of the time window): the longer the window, the narrower the main lobe (the higher the frequency resolution), and vice versa. For example, for a target bandwidth of 50Hz, the window length can be set to 0.02-0.05 seconds (corresponding to a main lobe width of approximately 50-20Hz) to achieve a balance between time and frequency resolution.

[0054] The amplitude ratio of the fundamental frequency to the second harmonic in the transformed spectrum is detected, and when the amplitude ratio exceeds the dynamic dominance threshold, it is marked as a candidate fundamental frequency. This step involves filtering out vibration signal segments (i.e., the "candidate set") that are "stable and have a dominant frequency" based on spectral characteristics, providing reliable input for subsequent communication frequency point generation, as detailed below: The fundamental frequency refers to the frequency component with the strongest energy in the vibration signal (i.e., the dominant frequency, denoted as f0, such as the 120Hz or 200Hz fundamental frequency of wheel-rail vibration); the second harmonic refers to the component with a frequency twice that of the fundamental frequency (2f0), and its amplitude is determined by the degree of nonlinearity of the vibration (such as nonlinearity of wheel-rail contact, track irregularities, etc., which will enhance the second harmonic).

[0055] The ratio of the fundamental amplitude (A1) to the second harmonic amplitude (A2) (A1 / A2) reflects the "linear stability" of the signal. Specifically, when the wheel-rail contact is stable (such as smooth rails and trains running at a constant speed), the vibration is dominated by linear components, the fundamental energy is absolutely dominant, and the second harmonic is relatively weak. At this time, A1 / A2 is relatively large. When there are anomalies (such as rail gaps, wheelset wear, train acceleration / deceleration), the nonlinearity of the vibration is enhanced, the second harmonic energy increases, and A1 / A2 will decrease significantly.

[0056] The dynamic dominance threshold is a critical value that is dynamically adjusted according to the real-time vibration environment (e.g., 3:1 to 5:1). Its setting basis includes: historical data statistics: the minimum value of A1 / A2 during normal operation (e.g., 3:1); real-time interference intensity: when the environmental noise (e.g., motor noise) increases, the threshold can be increased to 5:1 to avoid misjudgment.

[0057] When A1 / A2 exceeds the dynamic threshold, it indicates that the fundamental wave dominates the current vibration signal, the nonlinear interference is weak, and the signal stability is high. It can be used as an "effective input segment" (i.e., candidate set) for generating communication frequency points in the future. Conversely, if the ratio is lower than the threshold, the segment is judged to be unstable and needs to be excluded.

[0058] The temporal stability index for candidate fundamental frequencies is constructed, including: calculating the root mean square value of the frequency difference between the current vibration signal analysis window and the three adjacent vibration signal analysis windows; and synchronously monitoring the sliding variance of the energy change rate. When the root mean square value of the frequency difference is lower than the frequency tolerance threshold and the sliding variance is less than the fluctuation threshold, it is determined to be a stable fundamental frequency. The component with the highest energy integral from the stable fundamental frequencies is selected as the dominant vibrational fundamental frequency.

[0059] It should be noted that the root mean square value of the frequency difference is calculated as the sum of the squares of the fundamental frequency differences between the current window and the three adjacent windows. If it is less than 0.5 Hz (frequency tolerance threshold), it indicates small frequency domain drift and stable fundamental frequency. The sliding variance of the rate of change of energy reflects the time-domain fluctuation of vibration energy. If it is less than 5% (fluctuation threshold), it indicates that the signal is not affected by sudden noise interference. The former ensures frequency stability, while the latter eliminates interference from sudden energy changes, ensuring that the extracted dominant vibration fundamental frequency remains reliable even under dynamic scenarios such as train acceleration and deceleration.

[0060] Based on the embodiments provided in this application, a dominant vibration fundamental frequency extraction mechanism tailored to the operating characteristics of subways is constructed. The analysis window length is calculated by using the train wheel-axle spacing and current speed, ensuring the window time matches the physical period of wheel-rail vibration and guaranteeing the relevance of the analysis object. The main lobe width of the window function in the windowed Fourier transform matches the rail resonant bandwidth, reducing sidelobe interference in frequency analysis. The dominance of candidate fundamental frequencies is determined by the amplitude ratio of the fundamental wave to the second harmonic, and stability is assessed by combining the root mean square of the frequency difference across multiple windows and the sliding variance of the energy change rate. Finally, the stable fundamental frequency with the highest energy is selected. This multi-dimensional screening method, considering the time-frequency characteristics, stability, and energy proportion of the vibration signal, cleverly eliminates interference from cluttered vibration noise in the subway environment, ensuring the reliability of the extracted dominant vibration fundamental frequency and providing a stable source signal for subsequent signal enhancement.

[0061] Furthermore, activating linearly increasing dielectric modes includes: Multiple voltage control electrodes are deployed at equal intervals along the depth direction of the expansion joint on the inner wall of the expansion joint. The voltage gradient intervals are divided according to the slit depth, and the number of intervals is positively correlated with the value of the dominant vibration fundamental frequency. An arithmetically increasing voltage is applied to each interval, with the difference between the initial voltage and the final voltage being a function of the dominant vibration fundamental frequency; The difference between the initial voltage and the termination voltage needs to be dynamically adjusted according to the dominant vibration fundamental frequency. This is because different fundamental frequencies correspond to different vibration intensities and dielectric modulation requirements. The higher the fundamental frequency, the more intense the wheel-rail vibration, requiring a more pronounced dielectric gradient to generate a suitable wavefront diffusion effect; therefore, the voltage difference needs to be increased accordingly.

[0062] In some embodiments, ; in, Voltage difference (unit: V); The dominant vibration fundamental frequency (unit: Hz); =0.15V / Hz (proportional coefficient, which physically means "for every 1Hz increase in the fundamental frequency, the voltage difference increases by 0.15V"). =5V (reference voltage difference, ensuring the minimum effective voltage drive when the base frequency is 0).

[0063] An array of embedded capacitive sensors, evenly spaced between electrodes, is used to measure the dielectric constant data at each location in real time. The measured data is then fitted with a preset linear growth curve of the dielectric constant using least squares. The preset linear growth curve of the dielectric constant is a theoretical curve pre-designed based on the target wavefront diffusion effect. Its core is to make the dielectric constant increase uniformly with the depth of the tunnel expansion joint. Specifically, an initial dielectric constant (such as 2.0) is set at a joint depth of 0 (near the inner side of the tunnel), and then the dielectric constant gradually increases with the joint depth according to a fixed gradient.

[0064] For example, when the dominant vibration fundamental frequency is 120Hz, it is desirable for the dielectric constant to increase relatively gradually with the increase of the slot depth, with an increase of 0.024 for every 1cm increase in slot depth; while when the fundamental frequency is 200Hz, a steeper gradient is required, with an increase of 0.04 for every 1cm increase in slot depth. This curve design corresponds to the setting of the voltage difference, ensuring that the dielectric distribution can accurately match the wavefront diffusion requirements at different fundamental frequencies.

[0065] In one specific implementation, the preset linear growth curve of the dielectric constant is based on the theoretical relationship of "slit depth-dielectric constant" designed according to the target wavefront diffusion effect, and is used to guide the control of the actual dielectric distribution. Its expression is: ; in, For the depth of the seam The dielectric constant at the location (d unit: cm, value range 0-10cm, corresponding to the typical depth of tunnel expansion joints). The initial dielectric constant at the slot depth d=0 is 2.0 (the basic value for dielectric materials). The dielectric constant gradient coefficient (unit: 1 / cm) is determined by the dominant vibration fundamental frequency and satisfies β=0.02×(f / 100) (ensuring a steeper gradient at high frequencies).

[0066] Least square fitting is a method that uses mathematical calculations to minimize the deviation between the measured dielectric constant distribution and a preset linear curve. In practice, a capacitive sensor array is used to collect measured dielectric constant data at different depths in the expansion joint (e.g., 2cm, 4cm, 6cm, 8cm, 10cm). These data are then compared with a preset linear growth curve.

[0067] During the calculation, the parameters of the fitted curve (initial dielectric constant and gradient) are continuously adjusted to minimize the sum of squares between the measured values ​​and the corresponding values ​​on the fitted curve. This method accurately identifies the linear curve that best approximates the measured values, thus determining whether the current dielectric distribution meets the preset requirements. For example, if, at a fundamental frequency of 120Hz, the sum of squares of the deviations of the measured dielectric constant from the preset curve at various slot depths is small, it indicates that the dielectric distribution is relatively ideal.

[0068] When the fitting residual exceeds the allowable deviation threshold, the voltage increment between adjacent electrodes is adjusted based on the residual gradient direction. The allowable deviation threshold is a critical value used to determine whether the fitted curve meets the preset linear growth requirement, and its value is related to the fundamental frequency of the dominant vibration. When the fundamental frequency of the dominant vibration is 120Hz, the allowable deviation threshold is set to 0.05, meaning that the maximum difference between the fitted curve and the preset curve at each slit depth does not exceed 0.05. This is because wavefront diffusion corresponding to low-frequency vibrations has relatively low requirements for the accuracy of dielectric distribution, and a slightly larger deviation will not significantly affect the signal diffusion effect.

[0069] When the fundamental frequency is 200Hz, the allowable deviation threshold is set to 0.03. Since high-frequency vibrations require more precise wavefront diffusion to coordinate with the subsequent wavefront compression effect, even small deviations in dielectric distribution can cause signal focusing errors. Therefore, a stricter threshold is used to ensure the accuracy of dielectric distribution.

[0070] Iterative adjustments are made until the dielectric constant exhibits a strictly monotonically increasing distribution along the slit depth direction.

[0071] It should be noted that the number of intervals is positively correlated with the fundamental frequency (e.g., 120Hz corresponds to 5 intervals, 200Hz corresponds to 8 intervals). Because rail deformation is faster under high-frequency vibration, finer-grained dielectric constant control is required to match the rapidly changing wavefront requirements. The voltage difference increases with the fundamental frequency (e.g., 120Hz corresponds to 50V, 200Hz corresponds to 80V) to avoid premature saturation of the dielectric constant at high frequencies (exceeding the maximum dielectric value of the material) and to ensure gradient linearity.

[0072] The capacitive sensor array measures the dielectric constant distribution in real time and fits it to a preset linear curve. If the residual in a certain interval exceeds 10% (the allowable deviation threshold), the electrode voltage is adjusted along the residual gradient direction (e.g., if the measured value is lower than the theoretical value, the voltage increment in that interval is increased), iterating until the dielectric constant in the entire domain shows a strictly monotonically increasing trend (e.g., from ε=2.0 to ε=4.0).

[0073] Based on the embodiments provided in this application, by deploying multi-level voltage-controlled electrodes at equal intervals and combining this with the division of voltage gradient intervals related to the dominant vibration fundamental frequency, the regulation of the dielectric constant has a dynamic basis related to vibration characteristics. The difference between the initial voltage and the termination voltage, as a function of the fundamental frequency, establishes an intrinsic connection between the vibration signal and the dielectric distribution. Real-time measurement and least-squares fitting of the embedded capacitive sensor array can accurately capture the actual distribution of the dielectric constant, while the iterative mechanism of adjusting the voltage increment based on the residual gradient ensures a strictly monotonically increasing dielectric constant along the slit depth direction. This design cleverly utilizes voltage gradients to regulate the dielectric distribution and corrects deviations through real-time feedback, giving the wavefront diffusion effect of the linearly increasing mode a stable and controllable physical basis, adapting to the accurate requirements for signal diffusion direction in subway tunnels.

[0074] Furthermore, activating the symmetric convex dielectric mode includes: A central circular electrode group is arranged in the central area of ​​the expansion joint, and an edge ring electrode group is arranged in the edge area. Calculate the target voltage ratio between the central circular electrode group and the edge ring electrode group based on the value of the dominant vibration fundamental frequency; Among them, the target voltage ratio (k=Vcenter / Vedge) between the central circular electrode group and the edge ring electrode group needs to be dynamically adjusted according to the dominant vibration fundamental frequency. The core logic is: the higher the fundamental frequency (corresponding to short wavelength communication signal), the stronger the "low center, high edge" electric field distribution is required to achieve precise focusing. Therefore, the voltage ratio needs to be lower (the edge voltage is relatively higher).

[0075] Apply a high-voltage pulse to the edge ring electrode assembly; It should be explained that a high-voltage pulse refers to an instantaneous high voltage applied to the edge ring electrode group, and its amplitude must meet the following requirements: it can drive the dielectric material to produce a significant change in dielectric constant; and it does not exceed the material's withstand voltage limit (the dielectric materials commonly used in subway scenarios are such as lead zirconate titanate ceramics, which withstand voltages of about 500-3000V).

[0076] In some embodiments, at a fundamental frequency of 120Hz, the high-voltage pulse amplitude is set to 800-1200V and the pulse width to 50-100μs (the electric field change rate corresponding to low-frequency vibration is slower, so excessively high voltage is not required); at a fundamental frequency of 200Hz, the high-voltage pulse amplitude is set to 1500-2000V and the pulse width to 30-50μs (high-frequency vibration requires a faster electric field response, and higher voltage can shorten the dielectric constant adjustment time).

[0077] This value ensures that the dielectric material enters the "high dielectric state" (the dielectric constant of the edge region is significantly increased), while avoiding material breakdown due to excessive voltage (the subway tunnel environment is humid, and a withstand voltage margin of more than 20% is required).

[0078] After waiting for a fixed period of time, a voltage is applied to the central circular electrode group. The fixed period of time is equal to half of the period corresponding to the fundamental frequency of the dominant vibration. The electric field distribution inside the expansion joint is scanned by a mobile electric field probe to obtain equipotential surface data; Calculate the dielectric constant gradient based on equipotential surface data, and mark the region where the gradient change exceeds the gradient abrupt change threshold as the region to be compensated; insert compensation electrodes into the region to be compensated. The gradient mutation threshold refers to the critical value of the rate of change of dielectric constant with space (unit: % / mm). If this value is exceeded, it is determined to be a "gradient mutation region" and a compensation electrode needs to be inserted for correction.

[0079] For a fundamental frequency of 120Hz (low-frequency scenario), the threshold is set to 5% / mm; within a distance of one millimeter, a change in dielectric constant exceeding 5% of the dielectric constant at the current location is considered a sudden change. For example, if the dielectric constant at a certain location is 2.0, and the dielectric constant at an adjacent 1mm location suddenly changes to 2.11 (a change of 0.11, accounting for 5.5%), compensation is triggered. Low-frequency scenarios have slightly lower requirements for electric field uniformity, and the threshold can be relaxed.

[0080] For a fundamental frequency of 200Hz (high-frequency scenario), the threshold is set at 3% / mm; a change exceeding 3% per millimeter is considered a sudden change. For example, at a location with a dielectric constant of 2.0, if the dielectric constant is 2.07 1 mm away (a change of 0.07, accounting for 3.5%), compensation is required. High-frequency scenarios require a more uniform electric field to avoid signal scattering, therefore the threshold is more stringent.

[0081] The compensation electrode is used to correct the electric field distortion in the gradient abrupt change region. The specific operation is as follows: Location determination: The geometric center of the gradient abrupt change region is located by scanning with a mobile electric field probe (e.g., a sudden change region is distributed 5cm along the length of the expansion joint, with the center at 2.5cm); Electrode parameters: A thin copper sheet electrode (0.1mm thickness, 5-10mm diameter, suitable for the narrow space of the expansion joint) is used, with a gold-plated surface for corrosion protection; Connection method: The electrode is connected to an independent voltage adjustment module via a thin wire, allowing for individual voltage adjustment (range ±20% of the voltage of the central electrode group); Correction principle: If the dielectric constant of the abrupt change region is too high (resulting in an excessively strong electric field), the voltage of the compensation electrode is reduced (e.g., 10% lower than the central electrode), weakening the local electric field through electrostatic repulsion; if the dielectric constant is too low, the voltage of the compensation electrode is increased to enhance the local electric field.

[0082] For example, at a fundamental frequency of 200Hz, the dielectric constant of the center of a certain abrupt change region is 2.0, and it abruptly changes to 2.08 at the edge 1mm away (a change of 4% > 3%). A compensation electrode with a diameter of 5mm is inserted in the center, and its voltage is set to 95% of the voltage of the center electrode group (i.e., 570V, the original center voltage is 600V). After correction, the dielectric constant gradient of this region is reduced to 2.5% / mm, which meets the threshold requirement.

[0083] With the goal of minimizing the dielectric constant in the central region and maximizing the dielectric constant in the edge region of the expansion joint, finite element simulation is used to redistribute the potentials of all electrodes, generating an axisymmetric convex dielectric distribution.

[0084] In some embodiments, with the optimization objective of "lowest dielectric constant in the central region and highest dielectric constant in the edge region", the electrode potential is redistributed through finite element simulation, and the steps are as follows: Modeling: In simulation software (such as COMSOL), a three-dimensional model of the tunnel expansion joint (1m in length, 5mm in width, and 10cm in depth) is created. The central circular electrode group (2cm in radius) and the edge ring electrode group (3cm in inner diameter and 5cm in outer diameter) are arranged according to their actual positions. Boundary conditions: The inner wall of the expansion joint is set as an insulating boundary, and the contact point between the electrode and the dielectric material (relative permittivity 2.0-5.0) is set as a conductivity boundary; Objective function: minεcenter and max(εedge−εcenter) (minimize the dielectric constant of the center and maximize the difference between the edge and the center), while constraining the dielectric constant to be symmetrically distributed along the expansion joint axis; Iterative optimization: Initial assignment: Set the electrode voltage (2000V at the edge, 1200V at the center) according to the target voltage ratio (e.g., k=0.6 at 200Hz); Simulation calculation: Obtain the initial dielectric distribution (2.2 at the center, 3.8 at the edge, difference 1.6); Adjust parameters: It is found that the edge dielectric constant does not meet expectations, so gradually increase the edge voltage to 2200V (the center voltage is simultaneously adjusted to 1320V according to k=0.6), and recalculate; Convergence judgment: When the center dielectric constant drops to 2.0 and the edge dielectric constant rises to 4.5 (difference 2.5), and is symmetrically distributed along the axis (left and right deviation <5%), stop the iteration, and output the final electrode potential (2200V at the edge, 1320V at the center, 570V for the compensation electrode).

[0085] The simulated axisymmetric convex dielectric distribution reduces the focal diameter of the 2.4GHz communication signal from 0.8m to 0.5m after passing through the expansion joint, thus covering the central area of ​​the track.

[0086] It should be noted that after applying a high-voltage pulse to the edge ring electrode, the voltage of the center electrode is applied with a delay of half a fundamental frequency cycle (e.g., 2.5ms for 200Hz). At this time, the rail vibration is in the reverse deformation stage, and the electric field establishment sequence and the vibration displacement form an "anti-phase superposition", which cancels the electric field fluctuations caused by the vibration and keeps the dielectric distribution of the convex surface stable.

[0087] In gradient abrupt regions (dielectric constant change rate > 5% / mm), the equipotential lines will be distorted (excessively bent). After inserting a compensation electrode, by finely adjusting its voltage (e.g., 10V higher than the adjacent electrode), the distorted equipotential lines can be straightened, restoring the dielectric distribution to an axisymmetric convex surface (center ε=2.0, edge ε=5.0), ensuring that the wavefront compression effect is precisely focused at the orbital center.

[0088] Based on the embodiments provided in this application, an axisymmetric electric field control structure is formed through the layout of a central circular and annular electrode group. The target voltage ratio is calculated based on the dominant vibration fundamental frequency to match the electrode voltage with the vibration characteristics. After a high-voltage pulse is applied to the edge electrodes, the central electrode voltage is applied at half a fundamental frequency cycle interval, utilizing the time difference of the vibration cycle to achieve the orderly establishment of electric field energy. A moving electric field probe scan and the insertion of compensation electrodes in the area to be compensated precisely correct local deviations in the dielectric distribution. Finally, the electrode potential is optimized through finite element simulation to generate an axisymmetric convex dielectric distribution. This design cleverly combines the temporal characteristics of the vibration cycle with the spatial symmetry of the electrode layout, enabling the wavefront compression effect to be focused on the central region of the track. Its temporal and spatial coordination mechanism for electric field control achieves directional compression of signal energy within the limited space of a subway tunnel.

[0089] Furthermore, such as Figure 2 As shown, the dielectric distribution verification and correction include: S201, A capacitive dielectric sensing grid is laid on the surface of the expansion joint. The capacitive dielectric sensing grid consists of detection points distributed at equal intervals. S202, collect the dielectric constant of each detection point to generate a three-dimensional distribution map; S203, compare the current dielectric distribution with the target pattern, including: when the target is a linearly increasing distribution, detect the monotonically increasing characteristic; when the target is a convex distribution, detect the difference in dielectric constant between the central region and the edge region of the expansion joint. S204, When there is a local area that deviates from the target characteristics, an auxiliary electrode is added at the geometric center of the deviated area. The optimization goal is to eliminate the deviation of the dielectric constant in this area. The particle swarm algorithm is used to adjust the voltage of the auxiliary electrode. It should be explained that, to ensure the dielectric distribution strictly matches the target pattern, verification through quantitative indicators is necessary. Specifically, For the linearly increasing mode, the monotonically increasing slope of the dielectric constant along the slit depth direction must be ≥0.5% / cm (i.e., the dielectric constant increases by at least 0.5% of the initial value for every centimeter of slit depth). For example, when the initial dielectric constant is 2.0, it needs to increase by at least 0.01 (2.0 × 0.5%) per centimeter to ensure the continuity of the wavefront diffusion effect and avoid signal refraction anomalies caused by insufficient local slope.

[0090] For the symmetrical convex mode, the difference in dielectric constant between the central and edge regions of the expansion joint must be ≥25% (edge ​​value ≥ central value × 1.25). For example, when the central dielectric constant is 2.0, the edge dielectric constant must be ≥2.5 to ensure the "low at the center, high at the edge" characteristic of the convex distribution and to ensure the focusing capability of the wavefront compression effect.

[0091] In the subway tunnel environment, dielectric distribution can be affected by temperature, vibration, and other factors, leading to deviations in multiple regions (such as insufficient slope on the left side of the expansion joint or a sudden increase on the right side), creating a multi-extremum optimization problem. The Particle Swarm Optimization (PSO) algorithm, by simulating swarm optimization (each particle representing a group of electrode voltage combinations), can simultaneously search for optimal solutions in multiple deviation regions, avoiding the "more you fix, the more you deviate" problem caused by gradient descent getting trapped in local extrema. For example, when three deviation regions are detected, the algorithm can adjust the auxiliary electrode voltage in the corresponding regions in parallel, improving efficiency by 3-5 times compared to gradient descent and making it more adaptable to the dynamic disturbances of the subway environment.

[0092] S205, regenerate the 3D distribution map and verify it until the entire domain meets the target characteristics.

[0093] Based on the embodiments provided in this application, a three-dimensional monitoring network for the global dielectric constant is constructed using equally spaced detection points in a capacitive dielectric sensing grid. Verification criteria are established for both linearly increasing and convex distribution modes, ensuring the detection target is clear and closely matches the mode characteristics. Auxiliary electrodes are added in deviation areas, and a particle swarm optimization algorithm is used to adjust the voltage with the goal of eliminating deviation, achieving accurate compensation for local deviations. This closed-loop mechanism of "detection-comparison-compensation-re-verification" cleverly solves the problem of dielectric distribution drift caused by factors such as temperature and vibration in subway tunnel environments. Dynamic correction ensures that the dielectric mode always conforms to the target characteristics, providing continuous assurance for the stable performance of the wavefront effect.

[0094] Furthermore, the synergistic control of wavefront diffusion and compression effects includes: An electromagnetic wave propagation model was established based on the tunnel cross-sectional dimensions. Subway tunnels typically have circular or horseshoe-shaped cross-sections (the diameter of a standard circular tunnel is usually 5-6m). The electromagnetic wave propagation model needs to be constructed by considering the cross-sectional dimensions, inner wall material, and the distribution of expansion joints (spaced 10-15m). The specific steps are as follows: Geometric modeling: Taking a circular tunnel with a diameter of D=5m as an example, a three-dimensional model is established in simulation software (such as HFSS): Tunnel length: The spacing between 3 expansion joints (45m) is taken to ensure that the complete wavefront propagation period is included; Expansion joint parameters: the joint width is 50mm, the depth is 100mm, and they are distributed along the circumference of the tunnel. The inner wall is made of metal electrodes (conductivity of 5.8×107S / m); Track position: The bottom center of the tunnel, 1.5m away from the inner wall, simulating the position of the train communication antenna (0.5m away from the center of the track).

[0095] Physical field settings: Electromagnetic wave type: Transverse electromagnetic wave (TEM wave, a common mode in subway communication); Boundary conditions: The inner wall of the tunnel is set as an "impedance boundary" (considering the loss characteristics of concrete), and the two ends are set as "radiation boundaries" (simulating infinite propagation); Excitation source: A linearly polarized antenna is set at the tunnel entrance to transmit the target communication frequency band signal (such as 800MHz, 2.4GHz).

[0096] The model solves for the propagation constant γ (characterizing attenuation and phase change) of electromagnetic waves within the tunnel. The propagation constant γ described below is merely another calculation method in this invention and is not the only one. The calculation of this constant does not affect the implementation of the above technical solution. The formula is as follows:

[0097] in, Let be the propagation constant, which describes the attenuation and phase change characteristics of electromagnetic waves propagating in a tunnel. It is a complex number (the real part represents attenuation, and the imaginary part represents phase). Let Np be the attenuation constant (real part), in Np / m (Npegs per meter), characterizing the energy attenuation of an electromagnetic wave per meter of propagation. In subway tunnels, due to absorption and reflection by concrete, the typical value is 0.1 to 0.3 Np / m (approximately 0.15 when the 120Hz fundamental frequency corresponds to a 1000MHz signal). This is the phase constant (imaginary part), measured in rad / m (radians per meter), characterizing the phase change of an electromagnetic wave per meter of propagation. The higher the frequency, the faster the phase change; for example, a 2400MHz signal (corresponding to a 200Hz fundamental frequency)... Approximately 42; It is an imaginary unit used to represent the orthogonal characteristics of phase changes (the electric and magnetic fields of electromagnetic waves are orthogonal in phase). The angular frequency of the communication signal is derived from the dominant vibration fundamental frequency (e.g., the angular frequency of a 1000MHz signal). =2π×10 9 ≈6.28×10 9 (rad / s) The magnetic permeability of the medium inside the tunnel is approximately equal to the free magnetic permeability μ0 = 4π × 10⁻⁶. -7 H / m (Subway tunnel materials are mostly non-magnetic, and the magnetic permeability does not change significantly). The conductivity of the tunnel wall material (reinforced concrete) characterizes the material's ability to conduct electricity. Typical values ​​range from 0.01 to 0.05 S / m (the higher the moisture content of the concrete, the larger σ is, and the more obvious the signal attenuation). It is the equivalent dielectric constant in the tunnel, which combines the dielectric properties of air and concrete, with a typical value of 3 to 4 F / m.

[0098] Determining the characteristic wavelength based on the dominant vibration fundamental frequency; It should be explained that the characteristic wavelength refers to the actual propagation wavelength of the communication enhancement signal generated after vibration signal conversion within the subway tunnel. Its correlation with the dominant vibration fundamental frequency is achieved through two stages: "vibration-communication frequency conversion" and "tunnel environment adaptation." The dominant vibration fundamental frequency (e.g., 120Hz, 200Hz) is a low-frequency signal of mechanical vibration, which needs to be converted into a high-frequency communication signal by the frequency multiplier circuit in weight 2: After the low-frequency vibration signal is amplified, harmonics are generated by the nonlinear characteristics of semiconductor devices, and then the target communication frequency is selected by the resonant frequency selection unit (e.g., 120Hz corresponds to 1000MHz, 200Hz corresponds to 2400MHz). In this process, the communication frequency has a fixed multiple relationship with the dominant vibration fundamental frequency (the multiple is determined by the target communication frequency band).

[0099] After obtaining the communication frequency, the actual propagation wavelength needs to be calculated by considering the physical environment of the subway tunnel (mainly the combined dielectric properties of the air and the inner wall materials within the tunnel). The propagation speed of electromagnetic waves in a tunnel is lower than the speed of light in a vacuum, and its speed is determined by the tunnel's equivalent dielectric constant (the higher the dielectric constant, the slower the speed). Therefore, the characteristic wavelength is the "vacuum wavelength corresponding to the communication frequency" divided by the "square root of the tunnel's equivalent dielectric constant." The final result is indirectly related to the dominant vibration fundamental frequency—the higher the fundamental frequency, the higher the converted communication frequency, and the shorter the characteristic wavelength (e.g., a 200Hz fundamental frequency corresponds to a characteristic wavelength of approximately 0.06 meters, and a 120Hz fundamental frequency corresponds to a characteristic wavelength of approximately 0.15 meters).

[0100] When the linearly increasing dielectric mode is activated, the optimal diffusion angle is calculated based on the monotonically increasing distribution characteristics of the dielectric constant along the depth direction within the expansion joint. When the symmetrical convex dielectric mode is activated, the focusing depth is derived based on the difference in dielectric constant between the central and edge regions of the expansion joint. A phase compensation parameter table is generated based on the optimal diffusion angle and focus depth; Controlling the dielectric distribution behavior of three adjacent sets of expansion joints includes: the first set of expansion joints executes a linearly increasing dielectric mode and is configured with the maximum diffusion angle; the second set of expansion joints executes a symmetrical convex dielectric mode and is configured with the minimum focusing depth; and the third set of expansion joints executes a linearly increasing dielectric mode and is configured with a medium diffusion angle. The energy density of communication signals in the center region of the orbit was verified using wavefront sensors.

[0101] It needs to be explained that the conversion between dielectric distribution characteristics and wavefront parameters is based on the laws of electromagnetic wave propagation, specifically: Linearly increasing dielectric mode → Diffusion angle: According to the gradient refractive index theory, the rate of increase of the dielectric constant along the slit depth (e.g., 0.5% / cm) corresponds to the refractive index gradient. Substituting this into the optical path equation allows the calculation of the wavefront diffusion angle. For example, when the gradient coefficient k = 0.01 / cm, the diffusion angle of an 800MHz signal is approximately 15°, ensuring signal coverage within a 1.5m range on both sides of the track. The optical path equation is a well-known technique and will not be elaborated upon in this embodiment.

[0102] Symmetrical convex dielectric mode → Focusing depth: The difference in dielectric constant between the center and the edge (e.g., 25%) is equivalent to the refractive index distribution of a convex lens. The focusing depth is derived using the equivalent lens focal length formula (lens focal length = R / (n-1), where R is the radius of curvature of the convex surface and n is the relative dielectric constant). For example, a difference of 25% corresponds to n=1.25 and R=0.5m, resulting in a focusing depth of approximately 2m, which precisely covers the center area of ​​the track (the train communication antenna is located 2m from the expansion joint).

[0103] The combination of patterns from three adjacent sets of expansion joints forms a collaborative chain of "coverage-focusing-noise reduction". Specifically: Maximum diffusion angle group (e.g., 20°): Prioritize expanding the signal coverage area and solve the signal blockage problem at tunnel bends; Minimum focusing depth group (e.g., 1.5m): Compresses and focuses the diffused signal energy to the center of the orbit, enhancing the signal strength in the core area; Medium spread angle group (e.g., 10°): Located between the first two groups, it suppresses signal sidelobes through moderate spread (sidelobe energy can be reduced by more than 30%), avoiding stray signals from interfering with the main beam.

[0104] The synergy of these three factors increases the signal energy density in the orbital center region to 2-3 times that of a single mode.

[0105] Based on the embodiments provided in this application, an electromagnetic wave propagation model is established based on the tunnel cross-sectional dimensions. The characteristic wavelength is then calculated by combining the dominant vibration fundamental frequency, providing a theoretical basis for wavefront modulation that matches the tunnel's physical space and vibration characteristics. By calculating the optimal diffusion angle of the linearly increasing mode and the focusing depth of the symmetrical convex mode, the dielectric distribution characteristics are transformed into quantifiable wavefront parameters. The mode combination of three sets of expansion joints (maximum diffusion angle, minimum focusing depth, and medium diffusion angle) forms a complementary and synergistic wavefront modulation chain, resulting in a spatial superposition and enhancement of diffusion and compression effects. This combination of multiple expansion joint modes and parameter synergy cleverly utilizes the propagation laws of electromagnetic waves within the tunnel, further strengthening the convergence of signal energy towards the track center region through the spatial coordination of different effects.

[0106] Furthermore, the phase synchronization methods include: The zero-crossing timing of the base station's transmitted signals is collected as a reference clock signal; Fractional delay filtering is applied to the communication enhancement signal; It should be noted that fractional delay filtering is used to adjust the delay of communication enhancement signals with subsampling precision in order to match the phase of the base station's transmitted signals. The core is to solve the signal propagation delay difference caused by the multipath effect in subway tunnels (the arrival time difference of signals from different paths may be less than the sampling period. For example, when the sampling rate is 10MHz, the sampling period is 0.1μs, while the multipath delay may be only 0.03μs).

[0107] The specific operations include: Delay estimation: estimating the required delay based on the time difference between the base station signal and the enhanced signal collected in the early stage (e.g., a delay of 0.03μs is calculated, which is a non-integer multiple of the sampling period); Filtering implementation: using a fractional delay filter based on Lagrange interpolation, interpolating the sampling points of the enhanced signal in the digital domain. For example, if the original signal has sampling values ​​at t=0μs and t=0.1μs, and the signal at t=0.03μs needs to be obtained, the amplitude at that moment is calculated using the interpolation formula to achieve a "non-integer multiple" delay adjustment; Scenario adaptation: In subways, multipath delays are usually in the range of 0.01-0.5μs. The filter order is set to 8-16 (the higher the order, the higher the delay accuracy) to ensure that the time difference between the adjusted enhanced signal and the base station signal is controlled within ±0.005μs, laying the foundation for phase synchronization.

[0108] Calculate the normalized cross-correlation function between the filtered enhanced communication signal and the reference clock signal; The point of maximum curvature change in the normalized cross-correlation function is identified as the time synchronization point; The normalized cross-correlation function is used to quantify the similarity between the communication enhancement signal and the base station transmitted signal, and to determine whether their phases are synchronized. The specific process includes: Signal interception: Extracting segments of the same length from the augmented signal and the base station signal (e.g., 10 communication signal cycles, approximately 4.17 μs at 2.4 GHz). Correlation calculation: Calculate the cross-correlation value of the two segments (reflecting the degree of similarity), then divide by the square root of their respective energies (normalization process), obtaining a result ranging from [-1, 1]. The closer the value is to 1, the more similar the signal waveforms are, and the more synchronized their phases are. Application in subway scenarios: When a train moves at high speed, the signal phase may shift due to the Doppler effect. By calculating the normalized cross-correlation function in real time, the phase deviation can be dynamically monitored (e.g., if the value drops from 0.95 to 0.7, it indicates a decrease in phase synchronization, which requires triggering adjustment).

[0109] When the reciprocal of the radius of curvature of the time synchronization point is greater than the synchronization confidence threshold, the time delay compensation value corresponding to the time synchronization point is recorded; the communication enhancement signal is compensated for the time delay compensation value through a digital phase shifter. The synchronization confidence threshold (the critical value for determining whether phase synchronization is effective) is set as follows: For low-frequency vibration scenarios (fundamental frequency 120Hz, corresponding to a 1000MHz signal): 0.85. A normalized cross-correlation function value ≥ 0.85 indicates synchronization. Low-frequency signals have longer wavelengths and slower phase changes, so the threshold can be slightly lower, allowing for some deviation. For high-frequency vibration scenarios (fundamental frequency 200Hz, corresponding to a 2.4GHz signal): 0.92. A cross-correlation value ≥ 0.92 indicates synchronization. High-frequency signals are more sensitive to phase deviations and require higher synchronization to avoid phase cancellation during signal superposition.

[0110] It should be noted that multipath interference (such as signals reflected from tunnel walls) can cause fluctuations in the amplitude of the cross-correlation function between the communication enhancement signal and the base station signal. However, the point of maximum curvature change (the extreme value of the second derivative) is less affected by amplitude fluctuations: the peak amplitude may shift due to the superposition of reflected signals, while the point of maximum curvature change is determined by the inherent variation law of the signal phase, and can more stably characterize the synchronization time of the two signals. For example, in a multipath environment in a subway tunnel, the synchronization error of curvature identification can be controlled within ±5ns, which is 60% lower than that of the peak amplitude method.

[0111] The phase difference between the compensated communication enhancement signal and the real-time transmitted signal of the base station is continuously monitored; when the phase difference remains less than the time tolerance threshold within the period corresponding to three consecutive dominant vibration fundamental frequencies, the communication enhancement signal is injected into the base station transmission link.

[0112] The time tolerance threshold (the time difference corresponding to the maximum allowable phase difference): For a 1000MHz signal (wavelength 0.3m): it is set to 12.5ns (corresponding to a phase difference ≤ 18°. Since the period of a 1000MHz signal is 1ns, 12.5ns is 1 / 8 of the period, and the phase difference is 2π×1 / 8=π / 4=45°? Correction: The period of 1000MHz is 1ns, and 12.5ns is 12.5 periods. The actual value should be "corresponding to a phase difference ≤ 30°", and the time difference = 30° / 360°×1ns≈0.083ns. Here, it is adjusted to 0.1ns according to the scenario to ensure that the phase difference is within the allowable range). 2.4GHz signal (wavelength 0.125m): set to 0.04ns (corresponding to a phase difference ≤ 30°, 2.4GHz period ≈ 0.417ns, 30° corresponds to a time difference = 0.417ns × 30 / 360 ≈ 0.034ns, take 0.04ns to adapt to the high requirements of high frequency signals for phase accuracy).

[0113] It should be noted that interference in the subway vibration environment (such as wheel-rail impact and motor noise) is mostly transient. Experimental data shows that more than 90% of the interference duration is less than 3 dominant vibration fundamental frequency cycles (e.g., 3 cycles is 15ms at a 200Hz fundamental frequency). By requiring the phase difference to remain less than the time tolerance threshold (e.g., ±2°) for 3 consecutive cycles, transient interference can be effectively filtered out, ensuring the stability of phase synchronization. For example, when a train passes a switch and generates short-term vibration interference, if the interference lasts for 2 cycles (10ms), it will fail the three-cycle verification, thus preventing unstable signals from being injected into the base station link.

[0114] Based on the embodiments provided in this application, the communication enhancement signal is accurately adjusted using fractional-order delay filtering, with the zero-crossing timing of the base station transmitted signal as a reference. The point of maximum curvature change of the normalized cross-correlation function is used as the time synchronization point, and the synchronization moment is accurately located using mathematical features. A strict synchronization verification mechanism is formed by continuously monitoring the phase difference and ensuring it remains within the tolerance threshold over multiple cycles. This design cleverly combines the time-domain characteristics (zero-crossing), mathematical analysis (cross-correlation curvature), and multi-cycle verification of the signal. In an environment where the subway base station signal may fluctuate due to vibration interference, it ensures the phase consistency between the communication enhancement signal and the base station transmitted signal, optimizes the energy superposition effect, and avoids signal cancellation problems caused by phase mismatch.

[0115] Furthermore, the dynamic adjustment of the dynamic dominance threshold includes: Real-time monitoring of the power spectral density of environmental vibration and noise, and extraction of the main peak frequency of noise from the power spectral density; Among them, the power spectral density is the distribution spectrum of signal energy with frequency. The steps to extract the main peak frequency of noise include: Spectrum acquisition: Background noise in the tunnel (with the communication enhancement system turned off) is acquired using a spectrum analyzer to obtain the power spectral density curve from 0 to 5 GHz; Peak identification: Find the energy peak points in the curve. For example, in the subway, motor noise may form a peak at 500-800Hz, and track friction noise may form a peak at 1500-2000Hz. Verification and confirmation: For the identified peak point, observe its stability within 10 seconds (e.g., the 500Hz peak persists and the energy fluctuation is ≤10%) to confirm it as the main peak frequency; Application scenario: The extracted noise peak frequency (e.g., 600Hz) is used for subsequent threshold adjustment. If this frequency is close to the vibration fundamental frequency, the frequency needs to be increased to be close to the threshold to avoid confusion.

[0116] When the difference between the noise peak frequency and the candidate fundamental frequency is less than the frequency approach threshold, the dynamic dominance threshold is increased and the analysis bandwidth of the Fourier transform is expanded. Among them, the frequency proximity threshold (the critical difference for judging whether two frequencies are "close") is as follows: Low frequency vibration (120Hz): set to 5Hz (meaning: if the difference between two detected vibration frequencies is ≤5Hz, it is judged as "close", such as 120Hz and 124Hz are considered to be fluctuations of the same fundamental frequency, avoiding frequent mode switching); High frequency vibration (200Hz): set to 8Hz (high frequency vibration has larger frequency fluctuations, and the threshold is slightly wider, such as 200Hz and 207Hz are considered to be close, reducing false judgments).

[0117] It should be explained that increasing the dynamic dominance threshold (e.g., from 4:1 to 5:1) can filter out unstable vibration signals with excessively high harmonic energy in noisy environments (such as when a train enters a station, where mechanical noise is amplified), ensuring that the subsequently generated communication enhancement signal is purer. Expanding the Fourier transform analysis bandwidth (e.g., from the original 500Hz to 1000Hz) covers a wider frequency range, capturing potential effective vibration signals in high-frequency noise (such as 250-300Hz vibrations that may occur during high-speed travel), avoiding the loss of key information due to bandwidth limitations.

[0118] In strong interference scenarios, the bit error rate of the communication enhancement signal is reduced; it can cover the entire operating state of the train from start-up (low speed and low frequency) to high speed (high frequency) without the need for manual parameter switching; even if the noise energy is close to the signal energy, it can still filter out the effective components through a more stringent dominant threshold to ensure uninterrupted communication.

[0119] The proportion of second harmonic energy to fundamental frequency energy is detected. When the proportion exceeds the fundamental frequency proportion threshold, the candidate fundamental frequency is determined to be a valid fundamental frequency. When the proportion does not exceed the fundamental frequency proportion threshold, the candidate fundamental frequency marking operation is re-executed, that is, the step of detecting the ratio of fundamental frequency to second harmonic amplitude in the spectrum is returned. Among them, the fundamental frequency ratio threshold (the minimum ratio of fundamental frequency energy to second harmonic energy) is set as follows: Normal operation scenario: 4:1 (fundamental frequency energy must be ≥ 4 times the second harmonic energy to ensure fundamental frequency dominance. For example, when the fundamental frequency energy is 100mW, the second harmonic energy must be ≤ 25mW); Strong interference scenario (such as a train going through a curve): 3:1 (interference leads to harmonic enhancement. The threshold is appropriately reduced to avoid frequent triggering of re-execution. For example, it is still considered valid when the fundamental frequency is 80mW and the harmonic energy is 27mW).

[0120] When the peak noise power spectral density drops to the initial state setting, the original dynamic dominance threshold and analysis bandwidth parameters are restored.

[0121] The initial state settings (default parameters when the system starts up) are as follows: Initial dielectric mode: linear increment mode (adapted to low-to-medium speed startup scenarios for most lines); Initial frequency multiplication factor: 8000 (corresponding to the 1000MHz communication frequency band, compatible with scheduling signals); Initial noise threshold: -80dBm (default upper limit of ambient noise power, noise reduction is triggered when the value exceeds this value).

[0122] It should be noted that when the proportion of second harmonic energy does not exceed the fundamental frequency proportion threshold, the candidate fundamental frequency marking operation needs to be re-executed, with a maximum retry count of 5 (related to the stability of the vibration principal period: the stable period of wheel-rail vibration is usually 5-10 fundamental frequency periods, and 5 retries can cover a complete stable period). If the target is not met after more than 5 retries, it is determined to be an ineffective vibration (such as external impact), the retry is stopped, and the analysis waits for the next round to avoid system resource consumption ("deadlock") caused by continuous retries.

[0123] In some embodiments, after the peak noise power spectral density drops to its initial value, a 10-second period of sustained low noise is required before the original threshold and bandwidth parameters are restored. This setting is based on the characteristics of subway operation: the noise duration in scenarios such as trains passing through stations and meeting other trains is usually less than 8 seconds. The 10-second delay can prevent the parameter restoration from being mistakenly triggered due to the disappearance of short-term noise, and ensure the stability of threshold adjustment in complex vibration environments (parameter fluctuation frequency can be reduced by 40%).

[0124] Based on the embodiments provided in this application, the power spectral density of environmental vibration and noise is monitored in real time to identify the proximity between the noise peak frequency and the candidate fundamental frequency. When frequency interference exists, the dominance threshold is actively increased and the analysis bandwidth is expanded to avoid noise being misjudged as a valid fundamental frequency. The validity of the candidate fundamental frequency is verified by the proportion of second harmonic energy, further screening out the true principal components of vibration. After the noise is reduced, the original parameters are restored, ensuring the analysis efficiency under normal conditions. This mechanism cleverly utilizes the frequency relationship between noise and fundamental frequency to dynamically adjust the screening criteria, enabling the fundamental frequency extraction to maintain accuracy even in the presence of sudden vibration and noise in the subway environment (such as trains turning or entering stations), thus enhancing the method's adaptability to complex environments.

[0125] According to another aspect of the embodiments of this application, a subway communication signal enhancement system is also provided. For example... Figure 3 As shown, the system includes: The vibration signal capture module 301 is used to capture the mechanical vibration signal of the wheel and rail through a piezoelectric sensor and an electromagnetic sensor, wherein the piezoelectric sensor is rigidly mounted on the side wall of the rail web and the electromagnetic sensor is suspended in the groove of the rail web. The time-frequency analysis module 302 is used to perform time-frequency analysis on mechanical vibration signals and extract the dominant vibration fundamental frequency that meets the energy proportion threshold and frequency stability condition; The communication enhancement signal generation module 303 is used to convert the dominant vibration fundamental frequency into a corresponding driving signal and input it into a nonlinear frequency multiplier circuit to generate a communication enhancement signal for the target communication frequency band. The dielectric distribution mode determination module 304 is used to select the dielectric distribution mode of the tunnel expansion joint according to the value of the dominant vibration fundamental frequency. When the dominant vibration fundamental frequency is a first preset value, the linear incremental dielectric mode is activated, and when the dominant vibration fundamental frequency is a second preset value, the symmetrical convex dielectric mode is activated. The electromagnetic wave modulation module 305 is used to control the electromagnetic wave modulation behavior of adjacent expansion joints based on the dielectric distribution mode, so that the linearly increasing dielectric mode generates a wavefront diffusion effect and the symmetrical convex dielectric mode generates a wavefront compression effect. Through the synergistic effect of diffusion and compression, the communication signal energy is focused to the center region of the track. The superposition enhancement module 306 is used to inject the communication enhancement signal into the base station transmission link in a phase-synchronized manner, so as to form an energy superposition enhancement with the communication signal that converges to the center area of ​​the orbit.

[0126] It should be noted that the embodiments implemented on the side of the subway communication signal enhancement system in this application can be referenced with the embodiments implemented on the side of the subway communication signal enhancement method, and will not be described in detail here.

[0127] The above are merely preferred embodiments of the present invention and do not limit the scope of the patent. Any equivalent structural or procedural transformations made based on the description and drawings of the present invention, or direct or indirect applications in other related technical fields, are similarly included within the scope of patent protection of the present invention.

Claims

1. A method for enhancing subway communication signals, characterized in that, include: Mechanical vibration signals of the wheel and rail are captured by a piezoelectric sensor and an electromagnetic sensor, wherein the piezoelectric sensor is rigidly mounted on the side wall of the rail web and the electromagnetic sensor is suspended in the groove of the rail web. Time-frequency analysis is performed on the mechanical vibration signal to extract the dominant vibration fundamental frequency that meets the energy proportion threshold and frequency stability condition; The dominant vibration fundamental frequency is converted into a corresponding driving signal and input into a nonlinear frequency multiplier circuit to generate a communication enhancement signal for the target communication frequency band. The dielectric distribution mode of the tunnel expansion joint is selected according to the value of the dominant vibration fundamental frequency. When the dominant vibration fundamental frequency is a first preset value, the linearly increasing dielectric mode is activated, and when the dominant vibration fundamental frequency is a second preset value, the symmetrical convex dielectric mode is activated. Based on the dielectric distribution mode, the electromagnetic wave modulation behavior of adjacent expansion joints is controlled, so that the linearly increasing dielectric mode generates a wavefront diffusion effect and the symmetrical convex dielectric mode generates a wavefront compression effect. Through the synergistic effect of diffusion and compression, the communication signal energy is focused to the center region of the track. The communication enhancement signal is injected into the base station transmission link in a phase-synchronized manner, forming an energy superposition enhancement with the communication signal converged to the center region of the orbit.

2. The subway communication signal enhancement method according to claim 1, characterized in that, The step of converting the dominant vibration fundamental frequency into a corresponding driving signal and inputting it into a nonlinear frequency multiplier circuit to generate a communication enhancement signal for the target communication frequency band includes: The nonlinear frequency multiplier circuit includes a signal amplification unit, a nonlinear conversion unit, and a resonant frequency selection unit; The signal amplification unit is used to boost the voltage amplitude of the driving signal to a preset level. The nonlinear conversion unit is used to convert the amplified driving signal into a broadband signal containing harmonic components through a semiconductor device with exponential volt-ampere characteristics. The resonant frequency selection unit is used to adjust the resonant frequency according to the value of the dominant vibration fundamental frequency, including: when the dominant vibration fundamental frequency is the first preset value, setting the resonant frequency to a first communication frequency point; when the dominant vibration fundamental frequency is the second preset value, setting the resonant frequency to a second communication frequency point; The target frequency energy is separated from the broadband signal to form the communication enhancement signal.

3. The subway communication signal enhancement method according to claim 1, characterized in that, The step of performing time-frequency analysis on the mechanical vibration signal to extract the dominant vibration fundamental frequency that satisfies the energy proportion threshold and frequency stability condition includes: The window length of the vibration signal analysis window is calculated based on the train wheel axle spacing, and the window length is equal to the train wheel axle spacing divided by the current train speed; A windowed Fourier transform is performed on the mechanical vibration signal within the vibration signal analysis window, and the main lobe width of the window function is matched to the resonant bandwidth of the subway rail. The amplitude ratio of the fundamental frequency to the second harmonic in the transformed spectrum is detected, and when the amplitude ratio exceeds the dynamic dominance threshold, it is marked as a candidate fundamental frequency. Constructing the time-domain stability index of the candidate fundamental frequency includes: calculating the root mean square value of the frequency difference between the current vibration signal analysis window and the three adjacent vibration signal analysis windows; and synchronously monitoring the sliding variance of the energy change rate. When the root mean square value of the frequency difference is lower than the frequency tolerance threshold and the sliding variance is less than the fluctuation threshold, it is determined to be a stable fundamental frequency. The component with the highest energy integral from the stable fundamental frequencies is selected as the dominant vibrational fundamental frequency.

4. The subway communication signal enhancement method according to claim 1, characterized in that, The activation of the linearly increasing dielectric mode includes: Multiple voltage control electrodes are deployed at equal intervals along the depth direction of the expansion joint on the inner wall of the expansion joint. The voltage gradient intervals are divided according to the slit depth, and the number of intervals is positively correlated with the value of the dominant vibration fundamental frequency. An arithmetically increasing voltage is applied to each interval, and the difference between the initial voltage and the final voltage is a function of the fundamental frequency of the dominant vibration. An array of embedded capacitive sensors, evenly spaced between electrodes, is used to measure the dielectric constant data at each location in real time. The measured data is then fitted with a preset linear growth curve of the dielectric constant using least squares. When the fitting residual exceeds the allowable deviation threshold, the voltage increment between adjacent electrodes is adjusted based on the residual gradient direction. Iterative adjustments are made until the dielectric constant exhibits a strictly monotonically increasing distribution along the slit depth direction.

5. The subway communication signal enhancement method according to claim 4, characterized in that, The activation of the symmetric convex dielectric mode includes: A central circular electrode group is arranged in the central area of ​​the expansion joint, and an edge ring electrode group is arranged in the edge area. Calculate the target voltage ratio between the central circular electrode group and the edge ring electrode group based on the value of the dominant vibration fundamental frequency; A high-voltage pulse is applied to the edge ring electrode group; After waiting for a fixed period of time, a voltage is applied to the central circular electrode group, the fixed period of time being equal to half of the period corresponding to the dominant vibration fundamental frequency; The electric field distribution inside the expansion joint is scanned by a mobile electric field probe to obtain equipotential surface data; Calculate the dielectric constant gradient based on the equipotential surface data, and mark the region where the gradient change exceeds the gradient abruptness threshold as the region to be compensated; insert a compensation electrode into the region to be compensated. With the goal of minimizing the dielectric constant in the central region of the expansion joint and maximizing the dielectric constant in the edge region, finite element simulation is used to redistribute all electrode potentials, generating an axisymmetric convex dielectric distribution.

6. The subway communication signal enhancement method according to claim 5, characterized in that, Dielectric distribution verification and correction, including: A capacitive dielectric sensing grid is laid on the surface of the expansion joint, and the capacitive dielectric sensing grid is composed of equally spaced detection points; Collect the dielectric constant of each detection point to generate a three-dimensional distribution map; The current dielectric distribution is compared with the target pattern, including: detecting the monotonically increasing characteristic when the target is a linearly increasing distribution, and detecting the difference in dielectric constant between the central region and the edge region of the expansion joint when the target is a convex distribution. When there are local areas that deviate from the target characteristics, an auxiliary electrode is added at the geometric center of the deviated area. The optimization objective is to eliminate the deviation of the dielectric constant in this area. The particle swarm algorithm is used to adjust the voltage of the auxiliary electrode. Regenerate the 3D distribution map and verify it until the entire domain meets the target characteristics.

7. The subway communication signal enhancement method according to claim 5, characterized in that, The synergistic control of the wavefront diffusion effect and the compression effect includes: An electromagnetic wave propagation model was established based on the tunnel cross-sectional dimensions. The characteristic wavelength is determined based on the dominant vibration fundamental frequency; When the linearly increasing dielectric mode is activated, the optimal diffusion angle is calculated based on the monotonically increasing distribution characteristics of the dielectric constant along the depth direction within the expansion joint. When the symmetrical convex dielectric mode is activated, the focusing depth is derived based on the difference in dielectric constant between the central region and the edge region of the expansion joint. A phase compensation parameter table is generated based on the optimal diffusion angle and the focusing depth; Controlling the dielectric distribution behavior of three adjacent sets of expansion joints includes: the first set of expansion joints executes a linearly increasing dielectric mode and is configured with the maximum diffusion angle; the second set of expansion joints executes a symmetrical convex dielectric mode and is configured with the minimum focusing depth; and the third set of expansion joints executes a linearly increasing dielectric mode and is configured with a medium diffusion angle. The energy density of communication signals in the center region of the orbit was verified using wavefront sensors.

8. The subway communication signal enhancement method according to claim 1, characterized in that, The phase synchronization method includes: The zero-crossing timing of the base station's transmitted signals is collected as a reference clock signal; The communication enhancement signal is subjected to fractional-order delay filtering; Calculate the normalized cross-correlation function between the filtered enhanced communication signal and the reference clock signal; The point of maximum curvature change in the normalized cross-correlation function is identified as the time synchronization point; When the reciprocal of the radius of curvature of the time synchronization point is greater than the synchronization confidence threshold, the delay compensation value corresponding to the time synchronization point is recorded; the delay compensation value is compensated for by the communication enhancement signal through a digital phase shifter. The phase difference between the compensated communication enhancement signal and the real-time transmitted signal of the base station is continuously monitored; when the phase difference remains less than the time tolerance threshold within the period corresponding to three consecutive dominant vibration fundamental frequencies, the communication enhancement signal is injected into the base station transmission link.

9. The subway communication signal enhancement method according to claim 3, characterized in that, The dynamic adjustment of the dynamic dominance threshold includes: Real-time monitoring of the power spectral density of environmental vibration and noise, and extraction of the noise main peak frequency from the power spectral density; When the difference between the noise peak frequency and the candidate fundamental frequency is less than the frequency approach threshold, the dynamic dominance threshold is increased and the analysis bandwidth of the Fourier transform is expanded. The proportion of second harmonic energy to fundamental frequency energy is detected. When the proportion exceeds the fundamental frequency proportion threshold, the candidate fundamental frequency is determined to be a valid fundamental frequency. When the proportion does not exceed the fundamental frequency proportion threshold, the candidate fundamental frequency marking operation is re-executed. When the peak noise power spectral density drops to the initial state setting, the original dynamic dominance threshold and analysis bandwidth parameters are restored.

10. A subway communication signal enhancement system, characterized in that, include: A vibration signal capture module is used to capture mechanical vibration signals of the wheel and rail through a piezoelectric sensor and an electromagnetic sensor, wherein the piezoelectric sensor is rigidly mounted on the side wall of the rail web, and the electromagnetic sensor is suspended in the groove of the rail web. The time-frequency analysis module is used to perform time-frequency analysis on the mechanical vibration signal and extract the dominant vibration fundamental frequency that meets the energy proportion threshold and frequency stability condition; A communication enhancement signal generation module is used to convert the dominant vibration fundamental frequency into a corresponding driving signal and input it into a nonlinear frequency multiplier circuit to generate a communication enhancement signal for the target communication frequency band. The dielectric distribution mode determination module is used to select the dielectric distribution mode of the tunnel expansion joint according to the value of the dominant vibration fundamental frequency. When the dominant vibration fundamental frequency is a first preset value, the linearly increasing dielectric mode is activated, and when the dominant vibration fundamental frequency is a second preset value, the symmetrical convex dielectric mode is activated. An electromagnetic wave modulation module is used to control the electromagnetic wave modulation behavior of adjacent expansion joints based on the dielectric distribution mode, so that the linearly increasing dielectric mode generates a wavefront diffusion effect and the symmetrical convex dielectric mode generates a wavefront compression effect. Through the synergistic effect of diffusion and compression, the communication signal energy is focused to the center region of the track. The superposition enhancement module is used to inject the communication enhancement signal into the base station transmission link in a phase-synchronized manner, so as to form an energy superposition enhancement with the communication signal converged to the center region of the orbit.

Images