Fire door alarm information transmission device and method based on the Internet of Things

Abstract

The present invention discloses an Internet of Things-based fire door alarm information transmission device and method, which relates to the field of fire alarm technology and includes the following steps: after the fire emergency mode is activated, continuously scanning the wireless communication environment in the area where the fire door alarm device is deployed in multiple frequency bands, identifying the target communication protocol frequency band used by the fire door alarm device, and obtaining spectrum parameter information for each frequency band to form a set of communication frequency bands to be analyzed; based on a pre-established environmental wireless device frequency band database, extracting the operating frequency bands used by surrounding wireless devices. The present invention constructs spectrum mapping through spectrum scanning and environmental database, combines support vector machines to achieve intelligent interference prediction, and dynamically adjusts coding parameters based on interference intensity, thereby improving the transmission reliability of key alarm information in high-interference environments. The system has feedforward identification and closed-loop control capabilities, effectively solving the problems of slow response and unstable transmission in traditional systems.

Description

Technical Field

[0001] The present invention relates to the technical field of fire alarm technology, and in particular to a fire door alarm information transmission device and method based on the Internet of Things. Background Art

[0002] The IoT-based fire door alarm information transmission device is a fire emergency response system that integrates sensor monitoring, intelligent identification, and remote communication capabilities. By deploying sensing units such as temperature sensors, smoke detectors, and door status monitors (such as open / close sensors) on fire doors, the device collects real-time environmental data and determines the fire risk level. Once the set fire alarm threshold is triggered, the system automatically closes the fire door and simultaneously uploads the alarm information to a cloud platform or fire management center via a built-in wireless communication module (such as NB-IoT, LoRa, or 4G), enabling remote alarm and situation monitoring. This device can further integrate intelligent algorithms to predict and warn of fire spread trends, significantly improving fire prevention and control capabilities and response efficiency within buildings.

[0003] The existing technology has the following deficiencies:

[0004] Existing IoT fire door alarm systems typically use low-power wide-area communication protocols such as NB-IoT and LoRa for remote alarms. However, when deployed in buildings densely populated with wireless devices, such as elevator control systems, industrial WLANs, and wireless access control systems, where frequency bands overlap significantly, these devices are susceptible to cross-band electromagnetic coupling interference. During peak interference periods, the system communication link may experience transient packet loss or transmission delays, preventing timely upload of alarm information. If such interference coincides with a rapidly spreading fire, the system will be unable to quickly issue an alarm and initiate the closing of the fire doors.

[0005] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not form the prior art that is already known to a person of ordinary skill in the art. Summary of the Invention

[0006] The purpose of the present invention is to provide a fire door alarm information transmission device and method based on the Internet of Things. By introducing an interference frequency band identification mechanism combining spectrum scanning with an environmental database, a spectrum mapping relationship between the communication frequency band and the interference source frequency band is constructed, and a support vector machine model is combined to perform high-precision interference level prediction on the characteristic indicator vector group. When the electromagnetic coupling risk is detected, the coding parameter adaptive adjustment mechanism is automatically triggered, and the error correction capability is dynamically improved according to the interference intensity, ensuring that key alarm information can still be stably and reliably uploaded and drive linkage control operations in a high-interference environment, so as to solve the problems in the above-mentioned background technology.

[0007] In order to achieve the above object, the present invention provides the following technical solution: a method for transmitting fire door alarm information based on the Internet of Things, comprising the following steps:

[0008] After the fire emergency mode is activated, the wireless communication environment in the area where the fire door alarm device is deployed is continuously scanned in multiple frequency bands to identify the target communication protocol frequency band used by the fire door alarm device and obtain the spectrum parameter information of each frequency band to form a set of communication frequency bands to be analyzed;

[0009] Based on a pre-established database of ambient wireless device frequency bands, the system extracts the operating frequency bands used by surrounding wireless devices, identifies interference frequency bands that are adjacent to the target communication frequency band, and constructs a spectrum mapping matrix between the communication frequency band and the interference source frequency band.

[0010] Based on the spectrum mapping matrix, the feature engineering method is used to extract the characteristic indicators reflecting the coupling characteristics between the communication frequency band and the interference frequency band. The characteristic indicator vector group used for electromagnetic coupling interference judgment is constructed through the analyzed characteristic indicators.

[0011] The characteristic indicator vector group is input into a support vector machine model that has been pre-trained with historical interference data. The model outputs the current coupling interference level indicator, and intelligently predicts the electromagnetic coupling interference risk between the communication frequency band and the interference frequency band. It is also used to determine whether the communication link is currently in a high electromagnetic coupling interference state.

[0012] When the support vector machine model prediction results indicate the presence of high electromagnetic coupling interference, the coding parameter adaptive control mechanism is triggered, and the redundant coding strength is dynamically adjusted according to the intensity of electromagnetic coupling interference. During the peak period of coupling interference intensity, a transmission format with a higher error correction coding level is used for communication.

[0013] Preferably, the specific steps of performing multi-band continuous scanning on the wireless communication environment of the fire door alarm device deployment area to identify the target communication protocol frequency band used are as follows:

[0014] In the initial stage after the fire door alarm device is activated, the scanning range is set by the spectrum analyzer connected to the device to cover the frequency band used by the communication protocol;

[0015] Configure scanning parameters including scanning step size, dwell time and power threshold, start spectrum continuous scanning function and record real-time power spectrum density graph;

[0016] Analyze the collected spectrum data and identify the current active frequency band based on power peak, modulation characteristics or spectrum stability;

[0017] Frequency band attribution matching is performed based on the communication protocol characteristics to determine the target communication protocol frequency band used by the fire door alarm device for subsequent interference analysis and frequency band mapping.

[0018] Preferably, the interfering frequency band that is adjacent to the target communication frequency band is identified, and the specific steps are as follows:

[0019] Based on a pre-established database of ambient wireless device frequency bands, the operating frequency band parameters of known wireless devices in the fire door alarm system deployment area are extracted, including center frequency, bandwidth range, and operating mode.

[0020] Compare the identified target communication protocol frequency band, use its center frequency as a reference, and set the frequency adjacent judgment threshold as the interference judgment window;

[0021] Calculate the center frequency difference and bandwidth overlap ratio of all surrounding device frequency bands to screen out wireless device frequency bands whose center frequencies partially or completely overlap with the target frequency band;

[0022] The filtered frequency bands are recorded as potential interference source frequency bands, and their corresponding device types, interference weights, and overlapping properties are marked. They are used as the interference frequency band set input for constructing the spectrum mapping matrix for subsequent electromagnetic coupling analysis and interference intensity modeling.

[0023] Preferably, a feature engineering method is used to extract characteristic indicators reflecting the coupling characteristics between the communication frequency band and the interference frequency band, wherein the extracted indicators include the noise distribution asymmetry of the target frequency band in the equidistant frequency domains on the left and right sides and the dynamic overlapping ratio of the interference frequency band and the target frequency band in multiple time periods. Under the analysis window, after analyzing the extracted characteristic indicators, a carrier-noise asymmetry factor and a dynamic overlap factor are generated respectively, and a characteristic indicator vector group for electromagnetic coupling interference judgment is constructed through the carrier-noise asymmetry factor and the dynamic overlap factor.

[0024] Preferably, a characteristic index vector group consisting of a carrier-noise asymmetry factor and a dynamic overlap factor is input into a support vector machine model that has been pre-trained with historical interference data. The electromagnetic coupling coefficient is output by the model, and the electromagnetic coupling interference risk between the communication frequency band and the interference frequency band is intelligently predicted based on the electromagnetic coupling coefficient, and is used to determine whether the communication link is currently in a high electromagnetic coupling interference state.

[0025] Preferably, the electromagnetic coupling coefficient generated when the electromagnetic coupling interference risk between the communication frequency band and the interference frequency band is intelligently predicted by the support vector machine model is compared with a preset electromagnetic coupling risk threshold to determine whether the communication link is currently in a high electromagnetic coupling interference state. The specific determination steps are as follows:

[0026] If the electromagnetic coupling coefficient is greater than the electromagnetic coupling risk threshold, a risk signal is generated, and it is determined that the current communication link is in a high electromagnetic coupling interference state; if the electromagnetic coupling coefficient is less than or equal to the electromagnetic coupling risk threshold, a normal signal is generated, and it is determined that the current communication link is in a low-risk interference state and is communicating efficiently.

[0027] Preferably, in the analysis window, the specific steps of generating the carrier-noise asymmetry factor after analyzing the noise distribution asymmetry in the frequency domains equidistant on the left and right sides of the target frequency band are as follows:

[0028] Taking the center frequency of the target communication frequency band as the reference, set a frequency offset distance on both sides of it, determine the symmetrical sampling points, and randomly select a bandwidth sub-band area near each symmetrical point to calculate the noise power spectrum density integrated energy. The calculation expression is: ,in: represents the noise power spectral density at frequency f; is the sub-band width; and It is a left-right symmetrical frequency point defined relative to the center frequency of the target communication frequency band. 、 , is the center frequency of the target communication frequency band; and Respectively represent the noise energy of the left and right sub-bands, reflecting the interference intensity;

[0029] Obtaining the noise energy on the left and right sides and Then, the carrier-noise asymmetry factor is calculated using the hyperbolic function. The calculation expression is: ,in: is the carrier-noise asymmetry factor, which characterizes the noise power asymmetry at equidistant frequency domain positions on the left and right sides of the target communication frequency band; is the trend amplification coefficient, which is used to adjust the response sensitivity of the carrier-noise asymmetry factor to the asymmetry degree.

[0030] Preferably, in the analysis window, the specific steps of generating the dynamic overlap factor after analyzing the dynamic overlap ratio of the interference frequency band and the target frequency band in multiple time periods are as follows:

[0031] The analysis window is divided into N equally spaced sub-windows. For the i-th sub-window, the interference frequency band is defined as , the target communication frequency band is a fixed value , calculate the actual occupancy ratio of the interference frequency band to the target frequency band in each sub-window, and the calculation expression is: ,in: represents the actual occupancy ratio of the interference frequency band to the target frequency band in the i-th sub-window; and The upper and lower boundaries of the target communication frequency band are fixed; and are the upper and lower boundary frequencies of the interference frequency band in the i-th sub-window; Indicates the length of the frequency overlap interval used to calculate the interference frequency band and the target communication frequency band in the i-th subwindow;

[0032] The trend shift function is defined to characterize the disturbance difference enhancement value between the current sub-window and its adjacent sub-windows, and the dynamic overlap factor is calculated by the trend shift function. The calculation expression is: ,in: is the dynamic overlap factor, which is used to measure the frequency overlap strength between the interference frequency band and the target communication frequency band in multiple sub-windows; is the trend shift enhancement function of the i-th sub-window, which is used to asymmetrically amplify the mid-segment shift; is the trend sensitivity index, which controls the degree of response to mutations; and It represents the actual occupancy ratio of the interference frequency bands on both sides of the i-th sub-window to the target frequency band, which is used to construct the local trend response.

[0033] Preferably, when the communication link is in a high electromagnetic coupling interference state, the redundant coding strength is dynamically adjusted according to the electromagnetic coupling interference strength. The specific steps are as follows:

[0034] When it is detected that the electromagnetic coupling coefficient is greater than the electromagnetic coupling risk threshold, the system enters the activation state and calculates a coding gain factor for regulating the coding strength. The specific calculation expression is: ,in: is the electromagnetic coupling coefficient currently predicted by the support vector machine model, which represents the coupling strength between the target communication frequency band and the interference frequency band; is the electromagnetic coupling risk threshold, when Determined to be in a high electromagnetic coupling interference state; is the coding gain factor, which is used to adjust the redundancy ratio and coding level in the future; It is the basic coupling amplitude response factor, which determines the impact of exceeding the electromagnetic coupling risk threshold amplitude on the redundancy strength; is the steepness adjustment coefficient of the nonlinear activation curve, which is used to amplify the response of the electromagnetic coupling degree in the high value range; is the time derivative adjustment coefficient, which controls the weighted proportion of the electromagnetic coupling change trend; It is the time derivative of the electromagnetic coupling coefficient, reflecting the speed of interference rise or fall;

[0035] When the coding gain factor is obtained Then, the error correction coding configuration is dynamically adjusted according to the coding gain factor, including the redundancy ratio and coding level, to form a new coding strategy parameter group. The adjustment formula is as follows: ,in: represents the redundancy ratio; It is the system default basic redundancy ratio; is the error correction coding level; is the default encoding level; and is the gain mapping coefficient, which is used to control the sensitivity of redundancy enhancement and the magnitude of coding order improvement; The updated encoding parameter configuration set is applied to the communication unit encoder layer.

[0036] An Internet of Things-based fire door alarm information transmission device includes a communication frequency band perception module, an interference frequency band identification module, a characteristic index construction module, an interference risk intelligent judgment module, and a coding parameter adaptive control module;

[0037] The communication frequency band sensing module, after the fire emergency mode is activated, continuously scans the wireless communication environment in the area where the fire door alarm device is deployed across multiple frequency bands, identifies the target communication protocol frequency band used by the fire door alarm device, and obtains spectrum parameter information for each frequency band to form a set of communication frequency bands to be analyzed;

[0038] The interference frequency band identification module extracts the operating frequency bands used by surrounding wireless devices based on a pre-established database of ambient wireless device frequency bands, identifies interference frequency bands that are adjacent to the target communication frequency band, and constructs a spectrum mapping matrix between the communication frequency band and the interference source frequency band;

[0039] The characteristic index construction module uses feature engineering methods to extract characteristic indicators reflecting the coupling characteristics between the communication frequency band and the interference frequency band based on the spectrum mapping matrix. The characteristic index vector group used for electromagnetic coupling interference judgment is constructed based on the analyzed characteristic indicators.

[0040] The intelligent interference risk identification module inputs the characteristic index vector group into a support vector machine model pre-trained with historical interference data. The model outputs the current coupling interference level indicator, intelligently predicts the electromagnetic coupling interference risk between the communication frequency band and the interference frequency band, and is used to determine whether the communication link is currently in a high electromagnetic coupling interference state.

[0041] The coding parameter adaptive control module triggers the coding parameter adaptive control mechanism when the support vector machine model prediction results indicate the presence of high electromagnetic coupling interference. The redundant coding strength is dynamically adjusted according to the intensity of electromagnetic coupling interference. During the peak period of coupling interference intensity, a transmission format with a higher error correction coding level is used for communication.

[0042] In the above technical solution, the technical effects and advantages provided by the present invention are:

[0043] The present invention introduces an interference frequency band identification mechanism that combines spectrum scanning with an environmental database, constructs a spectrum mapping relationship between the communication frequency band and the interference source frequency band, and combines the support vector machine model to perform high-precision interference level prediction on the characteristic index vector group. When the electromagnetic coupling risk is detected, the coding parameter adaptive adjustment mechanism is automatically triggered, and the error correction capability is dynamically improved according to the interference intensity, ensuring that key alarm information can still be stably and reliably uploaded and drive linkage control operations in high-interference environments. Compared with existing static coding configurations or single anti-interference strategies, this method has the overall capabilities of feedforward identification and closed-loop control, and effectively solves the core problems of traditional fire communication in interference-intensive scenarios, such as "untimely identification, inaccurate response, and unstable transmission." BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction to the drawings required for use in the embodiments will be given below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0045] Figure 1 This is a flow chart of a method for transmitting fire door alarm information based on the Internet of Things in the present invention.

[0046] Figure 2 This is a module schematic diagram of a fire door alarm information transmission device based on the Internet of Things of the present invention. DETAILED DESCRIPTION

[0047] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, these example embodiments are provided so that the description of this disclosure will be thorough and complete and will fully convey the concepts of the example embodiments to those skilled in the art.

[0048] The present invention provides Figure 1 The method for transmitting fire door alarm information based on the Internet of Things includes the following steps:

[0049] After the fire emergency mode is activated, a spectrum analyzer is used to continuously scan the wireless communication environment in the area where the fire door alarm device is deployed in multiple frequency bands to identify the target communication protocol frequency band used by the fire door alarm device. Spectral parameter information such as the center frequency, bandwidth, and power spectrum density of each frequency band is obtained to form a set of communication frequency bands to be analyzed.

[0050] The specific steps for using a spectrum analyzer to continuously scan the wireless communication environment in the area where the fire door alarm device is deployed to identify the target communication protocol frequency band are as follows: First, in the initial stage after the fire door alarm device is activated, the spectrum analyzer connected to the device is used to set the scanning range to cover the frequency band range used by typical communication protocols (such as NB-IoT, LoRa, etc.) (such as 430 MHz–470 MHz, 780 MHz–960 MHz, etc.); second, the scanning parameters including the scanning step size, dwell time, and power threshold are configured, the spectrum continuous scanning function is enabled, and the real-time power spectrum density plot is recorded; then, the collected spectrum data is analyzed, and the current active frequency band is identified based on power peak, modulation characteristics, or spectrum stability; finally, the frequency band attribution is matched based on the communication protocol characteristics (such as the spreading factor bandwidth of LoRa and the subcarrier allocation of NB-IoT) to determine the target communication protocol frequency band used by the fire door alarm device for subsequent interference analysis and frequency band mapping.

[0051] This step provides spectrum data and interference risk modeling support for subsequent electromagnetic coupling interference identification and adaptive communication control. It serves as the sensing entry point and data source for the entire interference prediction and control mechanism. During fire emergency response mode, the wireless channel environment becomes instantly complex and highly dynamic due to rapidly rising ambient temperatures, frequent human activity, and the simultaneous activation of alarm communication functions by multiple devices. This makes it highly susceptible to communication band conflicts, signal blockage, and interference overlap. Continuously scanning the wireless communication environment across multiple frequency bands using a spectrum analyzer provides comprehensive, real-time visibility into the wireless energy distribution within the area, accurately identifying the target communication protocol band currently being used by the fire door alarm. Further extraction of key spectrum parameters such as center frequency, bandwidth, and power spectral density not only helps characterize the communication link usage but also provides the necessary raw input for subsequent frequency adjacency assessment with interference source bands, bandwidth overlap calculation, and power interference trend analysis. By forming a set of communication frequency bands to be analyzed, interference identification accuracy and response speed are effectively improved, ensuring the system's ability to accurately identify and rapidly control even in high-interference environments. This ensures that alarm information is promptly and reliably uploaded to the backend system and linked to control devices during a fire.

[0052] Based on a pre-established database of ambient wireless device frequency bands, the system extracts the operating frequency bands used by surrounding wireless devices such as elevator control systems, industrial WLANs, and security wireless units. It then identifies interference frequency bands adjacent to the target communication frequency bands and constructs a spectrum mapping matrix between the communication frequency bands and the interference source frequency bands.

[0053] Identifying interfering frequency bands that are frequency-adjacent to the target communication frequency band involves the following steps: First, based on a pre-established database of ambient wireless device frequency bands, the operating frequency band parameters of known wireless devices (such as elevator systems, industrial WLANs, wireless access control systems, and security cameras) within the fire door alarm deployment area are extracted, including center frequency, bandwidth range, and operating mode. Second, the identified target communication protocol frequency band (such as NB-IoT or LoRa) is compared with its center frequency, and a frequency adjacency threshold (e.g., ±20 MHz or ±10% bandwidth) is set as the interference determination window. Center frequency differences and bandwidth overlap ratio analysis are then performed on all surrounding device frequency bands to identify wireless device frequency bands whose center frequencies partially or completely overlap with the target frequency band (if the bandwidth overlap exceeds 20%, it is considered to have partial bandwidth overlap). Finally, these frequency bands are recorded as potential interference source bands, annotated with their corresponding device type, interference weight, and overlap properties, and used as the interference frequency band set input for constructing the spectrum mapping matrix for subsequent electromagnetic coupling analysis and interference intensity modeling.

[0054] This process ensures that the potential coupling interference risks caused by spectrum proximity can be fully captured, improving the refinement and completeness of spectrum resource perception.

[0055] This step primarily builds an interference correlation model within the IoT communication environment, providing accurate and structured spectrum data support for the subsequent identification, prediction, and communication control of electromagnetic coupling interference. In real-world building environments, fire door alarms are not the only wireless devices operating. The communication protocols they employ, such as NB-IoT and LoRa, often operate in adjacent or partially overlapping frequency bands with devices such as elevator control systems, industrial WLANs, security cameras, and wireless access control systems. Since these devices may collectively activate during a fire, multiple high-power signals in the same physical space can be activated simultaneously. This can easily cause coupling effects such as cross-modulation interference, bandwidth leakage, and frequency suppression between adjacent frequency bands, impacting the real-time transmission of critical alarm data. Therefore, this step systematically extracts the communication frequency bands of all active devices in the area by invoking a pre-established database of ambient wireless device frequency bands. It then performs frequency spacing analysis and bandwidth overlap determination with the target communication frequency band to identify frequency bands with potential interference risks. This then constructs a spectrum mapping matrix with a "communication frequency band → interference source frequency band" relationship, accurately characterizing the impact range and interference paths of interference sources on the target frequency band. This matrix serves as the core input for subsequent electromagnetic coupling modeling, interference intensity prediction, and redundant coding strategy adjustment. It plays a key role in bridging the spectrum structure and regulation mechanism, significantly improving the system's ability to perceive and respond to communication risks in multi-source interference scenarios.

[0056] Based on the spectrum mapping matrix, the feature engineering method is used to extract the characteristic indicators reflecting the coupling characteristics between the communication frequency band and the interference frequency band. The characteristic indicator vector group used for electromagnetic coupling interference judgment is constructed through the analyzed characteristic indicators.

[0057] A feature engineering method is used to extract characteristic indicators reflecting the coupling characteristics between the communication frequency band and the interference frequency band. The extracted indicators include the noise distribution asymmetry of the target frequency band in the equidistant frequency domains on the left and right sides and the dynamic overlap ratio of the interference frequency band and the target frequency band in multiple time periods. After analyzing the extracted characteristic indicators in the analysis window, the carrier-noise asymmetry factor and the dynamic overlap factor are generated respectively. The carrier-noise asymmetry factor and the dynamic overlap factor are used to construct a characteristic indicator vector group for electromagnetic coupling interference judgment.

[0058] The greater the asymmetry of the noise distribution between equally spaced left and right sides of the target frequency band, the higher the risk of electromagnetic coupling interference from one of the interfering frequency bands. Ideally, an interference-free spectrum environment should exhibit a nearly symmetrical noise floor distribution, meaning that the background noise power at equally spaced locations on both sides of the target frequency band should be roughly the same. However, when a high-power transmitter (such as an industrial WLAN or elevator wireless unit) is present in an adjacent frequency band, its spectral sideband leakage or modulation outward spread energy will penetrate toward the target frequency band, significantly increasing the noise power on that side and forming a carrier-to-noise asymmetry phenomenon. This asymmetry not only reflects energy offset in the frequency domain but also suggests that modulation crosstalk or frequency modulation drift at the physical layer may be acting across the target communication channel, potentially leading to increased bit error rates, packet loss, or transmission delays. Therefore, noise distribution asymmetry is a sensitive and directionally discriminative characteristic, serving as an important signal feature for assessing the strength of electromagnetic coupling interference.

[0059] In the analysis window, the specific steps for generating the carrier-noise asymmetry factor after analyzing the noise distribution asymmetry in the frequency domains equidistant on the left and right sides of the target frequency band are as follows:

[0060] Taking the center frequency of the target communication frequency band as the reference, set a frequency offset distance on both sides of it, determine the symmetrical sampling points, and randomly select a bandwidth sub-band area near each symmetrical point to calculate the noise power spectrum density integrated energy. The calculation expression is: ,in: represents the noise power spectral density at frequency f; is the sub-band width, which determines the size of the integration area; and It is a left-right symmetrical frequency point defined relative to the center frequency of the target communication frequency band. 、 , is the center frequency of the target communication frequency band; and Respectively represent the noise energy of the left and right sub-bands, reflecting the interference intensity;

[0061] The purpose of this step is to establish a preliminary directional perception basis for spectrum interference through the integrated energy difference in the left and right symmetrical regions, and to provide energy input characteristics for asymmetric modeling.

[0062] Obtaining the noise energy on the left and right sides and Finally, the carrier-noise asymmetry factor is calculated using a hyperbolic function to enhance the ability to perceive small differences while ensuring the continuity and boundedness of the carrier-noise asymmetry factor. The calculation expression of the carrier-noise asymmetry factor is: ,in: is the carrier-noise asymmetry factor, which characterizes the noise power asymmetry at equidistant frequency domain positions on the left and right sides of the target communication frequency band; is the trend amplification coefficient, which is used to adjust the response sensitivity of the carrier-noise asymmetry factor to the asymmetry degree; Hyperbolic tangent function, compressing the output value to , the closer the output value is to 1, the higher the degree of asymmetry;

[0063] The purpose of this step is to convert the left and right noise energy difference into a quantifiable and directionally clear carrier-noise asymmetry factor, providing a quantitative basis for the intelligent model to judge the communication interference trend and adjust the transmission strategy.

[0064] The carrier-to-noise asymmetry factor (CNAF) shows that, within the analysis window, the CNAF, generated by analyzing the asymmetry of the noise distribution in equally spaced frequency domains on the left and right sides of the target frequency band, indicates that a larger CNAF value indicates a higher risk of electromagnetic coupling interference between the communication and interference bands. Conversely, a smaller CNAF value indicates a lower risk of electromagnetic coupling interference between the communication and interference bands. This is because electromagnetic coupling interference typically propagates to adjacent frequency bands in the form of energy leakage or modulation crosstalk, manifesting as an abnormal increase in localized frequency-domain noise levels. This effect exhibits a pronounced spectral directionality, especially when a high-power interference source is present in only one frequency band. The CNAF quantifies this directional shift by performing a nonlinear, normalized comparison of the integrated noise energy on the left and right sides. Therefore, a larger CNAF value indicates more concentrated and intrusive interference, and a more significant impact on the communication link.

[0065] A high dynamic overlap ratio between the target frequency band and the interfering frequency band over multiple time periods generally indicates a greater risk of electromagnetic coupling interference between the communication and interfering frequency bands. This is because in complex wireless environments, interference sources often exhibit non-persistent transmission behavior, such as bursty transmissions, periodic beam switching, or frequency drift caused by temperature drift. If their spectral energy repeatedly enters the target communication frequency band or its sidebands over multiple time periods, this indicates a combination of stability and persistence. This "high-frequency dynamic overlap" significantly increases modulation crosstalk, frame error rates, and errored packet retransmission probability between signals. This is particularly true for low-power protocols such as LoRa or NB-IoT, whose link reliability is highly sensitive to spectrum cleanliness. Therefore, a higher overlap ratio indicates a greater risk of electromagnetic coupling interference. The dynamic overlap ratio not only captures frequency domain overlap but also incorporates temporal trends. It is an important precursor parameter for identifying the evolution of weak coupling into strong coupling and has a high discriminant value in the spectrum mapping matrix.

[0066] In the analysis window, the specific steps for generating a dynamic overlap factor after analyzing the dynamic overlap ratio between the interference frequency band and the target frequency band in multiple time periods are as follows:

[0067] The analysis window is divided into N equally spaced sub-windows (e.g. one spectrum snapshot per second). For the i-th sub-window, the interference frequency band is defined as , the target communication frequency band is a fixed value , calculate the actual occupancy ratio of the interference frequency band to the target frequency band in each sub-window, and the calculation expression is: ,in: represents the actual occupancy ratio of the interference frequency band to the target frequency band in the i-th sub-window; and The upper and lower boundaries of the target communication frequency band are fixed; and is the upper and lower boundary frequencies of the interference frequency band in the i-th sub-window, which may change dynamically due to frequency modulation or drift; Indicates the length of the frequency overlap interval used to calculate the interference frequency band and the target communication frequency band in the i-th subwindow;

[0068] The purpose of this step is to construct a multi-period overlapping rate series , forming a perception of the temporal distribution of coupling strength, which is used for the subsequent construction of dynamic overlap factors.

[0069] In order to identify whether the interference has a trend of "stable suppression" or "periodic intrusion" in multiple sub-windows, a trend shift function is defined to characterize the disturbance difference enhancement value between the current sub-window and its adjacent sub-windows, and the dynamic overlap factor is calculated by the trend shift function. The calculation expression is: ,in: is the dynamic overlap factor, which is used to measure the frequency overlap strength between the interference frequency band and the target communication frequency band in multiple sub-windows; is the trend shift enhancement function of the i-th sub-window, which is used to asymmetrically amplify the mid-segment shift; is the trend sensitivity index, which controls the degree of response to sudden changes (e.g., a value of 1.5 or 2); and It represents the actual occupation ratio of the interference frequency bands on both sides of the i-th sub-window to the target frequency band, which is used to construct the local trend response; this construction uses the product form to suppress the “stationary sequence” internally (if , the dynamic overlap factor is close to zero), and the “local mutation” is exponentially enhanced (such as burst suppression and short-term drift).

[0070] The purpose of this step is to comprehensively evaluate the nonlinear coupling trend of the interference frequency band on the time axis and enhance the sensitivity to irregular intrusion behavior through the asymmetric structure. The larger the dynamic overlap factor, the more frequent discontinuous and unbalanced suppression behavior in the interference frequency band, posing a substantial threat to the stability of the communication link.

[0071] The dynamic overlap factor, generated by analyzing the dynamic overlap ratio between the interfering and target frequency bands over multiple time periods within the analysis window, indicates that a higher value indicates a higher risk of electromagnetic coupling interference between the communication and interfering frequency bands. Conversely, a lower value indicates a lower risk of electromagnetic coupling interference between the communication and interfering frequency bands. This is because the dynamic overlap factor quantifies the dynamic coupling behavior between the interfering and target frequency bands by analyzing the trend, concentration, and disturbance enhancement of the frequency overlap ratio sequence over multiple time periods within the analysis window. A high dynamic overlap factor indicates that the interfering frequency band frequently and intensely intrudes into or approaches the target frequency band over time. Its overlapping behavior is no longer random or sporadic, but rather exhibits a stable or periodic intrusion trend. This type of coupling often leads to serious consequences such as modulation crosstalk, frame loss, and communication link instability, resulting in a high-risk interference state. Conversely, a low dynamic overlap factor indicates that the interfering frequency band only occasionally or briefly approaches the target frequency band, with sparse overlap and low disturbance intensity. The system can maintain normal communication through conventional error correction mechanisms, thus resulting in a lower risk of electromagnetic coupling interference. The size of the dynamic overlap factor essentially reflects the combined impact of spatial frequency domain overlap and time change trend, and is therefore an effective indicator for judging the strength of coupling interference risk.

[0072] The characteristic indicator vector group is input into a support vector machine model that has been pre-trained with historical interference data. The model outputs the current coupling interference level indicator, and intelligently predicts the electromagnetic coupling interference risk between the communication frequency band and the interference frequency band. It is also used to determine whether the communication link is currently in a high electromagnetic coupling interference state.

[0073] The characteristic indicator vector group composed of the carrier-noise asymmetry factor and the dynamic overlap factor is input into the support vector machine model that has been pre-trained with historical interference data. The electromagnetic coupling coefficient is output by the model. Based on the electromagnetic coupling coefficient, the electromagnetic coupling interference risk between the communication frequency band and the interference frequency band is intelligently predicted, and it is used to determine whether the communication link is currently in a high electromagnetic coupling interference state.

[0074] A support vector machine (SVM) model pre-trained with historical interference data is trained using a large amount of labeled wireless communication interference sample data before system deployment or during the initialization of communication security policies. This allows the SVM to automatically determine the electromagnetic coupling interference risk between the communication and interference frequency bands. In this context, the SVM model's training input is a set of feature indicator vectors consisting of a "carrier-noise asymmetry factor" and a "dynamic overlap factor," and the training output is the interference level label corresponding to each vector (e.g., "no interference," "mild coupling interference," "moderate coupling interference," "heavy coupling interference," etc.). As a classic two- or multi-class classification algorithm, the basic idea of ​​a support vector machine is to find an optimal hyperplane in the feature space that optimally separates each class of samples on either side of the plane. Because electromagnetic interference data often exhibits nonlinear distribution characteristics, the SVM model uses a kernel function (such as an RBF kernel or a polynomial kernel) to map the original feature space into a high-dimensional space, achieving linear separability in the high-dimensional feature space. This output then accurately reflects the degree of coupling risk and serves as a discriminant.

[0075] During the training process, a historical sample library containing various scenarios, multiple communication protocol frequency bands, and various interference sources must first be constructed. Each sample contains two core characteristic parameters: the carrier-to-noise asymmetry factor and the dynamic overlap factor. The carrier-to-noise asymmetry factor measures the difference in noise power between the left and right adjacent frequency regions of the target communication band, that is, the degree of noise asymmetry between the left and right frequency bands. If the communication band is continuously interfered with by one frequency band, its noise environment will exhibit significant asymmetry, and the carrier-to-noise asymmetry factor will significantly deviate from zero. This characteristic can indicate the presence of directional coupling interference, such as the unilateral leakage suppression effect of elevator wireless modules or access control systems on the communication band in a specific direction. The dynamic overlap factor describes the dynamic rate of spectral overlap between the interference band and the communication band over time, that is, the probability of the interference band repeatedly invading the communication band over multiple time periods. A high dynamic overlap factor indicates that the interference source is not sporadic but rather periodic, indicating a risk of temporally stable coupling. These two parameters together construct a two-dimensional feature space, in which each sample point represents the coupling state of a certain frequency band pair at a certain moment in a specific environment.

[0076] After these feature samples are fed into a support vector machine for training, the machine learns how to delineate the boundaries between different levels of electromagnetic coupling in the two-dimensional carrier-to-noise asymmetry factor-dynamic overlap factor space. Once trained, the model is ready for real-time prediction: the system feeds the currently acquired carrier-to-noise asymmetry factor and dynamic overlap factor into the trained model as input vectors. The model then outputs a normalized electromagnetic coupling coefficient, which can be set between 0 and 1. This coefficient represents the current coupling strength between the communication band and the interference band.

[0077] The biggest advantages of this mechanism are:

[0078] Ability to structure dynamic and time-varying interference behavior data into learnable patterns;

[0079] The classification capability of support vector machines is used to quickly and accurately judge the multi-dimensional interference situation;

[0080] Leverage physical layer characteristics (such as carrier-noise asymmetry factor and dynamic overlap factor) to establish an early prediction model for communication link quality degradation.

[0081] Implementing closed-loop control logic from interference detection to interference mitigation and regulation enhances the system's communication robustness in high-risk situations like fire. Therefore, the "support vector machine model pre-trained with historical interference data" not only serves as the core engine for intelligent prediction, but also serves as the key to intelligently sensing and controlling the entire process from "electromagnetic coupling risk to regulation and response." This model demonstrates strong adaptability and generalization in engineering practice, making it suitable for IoT secure communication systems across different frequency bands, protocols, and deployment scenarios.

[0082] The electromagnetic coupling coefficient generated by the intelligent prediction of electromagnetic coupling interference risk between the communication frequency band and the interference frequency band using the support vector machine model is compared with a preset electromagnetic coupling risk threshold (for example, 0.7 or 0.8) to determine whether the communication link is currently in a high electromagnetic coupling interference state. The specific determination steps are as follows:

[0083] If the electromagnetic coupling coefficient is greater than the electromagnetic coupling risk threshold, a risk signal is generated, and it is determined that the current communication link is in a high electromagnetic coupling interference state; if the electromagnetic coupling coefficient is less than or equal to the electromagnetic coupling risk threshold, a normal signal is generated, and it is determined that the current communication link is in a low-risk interference state and is communicating efficiently.

[0084] When the support vector machine model prediction results indicate the presence of high electromagnetic coupling interference, the coding parameter adaptive control mechanism is triggered to dynamically adjust the redundant coding strength according to the electromagnetic coupling interference intensity. During the peak period of coupling interference intensity, a transmission format with a higher error correction coding level is used to improve data transmission reliability.

[0085] When the communication link is in a high electromagnetic coupling interference state, the redundant coding strength is dynamically adjusted according to the electromagnetic coupling interference intensity. The specific steps are as follows:

[0086] When the electromagnetic coupling coefficient is detected to be greater than the electromagnetic coupling risk threshold, the system enters the activation state and calculates a coding gain factor for regulating the coding strength. The coding gain factor uses the magnitude of the coupling strength exceeding the threshold as the core driving factor and introduces a nonlinear activation modulation function for calculation in combination with the time variation trend. The specific calculation expression is: ,in: is the electromagnetic coupling coefficient currently predicted by the support vector machine model, which represents the coupling strength between the target communication frequency band and the interference frequency band; is the electromagnetic coupling risk threshold, when Determined to be in a high electromagnetic coupling interference state; is the coding gain factor, which is used to adjust the redundancy ratio and coding level in the future; It is the basic coupling amplitude response factor, which determines the impact of exceeding the electromagnetic coupling risk threshold amplitude on the redundancy strength; is the steepness adjustment coefficient of the nonlinear activation curve, which is used to amplify the response of the electromagnetic coupling degree in the high value range; is the time derivative adjustment coefficient, which controls the weighted proportion of the electromagnetic coupling change trend (such as interference surge); It is the time derivative of the electromagnetic coupling coefficient, reflecting the speed of interference rise or fall;

[0087] This step combines the static coupling strength and the interference variation trend to generate a quantifiable and continuously adjustable coding gain factor. , serving as the core driver for subsequent encoding parameter updates. By introducing a nonlinear activation function (in the form of Softplus), the control mechanism is made insensitive to small fluctuations and responds quickly to large disturbances, enhancing system robustness.

[0088] When the coding gain factor is obtained Then, the error correction coding configuration is dynamically adjusted according to the coding gain factor, including the redundancy ratio and coding level, to form a new coding strategy parameter group. The adjustment formula is as follows: ,in: Indicates the redundancy ratio (ratio of redundant bits to valid bits). Increasing this value can improve fault tolerance. The default basic redundancy ratio of the system is usually 0.1~0.3; is the error correction coding level, which is usually mapped to the LDPC code depth, convolutional code constraint length or Turbo code structure complexity; It is the default coding level, indicating the lowest fault-tolerant configuration of the system in a non-interference state; and is the gain mapping coefficient, which is used to control the sensitivity of redundancy enhancement and the magnitude of coding order improvement; The updated encoding parameter configuration set is applied to the communication unit encoder layer;

[0089] This step converts the coding gain factor into actual communication parameter adjustment actions, forming a nonlinear adaptive coding control mechanism. This ensures that when the interference intensity is higher and the changes are more drastic, a stronger error correction strategy (such as a higher LDPC level and a larger convolution kernel) is adopted, thereby improving the system's data integrity and transmission reliability under high-coupling interference conditions.

[0090] The core function of this step is to build an intelligent, adaptive anti-interference mechanism based on dynamic interference intensity identification to ensure the transmission reliability and system stability of IoT communication links in high electromagnetic coupling interference environments. When the support vector machine model prediction results indicate a high electromagnetic coupling risk between the current communication frequency band and the interference frequency band, it indicates that the communication link is facing strong interference caused by factors such as frequency proximity, bandwidth overlap, or energy leakage. Continuing to maintain the existing low redundancy and low error correction level configuration at this time will easily lead to serious problems such as communication packet loss, increased bit flip rate, and control command lag. This can lead to consequences such as the inability to upload fire warning information in a timely manner and the failure of fire door linkage response. Therefore, by triggering the adaptive coding parameter control mechanism to sense and respond to changes in electromagnetic interference intensity in real time and dynamically adjust the redundancy coding strength and error correction coding level, it can not only effectively improve error recovery capabilities and reduce the retransmission probability, but also extend the effective communication window and improve channel utilization efficiency. When the interference intensity reaches its peak, the system automatically switches to a higher-level error correction format (such as enhanced LDPC code, deep convolutional code, etc.) to further enhance the fault tolerance to instantaneous signal pollution and data distortion, ensuring that critical alarm data can still be transmitted to the target node stably, accurately and securely in extreme electromagnetic environments. Realizing closed-loop robust communication control from "interference identification" to "control response" is a key step to ensure the reliable operation of the intelligent fire protection system.

[0091] This IoT-based fire door alarm information transmission method enables real-time perception, intelligent identification, and dynamic interference mitigation control of multi-source wireless device interference in complex building environments, significantly improving the system's communication robustness and response reliability under extreme conditions such as fires. This method utilizes an interference frequency band identification mechanism that combines spectrum scanning with an environmental database to construct a spectral mapping relationship between the communication frequency band and the interference source frequency band. Furthermore, it employs a support vector machine model to accurately predict the interference level of characteristic indicator vectors. Upon detecting electromagnetic coupling risks, it automatically triggers an adaptive coding parameter adjustment mechanism, dynamically improving error correction capabilities based on interference intensity, ensuring reliable upload of critical alarm information and driving coordinated control operations even in high-interference environments. Compared to existing static coding configurations or single anti-interference strategies, this method combines feedforward identification and closed-loop control, effectively addressing the core challenges of traditional fire communication systems in high-interference scenarios, such as untimely identification, inaccurate response, and unstable transmission. This approach offers significant engineering value and potential for widespread adoption.

[0092] The present invention provides Figure 2 The fire door alarm information transmission device shown in the figure based on the Internet of Things includes a communication frequency band perception module, an interference frequency band identification module, a characteristic index construction module, an interference risk intelligent judgment module and a coding parameter adaptive control module;

[0093] The communication frequency band sensing module, after the fire emergency mode is activated, continuously scans the wireless communication environment in the area where the fire door alarm device is deployed across multiple frequency bands, identifies the target communication protocol frequency band used by the fire door alarm device, and obtains spectrum parameter information for each frequency band to form a set of communication frequency bands to be analyzed;

[0094] The interference frequency band identification module extracts the operating frequency bands used by surrounding wireless devices based on a pre-established database of ambient wireless device frequency bands, identifies interference frequency bands that are adjacent to the target communication frequency band, and constructs a spectrum mapping matrix between the communication frequency band and the interference source frequency band;

[0095] The characteristic index construction module uses feature engineering methods to extract characteristic indicators reflecting the coupling characteristics between the communication frequency band and the interference frequency band based on the spectrum mapping matrix. The characteristic index vector group used for electromagnetic coupling interference judgment is constructed based on the analyzed characteristic indicators.

[0096] The intelligent interference risk identification module inputs the characteristic index vector group into a support vector machine model pre-trained with historical interference data. The model outputs the current coupling interference level indicator, intelligently predicts the electromagnetic coupling interference risk between the communication frequency band and the interference frequency band, and is used to determine whether the communication link is currently in a high electromagnetic coupling interference state.

[0097] The coding parameter adaptive control module triggers the coding parameter adaptive control mechanism when the support vector machine model prediction results indicate the presence of high electromagnetic coupling interference. The redundant coding strength is dynamically adjusted according to the intensity of electromagnetic coupling interference. During the peak period of coupling interference intensity, a transmission format with a higher error correction coding level is used for communication.

[0098] An embodiment of the present invention provides a method for transmitting fire door alarm information based on the Internet of Things, which is realized by the above-mentioned fire door alarm information transmission device based on the Internet of Things. The specific method and process of the fire door alarm information transmission device based on the Internet of Things are detailed in the embodiment of the above-mentioned method for transmitting fire door alarm information based on the Internet of Things, and will not be repeated here.

[0099] The above formulas are all dimensionless and numerical calculations. The formulas are obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters in the formulas are set by technicians in this field according to actual conditions.

[0100] The above description is merely illustrative of certain exemplary embodiments of the present invention. It goes without saying that those skilled in the art will be able to modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the above drawings and description are illustrative in nature and should not be construed as limiting the scope of protection of the claims.

[0101] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

Claims

1. A fire door alarm information transmission method based on the Internet of Things, characterized in that: The following steps are involved: After the fire emergency mode is activated, the wireless communication environment in the area where the fire door alarm device is deployed is continuously scanned in multiple frequency bands to identify the target communication protocol frequency band used by the fire door alarm device and obtain the spectrum parameter information of each frequency band to form a set of communication frequency bands to be analyzed; Based on a pre-established database of ambient wireless device frequency bands, the system extracts the operating frequency bands used by surrounding wireless devices, identifies interference frequency bands that are adjacent to the target communication frequency band, and constructs a spectrum mapping matrix between the communication frequency band and the interference source frequency band. Based on the spectrum mapping matrix, the feature engineering method is used to extract the characteristic indicators reflecting the coupling characteristics between the communication frequency band and the interference frequency band. The characteristic indicator vector group used for electromagnetic coupling interference judgment is constructed through the analyzed characteristic indicators. The characteristic indicator vector group is input into a support vector machine model that has been pre-trained with historical interference data. The model outputs the current coupling interference level indicator, and intelligently predicts the electromagnetic coupling interference risk between the communication frequency band and the interference frequency band. It is also used to determine whether the communication link is currently in a high electromagnetic coupling interference state. When the support vector machine model predicts the presence of high electromagnetic coupling interference, the adaptive control mechanism of the coding parameters is triggered, dynamically adjusting the redundant coding strength according to the intensity of the electromagnetic coupling interference. During peak periods of coupling interference intensity, a transmission format with a higher error correction coding level is used for communication. When the communication link is in a high electromagnetic coupling interference state, the redundant coding strength is dynamically adjusted according to the electromagnetic coupling interference intensity. The specific steps are as follows: When it is detected that the electromagnetic coupling coefficient is greater than the electromagnetic coupling risk threshold, the system enters the activation state and calculates a coding gain factor for regulating the coding strength. The specific calculation expression is: ,in: is the electromagnetic coupling coefficient currently predicted by the support vector machine model, which represents the coupling strength between the target communication frequency band and the interference frequency band; is the electromagnetic coupling risk threshold, when Determined to be in a high electromagnetic coupling interference state; is the coding gain factor, which is used to adjust the redundancy ratio and coding level in the future; It is the basic coupling amplitude response factor, which determines the impact of exceeding the electromagnetic coupling risk threshold amplitude on the redundancy strength; is the steepness adjustment coefficient of the nonlinear activation curve, which is used to amplify the response of the electromagnetic coupling degree in the high value range; is the time derivative adjustment coefficient, which controls the weighted proportion of the electromagnetic coupling change trend; It is the time derivative of the electromagnetic coupling coefficient, reflecting the speed of interference rise or fall; When the coding gain factor is obtained Then, the error correction coding configuration is dynamically adjusted according to the coding gain factor, including the redundancy ratio and coding level, to form a new coding strategy parameter group. The adjustment formula is as follows: ,in: represents the redundancy ratio; It is the system default basic redundancy ratio; is the error correction coding level; It is the default encoding level; and is the gain mapping coefficient, which is used to control the sensitivity of redundancy enhancement and the magnitude of coding order improvement; The updated encoding parameter configuration set is applied to the communication unit encoder layer.

2. The method for transmitting fire door alarm information based on the Internet of Things according to claim 1, characterized in that: The specific steps for performing a multi-band continuous scan of the wireless communication environment in the area where the fire door alarm system is deployed to identify the target communication protocol frequency band used are as follows: In the initial stage after the fire door alarm device is activated, the scanning range is set by the spectrum analyzer connected to the device to cover the frequency band used by the communication protocol; Configure scanning parameters including scanning step size, dwell time and power threshold, start spectrum continuous scanning function and record real-time power spectrum density graph; Analyze the collected spectrum data and identify the current active frequency band based on power peak, modulation characteristics or spectrum stability; Frequency band attribution matching is performed based on the communication protocol characteristics to determine the target communication protocol frequency band used by the fire door alarm device for subsequent interference analysis and frequency band mapping.

3. The method for transmitting fire door alarm information based on the Internet of Things according to claim 1, characterized in that: Identify the interfering frequency bands that are adjacent to the target communication frequency band. The specific steps are as follows: Based on a pre-established database of ambient wireless device frequency bands, the operating frequency band parameters of known wireless devices in the fire door alarm system deployment area are extracted, including center frequency, bandwidth range, and operating mode. Compare the identified target communication protocol frequency band, use its center frequency as a reference, and set the frequency adjacent judgment threshold as the interference judgment window; Calculate the center frequency difference and bandwidth overlap ratio of all surrounding device frequency bands to screen out wireless device frequency bands whose center frequencies partially or completely overlap with the target frequency band; The filtered frequency bands are recorded as potential interference source frequency bands, and their corresponding device types, interference weights, and overlapping properties are marked. They are used as the interference frequency band set input for constructing the spectrum mapping matrix for subsequent electromagnetic coupling analysis and interference intensity modeling.

4. The method for transmitting fire door alarm information based on the Internet of Things according to claim 1, characterized in that: A feature engineering method is used to extract characteristic indicators reflecting the coupling characteristics between the communication frequency band and the interference frequency band. The extracted indicators include the noise distribution asymmetry of the target frequency band in the equidistant frequency domains on the left and right sides and the dynamic overlap ratio of the interference frequency band and the target frequency band in multiple time periods. After analyzing the extracted characteristic indicators in the analysis window, the carrier-noise asymmetry factor and the dynamic overlap factor are generated respectively. The carrier-noise asymmetry factor and the dynamic overlap factor are used to construct a characteristic indicator vector group for electromagnetic coupling interference judgment.

5. The method for transmitting fire door alarm information based on the Internet of Things according to claim 4, characterized in that: The characteristic indicator vector group composed of the carrier-noise asymmetry factor and the dynamic overlap factor is input into the support vector machine model that has been pre-trained with historical interference data. The electromagnetic coupling coefficient is output by the model. Based on the electromagnetic coupling coefficient, the electromagnetic coupling interference risk between the communication frequency band and the interference frequency band is intelligently predicted, and it is used to determine whether the communication link is currently in a high electromagnetic coupling interference state.

6. The method for transmitting fire door alarm information based on the Internet of Things according to claim 5, characterized in that: The electromagnetic coupling coefficient generated by the support vector machine model when intelligently predicting the electromagnetic coupling interference risk between the communication frequency band and the interference frequency band is compared with the pre-set electromagnetic coupling risk threshold to determine whether the communication link is currently in a high electromagnetic coupling interference state. The specific determination steps are as follows: If the electromagnetic coupling coefficient is greater than the electromagnetic coupling risk threshold, a risk signal is generated, and it is determined that the current communication link is in a high electromagnetic coupling interference state; if the electromagnetic coupling coefficient is less than or equal to the electromagnetic coupling risk threshold, a normal signal is generated, and it is determined that the current communication link is in a low-risk interference state and is communicating efficiently.

7. The method for transmitting fire door alarm information based on the Internet of Things according to claim 4, characterized in that: In the analysis window, the specific steps for generating the carrier-noise asymmetry factor after analyzing the noise distribution asymmetry in the frequency domains equidistant on the left and right sides of the target frequency band are as follows: Taking the center frequency of the target communication frequency band as the reference, set a frequency offset distance on both sides of it, determine the symmetrical sampling points, and randomly select a bandwidth sub-band area near each symmetrical point to calculate the noise power spectrum density integrated energy. The calculation expression is: ,in: represents the noise power spectral density at frequency f; is the sub-band width; and It is a left-right symmetrical frequency point defined relative to the center frequency of the target communication frequency band. 、 , is the center frequency of the target communication frequency band; and Respectively represent the noise energy of the left and right sub-bands, reflecting the interference intensity; Obtaining the noise energy on the left and right sides and Then, the carrier-noise asymmetry factor is calculated using the hyperbolic function. The calculation expression is: ,in: is the carrier-noise asymmetry factor, which characterizes the noise power asymmetry at equidistant frequency domain positions on the left and right sides of the target communication frequency band; is the trend amplification coefficient, which is used to adjust the response sensitivity of the carrier-noise asymmetry factor to the asymmetry degree.

8. The method for transmitting fire door alarm information based on the Internet of Things according to claim 4, characterized in that: In the analysis window, the specific steps for generating a dynamic overlap factor after analyzing the dynamic overlap ratio between the interference frequency band and the target frequency band in multiple time periods are as follows: The analysis window is divided into N equally spaced sub-windows. For the i-th sub-window, the interference frequency band is defined as , the target communication frequency band is a fixed value , calculate the actual occupancy ratio of the interference frequency band to the target frequency band in each sub-window, and the calculation expression is: ,in: represents the actual occupancy ratio of the interference frequency band to the target frequency band in the i-th sub-window; and The upper and lower boundaries of the target communication frequency band are fixed; and are the upper and lower boundary frequencies of the interference frequency band in the i-th sub-window; Indicates the length of the frequency overlap interval used to calculate the interference frequency band and the target communication frequency band in the i-th subwindow; The trend shift function is defined to characterize the disturbance difference enhancement value between the current sub-window and its adjacent sub-windows, and the dynamic overlap factor is calculated by the trend shift function. The calculation expression is: ,in: is the dynamic overlap factor, which is used to measure the frequency overlap strength between the interference frequency band and the target communication frequency band in multiple sub-windows; is the trend shift enhancement function of the i-th sub-window, which is used to asymmetrically amplify the mid-segment shift; is the trend sensitivity index, which controls the degree of response to mutations; and It represents the actual occupancy ratio of the interference frequency bands on both sides of the i-th sub-window to the target frequency band, which is used to construct the local trend response.

9. An Internet of Things-based fire door alarm information transmission device, used to implement the Internet of Things-based fire door alarm information transmission method according to any one of claims 1 to 8, characterized in that: It includes a communication frequency band perception module, an interference frequency band identification module, a characteristic index construction module, an interference risk intelligent judgment module, and a coding parameter adaptive control module; The communication frequency band sensing module, after the fire emergency mode is activated, continuously scans the wireless communication environment in the area where the fire door alarm device is deployed across multiple frequency bands, identifies the target communication protocol frequency band used by the fire door alarm device, and obtains spectrum parameter information for each frequency band to form a set of communication frequency bands to be analyzed; The interference frequency band identification module extracts the operating frequency bands used by surrounding wireless devices based on a pre-established database of ambient wireless device frequency bands, identifies interference frequency bands that are adjacent to the target communication frequency band, and constructs a spectrum mapping matrix between the communication frequency band and the interference source frequency band; The characteristic index construction module uses feature engineering methods to extract characteristic indicators reflecting the coupling characteristics between the communication frequency band and the interference frequency band based on the spectrum mapping matrix. The characteristic index vector group used for electromagnetic coupling interference judgment is constructed based on the analyzed characteristic indicators. The intelligent interference risk identification module inputs the characteristic index vector group into a support vector machine model pre-trained with historical interference data. The model outputs the current coupling interference level indicator, intelligently predicts the electromagnetic coupling interference risk between the communication frequency band and the interference frequency band, and is used to determine whether the communication link is currently in a high electromagnetic coupling interference state. The coding parameter adaptive control module triggers the coding parameter adaptive control mechanism when the support vector machine model prediction results indicate the presence of high electromagnetic coupling interference. The redundant coding strength is dynamically adjusted according to the intensity of electromagnetic coupling interference. During the peak period of coupling interference intensity, a transmission format with a higher error correction coding level is used for communication.

Images