4G communication module data transmission method

Abstract

The invention discloses a 4G communication module data transmission method, which relates to the technical field of data transmission, and comprises the following steps: in the initial stage of data transmission, splitting a data file to be transmitted according to a preset fixed-length data unit so as to adapt to the bandwidth limitation of a 4G wireless channel. According to the method, the wireless signal quality is monitored in real time, the signal change trend is intelligently evaluated in combination with a machine learning model, and the splitting length of the data unit is dynamically adjusted. When the signal quality is poor, the system shortens the splitting length, reduces the data packet loss and retransmission rate, and improves the transmission reliability; when the signal quality is good, the splitting length is dynamically increased, the protocol overhead is reduced, and the throughput and the transmission efficiency are improved. The system extracts key characteristic values, generates signal quality change indexes and accurately divides signal states, limitation of a traditional static threshold value is avoided, it is ensured that a transmission strategy is higher in adaptability and stability, and the data transmission performance of a 4G communication module is effectively improved.

Description

Technical Field

[0001] The present invention relates to the technical field of data transmission, and in particular to a 4G communication module data transmission method. Background Art

[0002] 4G communication module data transmission refers to the data transmission process carried out through the fourth generation of mobile communication technology (4G). Compared with the previous generation 3G technology, the 4G network provides higher data transmission rate, lower latency and wider coverage. In the 4G communication module, data transmission is mainly carried out between the terminal device and the base station through wireless signals, involving the modulation and demodulation of radio waves, signal encoding and decoding, and data segmentation and reorganization. Users can achieve fast Internet access, high-definition video calls, online games, file downloads and other services through the 4G communication module. These modules are often embedded in various devices, such as mobile phones, routers, smart hardware, etc., so that devices can exchange data through mobile networks without Wi-Fi. 4G's high-speed data transmission not only improves the daily communication experience, but also provides basic network support for applications such as the Internet of Things and smart cities.

[0003] The prior art has the following deficiencies:

[0004] During the data transmission process of the 4G communication module, due to the limited bandwidth of the wireless signal, the data cannot be transmitted through the channel at one time, so the data file must be split into multiple small data packets for transmission one by one. This process is called segmented transmission. Existing 4G communication technology usually uses fixed-length data units for splitting to simplify the management and scheduling of data transmission. However, when the wireless signal quality is poor, continuing to use fixed-length data units for transmission may lead to serious consequences. In the case of poor signal quality, the fixed-length data unit cannot be adaptively adjusted according to the current network status, resulting in each data packet still being transmitted according to the established size, and unable to effectively cope with signal fluctuations. This not only prolongs the transmission time of each data packet and reduces the transmission efficiency, but also increases the time for retransmission and waiting for confirmation, thereby exacerbating the delay. For delay-sensitive applications (such as VoIP, high-definition video calls, etc.), this increase in delay may lead to a decrease in voice or video quality, and even problems such as audio and video asynchrony or call interruption. For large-scale data transmission (such as file downloads), high latency and low throughput will significantly slow down the transmission speed, prolong the user's waiting time, and greatly affect the user experience.

[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 constitute the prior art that is already known to one of ordinary skill in the art. Summary of the invention

[0006] The purpose of the present invention is to provide a 4G communication module data transmission method, which realizes dynamic adjustment of the data unit split length by real-time monitoring of wireless signal quality and intelligent evaluation of signal change trend in combination with a machine learning model. When the signal quality is poor, the system shortens the split length, reduces packet loss and retransmission rate, and improves transmission reliability; when the signal quality is good, the split length is dynamically lengthened to reduce protocol overhead, improve throughput and transmission efficiency. The system extracts key eigenvalues ​​(such as instantaneous channel gain and carrier synchronization error), generates a signal quality change index, accurately divides the signal state, avoids the limitations of traditional static thresholds, ensures that the transmission strategy is more adaptable and stable, and effectively improves the data transmission performance of the 4G communication module 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 4G communication module data transmission method, comprising the following steps:

[0008] In the initial stage of data transmission, the data file to be transmitted is split into pre-set fixed-length data units to adapt to the bandwidth limitation of the 4G wireless channel;

[0009] During the data splitting process, the quality information of the current wireless signal is monitored and obtained in real time, changes in signal quality are detected in a timely manner, and potential signal quality problems are identified;

[0010] Preprocess the acquired wireless signal quality information, extract key features reflecting the current wireless signal quality changes from the preprocessed signal quality data, provide strong support for intelligent evaluation, analyze the extracted features under the detection window, and quantify the changes in wireless signal quality through the analyzed features;

[0011] The quantized features are input into a pre-trained machine learning model, and the machine learning model is used to intelligently evaluate the trend of the quantized signal quality changes to determine the current signal quality status;

[0012] Based on the evaluation results of the machine learning model, the wireless signal quality is divided into two categories: high-quality signals and low-quality signals;

[0013] When a wireless signal is classified as a low-quality signal, the split length of the data unit is shortened according to the evaluation results of the machine learning model, the amount of information in each data packet is reduced, the probability of data packet loss and damage during transmission is reduced, and the reliability and stability of transmission are improved;

[0014] When the wireless signal is classified as a high-quality signal, the data unit split length is dynamically lengthened according to the length of the data to be transmitted, thereby reducing protocol overhead, increasing throughput, making full use of available bandwidth, and improving overall data transmission efficiency.

[0015] Preferably, key features reflecting the current changes in wireless signal quality are extracted from the preprocessed signal quality data, the extracted features include instantaneous changes in channel gain and the degree of carrier synchronization error, the instantaneous changes in channel gain and the degree of carrier synchronization error are analyzed under the detection window, and instantaneous channel gain reference values ​​and carrier synchronization error reference values ​​are generated respectively, and the changes in wireless signal quality are quantified by the instantaneous channel gain reference values ​​and the carrier synchronization error reference values.

[0016] Preferably, the specific steps of analyzing the instantaneous change of the channel gain in the detection window and generating the instantaneous channel gain reference value are as follows:

[0017] Calculate the instantaneous change of channel gain. By calculating the ratio of the current signal strength to the previous signal strength, determine the signal gain fluctuation. The calculation expression is as follows:

[0018]

[0019] , where ΔG inst is the instantaneous change of channel gain, G current is the signal strength at the current moment, G previous is the signal strength at the previous moment, max(G current , G previous ) is the maximum value of the signal strength at the current moment and the previous moment;

[0020] Next, the fluctuation frequency of the channel gain is quantified to reveal the frequency of drastic changes in the signal. The quantification formula is as follows:

[0021]

[0022] , where F change is the frequency at which the channel gain changes, G i is the signal strength at the i-th moment, G i+1 is the signal strength at the i+1th moment, which is equal to G i The signal strength at the next adjacent moment, γ is the set change threshold, N is the detection window length, and 1(·) is the indicator function;

[0023] To further enhance the sensitivity to changes in signal quality, a nonlinear mapping is introduced to weight changes in channel gain, as follows:

[0024]

[0025] , where ΔG enhanced is the channel gain change enhanced by nonlinear mapping, α is the nonlinear amplification factor, β is the gain fluctuation threshold, 1(|G i -G i+1 |>β) is the indicator function;

[0026] Finally, by substituting the instantaneous change in channel gain ΔG inst , the frequency of channel gain change F change And the channel gain change ΔG enhanced by nonlinear mapping enhanced Combined, the instantaneous channel gain reference value is generated, and the generation formula is as follows:

[0027] ICGI=(ΔG inst ·F change ) δ ΔG enhanced

[0028] , where ICGI is the instantaneous channel gain reference value and δ is the adjustment factor.

[0029] Preferably, the specific steps of analyzing the carrier synchronization error degree in the detection window and generating the carrier synchronization error reference value are as follows:

[0030] During the signal reception process, the carrier synchronization error is quantified by measuring the frequency and phase errors of the received signal. The carrier synchronization error calculation expression is as follows: Δφ=φ r -φ s , where Δφ is the carrier synchronization error, φ r is the received signal phase, φ s is the ideal transmitted signal phase;

[0031] Next, in order to quantify the degree of error fluctuation, the error fluctuation amount is defined to measure the dynamic change of the phase error. The formula is as follows:

[0032]

[0033] , where Δφ p is the phase error at time point p, Δφ p-1 is the phase error at the moment before time point p, w p is the weight of each time point, N is the detection window length, Ω e is the error fluctuation;

[0034] In the analysis of carrier synchronization error, while considering the phase deviation, we should also pay attention to the deviation between the received signal frequency and the reference frequency. The frequency deviation can be quantified using the frequency domain analysis method. The formula is as follows:

[0035]

[0036] , where Δf is the signal frequency deviation, exp(-j·2πf p ) is a complex exponential function used to map the phase error to the frequency domain, j is an imaginary unit, and f pis the frequency component corresponding to time point p, and 2π is a mathematical constant representing a complete arc of a cycle;

[0037] The carrier synchronization error is not only a linear change, but also affected by nonlinear factors. In order to correct the nonlinear error, a nonlinear filter is used for compensation and the nonlinear error compensation factor is defined. The formula is as follows:

[0038]

[0039] , where η is the nonlinear error compensation factor, λ p is the adjustment coefficient;

[0040] Comprehensive error fluctuation Ω e , signal frequency deviation Δf and nonlinear error compensation factor η, generate the carrier synchronization error reference value, the generation formula is as follows:

[0041]

[0042] , where CSEI is the carrier synchronization error reference value.

[0043] Preferably, the quantized instantaneous channel gain reference value and carrier synchronization error reference value are input into a pre-learned machine learning model, a signal quality change index is generated through the machine learning model, and the signal quality change trend after quantization is intelligently evaluated through the signal quality change index to determine the current signal quality status.

[0044] Preferably, the quantified signal quality change trend is intelligently evaluated through a pre-learned machine learning model, and the signal quality change index generated when judging the current signal quality state is compared and analyzed with a pre-set signal quality change index reference threshold, and the wireless signal quality is divided, and the division steps are as follows:

[0045] If the signal quality change index is greater than a preset signal quality change index reference threshold, the wireless signal quality is classified as a low quality signal;

[0046] If the signal quality change index is less than or equal to a preset signal quality change index reference threshold, the wireless signal quality is classified as a high-quality signal.

[0047] Preferably, when the wireless signal is classified as a low-quality signal, the specific steps of shortening the split length of the data unit according to the evaluation result of the machine learning model to improve the reliability and stability of the transmission are as follows:

[0048] When the wireless signal is classified as a low-quality signal, the split length of the data unit is dynamically adjusted according to the result of the signal quality change index AQCVI. When the signal quality change index AQCVI is greater than the preset signal quality change index reference threshold, the data unit is split into smaller fixed-length data packets to adapt to the quality fluctuation of the current channel. The adjusted split length is calculated by the following formula:

[0049]

[0050] , where L new，short is the shortened data unit split length, L base is the preset data unit split length, AQCVI ref is the signal quality change index reference threshold, C fade is the channel fading factor, B current is the current available bandwidth, ω and are all exponential adjustment factors. ω is used to control the influence of the signal quality change index AQCVI on the split length. Used to control the impact of channel fading and bandwidth on the split length.

[0051] Preferably, when the wireless signal is classified as a high-quality signal, the specific steps of dynamically lengthening the data unit split length according to the length of the data to be transmitted to improve the overall data transmission efficiency are as follows:

[0052] When the wireless signal is divided into high-quality signals, the split length of the data unit is dynamically adjusted and lengthened according to the amount of data to be transmitted and the current network status, thereby improving data transmission efficiency, reducing protocol overhead, and making full use of available bandwidth resources. A refined dynamic adjustment model is constructed to dynamically adjust the split length. The calculation expression for lengthening the split length of the data unit is as follows:

[0053]

[0054] , where L new，long is the length of the data unit after the extension, B max is the maximum available bandwidth of the channel, B current is the current available bandwidth, D current is the current amount of data to be transmitted, D max is the maximum amount of data supported to be transmitted, θ, μ and ρ are all exponential adjustment factors, θ is used to control the influence of the signal quality change index AQCVI on the lengthened split length, μ is used to control the influence of bandwidth utilization on the split length, and ρ is used to control the influence of the data amount on the lengthened split length.

[0055] In the above technical solution, the technical effects and advantages provided by the present invention are:

[0056] The present invention solves the problem of low transmission efficiency and high retransmission rate of fixed-length data units in traditional 4G communication technology under signal fluctuation environment by real-time monitoring of wireless signal quality and dynamically adjusting the split length of data units according to signal changes. When the signal quality is poor, the system shortens the split length of the data unit and reduces the amount of information in each data packet, thereby reducing the probability of data packet loss and damage during transmission, reducing the number of retransmissions, and improving the reliability and stability of transmission; when the signal quality is good, the system dynamically lengthens the split length of the data unit and reduces the number of data packets, thereby reducing protocol overhead and improving the throughput and overall efficiency of data transmission. This dynamic adjustment mechanism makes full use of the available bandwidth, making data transmission more adaptable to complex wireless environments, ensuring the efficiency and stability of transmission under different signal conditions, and improving user experience.

[0057] The present invention introduces a machine learning model to intelligently evaluate the changing trend of wireless signal quality, thereby achieving precise adjustment of the transmission strategy. In the process of signal quality monitoring, key characteristic values ​​that can reflect signal fluctuations, such as instantaneous channel gain and carrier synchronization error, are extracted, and these characteristics are quantified as reference values ​​and input into the machine learning model. Through the signal quality change index generated by intelligent evaluation, the system can dynamically judge the current signal state, divide the signal into high quality and low quality categories, and adjust the data unit splitting strategy accordingly. This dynamic division method based on machine learning avoids the limitations of traditional static threshold setting, makes the transmission strategy more adaptable and accurate, thereby effectively improving the data transmission efficiency and stability of the 4G communication module in a complex network environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced 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.

[0059] Figure 1 This is a method flow chart of a 4G communication module data transmission method of the present invention. DETAILED DESCRIPTION

[0060] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of 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 the present disclosure will be more comprehensive and complete, and the concept of the example embodiments will be fully conveyed to those skilled in the art.

[0061] The present invention provides Figure 1A 4G communication module data transmission method shown includes the following steps:

[0062] In the initial stage of data transmission, the data file to be transmitted is split into pre-set fixed-length data units to adapt to the bandwidth limitation of the 4G wireless channel;

[0063] Before data transmission begins, large files or data streams are split into several fixed-size data blocks (data units) according to the requirements of the network protocol or communication system. The size of each data unit is set in advance, usually determined by the transmission protocol (such as IP protocol or TCP protocol) or the bandwidth limitation of the wireless channel. The purpose of this splitting method is to ensure that data can be effectively transmitted through the wireless channel, because the bandwidth of the wireless channel is limited and cannot carry too much data at one time. By splitting data files into data units of appropriate sizes, network congestion and packet loss caused by overly large data packets can be avoided during transmission, while simplifying the data transmission and management process. This method is a standardized processing method that aims to improve the efficiency and reliability of data transmission, especially in a 4G network environment with limited bandwidth.

[0064] During the data splitting process, the quality information of the current wireless signal is monitored and obtained in real time, changes in signal quality are detected in a timely manner, and potential signal quality problems are identified;

[0065] In order to dynamically respond to changes in wireless signal quality, it is necessary to continuously monitor key indicators such as signal strength, noise level, and interference level of the wireless channel. This information can be obtained through the signal receiving module of the terminal device or the feedback mechanism provided by the base station. Real-time acquisition of signal quality information enables the system to understand the status of the current network environment in a timely manner, so as to make necessary adjustments according to the actual situation during the data transmission process. This process is the basis for achieving adaptive data transmission, ensuring that the transmission strategy can be optimized under different signal conditions and improving the overall transmission performance.

[0066] Preprocess the acquired wireless signal quality information, extract key features reflecting the current wireless signal quality changes from the preprocessed signal quality data, provide strong support for intelligent evaluation, analyze the extracted features under the detection window, and quantify the changes in wireless signal quality through the analyzed features;

[0067] The preprocessing step includes operations such as filtering, denoising, and normalization to improve the accuracy and reliability of signal quality data. Filtering can remove high-frequency noise and interfering signals; normalization converts signal quality indicators of different dimensions to the same scale to facilitate subsequent feature extraction and analysis. In addition, preprocessing may also involve smoothing and interpolation of data to fill missing values ​​or process incomplete data. The purpose of this stage is to ensure that the data input into the machine learning model is high-quality and representative, thereby improving the accuracy and reliability of model evaluation.

[0068] The key features reflecting the current changes in wireless signal quality are extracted from the preprocessed signal quality data. The extracted features include instantaneous changes in channel gain and the degree of carrier synchronization error. The instantaneous changes in channel gain and the degree of carrier synchronization error are analyzed under the detection window to generate instantaneous channel gain reference values ​​and carrier synchronization error reference values, respectively. The changes in wireless signal quality are quantified by the instantaneous channel gain reference values ​​and the carrier synchronization error reference values.

[0069] A sudden and significant drop in channel gain usually indicates that the current wireless signal quality is poor. Channel gain reflects the degree of attenuation of the signal during transmission, that is, the ratio of signal strength to the source. If the channel gain drops suddenly and significantly, it means that the signal has suffered a large attenuation or other adverse factors during transmission. Common reasons include signal shielding, multipath effects, interference, or rapid changes in the wireless link. For example, when the receiving device is far away from the base station or encounters an obstruction, the attenuation of the signal will increase, resulting in a significant drop in channel gain. In addition, rapid multipath fading can also cause instantaneous fluctuations in signal gain, especially in urban or indoor environments, where the signal reaches the receiving end through multiple paths, and the phase of some path signals will be reversed, causing the strength to weaken. A sharp drop in signal gain usually means that the receiving end cannot stably receive a signal of sufficient strength, which will affect the correct reception and decoding of the data, increase the packet loss rate, and may cause a decrease in communication quality. Therefore, a sudden and significant drop in channel gain is an important signal of poor wireless signal quality and unstable link, indicating that data transmission may encounter greater difficulties, such as delays, increased bit error rates, or transmission failures.

[0070] The specific steps for analyzing the instantaneous change of channel gain in the detection window and generating the instantaneous channel gain reference value are as follows:

[0071] Calculate the instantaneous change of channel gain. By calculating the ratio of the current signal strength to the previous signal strength, the signal gain fluctuation is determined. The core is to capture the rapid change of the signal over time. The calculation expression is as follows:

[0072]

[0073] , where ΔG inst is the instantaneous change of channel gain, which indicates the change amplitude of channel gain between two adjacent time points. current is the signal strength at the current moment, indicating the strength of the wireless signal received at a certain moment within the detection window, G previous is the signal strength at the previous moment, indicating the signal strength received at the moment before the current moment, max(G current , G previous ) is the maximum value of the signal strength at the current moment and the previous moment, and is used to normalize the change in signal gain. By normalizing the change amplitude, the influence of the absolute value of the signal strength can be eliminated, and the focus can be placed on the relative change;

[0074] This step calculates the difference between the current and previous signal strengths and normalizes it with the larger signal strength to obtain the gain change amplitude. When this value is large, it means that the signal strength has fluctuated dramatically in a short period of time, which usually means that the channel conditions have deteriorated.

[0075] Next, the fluctuation frequency of the channel gain is quantified, that is, the frequency of channel gain changes within the detection window (such as the number of fluctuations or the amplitude range), which reveals the frequency of drastic signal changes. A channel with a high fluctuation frequency usually means poor signal quality. The quantification formula is as follows:

[0076]

[0077] , where F change It is the frequency of channel gain change, indicating the number of times the channel gain change exceeds the set threshold within the detection window, reflecting the frequency of signal gain fluctuation. i is the signal strength at the i-th moment, G i+1 is the signal strength at the i+1th moment, which is equal to G i The signal strength at the next adjacent moment, γ is the set change threshold, which is used to distinguish normal signal fluctuations from abnormal fluctuations, N is the detection window length, 1(·) is the indicator function, when the change in signal gain |G i -G i+1 | is greater than the threshold γ, the function value is 1, indicating that the signal gain has changed significantly at this moment (i.e., the fluctuation is large); otherwise, it is 0, indicating that the signal gain has not changed significantly at this moment and has not exceeded the predetermined range of change;

[0078] By counting the number of times the signal gain change exceeds the set threshold, the frequency of channel gain change is obtained. change A low value indicates that the signal fluctuates more frequently, which usually means poor signal quality because high-frequency fluctuations may be caused by interference or multipath effects.

[0079] To further enhance the sensitivity to changes in signal quality, a nonlinear mapping is introduced to weight changes in channel gain, which helps to highlight signal changes with larger amplitudes so that a quick response can be made when the signal quality is extremely poor. The formula is as follows:

[0080]

[0081] , where ΔG enhanced It is the channel gain change enhanced by nonlinear mapping, which is used to reflect the amplitude of channel gain fluctuation. After nonlinear weighting, it can highlight the larger fluctuation. α is the nonlinear amplification factor, which is used to nonlinearly enhance the gain change. It is usually set to α>1. β is the gain fluctuation threshold, which is used to screen the threshold of significant gain fluctuation. Only when the gain change at adjacent moments exceeds β, it will be included in the final calculation. 1(|G i -G i+1 |>β) is the indicator function, when the gain change at adjacent moments |G i -G i+1 When | is greater than the threshold β, the value is 1, otherwise it is 0;

[0082] Indicator function 1(|G i -G i+1 |>β) is used to determine whether the condition is true and output the result:

[0083] Value is 1: When the condition |G i -G i+1 When |>β holds, the change in channel gain |G i -G i+1 |When the value exceeds the set threshold β, the indicator function returns a value of 1, indicating that the change is significant and should be calculated as an important fluctuation in the channel gain.

[0084] Value is 0: When the condition |G i -G i+1 When |≤β does not hold, the change in channel gain |G i -G i+1 |If the threshold β is not exceeded, the indicator function returns a value of 0, indicating that the change is not significant and will not be included in the calculation.

[0085] At this stage, nonlinear enhancement is used to significantly amplify the effects of large gain changes, making them have a larger impact on the final index value. This step helps increase sensitivity to large signal fluctuations, especially when the signal quality is very poor.

[0086] Finally, by substituting the instantaneous change in channel gain ΔG inst , the frequency of channel gain change F change And the channel gain change ΔG enhanced by nonlinear mappingenhanced Combined with the above, the instantaneous channel gain reference value is generated. The channel gain reference value can comprehensively reflect the quality and stability of the signal and is an important indicator for signal quality analysis. The generation formula is as follows:

[0087] ICGI=(ΔG inst ·F change ) δ ΔG enhanced

[0088] , where ICGI is the instantaneous channel gain reference value and δ is the adjustment factor used to adjust the weight of instantaneous change and fluctuation frequency in the final index.

[0089] The final instantaneous channel gain reference value ICGI is obtained by integrating the instantaneous change of the channel gain, the fluctuation frequency, and the change after nonlinear enhancement. When the instantaneous channel gain reference value ICGI is large, it indicates that the signal quality is poor, because it indicates that the signal fluctuation amplitude is large, the change is frequent, and the gain changes dramatically. On the contrary, when the instantaneous channel gain reference value ICGI is small, it indicates that the signal quality is good, the signal gain change is stable, and the transmission environment is stable.

[0090] The larger the instantaneous channel gain reference value generated after analyzing the instantaneous change of channel gain under the detection window, the worse the current wireless signal quality is. Conversely, the better the signal quality is. This is because the instantaneous channel gain reference value reflects the fluctuation and change amplitude of the signal strength. The stability of the channel is usually evaluated by monitoring the change of the signal in a short period of time. If the channel gain performance value is large, it usually means that the signal has experienced a large attenuation or instability during the transmission process, which may be caused by factors such as obstruction, interference, and multipath effects. Large fluctuations in signal gain indicate that the signal strength received by the receiving end is unstable, resulting in poor signal quality and reduced reliability of data transmission. If the instantaneous channel gain reference value is small, it means that the signal strength is relatively stable, the channel quality is good, the signal can be transmitted stably, and the probability of data packet loss and bit error is low.

[0091] An increase in carrier synchronization error usually indicates that the current wireless signal quality is poor. Carrier synchronization is the process of ensuring that the frequency and phase of the transmitter and receiver are aligned in a wireless communication system. Its purpose is to ensure accurate decoding of the signal and prevent information loss or errors caused by frequency offset or phase deviation. If the carrier synchronization error increases, it means that the synchronization accuracy between the receiver and the transmitter has decreased, which is usually caused by the following factors:

[0092] Signal attenuation: When a wireless signal encounters an obstacle during propagation or is far away from a base station, the signal strength attenuates, which may result in a large frequency difference between the received signal and the expected signal, leading to increased synchronization errors.

[0093] Multipath effect: In a complex wireless propagation environment, signals may propagate to the receiving end through multiple paths, resulting in increased signal phase differences on different paths, which will affect the accurate synchronization of the carrier.

[0094] Interference: Interference from external wireless signal sources or co-channel interference can change the frequency or phase of the received signal, making carrier synchronization more difficult. Therefore, an increase in carrier synchronization error directly reflects a decrease in signal quality, indicating that signal transmission is unstable, which may lead to errors or loss of data transmission, affecting system reliability and performance.

[0095] The specific steps of analyzing the carrier synchronization error degree in the detection window and generating the carrier synchronization error reference value are as follows:

[0096] During the signal reception process, the carrier synchronization error is quantified by measuring the frequency and phase errors of the received signal. The carrier synchronization error calculation expression is as follows: Δφ=φ r -φ s , where Δφ is the carrier synchronization error, reflecting the deviation between the phase of the received signal and the ideal phase, φ r is the received signal phase, which refers to the phase information of the signal demodulated by the receiving end, φ s is the ideal transmission signal phase, which refers to the phase of the signal sent by the transmitter;

[0097] This step represents the phase deviation between the received signal and the ideal signal. The phase of the received signal φ r is obtained through the demodulation process, and the ideal phase φ s It is determined based on the reference signal of the transmitting end.

[0098] Next, in order to quantify the degree of error fluctuation, the error fluctuation amount is defined to measure the dynamic change of the phase error. The formula is as follows:

[0099]

[0100] , where Δφ p is the phase error at time point p, Δφ p-1 is the phase error at the moment before time point p, w p is the weight of each time point, which is used to emphasize the error fluctuation at certain moments, N is the detection window length, Ω e It is the error fluctuation amount, which is an indicator used to quantify the change of phase error;

[0101] This quantization process reflects the stability of signal synchronization by calculating the cumulative value of each error change.

[0102] In the analysis of carrier synchronization error, while considering the phase deviation, we should also pay attention to the deviation between the received signal frequency and the reference frequency, because the frequency deviation directly affects the demodulation of the signal and the correctness of the data. The frequency domain analysis method is used to quantify the frequency deviation. The formula is as follows:

[0103]

[0104] , where Δf is the signal frequency deviation, that is, the deviation between the received signal frequency and the reference frequency, exp(-j·2πf p ) is a complex exponential function used to map the phase error to the frequency domain, j is an imaginary unit, and f p is the frequency component corresponding to time point p, 2π is a mathematical constant representing a complete arc of a cycle, and the frequency component f p Convert from Hertz (number of cycles) to angular frequency (radians) to make the formula units consistent and accurately represent periodic signals;

[0105] Through this frequency domain analysis, the frequency deviation distribution of the carrier synchronization error can be obtained, thereby evaluating the frequency stability of the signal. When the frequency deviation increases, it means that the signal quality is reduced and the frequency synchronization difference is aggravated, which will directly affect the reliability of communication.

[0106] The carrier synchronization error is not only a linear change, but also affected by nonlinear factors, such as multipath effect and environmental changes. In order to correct the nonlinear error, a nonlinear filter is used for compensation and a nonlinear error compensation factor is defined. The formula is as follows:

[0107]

[0108] , where η is the nonlinear error compensation factor, which indicates the degree of nonlinear correction of synchronization error, λ p It is the adjustment coefficient, which is used to control the intensity of nonlinear compensation. The size of this coefficient determines the degree of correction of different error values;

[0109] Through the nonlinear compensation factor, the influence of the nonlinear error in the signal on the carrier synchronization can be adjusted to make the overall error reasonable and enhance the stability of signal demodulation.

[0110] Comprehensive error fluctuation Ω e , signal frequency deviation Δf and nonlinear error compensation factor η, generate the carrier synchronization error reference value, the generation formula is as follows:

[0111]

[0112] , where CSEI is the carrier synchronization error reference value.

[0113] The carrier synchronization error reference value combines the phase and frequency synchronization errors of the signal, taking into account nonlinear factors, and finally obtains a comprehensive error index. The larger the carrier synchronization error reference value, the more serious the carrier synchronization error and the worse the signal quality; conversely, it means that the signal synchronization accuracy is higher and the signal quality is better.

[0114] The larger the carrier synchronization error reference value generated after analyzing the carrier synchronization error degree under the detection window, the worse the current wireless signal quality is. Conversely, the better the signal quality is. The carrier synchronization error refers to the frequency and phase deviation between the receiving end and the transmitting end, which reflects the accuracy of signal synchronization. Under the monitoring window, when the carrier synchronization error degree increases, it usually means that the error of signal synchronization is also increasing, which means that the quality of the signal is deteriorating, which may be caused by factors such as signal attenuation, interference or multipath effect. The increase in the carrier synchronization error reference value usually indicates that the phase and frequency of the signal are unstable, and the synchronization between the receiving end and the transmitting end is difficult, which will cause data decoding errors and information loss, thereby affecting the reliability and quality of data transmission. On the contrary, if the carrier synchronization error reference value is small, it indicates that the synchronization accuracy of the signal is high, the frequency and phase of the receiving end and the transmitting end are basically aligned, the signal quality is good, and the transmission process is stable and reliable.

[0115] The quantized features are input into a pre-trained machine learning model, and the machine learning model is used to intelligently evaluate the trend of the quantized signal quality changes to determine the current signal quality status;

[0116] A pre-learned machine learning model refers to a machine learning model that has learned the law and characteristics of signal quality changes by training on historical data. Specifically, this model is learned in a training phase using a large amount of wireless signal quality data (such as signal strength, signal-to-noise ratio, bit error rate, etc.) and corresponding labels (such as good or poor signal quality). Through these historical data, the model can identify the relationship between signal quality and different environmental factors, and learn how to predict or judge the state of signal quality based on the input signal characteristics. The training process usually involves optimizing the algorithm using a data set. Common machine learning algorithms include supervised learning methods (such as support vector machines, decision trees, random forests, neural networks, etc.). These algorithms can extract patterns from training data and apply them to new, unseen data for effective prediction and classification.

[0117] In this way, the machine learning model can intelligently evaluate the changing trend of signal quality after inputting the quantized signal features, and judge the current signal quality status based on the output results of the model. For example, when new signal data is input, the machine learning model will analyze the characteristics of the current signal based on the rules learned from historical data to determine whether the signal is in a high-quality or low-quality state. The advantage of this method is that it can automatically adapt to complex changes in the signal environment without relying on manually set rules or thresholds, thereby improving the accuracy and flexibility of the evaluation. In addition, the scalability of the machine learning model also enables it to continuously optimize and improve the prediction performance as more signal quality data is collected. Therefore, the pre-learned machine learning model can realize dynamic signal quality evaluation based on real-time data, helping the system to respond more intelligently and promptly.

[0118] The quantized instantaneous channel gain reference value and carrier synchronization error reference value are input into the pre-learned machine learning model, and a signal quality change index is generated through the machine learning model. The signal quality change index is used to intelligently evaluate the quantized signal quality change trend to determine the current signal quality status.

[0119] The machine learning model is not limited here, and any machine learning model that can perform a comprehensive analysis of the instantaneous channel gain reference value ICGI and the carrier synchronization error reference value CSEI to generate a signal quality change index AQCVI is acceptable. To implement the technical solution of the present invention, the present invention provides a specific implementation method;

[0120] The signal quality variation index AQCVI generation formula is as follows:

[0121]

[0122] , where k1 and k2 are preset proportional coefficients of the instantaneous channel gain reference value ICGI and the carrier synchronization error reference value CSEI, respectively, and both k1 and k2 are greater than 0.

[0123] It can be seen from the signal quality change index that the larger the instantaneous channel gain reference value generated after analyzing the instantaneous change of the channel gain under the detection window, the larger the carrier synchronization error reference value generated after analyzing the carrier synchronization error degree under the detection window, indicating that the signal quality change trend after quantization is intelligently evaluated through the pre-learned machine learning model. The larger the signal quality change index generated when judging the current signal quality status, the greater the probability that the current signal is a low-quality signal. Conversely, the smaller the probability that the current signal is a low-quality signal.

[0124] The preset proportionality coefficients (k1 and k2) are constant coefficients used in the machine learning model to adjust the signal quality change calculation formula. In the formula, k1 and k2 correspond to the contribution of the channel gain reference value ICGI and the carrier synchronization error reference value CSEI to the signal quality change. Specifically, these coefficients are used to control the influence of each signal quality indicator (ICGI and CSEI) in the final calculation result.

[0125] The purpose of the preset proportional coefficient is to ensure that the model can flexibly adjust the influence of different indicators according to the characteristics of different signal quality changes. For example, if k1 is large, it means that the change of the channel gain reference value ICGI has a more significant impact on the signal quality change; conversely, if k2 is large, it means that the change of the carrier synchronization error reference value CSEI has a greater impact on the signal quality change. This preset coefficient can be adjusted according to the actual signal environment and the training results of the model, so that the model is more suitable for specific wireless signal quality analysis tasks, thereby improving prediction accuracy and response speed.

[0126] Based on the evaluation results of the machine learning model, the wireless signal quality is divided into two categories: high-quality signals and low-quality signals;

[0127] The quantified signal quality change trend is intelligently evaluated through the pre-learned machine learning model. The signal quality change index generated when judging the current signal quality status is compared with the pre-set signal quality change index reference threshold to divide the wireless signal quality. The division steps are as follows:

[0128] If the signal quality change index is greater than a preset signal quality change index reference threshold, the wireless signal quality is classified as a low quality signal;

[0129] If the signal quality change index is less than or equal to a preset signal quality change index reference threshold, the wireless signal quality is classified as a high-quality signal.

[0130] High-quality signals refer to signals with good strength and stability in wireless communications, which can provide clear and reliable connections and data transmission. Low-quality signals are characterized by large signal attenuation, severe noise interference, and unstable signals, which cannot meet normal data transmission needs.

[0131] When a wireless signal is classified as a low-quality signal, the split length of the data unit is shortened according to the evaluation results of the machine learning model, the amount of information in each data packet is reduced, the probability of data packet loss and damage during transmission is reduced, and the reliability and stability of transmission are improved;

[0132] When a wireless signal is classified as a low-quality signal, the specific steps for shortening the split length of the data unit and improving the reliability and stability of transmission are as follows:

[0133] When the wireless signal is classified as a low-quality signal, the split length of the data unit is dynamically adjusted according to the result of the signal quality change index AQCVI. When the signal quality change index AQCVI is greater than the preset signal quality change index reference threshold, the data unit is split into smaller fixed-length data packets to adapt to the quality fluctuations of the current channel. This adjustment helps reduce the risk of data packet loss and damage and improve the stability of data transmission. The adjusted split length is calculated by the following formula:

[0134]

[0135] , where L new，short is the shortened data unit split length, L base is the preset data unit split length, AQCVI ref is the signal quality change index reference threshold, C fade is the channel fading factor, which reflects the signal attenuation of the current channel. current is the current available bandwidth, which is used to measure the transmission capacity of the channel, ω and Both are exponential adjustment factors. ω is used to control the influence of the signal quality change index AQCVI on the split length. By adjusting the value of ω, the sensitivity of signal fluctuation to the split length is balanced. It is used to control the influence of channel fading and bandwidth on the split length by adjusting The value of controls the weight of the impact of channel fading and bandwidth on the data split length.

[0136] The purpose of this step is to improve the reliability and stability of data transmission by shortening the split length of the data unit and reducing the amount of information in each data packet when the wireless signal quality is poor. Specifically, when the machine learning model evaluates that the signal quality is low, the stability of the channel and the reliability of data transmission are usually reduced, and the data packets are prone to loss or damage during transmission. By splitting the data packets into smaller pieces, the amount of information carried by each data packet when it is transmitted in the channel can be reduced, making it easier for each data packet to be successfully transmitted, especially in environments with poor signal quality. Doing so not only reduces the need for retransmission, but also effectively reduces the risk of data loss due to network instability, thereby improving the stability of data transmission. In addition, the transmission of small data packets can also reduce the impact of signal fluctuations, ensuring that the integrity and correctness of the data can be guaranteed as much as possible in harsh network environments.

[0137] When the wireless signal is classified as a high-quality signal, the data unit split length is dynamically lengthened according to the length of the data to be transmitted, reducing protocol overhead, increasing throughput, making full use of available bandwidth, and improving overall data transmission efficiency;

[0138] When the wireless signal is classified as a high-quality signal, the specific steps of dynamically lengthening the data unit split length according to the length of the data to be transmitted and improving the overall data transmission efficiency are as follows:

[0139] When the wireless signal is divided into high-quality signals, the split length of the data unit is dynamically adjusted and lengthened according to the amount of data to be transmitted and the current network status, thereby improving data transmission efficiency, reducing protocol overhead, and making full use of available bandwidth resources. To achieve this goal, key parameters such as network delay, channel capacity, and amount of data to be transmitted are introduced to build a refined dynamic adjustment model to dynamically adjust the split length. This optimization method can not only effectively adapt to changes in the network environment, but also significantly improve the overall data throughput and transmission performance, ensuring the maximum utilization of resources. The calculation expression for lengthening the split length of the data unit is as follows:

[0140]

[0141] , where L new，long is the length of the data unit after the extension, B max is the maximum available bandwidth of the channel, reflecting the theoretical maximum transmission capacity of the channel, B current is the current available bandwidth, D current is the current amount of data to be transmitted, reflecting the total amount of data that needs to be transmitted, D max is the maximum amount of data to be transmitted, representing the maximum amount of data that can be effectively processed under good network conditions. θ, μ, and ρ are all exponential adjustment factors. θ is used to control the influence of the signal quality change index AQCVI on the lengthening of the split length. By adjusting θ, the sensitivity of signal fluctuations to the increase in split length can be balanced. μ is used to control the influence of bandwidth utilization on the split length, and adjust the weight of bandwidth utilization on the increase in split length. ρ is used to control the influence of data volume on the lengthening of the split length, and adjust the weight of the data volume factor on the increase in split length.

[0142] The purpose of this step is to dynamically adjust the data unit split length according to the changes in the quality of the wireless signal, so as to reduce the protocol overhead and improve the data transmission efficiency in a high-quality signal environment. When the wireless signal quality is good, the signal stability is high, the data transmission is not easily affected by attenuation or interference, and the utilization rate of the network bandwidth is high. In this case, it is no longer mandatory to use fixed-length data units when splitting data, but the size of the split unit is dynamically adjusted according to the actual length of the data to be transmitted. By lengthening the length of the data unit, the header overhead of each data packet can be reduced, thereby reducing the protocol overhead and the additional burden during the transmission process. In addition, reducing the number of data packets can also reduce the management and scheduling complexity during the transmission process, improve the overall throughput, and avoid frequent packet header information transmission. Ultimately, this adjustment can make more efficient use of available bandwidth, achieve higher transmission rates and overall data transmission efficiency, especially when the signal quality is good, which helps to improve user experience and optimize the use of network resources.

[0143] The present invention solves the problem of low transmission efficiency and high retransmission rate of fixed-length data units in traditional 4G communication technology under signal fluctuation environment by real-time monitoring of wireless signal quality and dynamically adjusting the split length of data units according to signal changes. When the signal quality is poor, the system shortens the split length of the data unit and reduces the amount of information in each data packet, thereby reducing the probability of data packet loss and damage during transmission, reducing the number of retransmissions, and improving the reliability and stability of transmission; when the signal quality is good, the system dynamically lengthens the split length of the data unit and reduces the number of data packets, thereby reducing protocol overhead and improving the throughput and overall efficiency of data transmission. This dynamic adjustment mechanism makes full use of the available bandwidth, making data transmission more adaptable to complex wireless environments, ensuring the efficiency and stability of transmission under different signal conditions, and improving user experience.

[0144] The present invention introduces a machine learning model to intelligently evaluate the changing trend of wireless signal quality, thereby achieving precise adjustment of the transmission strategy. In the process of signal quality monitoring, key characteristic values ​​that can reflect signal fluctuations, such as instantaneous channel gain and carrier synchronization error, are extracted, and these characteristics are quantified as reference values ​​and input into the machine learning model. Through the signal quality change index generated by intelligent evaluation, the system can dynamically judge the current signal state, divide the signal into high quality and low quality categories, and adjust the data unit splitting strategy accordingly. This dynamic division method based on machine learning avoids the limitations of traditional static threshold setting, makes the transmission strategy more adaptable and accurate, thereby effectively improving the data transmission efficiency and stability of the 4G communication module in a complex network environment.

[0145] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0146] The above description is only by way of illustration of certain exemplary embodiments of the present invention. It is undoubted that those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the above drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

[0147] It should be noted that, in this article, if there are relational terms such as first and second, etc., they are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0148] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0149] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0150] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0151] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0152] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0153] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0154] The above description is only by way of illustration of certain exemplary embodiments of the present invention. It is undoubted that those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the above drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

Claims

1. A 4G communication module data transmission method, characterized in that: The following steps are involved: In the initial stage of data transmission, the data file to be transmitted is split into pre-set fixed-length data units to adapt to the bandwidth limitation of the 4G wireless channel; During the data splitting process, the quality information of the current wireless signal is monitored and obtained in real time, changes in signal quality are detected in a timely manner, and potential signal quality problems are identified; Preprocess the acquired wireless signal quality information, extract key features reflecting the current wireless signal quality changes from the preprocessed signal quality data, provide strong support for intelligent evaluation, analyze the extracted features under the detection window, and quantify the changes in wireless signal quality through the analyzed features; The quantized features are input into a pre-trained machine learning model, and the machine learning model is used to intelligently evaluate the trend of the quantized signal quality changes to determine the current signal quality status; Based on the evaluation results of the machine learning model, the wireless signal quality is divided into two categories: high-quality signals and low-quality signals; When a wireless signal is classified as a low-quality signal, the split length of the data unit is shortened according to the evaluation results of the machine learning model, the amount of information in each data packet is reduced, the probability of data packet loss and damage during transmission is reduced, and the reliability and stability of transmission are improved; When the wireless signal is classified as a high-quality signal, the data unit split length is dynamically lengthened according to the length of the data to be transmitted, thereby reducing protocol overhead, increasing throughput, making full use of available bandwidth, and improving overall data transmission efficiency.

2. A 4G communication module data transmission method according to claim 1, characterized in that: The key features reflecting the current changes in wireless signal quality are extracted from the preprocessed signal quality data. The extracted features include instantaneous changes in channel gain and the degree of carrier synchronization error. The instantaneous changes in channel gain and the degree of carrier synchronization error are analyzed under the detection window to generate instantaneous channel gain reference values ​​and carrier synchronization error reference values, respectively. The changes in wireless signal quality are quantified by the instantaneous channel gain reference values ​​and the carrier synchronization error reference values.

3. A 4G communication module data transmission method according to claim 2, characterized in that: The specific steps for analyzing the instantaneous change of channel gain in the detection window and generating the instantaneous channel gain reference value are as follows: Calculate the instantaneous change of channel gain. By calculating the ratio of the current signal strength to the previous signal strength, determine the signal gain fluctuation. The calculation expression is as follows: In the formula, ΔG inst is the instantaneous change of channel gain, G current is the signal strength at the current moment, G previous is the signal strength at the previous moment, max(G current , G previous ) is the maximum value of the signal strength at the current moment and the previous moment; Next, the fluctuation frequency of the channel gain is quantified to reveal the frequency of drastic changes in the signal. The quantification formula is as follows: In the formula, F change is the frequency at which the channel gain changes, G i is the signal strength at the i-th moment, G i+1 is the signal strength at the i+1th moment, which is equal to G i The signal strength at the next adjacent moment, γ is the set change threshold, N is the detection window length, and 1(·) is the indicator function; To further enhance the sensitivity to changes in signal quality, a nonlinear mapping is introduced to weight changes in channel gain, as follows: In the formula, ΔG enhanced is the channel gain change enhanced by nonlinear mapping, α is the nonlinear amplification factor, β is the gain fluctuation threshold, 1(|G i -G i+1 |>δ) is the indicator function; Finally, by substituting the instantaneous change in channel gain ΔG inst , the frequency of channel gain change F change And the channel gain change ΔG enhanced by nonlinear mapping enhanced Combined, the instantaneous channel gain reference value is generated, and the generation formula is as follows: ICGI=(ΔG inst ·F change ) δ ·ΔG enhanced Where ICGI is the instantaneous channel gain reference value and δ is the adjustment factor.

4. A 4G communication module data transmission method according to claim 2, characterized in that: The specific steps of analyzing the carrier synchronization error degree in the detection window and generating the carrier synchronization error reference value are as follows: During the signal reception process, the carrier synchronization error is quantified by measuring the frequency and phase errors of the received signal. The carrier synchronization error calculation expression is as follows: Δφ=φ r -φ s , where Δφ is the carrier synchronization error, φ r is the received signal phase, φ s is the ideal transmitted signal phase; Next, in order to quantify the degree of error fluctuation, the error fluctuation amount is defined to measure the dynamic change of the phase error. The formula is as follows: In the formula, Δφ p is the phase error at time point p, Δφ p-1 is the phase error at the moment before time point p, w p is the weight of each time point, N is the detection window length, Ω e is the error fluctuation; In the analysis of carrier synchronization error, while considering the phase deviation, we should also pay attention to the deviation between the received signal frequency and the reference frequency. The frequency deviation can be quantified using the frequency domain analysis method. The formula is as follows: Where Δf is the signal frequency deviation, exp(-j·2πf p ) is a complex exponential function used to map the phase error to the frequency domain, j is an imaginary unit, and f p is the frequency component corresponding to time point p, and 2π is a mathematical constant representing a complete arc of a cycle; The carrier synchronization error is not only a linear change, but also affected by nonlinear factors. In order to correct the nonlinear error, a nonlinear filter is used for compensation and the nonlinear error compensation factor is defined. The formula is as follows: Where η is the nonlinear error compensation factor, λ p is the adjustment coefficient; Comprehensive error fluctuation Ω e , signal frequency deviation Δf and nonlinear error compensation factor η, generate the carrier synchronization error reference value, the generation formula is as follows: Where CSEI is the carrier synchronization error reference value.

5. A 4G communication module data transmission method according to claim 2, characterized in that: The quantized instantaneous channel gain reference value and carrier synchronization error reference value are input into the pre-learned machine learning model, and a signal quality change index is generated through the machine learning model. The signal quality change index is used to intelligently evaluate the quantized signal quality change trend to determine the current signal quality status.

6. A 4G communication module data transmission method according to claim 5, characterized in that: The quantified signal quality change trend is intelligently evaluated through the pre-learned machine learning model. The signal quality change index generated when judging the current signal quality status is compared with the pre-set signal quality change index reference threshold to divide the wireless signal quality. The division steps are as follows: If the signal quality change index is greater than a preset signal quality change index reference threshold, the wireless signal quality is classified as a low quality signal; If the signal quality change index is less than or equal to a preset signal quality change index reference threshold, the wireless signal quality is classified as a high-quality signal.

7. A 4G communication module data transmission method according to claim 6, characterized in that: When a wireless signal is classified as a low-quality signal, the specific steps for shortening the split length of the data unit and improving the reliability and stability of transmission are as follows: When the wireless signal is classified as a low-quality signal, the split length of the data unit is dynamically adjusted according to the result of the signal quality change index AQCVI. When the signal quality change index AQCVI is greater than the preset signal quality change index reference threshold, the data unit is split into smaller fixed-length data packets to adapt to the quality fluctuation of the current channel. The adjusted split length is calculated by the following formula: Where, L new，short is the shortened data unit split length, L base is the preset data unit split length, AQCVI ref is the signal quality change index reference threshold, C fade is the channel fading factor, B current is the current available bandwidth, ω and are all exponential adjustment factors. ω is used to control the influence of the signal quality change index AQCVI on the split length. Used to control the impact of channel fading and bandwidth on the split length.

8. A 4G communication module data transmission method according to claim 6, characterized in that: When the wireless signal is classified as a high-quality signal, the specific steps of dynamically lengthening the data unit split length according to the length of the data to be transmitted and improving the overall data transmission efficiency are as follows: When the wireless signal is divided into high-quality signals, the split length of the data unit is dynamically adjusted and lengthened according to the amount of data to be transmitted and the current network status, thereby improving data transmission efficiency, reducing protocol overhead, and making full use of available bandwidth resources. A refined dynamic adjustment model is constructed to dynamically adjust the split length. The calculation expression for lengthening the split length of the data unit is as follows: Where, L new，long is the length of the data unit after the extension, B max is the maximum available bandwidth of the channel, B current is the current available bandwidth, D current is the current amount of data to be transmitted, D max is the maximum amount of data supported to be transmitted, θ, μ and ρ are all exponential adjustment factors, θ is used to control the influence of the signal quality change index AQCVI on the lengthened split length, μ is used to control the influence of bandwidth utilization on the split length, and ρ is used to control the influence of the data amount on the lengthened split length.

Images