Remote control message sending method and system

Abstract

The invention discloses a remote control message sending method and system, relates to the technical field of channel detection, and is used for solving the problem of large error when a maximum likelihood estimation algorithm is used for measuring a complex channel, and the method comprises the steps of collecting project data and channel data, analyzing measurement requirements according to the project data, analyzing channel measurement difficulty by using the channel data, and sending a remote control message. Selecting a subspace method estimation algorithm, a maximum likelihood estimation algorithm or a hybrid estimation algorithm according to the measurement requirements and the measurement difficulty, respectively collecting different feedback data according to the selected different algorithms, modulating or optimizing the algorithms according to the different feedback data, and if the subspace method estimation is selected, selecting the maximum likelihood estimation algorithm or the hybrid estimation algorithm; if yes, performing dimension reduction processing on the signal subspace or the noise subspace by using principal component analysis; if the maximum likelihood estimation is selected, the digital modulation technology is switched for simulation, and the technology for obtaining the optimal signal-to-noise ratio through simulation is selected for implementation; if the hybrid estimation algorithm is selected, the optimal switching time point of the algorithm is searched, so that the error of the measurement channel is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of channel detection, and more particularly, to a remote control message sending method and system. Background Art

[0002] Channel detection technology is a very important part of communication systems. It is mainly used to detect and evaluate channel status and provide reliable channel quality feedback for upper-layer protocols. Channel detection technology applied to wireless remote sensing and telemetry can monitor channel noise and interference levels and promptly detect and locate interference sources.

[0003] The existing technology has the following deficiencies:

[0004] In the past, maximum likelihood estimation was widely used in quantum tunneling communication for infinite remote sensing and telemetry. However, maximum likelihood estimation is an offline channel estimation algorithm and is not suitable for channels with large dynamic changes and high complexity. When the measurement channel is more special, the offline channel estimation algorithm has higher measurement costs and greater measurement errors. Summary of the Invention

[0005] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides a remote control message sending method and system, which selects a suitable channel estimation algorithm by analyzing the enterprise's channel measurement requirements and channel measurement difficulty to solve the problems raised in the above-mentioned background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A remote control information transmission system includes: a data acquisition module, an algorithm selection module and a feedback implementation module, and signal connections between the modules;

[0008] The data acquisition module is used to collect project data and obtain real-time channel data, send the collected project data and channel data to the algorithm selection module for processing and analysis, receive the marking signal, collect the corresponding feedback data according to the marking signal, and send the feedback data and marking signal to the feedback implementation module;

[0009] After receiving the project data and channel data, the algorithm selection module performs preprocessing. It uses the LOF algorithm to analyze the channel measurement requirements based on the project data and the MPC algorithm to analyze the channel measurement difficulty based on the channel data. The algorithm selection module automatically selects the measurement estimation algorithm based on the channel measurement requirements and the channel measurement difficulty and marks it. After marking, it transmits the marking signal back to the data acquisition module.

[0010] The feedback implementation module receives the marking signal and feedback data, confirms the algorithm that needs to be adjusted and optimized based on the marking signal, and optimizes the algorithm process or analyzes the best implementation measures based on the feedback data.

[0011] In a preferred embodiment, the algorithm selection module performs the following steps to pre-process the project data and channel data:

[0012] The algorithm selection module processes the enterprise's project data as project samples, where the project data includes the software system life cycle and the software system node deployment density. The enterprise software system life cycle and the corresponding software system node deployment density are normalized. The normalized results of the project sample enterprise software system life cycle are merged into a life cycle dataset, and the normalized results of the corresponding software system node deployment density are merged into a deployment density dataset.

[0013] The channel data includes the power density of the interference signal in the channel and the channel coherence time. The algorithm selection module selects a period of time as the sample time and sets multiple detection points within the sample time. At each detection point, the power density of the interference signal in the channel and the channel coherence time are recorded and merged into a power density data set and a coherence time data set respectively, and the time interval between each detection point is marked.

[0014] In a preferred embodiment, the algorithm selection module uses the LOF algorithm to analyze the channel measurement requirements. The specific steps are as follows:

[0015] Calculate the local outlier factor: Randomly select a data point from the lifecycle dataset, normalize the corresponding software system node deployment density in the deployment density dataset, and calculate the normalized result. Then, sum the randomly selected data point from the lifecycle dataset with the normalized result of the corresponding software system node deployment density to obtain the local outlier factor.

[0016] Calculate the local outlier factor deviation coefficient: Arrange the calculated local outlier factors from small to large, calculate the standard deviation, and use the standard deviation to calculate the local outlier factor deviation coefficient;

[0017] Set the measurement requirement threshold: When the deviation coefficient of a local outlier factor exceeds the deviation threshold, the corresponding local outlier factor will be eliminated, and the remaining local outlier factor will be used as the measurement requirement threshold;

[0018] Analyze channel measurement requirements: If the local outlier factor exceeds the measurement requirement threshold, the current enterprise channel measurement requirement is judged to be high; otherwise, the current enterprise channel measurement requirement is judged to be low.

[0019] In a preferred embodiment, the algorithm selection module analyzes the channel measurement difficulty using the MPC algorithm. The specific steps are as follows:

[0020] Constructing a dynamic equation: Select the interference signal power density or channel coherence time recorded at each detection point within the sample time, construct a dynamic equation based on the time interval t between each detection point, and use the dynamic equation to obtain and mark the calculation results.

[0021] Calculate the average value of the marking results: calculate the average value of the results obtained by using the dynamic equation to calculate the in-channel interference signal power or channel coherence time recorded at each detection point;

[0022] Repeat calculation: adjust the time interval t of each detection point, set N groups to calculate the average value of the marking results respectively;

[0023] Construct the control equation: Select the median of the average value of the marked results to construct the control equation: K = g(p, q), where K is the output of the control equation and g is the number of anomalies set;

[0024] Analyze the channel measurement difficulty: when K=1, it is judged that the channel measurement difficulty is high; when K=0, it is judged that the channel measurement difficulty is low.

[0025] In a preferred embodiment, in constructing the control equation, p and q are compared with the preset power density calculation result threshold and coherence time calculation result threshold. If p exceeds the power density calculation result threshold, it is marked as abnormal; if q is lower than the coherence time calculation result threshold, it is marked as abnormal, and g is set as the number of abnormalities. When the number of p and q abnormalities exceeds g, K is set to 1, otherwise K is set to 0.

[0026] In a preferred embodiment, the algorithm selection module enables the subspace estimation method to measure the channel in the following specific steps:

[0027] Step A1, constructing a channel model;

[0028] Step A2: define the subspace rule, perform SVD decomposition on the received signal, and separate two orthogonal subspaces;

[0029] Step A3, subspace estimation: collect multiple received signals to form a signal matrix, perform singular value decomposition on the signal matrix and solve the channel response matrix.

[0030] In a preferred embodiment, the algorithm selection module enables the maximum likelihood estimation measurement channel in the following specific steps:

[0031] Step B1: construct a channel model;

[0032] Step B2: Define a likelihood rule: Estimate the channel matrix based on the received signal, calculate the likelihood function of the channel matrix, and obtain the maximum likelihood estimate of the channel matrix by maximizing the likelihood function;

[0033] Step B3: Calculate the maximum likelihood estimate: denote the covariance matrix as σ 2 , the calculation formula for maximizing the likelihood function S(M|y) is: where ||y-Mx|| 2is the square of the Euclidean distance between the received signal and the predicted signal, C is the set constant term, and the derivative of S(M|y) is taken and set equal to 0 to calculate the maximum likelihood estimate

[0034] In a preferred embodiment, after receiving the signal to start the subspace estimation algorithm, the feedback implementation module performs dimensionality reduction processing using principal component analysis as follows:

[0035] Recording of received and transmitted signals: merging received signals into a received data set and transmitting signals into a transmitted data set and recording them;

[0036] Set the principal component interval: Arrange the data in the received or sent dataset from small to large, set the principal component threshold interval ratio, and classify the data into principal component data and secondary component data based on the principal component threshold interval ratio. After removing the secondary component data, use the principal component data as the parameters of the received signal vector and the sent signal vector in the subspace estimation algorithm.

[0037] After receiving the maximum likelihood estimation algorithm signal, the feedback real-time module records the channel signal power and noise power and calculates the signal-to-noise ratio, compares the signal-to-noise ratio with the signal-to-noise ratio threshold, and selects the M_QAM or M_PSK digital modulation technology based on the judgment result.

[0038] A remote control message sending method, used to implement the above-mentioned remote control message sending system, is characterized by comprising the following steps:

[0039] Step 1: Collect project data and channel data;

[0040] Step 2: Analyze channel measurement requirements using the LOF algorithm based on project data, analyze channel measurement difficulty using the MPC algorithm based on channel data, and select and mark a measurement estimation algorithm based on the comprehensive channel measurement requirements and channel measurement difficulty.

[0041] Step 3: Receive the marking signal and collect corresponding feedback data according to the marking signal;

[0042] Step 4: Identify the algorithms that need to be adjusted and optimized based on the marked signals and optimize the algorithm process or analyze the best implementation measures based on the feedback data.

[0043] The technical effects and advantages of the remote control information sending method and system of the present invention are as follows:

[0044] The present invention collects project data and channel data, analyzes measurement requirements according to the project data, analyzes channel measurement difficulty using the channel data, selects subspace estimation, maximum likelihood estimation or hybrid estimation algorithm based on the comprehensive measurement requirements and measurement difficulty, collects different feedback data according to different selections, and if the subspace estimation is used, calculates the estimation complexity coefficient according to the feedback data, and if the estimation complexity coefficient exceeds a threshold, performs dimensionality reduction processing on the signal subspace or the noise subspace using principal component analysis; if the maximum likelihood estimation is used, calculates the channel signal-to-noise ratio according to the feedback data, and if the signal-to-noise ratio is lower than the threshold, performs switching simulation on the two digital modulation technologies of M_QAM and M_PSK, and selects the technology that obtains the best signal-to-noise ratio through simulation for implementation; if the hybrid estimation algorithm is used, determines the error situation of the subspace estimation and maximum likelihood estimation channels according to the feedback data, and calculates the optimal switching time point between the subspace estimation and the maximum likelihood estimation according to the error situation. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 This is a structural diagram of a remote control information sending system according to the present invention.

[0046] Figure 2 This is a flow chart of a remote control information sending method of the present invention. DETAILED DESCRIPTION

[0047] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0048] The present invention collects project data and channel data, analyzes measurement requirements according to the project data, analyzes channel measurement difficulty using the channel data, selects subspace estimation, maximum likelihood estimation or hybrid estimation algorithm based on the comprehensive measurement requirements and measurement difficulty, collects different feedback data according to different selections, and if the subspace estimation is used, calculates the estimation complexity coefficient according to the feedback data, and if the estimation complexity coefficient exceeds a threshold, performs dimensionality reduction processing on the signal subspace or the noise subspace using principal component analysis; if the maximum likelihood estimation is used, calculates the channel signal-to-noise ratio according to the feedback data, and if the signal-to-noise ratio is lower than the threshold, performs switching simulation on the two digital modulation technologies of M_QAM and M_PSK, and selects the technology that obtains the best signal-to-noise ratio through simulation for implementation; if the hybrid estimation algorithm is used, determines the error situation of the subspace estimation and maximum likelihood estimation channels according to the feedback data, and calculates the optimal switching time point between the subspace estimation and the maximum likelihood estimation according to the error situation.

[0049] Example 1, a remote control information sending system, such as Figure 1As shown, it includes: a data acquisition module, an algorithm selection module and a feedback implementation module, and signal connections between each module.

[0050] The functions of each module are as follows:

[0051] The data acquisition module collects project data by accessing the enterprise management system and automatically obtains real-time channel data collected by the network analyzer. The project data includes the software system life cycle and the software system node deployment density, and the channel data includes the interference signal power density and channel coherence time in the channel. The data acquisition module sends the collected project data and channel data to the algorithm selection module for processing and analysis and receives the returned marking signal. According to the marking signal, the corresponding feedback data is collected and the collected feedback data and marking signal are sent to the feedback implementation module.

[0052] After receiving the project data and channel data, the algorithm selection module pre-processes them, analyzes the channel measurement requirements based on the project data using the LOF algorithm, and analyzes the channel measurement difficulty based on the channel data using the MPC algorithm. The algorithm selection module automatically integrates the channel measurement requirements and channel measurement difficulty to select the measurement estimation algorithm and mark it. The estimation algorithms include subspace estimation, maximum likelihood estimation and hybrid estimation algorithm. After marking, the marking signal is transmitted back to the data acquisition module.

[0053] The feedback implementation module confirms the algorithm that needs to be adjusted and optimized based on the marking signal and optimizes the algorithm process or analyzes the best implementation measures based on the received feedback data.

[0054] It should be noted that the enterprise management system realizes the automation and digitization of enterprise management by integrating the information and processes of various business departments within the enterprise. Project data can be collected in the enterprise management system. The network analyzer is an instrument commonly used to measure and analyze the network performance of communication equipment, etc. The network analyzer can be used to collect channel data. When the data acquisition module collects feedback data based on the marker signal, if the marker signal is estimated by the subspace method, the feedback data is the evaluation execution time; if the marker signal is estimated by the maximum likelihood method, the feedback data is the signal power and noise power; if the marker signal is a hybrid estimation algorithm, the feedback data is the real-time signal strength.

[0055] The specific steps for the algorithm selection module to preprocess project data and channel data are as follows:

[0056] The algorithm selection module processes a sufficient amount of enterprise project data as project samples, and normalizes the enterprise software system life cycle and the corresponding software system node deployment density. The normalization can be processed using the sigmoid normalization formula: X norm =1 / 1+e -X, where X is the life cycle of the software system in the project sample or the corresponding software system node deployment density data, X norm To normalize the corresponding data, the normalized results of the project sample enterprise software system life cycle and the corresponding software system node deployment density are merged into the life cycle dataset and deployment density dataset respectively.

[0057] The algorithm selection module automatically selects a period of time as the sample time and sets multiple detection points within the sample time. The real-time channel data sent by the data acquisition module is recorded at each detection point for the interference signal power density and channel coherence time in the channel and merged into a power density data set and a coherence time data set respectively. The time interval between each detection point is marked as t.

[0058] It should be noted that the selection of the number of project samples, the selection of sample time, and the number of detection points can be adjusted according to actual conditions. The normalization formulas involved in the above steps are not unique and will not be repeated here.

[0059] The algorithm selection module uses the LOF algorithm to analyze channel measurement requirements based on project data. The specific steps are as follows:

[0060] Calculation of local outlier factor: The algorithm selection module uses the life cycle dataset and deployment density dataset obtained after preprocessing to calculate the measurement coefficient. A data point in the life cycle dataset is randomly selected and marked as a. The result of normalizing the deployment density of the corresponding software system node in the deployment density dataset is marked as b. The local outlier factor is calculated using the weighted formula: c = a + b.

[0061] Calculate the local outlier factor deviation coefficient: Arrange the calculated local outlier factors from small to large, mark the calculated standard deviation as S, and the formula for calculating the local outlier factor deviation coefficient can be: C p =S*exp(S+1) 2 , where C p is the local outlier factor deviation coefficient, and exp is the exponential function form.

[0062] Set the measurement requirement threshold: Calculate the deviation coefficient of the local outlier factor and compare it with the preset deviation threshold. When the deviation coefficient of the local outlier factor exceeds the deviation threshold, the corresponding local outlier factor will be eliminated, and the local outlier factor that has not been eliminated will be used as the measurement requirement threshold.

[0063] Analyze channel measurement requirements: The algorithm selection module normalizes the current enterprise software system life cycle and the corresponding software system node deployment density using the same normalization method and calculates the local outlier factor. If the local outlier factor exceeds the measurement requirement threshold, the current enterprise channel measurement requirement is judged to be high; otherwise, the current enterprise channel measurement requirement is judged to be low.

[0064] It should be noted that the LOF algorithm is an unsupervised learning method for detecting outliers in a data set. It can be used to improve data accuracy and set corresponding thresholds. The deviation thresholds involved in the above steps are set by those skilled in the art based on actual conditions.

[0065] The algorithm selection module uses the MPC algorithm to record the dynamic changes of the interference signal power density and channel coherence time in the channel within the sample time according to the channel data, and then analyzes the abnormality of the channel data from the calculation results to analyze the difficulty of channel measurement. The specific steps are as follows:

[0066] Construct a dynamic equation: Select the interference signal power density recorded at each detection point within the sample time, and construct a dynamic equation based on the time interval t between each detection point. The equation can be: where x i (t) is the interference signal power in the channel recorded at the i-th detection point, u(t) is the control input, is x i (t) is the derivative with respect to time t, f is a function that describes the behavior of the power change of the interference signal in the channel, and can be adjusted according to the set control input. It should be explained that the control input u(t) determines the state change of the power density of the interference signal in the channel over a period of time in the future, and can be set to the difference average value of the power density of the interference signal in the channel recorded at each monitoring point.

[0067] calculate Average value: The average value of the interference signal power in the channel recorded at each detection point is calculated using the dynamic equation and recorded as

[0068] Repeat calculation: adjust the time interval t of each detection point and set N groups to calculate separately The jth calculated Similarly, the calculation result obtained by using the channel coherence time is marked as

[0069] Construct the governing equations: Select and The medians are marked as p and q, and the control equation is constructed: K = g(p, q), where K is the output result of the control equation and g is the number of set anomalies.

[0070] Analyze the channel measurement difficulty: when K=1, it is judged that the channel measurement difficulty is high; when K=0, it is judged that the channel measurement difficulty is low.

[0071] It should be noted that the MPC algorithm is a method of optimizing the system based on the dynamic system model and the prediction of the future state. It can describe the magnitude of an upcoming change in a data point and be used to predict the next trend of a state variable. When constructing the control equation, p and q are compared with preset thresholds for power density and coherence time calculations. If p exceeds the power density threshold, it is marked as an anomaly; if q falls below the coherence time threshold, it is also marked as an anomaly. g can be adjusted based on actual conditions. For example, if g is set to 2, which means that if both p and q are anomalies, K is set to 1.

[0072] The algorithm selection module selects the measurement estimation algorithm function settings as follows:

[0073] If it is judged that the current enterprise channel measurement requirements are high and the channel measurement difficulty is high, the hybrid estimation algorithm is enabled; if it is judged that the current enterprise channel measurement requirements are low and the channel measurement difficulty is low, the subspace estimation method is enabled; otherwise, the maximum likelihood estimation is enabled.

[0074] The algorithm selection module enables the subspace estimation method to measure the channel. The specific steps are as follows:

[0075] Step A1, constructing a channel model, assuming that the channel can be described by a linear discrete time-invariant model: y = Hx + n, where y is the received signal vector, x is the transmitted signal vector, H is the channel response matrix, and n is the noise vector.

[0076] Step A2, define the subspace rule. The received signal y can be expressed as a linear combination of the signal subspace and the noise subspace. The signal subspace is spanned by the column space of the channel response matrix H, and the noise subspace is spanned by the null space of H. By performing SVD decomposition on the received signal y, these two orthogonal subspaces can be separated.

[0077] Step A3, subspace estimation: Collect L received signal vectors y to form a matrix Y, perform singular value decomposition on Y: Y = UΣV^H, where U is the left singular matrix, Σ is the singular value matrix, and V^H is the right singular matrix. Divide the U matrix into signal subspaces U s and noise subspace U n , using the signal subspace U s The channel response matrix H is solved, for example, using the least squares method.

[0078] It should be noted that the singular value matrix Σ is a diagonal matrix obtained by the singular value decomposition SVD in the decomposition, the diagonal elements are the singular values of the original matrix, and the singular values are the square roots of the eigenvalues of the original matrix. The least squares method is a conventional algorithm and will not be given as an example here.

[0079] The specific steps for the algorithm selection module to enable maximum likelihood estimation measurement channel are as follows:

[0080] Step B1: Construct a channel model: y=Mx+n, where y is the received signal vector, x is the transmitted signal vector, M is the channel response matrix, and n is the noise vector.

[0081] Step B2: Define the likelihood rule: Estimate the unknown channel matrix M based on the received signal y and calculate the likelihood function S(M|y) of M. The maximum likelihood estimate of M is obtained by maximizing the likelihood function S(M|y).

[0082] Step B3: Calculate the maximum likelihood estimate Assume that the noise n follows a zero-mean Gaussian distribution and the covariance matrix is σ 2 , that is, the calculation formula of S(M|y) is: where ||y-Mx|| 2 is the square of the Euclidean distance between the received signal y and the predicted signal Mx, C is the set constant term, and the maximum likelihood estimate of M can be obtained by taking the derivative of S(M|y) and setting the derivative equal to 0.

[0083] It should be noted that the hybrid estimation algorithm sets different time periods during detection, switches between the subspace estimation method and the maximum likelihood estimation method to measure the channel respectively.

[0084] After receiving the signal to enable the subspace estimation algorithm, the feedback implementation module collects the algorithm execution time and compares it with the preset expected time threshold. If the subspace estimation algorithm execution time exceeds the expected time threshold, the principal component analysis method is used to perform dimensionality reduction processing as follows:

[0085] Recording of received and transmitted signals: Merges the received signal and the transmitted signal into a received data set and a transmitted data set respectively and records them.

[0086] Set the principal component interval: Arrange the data in the received and sent datasets from smallest to largest, set the principal component threshold interval ratio to z, and classify the data into principal component data and secondary component data based on the principal component threshold interval ratio z. After removing the secondary component data, use the principal component data as the parameters of the received and sent signal vectors in the subspace estimation algorithm. For example, if z is set to 0.3, the median of the received or sent dataset is selected, and the 30% of the data before and after the median is used as the principal component data.

[0087] It should be noted that the principal component threshold interval ratio z is set according to the preset expected time threshold. The principal component analysis method is used to filter data, reduce the algorithm difficulty, and save the running time to below the expected time threshold.

[0088] After receiving the maximum likelihood estimation algorithm signal, the real-time feedback module automatically records the channel signal power and noise power and calculates the signal-to-noise ratio (SNR). Based on the SNR, it selects either M_QAM or M_PSK digital modulation. The SNR is calculated using the recorded channel signal power and noise power using the following formula: SNR = Channel signal power / Noise power. If the calculated SNR exceeds the SNR threshold, M_QAM digital modulation is selected for signal reception; otherwise, M_PSK is selected. The SNR threshold can be set based on expert experience or multiple experiments. For example, setting the SNR threshold to 0.5 is not discussed here.

[0089] It should be noted that M_QAM and M_PSK are two conventional digital modulation technologies. Among them, M_QAM has higher requirements on the signal-to-noise ratio, but can provide higher data transmission speeds. M_PSK has lower requirements on the signal-to-noise ratio, but its spectrum efficiency is not as high as M_QAM. Spectral efficiency refers to data transmission speed.

[0090] When the feedback implementation module receives the signal to enable the hybrid estimation algorithm, it combines the characteristics of the subspace estimation method and the maximum likelihood estimation method to find the switching algorithm point. The feedback implementation module sets a period of time as the simulation time to simulate and measure the channel using the two algorithms respectively and obtains the execution time, channel signal power and noise power of the two algorithms in real time. The signal-to-noise ratio is calculated using the channel signal power and noise power as the signal-to-noise coefficient. If the execution time of the subspace estimation method exceeds the time coefficient threshold, the switching algorithm is the maximum likelihood estimation method, and the signal-to-noise ratio is calculated using the channel signal power and noise power as the signal-to-noise coefficient. If the signal-to-noise coefficient of the maximum likelihood estimation method is lower than the signal-to-noise coefficient threshold, the switching algorithm is the subspace estimation method.

[0091] It should be noted that the setting of the time coefficient threshold and the signal-to-noise coefficient threshold needs to meet the requirement that when the signal-to-noise coefficient of the maximum likelihood estimation method exceeds the signal-to-noise coefficient threshold, the execution time of the subspace estimation method is lower than the time coefficient threshold. When the above conditions are met, switching can be performed. The time coefficient threshold and the signal-to-noise coefficient threshold are set by professional technicians and are not analyzed here. In addition, under low signal-to-noise ratio conditions, the subspace estimation method can maintain a higher estimation accuracy, which is better than the maximum likelihood estimation method. Under high signal-to-noise ratio conditions, the maximum likelihood estimation method can often give more accurate channel estimation results.

[0092] Example 2, a remote control message sending method, such as Figure 2 As shown, the following steps are included:

[0093] Step 1: Collect project data and channel data.

[0094] Step 2: Use the LOF algorithm to analyze the channel measurement requirements based on the project data. Use the MPC algorithm to analyze the channel measurement difficulty based on the channel data. Select the measurement estimation algorithm based on the comprehensive channel measurement requirements and channel measurement difficulty and mark it.

[0095] Step 3: Receive the marker signal and collect corresponding feedback data based on the marker signal.

[0096] Step 4: Identify the algorithms that need to be adjusted and optimized based on the marked signals and optimize the algorithm process or analyze the best implementation measures based on the feedback data.

[0097] The above embodiments may be implemented in whole or in part through software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments may be implemented in whole or in part in the form of a computer program product.

[0098] Those skilled in the art will appreciate that the modules 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 of the technical solution and the invention constraints. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0099] In addition, each functional module in each embodiment of the present application may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.

[0100] 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.

[0101] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A remote control message sending system, comprising a data acquisition module, an algorithm selection module, and a feedback implementation module, with signal connections between the modules; The data acquisition module is used to collect project data and obtain real-time channel data, send the collected project data and channel data to the algorithm selection module for processing and analysis, receive the marking signal, collect the corresponding feedback data according to the marking signal, and send the feedback data and marking signal to the feedback implementation module; After receiving the project data and channel data, the algorithm selection module performs preprocessing. It uses the LOF algorithm to analyze the channel measurement requirements based on the project data and the MPC algorithm to analyze the channel measurement difficulty based on the channel data. The algorithm selection module automatically selects the measurement estimation algorithm based on the channel measurement requirements and the channel measurement difficulty and marks it. After marking, it transmits the marking signal back to the data acquisition module. The feedback implementation module receives the marking signal and feedback data, confirms the algorithm that needs to be adjusted and optimized based on the marking signal, and optimizes the algorithm process or analyzes the best implementation measures based on the feedback data.

2. A remote control message sending system according to claim 1, characterized in that: The specific steps for the algorithm selection module to preprocess project data and channel data are as follows: The algorithm selection module processes the enterprise's project data as project samples, where the project data includes the software system life cycle and the software system node deployment density. The enterprise software system life cycle and the corresponding software system node deployment density are normalized. The normalized results of the project sample enterprise software system life cycle are merged into a life cycle dataset, and the normalized results of the corresponding software system node deployment density are merged into a deployment density dataset. The channel data includes the power density of the interference signal in the channel and the channel coherence time. The algorithm selection module selects a period of time as the sample time and sets multiple detection points within the sample time. At each detection point, the power density of the interference signal in the channel and the channel coherence time are recorded and merged into a power density data set and a coherence time data set respectively, and the time interval between each detection point is marked.

3. A remote control message sending system according to claim 1, characterized in that: The algorithm selection module uses the LOF algorithm to analyze channel measurement requirements. The specific steps are as follows: Calculate the local outlier factor: Randomly select a data point from the lifecycle dataset, normalize the corresponding software system node deployment density in the deployment density dataset, and calculate the normalized result. Then, sum the randomly selected data point from the lifecycle dataset with the normalized result of the corresponding software system node deployment density to obtain the local outlier factor. Calculate the local outlier factor deviation coefficient: Arrange the calculated local outlier factors from small to large, calculate the standard deviation, and use the standard deviation to calculate the local outlier factor deviation coefficient; Set the measurement requirement threshold: When the deviation coefficient of a local outlier factor exceeds the deviation threshold, the corresponding local outlier factor will be eliminated, and the remaining local outlier factor will be used as the measurement requirement threshold; Analyze channel measurement requirements: If the local outlier factor exceeds the measurement requirement threshold, the current enterprise channel measurement requirement is judged to be high; otherwise, the current enterprise channel measurement requirement is judged to be low.

4. A remote control message sending system according to claim 1, characterized in that: The algorithm selection module analyzes the channel measurement difficulty using the MPC algorithm. The specific steps are as follows: Constructing a dynamic equation: Select the interference signal power density or channel coherence time recorded at each detection point within the sample time, construct a dynamic equation based on the time interval t between each detection point, and use the dynamic equation to obtain and mark the calculation results. Calculate the average value of the marking results: calculate the average value of the results obtained by using the dynamic equation to calculate the in-channel interference signal power or channel coherence time recorded at each detection point; Repeat calculation: adjust the time interval t of each detection point, set N groups to calculate the average value of the marking results respectively; Construct the control equation: Select the median of the average value of the marked results to construct the control equation: K = g(p, q), where K is the output of the control equation and g is the number of anomalies set; Analyze the channel measurement difficulty: when K=1, it is judged that the channel measurement difficulty is high; when K=0, it is judged that the channel measurement difficulty is low.

5. A remote control message sending system according to claim 4, characterized in that: In constructing the control equation, p and q are compared with the preset thresholds of the power density calculation result and the coherence time calculation result. If p exceeds the threshold of the power density calculation result, it is marked as abnormal; If q is lower than the threshold of the coherence time calculation result, it is marked as an anomaly and g is set to the number of anomalies. If the number of anomalies of p and q exceeds g, K is set to 1, otherwise K is set to 0.

6. A remote control message sending system according to claim 1, characterized in that: The algorithm selection module enables the subspace estimation method to measure the channel. The specific steps are as follows: Step A1, constructing a channel model; Step A2: define the subspace rule, perform SVD decomposition on the received signal, and separate two orthogonal subspaces; Step A3, subspace estimation: collect multiple received signals to form a signal matrix, perform singular value decomposition on the signal matrix and solve the channel response matrix.

7. A remote control message sending system according to claim 1, characterized in that: The specific steps for the algorithm selection module to enable maximum likelihood estimation measurement channel are as follows: Step B1: construct a channel model; Step B2: Define a likelihood rule: Estimate the channel matrix based on the received signal, calculate the likelihood function of the channel matrix, and obtain the maximum likelihood estimate of the channel matrix by maximizing the likelihood function; Step B3: Calculate the maximum likelihood estimate: denote the covariance matrix as σ 2 , the calculation formula for maximizing the likelihood function S(M|y) is: where ||y-Mx|| 2 is the square of the Euclidean distance between the received signal and the predicted signal, C is the set constant term, and the derivative of S(M|y) is taken and set equal to 0 to calculate the maximum likelihood estimate 8. A remote control message sending system according to claim 1, characterized in that: After receiving the signal to start the subspace estimation algorithm, the feedback implementation module uses the principal component analysis method to perform dimensionality reduction processing as follows: Recording of received and transmitted signals: merging received signals into a received data set and transmitting signals into a transmitted data set and recording them; Set the principal component interval: Arrange the data in the received or sent dataset from small to large, set the principal component threshold interval ratio, and classify the data into principal component data and secondary component data based on the principal component threshold interval ratio. After removing the secondary component data, use the principal component data as the parameters of the received signal vector and the sent signal vector in the subspace estimation algorithm. After receiving the maximum likelihood estimation algorithm signal, the feedback real-time module records the channel signal power and noise power and calculates the signal-to-noise ratio, compares the signal-to-noise ratio with the signal-to-noise ratio threshold, and selects the M_QAM or M_PSK digital modulation technology based on the judgment result.

9. A remote control message sending method, based on a remote control message sending system according to any one of claims 1 to 8, characterized in that: The following steps are involved: Step 1: Collect project data and channel data; Step 2: Analyze channel measurement requirements using the LOF algorithm based on project data, analyze channel measurement difficulty using the MPC algorithm based on channel data, and select and mark a measurement estimation algorithm based on the comprehensive channel measurement requirements and channel measurement difficulty. Step 3: Receive the marking signal and collect corresponding feedback data according to the marking signal; Step 4: Identify the algorithms that need to be adjusted and optimized based on the marked signals and optimize the algorithm process or analyze the best implementation measures based on the feedback data.

Images