129 results about "Viterbi algorithm" patented technology

The invention discloses a method and a system for evaluating the running risk of an automatic driving taxi, and relates to the field of intelligent traffic and automatic driving. Aiming at the problems of insufficient dynamic traffic environment adaptability, rigidity of space decision, lack of vehicle operation rule compatibility and the like in the prior art, the evaluation method comprises the following steps: dynamically dividing an urban road network into hexagonal honeycomb units, adjusting the honeycomb radius according to a real-time travel demand, and selecting key nodes; accessing traffic flow density, conflict event number, communication delay and environment parameters in real time to construct a key node risk index; a hidden Markov chain model is adopted to simulate behavior response of an automatic driving taxi at a key node, an S1 safety state and an S2 risk state are defined, and a Viterbi algorithm is combined to deduce a hidden state probability; and finally, comprehensively judging the risk level according to the risk index and the state probability. According to the method, through multi-source data fusion and scene behavior modeling, dynamic evaluation and control of the operation risk of the automatic driving taxi are realized.

The invention discloses a motor insulation breakdown voltage prediction method based on a continuous hidden Markov model, and the method comprises the steps: taking a plurality of groups of obtained experimental data of an insulating material of a motor in different motor operation time periods as a training set, and enabling the experimental data to comprise capacitance, insulation resistance, dielectric loss, maximum discharge capacity and breakdown voltage; constructing a continuous hidden Markov model by taking the test parameters as an observation sequence and taking the discretized breakdown voltage as a hidden state; performing iterative calculation on parameters of the continuous hidden Markov model through a Baum-Welch algorithm based on the training set, and determining optimal parameters of the continuous hidden Markov model; and based on the continuous hidden Markov model of the optimal parameters, determining a hidden state sequence with the maximum probability under the observation sequence at the current moment through a Viterbi algorithm, and determining the breakdown voltage at the current moment based on the hidden state sequence with the maximum probability. According to the invention, the breakdown voltage of the motor insulating material in the motor operation process can be effectively predicted.

The invention relates to the technical field of speech recognition, and discloses an out-of-set word rejection recognition method and device, equipment and a medium, and the method comprises the steps: carrying out the forced alignment of a current speech segment recognition result and a target command word, and obtaining an alignment path, and the process is based on dynamic planning and a Viterbi algorithm, and is combined with a CTC criterion; determining a key interval according to the alignment path, and judging and counting wrong line counts according to two types of alignment results of a bank phoneme and a target command word phoneme; positioning a key phoneme pair containing position association in combination with the alignment path and the phoneme sequence characteristics of the target command word; calculating the position distance of the key phoneme pair and judging whether the position distance is abnormal; and if the error line count exceeds the set threshold value or the position distance is abnormal, determining the word as an out-of-set word and refusing to identify, otherwise, determining the word as an effective command word. The method does not need to train a model again, is adaptive to low-resource equipment, is simple in calculation, is small in resource occupation, can precisely recognize the similar set external words, reduces the error recognition rate, improves the voice interaction accuracy, and meets the precise control demands of a smart home.

The invention discloses a signaling road network matching method based on a heterogeneous graph convolutional network and an attention mechanism, and belongs to the crossing field of mobile communication and intelligent traffic technologies. According to the method, a multi-relation graph containing a road segment set and a base station sampling point set is constructed, wherein an edge set covers associated edges of base station sampling points and road segments, topological edges of the road segments and time edges of the base station sampling points; performing feature extraction on the multi-relation graph by adopting a 3-5-layer heterogeneous graph convolutional network to obtain low-dimensional vector representation of nodes; dynamically calculating an observation probability and a transition probability based on an attention mechanism; and performing path search by using an improved Viterbi algorithm in combination with a backtracking mechanism and a pruning strategy, and outputting a path with the highest score as a matching result. The method can effectively solve the problems of poor adaptability to complex road networks, insufficient robustness to high positioning error signaling data, low precision caused by matching probability immobilization and the like in the prior art, significantly improves the accuracy and robustness of signaling road network matching in complex scenes, and can be applied to the fields of traffic flow monitoring, travel mode analysis, urban planning and the like.

Systems and methods detect and prevent vishing attacks through an integrated framework combining SIP header customization, STIR / SHAKEN frameworks, AI / ML analysis, and real-time speech analysis using the Viterbi algorithm. The system begins with call initiation, embedding authentication information in the SIP header. The SIP data is transmitted and verified using STIR / SHAKEN frameworks, ensuring the authenticity of the caller's identity. Verified data is cross-referenced with third-party databases and analyzed by an AI / ML engine to detect anomalies. If potential fraud is detected, the call is blocked, and the customer is notified. Calls that pass initial checks are further analyzed using the Viterbi algorithm, which converts speech to text and identifies suspicious patterns. An anomaly pattern detector processes the converted text to detect vishing indicators, terminating the call if a match is found. This multi-layered approach ensures robust protection against vishing, enhancing the security and reliability of voice communications while safeguarding users from fraud.

The invention relates to a non-intrusive load decomposition method based on a hidden Markov model, and the non-intrusive load decomposition method based on the hidden Markov model is characterized in that the hidden Markov model is expanded into a factor hidden Markov model which is applied to non-intrusive load monitoring; wherein the task of non-intrusive load monitoring corresponds to a parameter learning problem and a state inference problem in the hidden Markov model; a public data set is used for experimenting a load decomposition method based on a hidden Markov model to confirm a decomposition effect, the hidden Markov model (HMM) is used for solving a non-intrusive load decomposition problem, a Viterbi algorithm is adopted for step-by-step solving, the Viterbi algorithm searches for an optimal state sequence by traversing all state transition conditions, and a non-intrusive load decomposition problem is solved. And applying dynamic programming to efficiently solve the state sequence with the maximum probability, and speculating the load state condition at the t + 1 moment according to the load condition at the t moment and the total power at the t + 1 moment in the step-by-step decomposition process.

The invention discloses a self-defined voice wake-up word detection method based on a finite-state machine. The method comprises the following steps: training a Transform network model by using an open-source Chinese voice data set on the Internet; performing framing windowing processing on the input voice signal, and extracting corresponding features; the extracted features are input into a Transform network model, and a frame probability tensor is generated; configuring target wake-up word content, obtaining an initial and final pinyin sequence corresponding to the target wake-up word, and constructing a finite-state machine decoding network; processing the frame probability tensor to generate a new frame probability tensor; inputting the new frame probability tensor into a finite-state machine decoding network, operating a Viterbi algorithm, and searching an optimal phoneme path; and calculating an average value of all phoneme state scores, comparing the average value with a preset threshold value, and activating the wake-up word when the average value exceeds the preset threshold value. According to the method, the Transform model and the finite-state machine technology are combined, self-defined voice wake-up word detection is achieved, and the method has the advantages of being flexible in configuration, low in power consumption operation, high in precision and high in robustness.

The invention provides a visual positioning method and device for a tunnel scene and a medium, and belongs to the field of visual localization, and the method comprises the steps: obtaining image data of a tunnel region shot by an imaging assembly on a test vehicle; the image data comprises a color image and a depth image; acquiring motion data acquired by an inertial sensor on the test vehicle; based on an oFAST algorithm, an FREAK algorithm, a non-maximum suppression algorithm and a random sampling consistency algorithm, performing optimization processing on the feature points extracted from the color image to obtain a uniformly distributed feature point set; based on a Viterbi algorithm and a second-order hidden Markov model, screening out a reference image sequence according to the color image, the depth image and the motion data; based on the improved target detection model, tunnel scene features in the tunnel are extracted according to the color image; and obtaining the current pose of the test vehicle according to the tunnel scene features, the reference image sequence and the depth image. According to the invention, the positioning precision in the tunnel scene is improved.

The invention discloses an online detection system and method for optical indexes of a hollow-core optical fiber, and relates to the technical field of optical performance detection of optical fibers, and the system comprises an optical fiber connection module, an optical performance detection module, a production linkage module and a data interaction module. The method comprises the following steps: welding the inner end of a production optical fiber with a system at one time; the industrial personal computer reads PLC production data and generates a test trigger condition; when the conditions are met, controlling the OTDR to test the optical indexes of the optical fiber; carrying out segment processing on the OTDR curve by using an improved Viterbi algorithm so as to identify fault features; performing automatic environment compensation based on an attenuation reference model established by the LSTM neural network; and finally, uploading the analysis data to a manufacturing execution system. According to the invention, on-line, real-time and accurate detection of the optical indexes of the hollow-core fiber is realized, traditional multi-time welding is replaced by one-time welding, the detection efficiency is greatly improved, and the problems of poor sealing performance, low efficiency and low data reliability in the prior art are solved.

The invention relates to a rotary machine composite fault detection method and system under a time-varying working condition, and belongs to the technical field of rotary machine fault diagnosis. The method comprises the steps of collecting a non-stationary vibration signal under a variable speed working condition, fusing a third-order Renyi entropy and kinematics prior to construct an adaptive time-frequency enhancement model, and outputting weighted time-frequency distribution; global optimization is carried out through a priori constraint Viterbi algorithm, and an instantaneous frequency ridge line is extracted under the condition of no rotating speed sensor; performing angular domain resampling by taking the ridge line as a phase reference, and converting a non-stationary signal into an angular domain stationary signal; constructing an extended difference mode decomposition convex optimization model, and iteratively solving an optimal difference spectrum; and determining a frequency spectrum segmentation threshold value based on self-adaptive change point analysis, performing decoupling separation and time domain reconstruction on the optimal difference spectrum, and completing composite fault diagnosis through feature comparison. The method does not need an external tachometer, can adaptively extract and separate the composite fault features under the variable speed working condition, is high in diagnosis accuracy, and is good in engineering application value.

The invention provides a method, device and equipment for promoting retention through interaction scene prediction and a storage medium, and the method comprises the steps: obtaining a current multi-modal interaction data stream between users in an interaction scene, carrying out the preprocessing of noise filtering, framing processing, size normalization, time aggregation and the like, extracting emotion features, and generating a real-time multi-dimensional observation vector sequence. The sequence captures complementarity and time sequence dependence of multi-modal information, and overcomes limitation of single-modal static analysis. The observation vector sequence is input into a pre-trained hidden Markov model, and the model is trained based on historical sequences and defines a hidden state set, an initial probability, a transfer matrix and an emission probability; the optimal emotional state path at the current moment is obtained through Viterbi algorithm reasoning, and probability modeling and dynamic prediction of emotion transfer uncertainty are achieved. The current emotion state is extracted from the path, the mimicry representation of the virtual pet is driven to be displayed on the interactive interface, visual feedback is formed, emotion connection is enhanced, and the user retention rate is increased.

The invention discloses a business state prediction method and device, electronic equipment and a storage medium. The system comprises the following steps: acquiring a log historical data set of a system and an initial observation probability of each log; constructing an initial hidden Markov model; according to a forward algorithm module, the initial observation probability and the model initial parameters, calculating the observation anomaly probability of any log observation sequence to obtain an anomaly probability distribution result of the log; performing weighted fusion on the abnormal probability distribution result, and determining the service state sequence length of the system according to the weighted abnormal probability distribution result; acquiring an initial service state of the system; according to the Viterbi algorithm module, the model initial parameter and the service state sequence length, performing backtracking analysis on the initial service state to obtain a target service state sequence of the system; through the target hidden Markov model, fault recovery and intervention are brought into the event and even before the event from post-event remedy, and the maintenance capability and the service continuity guarantee capability of the system are improved.

The invention relates to the technical field of nonferrous metal beneficiation intelligent control, and discloses a lead zinc ore flotation process state evolution modeling method. The method comprises the following steps: collecting multi-source heterogeneous data of a flotation process, preprocessing the multi-source heterogeneous data, then obtaining a plurality of key features, and constructing a feature matrix according to the plurality of key features; constructing a hidden Markov model based on the hidden state of the flotation process, inputting the feature matrix to the hidden Markov model, optimizing parameters of the hidden Markov model through a Baum-Welch algorithm, and screening evolution models in a training process in combination with a Bayesian information criterion; and acquiring real-time multi-source heterogeneous data of the flotation process and acquiring a corresponding feature matrix, inputting the feature matrix into the evolution model, and acquiring an optimal process state sequence in combination with a dimensional bit algorithm. The problem that an existing beneficiation model cannot effectively and accurately predict the flotation process state is solved.

The invention provides a bearing temperature vibration signal fault evolution path intelligent identification method and system, and relates to the technical field of fault identification, and the method comprises the steps: carrying out the collection and time sequence alignment of bearing temperature, vibration signals and working condition parameters, and constructing a multi-source heterogeneous data matrix; multi-dimensional features are extracted after self-adaptive noise reduction processing; constructing a fault evolution mechanism causal association network by adopting a causal inference model; performing time sequence clustering based on the network to divide health state clusters, and establishing a state transition model by using a hidden Markov model; and finally, decoding by using a Viterbi algorithm to obtain an optimal fault evolution path. The method can effectively identify the bearing fault evolution law and improve the prediction precision.

The invention relates to an MIMO windowing OFDM communication system in a high-speed mobile scene and a space diversity utilization method thereof, and the method comprises the steps: dividing a system bandwidth into K subcarriers, and transmitting a signal through M transmitting antennas; modulating a frequency domain signal transmitted by a # imgabs0 # transmitting antenna through OFDM (Orthogonal Frequency Division Multiplexing) to obtain a time domain signal, multiplying the time domain signal by a window function and adding a cyclic prefix; coding the transmitting antenna signal according to the space-frequency repetition code, and transmitting the transmitting antenna signal to a receiving end through a dual-selective channel; receiving signals through N receiving antennas at a receiving end; decoding the received signal by adopting an MIMO (Multiple Input Multiple Output) frequency-varying Viterbi decoding scheme; and iteratively executing solution by using a surround viterbi algorithm in the decoding process until the result converges or reaches the maximum iteration times, thereby obtaining a decoding result and realizing space diversity utilization. By adopting the method, a windowing orthogonal frequency division multiplexing scheme can be expanded to a multi-input multi-output system in a high-speed moving scene, and space diversity is utilized.

The invention discloses an SOQPSK demodulation performance monitoring method and system based on upper computer visual feedback. An SOQPSK signal is demodulated by adopting a maximum likelihood sequence detection method based on a Viterbi algorithm; removing the influence of a phase pulse function through matched filtering processing, extracting I / Q data of a fixed sampling point in each symbol period, carrying out time domain superposition, and generating eye pattern data in real time; based on the mapping relation between the grid state variable of the SOQPSK demodulator and the CPM phase state, the maximum survival state in the Viterbi algorithm is used as a phase value, and constellation diagram data is generated in real time in combination with the survival path measurement increment; and packaging the eye diagram data and the constellation diagram data to an upper computer for display, performing performance parameter evaluation and feeding back to a demodulator. The FPGA generates eye diagram and constellation diagram data in real time and sends the eye diagram and constellation diagram data to the upper computer for display and performance evaluation, and visual performance monitoring is provided. And parameters of the demodulator are adjusted in real time, so that a closed loop is integrally formed.

The application belongs to the technical field of underwater acoustic, and discloses a line spectrum detection front tracking method and a multi-target distinguishing method based on signal space transformation. The track energy accumulation value on the slice obtained after signal space transformation is constructed as a cost function in a particle swarm parameter optimization algorithm, so that the complexity of slice search in the transformation space is reduced, and the calculation amount is greatly reduced. Meanwhile, a hidden Markov model is established on the slice, and a Viterbi algorithm is used to track the track, compared with the method of using a Radon algorithm to integrate on the slice, the selection of the integration path is more accurate, and the processing gain is improved. The method improves the detection probability, tracking accuracy and parameter optimization speed of the uniform motion target under low signal-to-noise ratio. The searched parameters can be used as the characteristics of the target to distinguish the target, and multi-target distinguishing is realized.

This application proposes a GPS-based method, device, and application for calculating real-time bus traffic conditions, including the following steps: S00, obtaining vehicle GPS data and road segment data for the current period, using the vehicle's road mileage as the actual mileage for the current period; wherein the road segment data is data consisting of a series of connected coordinate points; S10, performing exception processing on the vehicle GPS data to obtain valid vehicle GPS data; S20, using the Viterbi algorithm to match each GPS point to the coordinate point based on the distance between the GPS point in the vehicle GPS data and each coordinate point in the road segment data, and calculating the road segment speed for the current period based on the matching results; S30, using the average speed of each vehicle passing through the road segment as the average speed of the current road segment and outputting it. This application can produce a more accurate analysis of real-time bus traffic conditions.

The invention relates to a robot positioning method and device in a degraded tunnel scene, and belongs to the technical field of robot positioning.The robot positioning method in the degraded tunnel scene comprises the steps that prior path information and road sign structure position information are obtained from a tunnel design drawing, constructing a path likelihood model, a relative distance constraint model and a pose constraint model; fusing the path likelihood model, the relative distance constraint model and the pose constraint model based on a particle filtering algorithm to obtain an odometer pose calculation result of the robot; extracting a multi-fine-grained feature from a tunnel side view image collected by a camera arranged on the robot, and carrying out loopback detection based on a Viterbi algorithm and the multi-fine-grained feature to obtain a loopback detection pose calculation result of the robot; and determining the optimal estimation of the pose of the robot based on the odometer pose calculation result and the loop-back detection pose calculation result. According to the invention, high-precision positioning of the robot in a degraded tunnel scene can be realized.

The invention relates to the technical field of reservoir characterization, in particular to a reservoir characterization method based on a Viterbi algorithm. The method comprises the following steps: step 1, describing a spatial change mode of a prior probability of a reservoir facies by adopting a Markov chain, and constructing a reservoir parameter joint probability distribution model with longitudinal continuity; 2, determining a maximum posteriori estimation problem of reservoir parameters by combining the reservoir parameter joint probability distribution model according to a maximum posteriori estimation theory; and step 3, based on a Viterbi algorithm, developing and solving the problem in the step 2, and establishing a corresponding reservoir characterization model. According to the method, on the basis of the Viterbi algorithm, rapid estimation of multi-target parameters of the hidden Markov model is achieved, a brand-new reservoir characterization model is established for the actual problem of quantitative prediction of reservoir properties, longitudinal geological continuity is introduced, and synchronous prediction of multiple reservoir parameters is achieved.

The application provides a bearing temperature and vibration signal fault evolution path intelligent identification method and system, relates to the technical field of fault identification, and comprises the following steps: collecting and time-aligning bearing temperature, vibration signals and working condition parameters to construct a multi-source heterogeneous data matrix; extracting multi-dimensional features after adaptive noise reduction processing; constructing a fault evolution mechanism causal correlation network by using a causal inference model; performing time series clustering to divide health state clusters based on the network and establishing a state transition model by using a hidden Markov model; and finally applying a Viterbi algorithm to decoding to obtain an optimal fault evolution path. The method can effectively identify the bearing fault evolution law and improve the prediction accuracy.

The invention relates to the field of algorithms, in particular to a Viterbi algorithm searching method and device, a medium and a program product, and the method comprises the steps: respectively calculating the cost from each state at an n moment to a state 1 at an (n + 1) moment, and obtaining an optimal preorder state; calculating the cost from the state 1 at the n moment to the state 2 to k at the n + 1 moment; respectively calculating the cost from the state 2 at the n moment to the state 2 at the k to the n + 1 moment to obtain an optimal preorder state; respectively calculating the cost from the state 2 at the n moment to the state 3 at the n + 1 moment to the k; and so on, the optimal preorder state of all the states at the n + 1 moment is found. According to the search method, the number of times of search calculation is not increased, but the memory access mode and the cache efficiency are greatly improved, and an experimental structure shows that the calculation time of a new search order is reduced by 30%.

This invention discloses an automatic construction method for catenary support equipment archives based on multi-frame visual sequence alignment, belonging to the field of railway operation and maintenance technology. The method includes: acquiring trackside inspection videos; segmenting the frame data sequence based on the area change trajectory of the pole boundary frame to obtain several pole trajectories compressed into candidate pole nodes; extracting representative frames, kilometer marker area characters, license plate area characters, and pole type labels to obtain a candidate pole node sequence arranged by time; acquiring pre-stored real structured trackside data; constructing a key point dictionary based on the uniqueness of kilometer markers; generating a real pole number sequence arranged in line order; scoring all video frames covered by the candidate pole nodes frame by frame; using the Viterbi algorithm to solve the globally optimal alignment path from the candidate pole node sequence to the real pole number sequence; and generating catenary support equipment archives. This invention solves the problems of low accuracy in catenary support equipment archive construction, inability to effectively handle missed detections, false detections, and disordered sequences.

A multi-source fusion positioning method for underground pipeline network robots based on factor graph weight adaptation, which is a combined positioning method for inertial navigation positioning, extremely low frequency positioning, radio frequency positioning, and multi-source sensors of underground pipeline network maps. First, according to the INS / UWB / ELF measurement and error models, INS / UWB / ELF factor nodes and their cost functions are constructed. Through the measurement errors at each moment, the weights of each sensor in the factor graph are dynamically updated and adjusted, and a sliding window constraint is applied to reduce the dimension of the navigation prior information. According to the optimal estimation sequence of the preliminary fusion, a hidden Markov model is established and solved using the Viterbi algorithm for map matching to obtain the multi-source fusion positioning result, realizing high-precision positioning of the underground pipeline network robot underground.

The invention relates to the technical field of radio signal processing, in particular to an unmanned aerial vehicle flight path positioning system based on radio detection. The method comprises the steps that a signal acquisition unit acquires radio detection information propagated in a space based on an antenna capture module; the information processing unit is based on the radio detection information acquired by the signal acquisition unit and performs data processing on the radio detection information; the signal demodulation unit restores instructions and state information sent by the unmanned aerial vehicle through a Viterbi algorithm based on the radio detection information processed by the information processing unit; the detection positioning unit determines the position and the flight trajectory of the unmanned aerial vehicle based on the trajectory positioning module and the position positioning module. Radio detection information in a space is accurately collected through the antenna capturing module, angle measurement is performed and spatial resolution is improved in combination with a multi-antenna array technology, and high-precision positioning of the position of the unmanned aerial vehicle can be realized.

The application provides a farmland feature extraction method based on texture analysis, and belongs to the technical field of remote sensing data processing.The application extracts multi-direction and multi-scale gray level co-occurrence matrix texture feature parameters by constructing a block integral graph to form an initial texture feature vector, constructs a Riemann kernel function matrix by using Riemann geodesic distance calculation, and realizes texture feature dimension reduction by principal component projection in a Riemann manifold tangent space.The reduced texture feature vector is encoded as a genetic algorithm chromosome individual, and the optimal texture feature subset is obtained by fitness evaluation and evolution operation.The optimal texture feature subset of multi-temporal images is used to construct a hidden Markov model, and the optimal state sequence path is searched by using the Viterbi algorithm to realize farmland feature time sequence recognition.After morphological post-processing, farmland feature data is output, and the problem that high-dimensional feature redundancy leads to low feature classification accuracy during farmland remote sensing image texture feature extraction is solved.

The present application belongs to the technical field of radar, and particularly relates to a bird wing-flapping frequency extraction method based on Doppler radial acceleration. The present application firstly tracks the instantaneous frequency trajectory of the time-frequency analysis result by using the Viterbi algorithm, dynamically searches for the best path, and thus effectively extracts the signal instantaneous frequency; then performs differential processing, converts the instantaneous Doppler information into Doppler radial acceleration; and finally sets an adaptive threshold, counts the wing-flapping times of the bird, and thus realizes the extraction of the wing-flapping frequency of the bird. The present application proposes a bird wing-flapping frequency extraction method based on Doppler radial acceleration, which provides an effective means for the extraction of the wing-flapping frequency of the bird. Compared with the existing bird wing-flapping frequency extraction method, the present method can greatly improve the success rate and precision of the extraction of the wing-flapping frequency of the bird under the condition of low signal-to-noise ratio, and will be helpful for the bird target recognition of the radar.

The present invention discloses a fiber optic nonlinear equalization method, system, device and storage medium, belonging to the field of optical transmission technology. The method includes: obtaining the M-QAM signal sequence received at the receiving end; inputting the M-QAM signal sequence into a trained bidirectional long short-term memory neural network based on a memristor array to output a score matrix; using the Viterbi algorithm to obtain the optimal label sequence corresponding to each input feature sequence according to the score matrix and the training score matrix obtained during the training process. The intermediate label values of all the optimal label sequences form the label values of the M-QAM signal sequence, and the data categories corresponding to the label values are subjected to constellation mapping to obtain the original signal sequence transmitted by the transmitting end, thus completing the nonlinear equalization. The present invention uses a memristor array to implement synaptic weights, greatly reducing the complexity of the network, and uses the bidirectional feature to process the current input information by combining the information of the previous and subsequent sequences, improving the signal quality and transmission performance.

Disclosed are an automatic model parameter selection method and system based on hidden Markov. The method comprises the following steps: in response to a selection instruction input by a user in a visual operation interface, determining selected areas in the model and constructing a selected area set; constructing a hidden state set, wherein the hidden state set comprises adjustable parameters of the model; obtaining a state transition probability matrix and an observation probability matrix; and on the basis of the selected areas and the acquired state transition probability matrix and observation probability matrix, calculating an implicit state sequence by means of a Viterbi algorithm to obtain a predicted parameter combination. According to the technical solution of the present application, on the basis of a hidden Markov algorithm, expert experience is solidified in the state transition matrix and the observation probability matrix, and after a user selects some areas, a parameter combination can be automatically predicted and prompted, and the user is guided to use, thereby reducing the experience threshold for use of the user, and improving the efficiency of parameter selection of technicians.

The present invention provides a multi-private domain visitor portrait sharing and privacy protection routing method based on federated learning, which relates to the technical field of data sharing and privacy protection, including: adding mixed noise to the original visitor portrait data for privacy protection by calculating feature sensitivity and allocating differential privacy budget, encrypting feature vectors by fully homomorphic encryption based on ideal lattice, performing threshold cipher segmentation by using Viterbi algorithm, distributing sub-key components to each private domain environment, constructing a hybrid network structure to extract features, performing federated optimization based on data quality and historical contribution, updating the feature extraction network, performing feature dimensionality reduction by using dual autoencoders, extracting key features and generating user feature fingerprints, grouping users by calculating feature fingerprint similarity, and realizing secure sharing of user similarity information and original visitor portrait data in combination with multi-party secure computing and hierarchical data access strategy.