436results about "Modulation type identification" patented technology

The invention discloses an unmanned aerial vehicle signal identification method and system in a complex electromagnetic environment, and belongs to the field of unmanned aerial vehicles. According to the invention, the form of the liquid metal channel of the liquid metal antenna is adjusted to adapt to the target frequency band, and regional electromagnetic signals are collected in all directions. And generating spatial distribution data including the unmanned aerial vehicle and the interference direction based on the phase difference and preset features. The deformation of the liquid metal droplets is regulated and controlled through the electrowetting voltage, so that the antenna main lobe is aligned with the unmanned aerial vehicle, and the side lobe suppression area covers an interference source. Then, a target direction signal set is collected, and time domain fluctuation, frequency domain distribution and spatial domain direction multi-dimensional features are extracted; and inputting the multi-dimensional features into a joint probability model constructed by a Bayesian network, and outputting posterior probability information. Markov chain Monte Carlo self-adaptive sampling is utilized, the classification confidence coefficient is calculated through iterative convergence, the signal category is judged according to a threshold value, and the unmanned aerial vehicle signal recognition accuracy can be improved in the complex electromagnetic environment.

Accordingly, there are disclosed herein receivers and receiving methods that provide a graceful transition from PAM2 to PAM4 signaling. One illustrative method includes: negotiating a link speed having PAM4 signaling; performing adaption of at least one gain or filter coefficient during PAM2 signaling; switching to PAM4 detection before receiving PAM4 signaling; disabling said adaptation before said switching to PAM4 detection; detecting PAM4 signaling using at least one statistic of detected PAM4 symbols; and enabling said adaptation after PAM4 signaling is detected. Another illustrative method includes: negotiating a link speed having PAM4 signaling; adapting at least one of gain and filter coefficients during PAM2 signaling; monitoring for a change in at least one signal characteristic while performing PAM2 detection; and transitioning to PAM4 detection after detecting said change.

The invention discloses an unmanned aerial vehicle radio communication frequency band countering method for automatic frequency identification and frequency writing, and relates to the technical field of unmanned aerial vehicle interference, and the method comprises a frequency band identification module which is used for scanning a 300MHz-6GHz frequency band in real time and identifying an enemy unmanned aerial vehicle communication signal; the signal analysis module is connected with the frequency band identification module, comprises a digital signal processor and an FPGA (Field Programmable Gate Array), and is used for analyzing signal modulation types and spectrum characteristics; the automatic frequency writing module is used for adjusting an interference frequency band through a direct digital frequency synthesis technology according to an analysis result; the multi-frequency-band interference transmitter supports synchronous interference in frequency bands of 1.2 GHz, 2.4 GHz, 5.2 GHz and 5.8 GHz; the intelligent control terminal integrates an adaptive interference algorithm, and the response time is less than or equal to 1 second. According to the unmanned aerial vehicle radio communication frequency band countering method capable of automatically identifying and writing the frequency, the radio communication frequency band used by the enemy unmanned aerial vehicle is automatically identified, and the corresponding frequency band is written in real time for accurate interference, so that the interference effect of an interference system facing the enemy unmanned aerial vehicle is improved.

The invention discloses a satellite signal modulation recognition method based on an enhanced multi-scale feature attention network, and the method comprises the steps: obtaining a to-be-recognized satellite signal, inputting the to-be-recognized satellite signal into a trained enhanced multi-scale feature attention network, and obtaining a satellite signal modulation recognition result. The training process of the enhanced multi-scale feature attention network comprises the following steps: acquiring a satellite communication data set, and performing sample expansion to obtain an augmented sample; performing noise removal processing and feature extraction on the augmented sample to obtain denoised data features; inputting the de-noised data features into a global context sensing module, extracting modulation features of different frequencies and time scales, and performing feature fusion on the modulation features and the de-noised data features to obtain fusion features; and performing feature reconstruction and channel recovery on the fusion features, inputting the fusion features into a Softmax classifier, performing classification prediction, calculating classification loss, and optimizing network parameters through the classification loss. Under a complete blind recognition scene, the accuracy of a satellite signal modulation recognition result can be obviously improved.

The invention discloses a semi-supervised learning method based on a self-adaptive threshold value, belongs to the field of radio communication, and aims to extract general semantic features of signals through comparative learning so as to improve generalization of downstream modulation identification tasks. A self-adaptive threshold mechanism is introduced to generate a high-quality pseudo tag, and confirmation deviation is reduced through additional category information, so that the robustness of the model is improved; besides, a hierarchical encoder capable of learning Fourier filtering is designed to capture multi-scale semantic features of large-scale long-sequence signals, improve the calculation efficiency of a model and research and verify an algorithm based on a plurality of large-scale data sets, so that the accuracy of automatic modulation recognition is remarkably improved, and particularly, under the condition of less label data, the accuracy of automatic modulation recognition is greatly improved. Compared with a mainstream method, the method has higher generalization ability and robustness, and can effectively cope with signal diversity and noise interference in a real communication environment.

The invention discloses a radio modulation signal identification method and system based on a hybrid neural network, and the method comprises the steps: obtaining a radio modulation signal data set, and dividing the radio modulation signal data set into a training set, a test set and a verification set according to a proportion; the method comprises the following steps of: constructing a hybrid neural network model on the basis of a frequency interaction module (FIM), an interaction gate Transform and a Bi-LSTM (Bidirectional Long Short Term Memory); and training and verifying the hybrid neural network model through the training set and the verification set to obtain an identification model, and inputting the test set into the identification model to complete identification of the radio modulation signal in the test set. Denoising is carried out through FIE, the global feature extraction capability of Transform and the time sequence feature extraction capability of Bi-LSTM are effectively combined, the recognition precision is effectively improved through reasonable design, meanwhile, the robustness of noise influence is enhanced, and the model can complete accurate recognition and classification of modulation signals in a complex noise environment.

Apparatus and methods for identifying a wireless signal-emitting device are disclosed. The apparatus is configured to sense and measure wireless communication signals from signal-emitting devices in a spectrum. The apparatus is operable to automatically detect a signal of interest from the wireless signal-emitting device and create a signal profile of the signal of interest; compare the signal profile with stored device signal profiles for identification of the wireless signal-emitting device; and calculate signal degradation data for the signal of interest based on information associated with the signal of interest in a static database including noise figure parameters of a wireless signal-emitting device outputting the signal of interest. The signal profile of the signal of interest, profile comparison result, and signal degradation data are stored in the apparatus.

The invention relates to a signal-to-noise ratio adaptive signal modulation identification method. The method comprises the following steps: inputting IQ information and a signal-to-noise ratio (SNR) estimated value of a signal into a signal modulation identification model, and identifying to obtain a modulation mode of the signal; the signal modulation identification model is an MCLDNN of which the feature fusion module is provided with an SNR attention module; the SNR attention module can receive the SNR estimated value and obtain a signal-to-noise ratio attention weight according to the SNR estimated value; receiving a signal multi-channel fusion feature map output by the multi-channel fusion convolutional layer of the MCLDNN; weighting the signal multi-channel fusion feature map by using a signal-to-noise ratio attention weight, connecting the weighted signal multi-channel fusion feature map with a signal multi-channel fusion feature map residual error, and outputting a multi-channel fusion feature combined with SNR attention; and the multi-channel fusion features combined with the SNR attention are input to a time sequence module of the MCLDNN.

The invention relates to the technical field of dual-mode interphones, and discloses an AI-based dual-mode interphone satellite communication noise reduction method and system, and the method comprises the steps: carrying out the orbit equation solving of satellite ephemeris data, and obtaining a satellite elevation angle and a channel fading parameter; receiving a first voice signal through a dual-mode interphone, and identifying a modulation mode according to the first voice signal; calculating a first noise reduction mask according to the channel fading parameter and the modulation mode; compressing the standard noise reduction neural network to form a lightweight noise reduction model; and starting an FM spectrum subtraction branch or a TETRA filtering branch in the lightweight noise reduction model according to the modulation mode, and performing frequency domain processing on the first noise reduction mask and the first voice signal to obtain a second voice signal, thereby providing stable and reliable real-time voice noise reduction capability for the portable dual-mode interphone in a dynamic satellite environment.

The invention relates to the technical field of data intelligent testing and traceability, and provides a civil aviation radio parameter intelligent testing and traceability method and system, and the method comprises the steps: obtaining target spectrum feature data obtained by scanning a target frequency band through a phased-array antenna array in a dynamic spectrum interference environment, and adjusting the phased-array antenna array; and converting a civil aviation radio receiving signal in the target spectrum data after anti-interference optimization into time-frequency domain data, generating a signal spectrum, generating a modulation parameter characteristic and a frequency deviation parameter characteristic, and generating a civil aviation radio parameter traceability result containing an equipment identity identifier and a parameter deviation value. According to the invention, through optimization adjustment and signal feature analysis of the phased-array antenna in a dynamic spectrum interference environment, the problem of unstable anti-interference effect in the prior art is solved; the intelligent level and the practicability of the whole system are not high, and intelligent testing and accurate traceability of civil aviation radio parameters are achieved.

The invention relates to an automatic modulation identification method, device and equipment based on a lightweight neural network, and belongs to the technical field of wireless communication. The method comprises the following steps: constructing a multi-stream lightweight global-local collaborative network, firstly adopting a multi-stream input architecture in the network, and performing fusion processing by taking parallel joint I / Q streams, independent I streams and independent Q streams as inputs, thereby effectively reducing feature preprocessing redundancy; secondly, a reverse residual module is introduced, parameter quantity is greatly reduced, and feature extraction efficiency is improved; and finally, developing a lightweight global-local collaboration module, and carrying out modulation type identification and classification by utilizing a classification module after local feature extraction and global relation modeling are deeply fused. According to the method, higher modulation type recognition accuracy can be achieved with lower calculation complexity, and higher robustness is shown in a low signal-to-noise ratio environment.

The invention belongs to the technical field of signal processing, and provides a blind demodulation method and device for a wireless signal, and the method comprises the steps: screening out a candidate modulation type from a plurality of modulation types according to the time domain features in the wireless signal; for each candidate modulation type, estimating a target parameter of the IQ data in the wireless signal corresponding to the candidate modulation type to obtain a target parameter estimation result; performing symbol synchronization processing on the target parameter estimation result to obtain a symbol synchronization point; calculating EVM (Error Vector Magnitude) corresponding to the candidate modulation type according to the symbol synchronization point, and determining the candidate modulation type corresponding to a minimum value as a target modulation type under the condition that the minimum value in the EVM corresponding to each candidate modulation type is smaller than or equal to an EVM threshold value; and demodulating the wireless signal based on the target modulation type to obtain a demodulation result. According to the method provided by the invention, the blind demodulation efficiency and accuracy of the composite modulation signal in a complex channel or a low signal-to-noise ratio environment are improved.

The invention belongs to the technical field of cognitive radio, and particularly relates to an open set modulation identification method based on time-frequency domain feature learning and fusion. The method comprises the following steps of: receiving a signal by adopting a receiving antenna through electromagnetic spectrum monitoring equipment, capturing a space electromagnetic radiation signal, converting the space electromagnetic radiation signal into an electric signal, amplifying the signal through a low-noise amplifier, performing frequency conversion and filtering through an analog receiver, and converting the received signal into an intermediate-frequency signal. Then, the signal is digitalized through AD sampling, and the intermediate frequency signal is converted into a baseband complex signal through frequency mixing and digital filtering in a digital receiver. Then performing time domain characterization calculation and frequency domain characterization calculation on the signal sampling data to form a plurality of time domain characterization vectors and a plurality of frequency domain characterization vectors; and then the vectors are sent to a trained deep neural network model, reasoning is carried out by using the network model, a reasoning result is post-processed, and a signal modulation identification result is finally obtained. According to the invention, the recognition accuracy of unknown signals is improved.

The invention relates to the technical field of wireless communication, and discloses a sensing radio front-end radio frequency signal detection system, which comprises a broadband radio frequency front-end module used for receiving radio frequency signals in an environment and supporting multi-band signal capture and anti-interference preprocessing; the analog-to-digital conversion module is used for converting the radio frequency signal into a digital signal by adopting an ADC (Analog to Digital Converter) with 14-bit resolution and 2GSPS sampling rate; the signal processing module is connected to the output end of the analog-to-digital conversion module and is used for performing spectrum analysis, feature extraction and signal detection on the digital signal and outputting a detection result; and a synchronous control bus. According to the method, energy detection, cyclostationary feature analysis and a deep learning classification algorithm are fused, a time-frequency domain joint analysis framework is constructed, and the time domain transient feature extraction capability of the 1D-CNN and the frequency domain long-range dependence modeling advantage of Transform are combined, so that high-precision recognition of complex modulation signals is realized, and the detection sensitivity and the anti-interference capability are improved.

The invention discloses a frequency spectrum monitoring and interference analysis method for power grid wireless communication, and particularly relates to the technical field of wireless frequency spectrum monitoring. Outputting an eddy current distribution parameter set based on the power frequency magnetic field intensity distribution data of the monitoring point; synchronously acquiring a frequency spectrum signal and a power frequency magnetic flux density signal; calculating high-frequency harmonic frequency point characteristics of secondary radiation generated by the frequency spectrum monitoring equipment in the power frequency magnetic field, and generating an eddy current characteristic spectral line graph; extracting a frequency band energy value matched with the eddy current characteristic spectral line diagram in the frequency spectrum signal, and calculating a pollution proportion of secondary radiation to the frequency spectrum signal in combination with the power frequency magnetic flux density signal; eliminating pollution signal components according to the pollution proportion, and outputting a real spectrum; and performing joint scanning on the real frequency spectrum, extracting modulation characteristics and spatial orientation of the residual interference signal, and generating an interference source positioning report. The problem of real spectrum signal distortion caused by eddy current secondary radiation in a power frequency magnetic field environment is solved, and the accuracy and reliability of spectrum monitoring and interference analysis are improved.

The invention discloses a sub-Nyquist joint broadband spectrum sensing and automatic modulation classification method based on deep learning. The method comprises the following steps: generating a generation data set for joint broadband spectrum sensing and automatic modulation classification; the method comprises the following steps: modeling a cognitive wireless network combined broadband spectrum sensing and automatic modulation classification system, and carrying out multi-coset preprocessing and analog low-pass filtering on a signal received by a cognitive wireless end; constructing a broadband spectrum sensing and automatic modulation classification neural network and training a model; and the local client combines the off-line broadband spectrum sensing training model and the off-line automatic modulation classification training model to online predict the spectrum occupation condition of the broadband signal and the modulation type of each occupied sub-band signal. According to the method, organic fusion of spectrum detection and modulation identification is realized, the processing efficiency and the intelligent degree of the system are remarkably improved, and the method has good robustness and high accuracy and is particularly suitable for a joint identification task of a multi-user broadband communication signal in a dynamic electromagnetic environment.

The invention discloses an adaptive modulation and radio frequency fingerprint combined identification method based on multi-task learning. A dual-task joint recognition model used in the method comprises a feature extraction expert network and a multi-head classifier. The feature extraction expert network comprises a plurality of shared convolutional neural network modules serving as expert networks, and further comprises two gating networks in one-to-one correspondence with two recognition tasks; the gating network obtains probabilities in one-to-one correspondence with the expert networks according to the input multi-label signals, and then performs weighted fusion with the output features of all the expert networks according to the probabilities; and the output features after weight fusion are sent to the corresponding classifiers to obtain recognition results of the corresponding tasks. According to the method, a multi-task architecture of a multi-gate hybrid expert network is adopted, two tasks of modulation mode and radio frequency fingerprint identification can be completed at the same time, meanwhile, technical means such as an attention mechanism and loss weight dynamic optimization are introduced, and the identification accuracy is improved.

The invention belongs to the technical field of radio electronics and signal processing, and particularly relates to a color noise environment signal frequency domain detection method and system based on spectrum envelope extraction, and the system comprises a signal acquisition unit, a signal conversion unit, a signal judgment unit and a signal output unit. The method is a research summary of a color noise environment signal frequency domain detection algorithm based on spectrum envelope extraction, and the algorithm is used for blind signal detection under the conditions of uneven noise spectrum and broadband reception, has good robustness, can be used for low signal-to-noise ratio occasions, and is high in signal detection accuracy.

Techniques are described that provide for waveform selection for uplink (UL) and / or downlink (DL) wireless transmissions based on one or more factors associated with the wireless transmission, a transmitter that is to transmit the wireless transmission, or combinations thereof. UL and DL transmissions may use one of a number of different waveforms, such as single-carrier waveforms that use a single carrier for information transmission of a wireless channel, and multi-carrier waveforms that use multiple carriers at different frequencies to transmit some of the bits on each channel. Multi-carrier transmission waveforms or single-carrier waveforms may be selected for wireless transmissions based on a bandwidth allocated for the transmissions, a capability of a transmitter that is to transmit the transmissions, or combinations thereof.

Disclosed are systems and techniques for wireless communications. For instance, a network entity may obtain a first plurality of bits from a first buffer, wherein a start position for obtaining the first plurality of bits from the first buffer is based on a redundancy version order. In some cases, the network entity may determine, based on a modulation order and a probability distribution associated with the first plurality of bits, at least one scaling factor for constellation mapping the first plurality of bits. In some examples, the network entity may map, based on the at least one scaling factor, the first plurality of bits to one or more symbols.

The invention relates to a multi-rate wireless frame adaptive architecture method and device, equipment and a storage medium, and the method comprises the steps: receiving channel state information, and determining pilot frequency configuration according to the channel state information; receiving service data to be transmitted, selecting a corresponding modulation mode based on the channel state information, and performing symbol modulation on the service data according to the modulation mode to obtain a service modulation symbol; carrying out frequency hopping distribution on the service modulation symbols based on the pilot frequency configuration, and carrying out frequency hopping transmission on the service modulation symbols according to the frequency hopping distribution; and receiving the frequency hopping data after the frequency hopping transmission, and performing data recovery on the frequency hopping data to obtain decoded service data. The method and the device have the effect of improving the adaptability of the multi-rate architecture.

A UE of a wireless communication network first receives a modulation and coding scheme index IMCS for demodulating a physical layer data channel of the network. The UE then demodulates the physical layer data channel with a non-square quadrature amplitude modulation (NS-QAM) having a modulation order and code rate corresponding to a spectral efficiency (SPEFF) substantially equal to a SPEFF of a square QAM (S-QAM) SPEFF indicted by the IMCS, where the NS-QAM has an effective channel code rate no larger than the code rate for which the UE may skip decoding a transport block in an initial received transmission.

The invention provides a semi-supervised signal modulation identification method based on multi-codebook discrete virtual adversarial training, and the method comprises the steps: carrying out the high-dimensional potential representation coding of input labeled and unlabeled sample signals through employing a vector quantization generative adversarial network, so as to obtain a continuous potential representation; mapping the continuous potential representation into a discrete vector sequence through a quantizer; virtual confrontation disturbance meeting the condition that the L2 norm does not exceed epsilon is applied to the unlabeled sample at the input end, and a disturbance sample is generated; respectively inputting a labeled sample, a label-free sample and a disturbance sample into a classifier, calculating cross entropy loss and KL divergence consistency loss, and jointly optimizing parameters of an encoder, a codebook and the classifier in a weighted sum form; therefore, the discrete potential representation is obtained through the vector quantization generative adversarial network, disturbance is constructed for the unlabeled sample at the input end, and joint optimization is performed in combination with the cross entropy and the KL divergence loss, so that the recognition accuracy and robustness under the conditions of few labels and low signal-to-noise ratio are improved.

The invention discloses a communication modulation signal identification method and device, equipment and a storage medium, and the method comprises the steps: collecting first signal information of a known transmitting end and second signal information of an unknown transmitting end, and generating a first sample set and a second sample set respectively; performing time domain sliding window processing and frequency domain sliding window processing on the first sample set and the second sample set in sequence through time-frequency analysis of a sliding window to obtain a first signal sample set and a second signal sample set, and converting characteristic values of two-dimensional time-frequency distribution of the first signal sample set and the second signal sample set into RGB images, enabling the size of the RGB image to meet a preset migration network model, updating a sample set, dividing the sample set, obtaining a training set and a test set required by the preset migration network model, setting parameters for training and testing until a target recognition rate is met, obtaining multi-dimensional fused fine features, and combining a migration learning network to obtain a target recognition rate. And the equipment identification accuracy is improved.

Some techniques described herein provide autonomous modification of a modulation and coding scheme (MCS) of a grant for a transmission by a user equipment UE. For example, the UE may receive a grant (e.g., a configured grant) indicating a resource allocation and a first MCS for a transmission on the resource allocation. The UE may use the resource allocation for a transmission of a number of transport blocks (e.g., one or more transport blocks), and may use a second MCS, different than the first MCS, for the transmission. In some examples, the second MCS may be a lower MCS, associated with a lower modulation scheme and / or code rate, than the first MCS. Thus, a per-resource-block power consumption of the transmission is decreased by using the second MCS, which provides power savings while reducing or eliminating padding of the uplink communication.

The present invention relates to a three-channel automatic modulation recognition method. When a single-input single-output communication system receives an unknown signal, the time domain signal is discretely sampled in the time domain to form a time domain discrete signal, and one-dimensional complex data is formed. The Fourier time-frequency domain transform obtains frequency domain data. The frequency domain data is combined with the time domain discrete signal to obtain time-frequency domain discrete signal data. The frequency data in the time-frequency domain discrete signal data is fed into the two-dimensional convolution layer of the time-space frequency fusion network model. The in-phase data and orthogonal data in the time domain data are respectively fed into the corresponding one-dimensional convolution layer. The model determines the modulation format type of the unknown signal based on the characteristics of various modulation formats. In a positive signal-to-noise ratio environment, the present invention greatly improves the recognition accuracy of the modulation type of unknown electromagnetic spectrum signals, and the training speed is several times faster than that of existing models. For high-dimensional modulated signals, it can better solve the confusion problem between similar signals and achieve higher recognition accuracy.

The invention discloses a modulation identification method based on microwave photon frequency-time mapping. The method comprises the following steps: firstly, performing microwave photon frequency-time mapping on a to-be-measured microwave signal, and mapping frequency information of the to-be-measured microwave signal to a time domain to obtain a time-frequency characteristic of the to-be-measured microwave signal; and inputting the time-frequency characteristics of the to-be-detected microwave signal into a pre-trained neural network model so as to identify the modulation form of the to-be-detected microwave signal. The invention also discloses a modulation identification device based on microwave photon frequency-time mapping. Compared with the prior art, on the basis of a microwave photon frequency-time mapping model, Fourier transform is directly completed through an optical device, a radio frequency signal is modulated to be processed in an optical domain, time delay characteristics of dispersion broadening and compression are combined, the picosecond-level time stretching precision is achieved, the instantaneous bandwidth limitation of a traditional spectrum analyzer is broken through, and the spectrum analysis precision is improved. Frequency spectrum processing of large-bandwidth signals can be completed in real time, computing resources are saved, and the real-time performance of modulation identification is improved.

The present application discloses an uplink data transmission method, apparatus, terminal and medium, which belongs to the field of communication technology. The uplink data transmission method of an embodiment of the present application includes: when a UE is configured to adopt a first waveform to transmit uplink data, the UE receives target downlink control information DCI from a network side device, and the target DCI is used to schedule target uplink data; when the indication information included in the target DCI meets a first preset condition, the UE adopts a second waveform to transmit the target uplink data; wherein the above-mentioned indication information is used to indicate: transmission parameters of the target uplink data.

The invention provides a blind identification and intelligent classification prediction method and system for a signal modulation mode, and relates to the technical field of signal processing, and the method comprises the steps: obtaining a multi-scale time domain feature through wavelet packet decomposition, obtaining an instantaneous frequency feature through Hilbert transform, and constructing a feature correlation matrix; blind identification of a signal modulation mode is realized by using a dynamic routing algorithm and a self-correction mechanism; distinguishing discrete and continuous modulation based on the feature matrix and the signal-to-noise ratio; and signal modulation mode prediction is realized by adopting hybrid filtering and layered Markov decision. According to the invention, the recognition accuracy and prediction stability of the signal modulation mode are improved.

The invention discloses a complex electromagnetic signal demodulation method based on multi-task learning, and relates to the technical field of electromagnetic signal demodulation, and the method comprises the steps: employing a ResNet-18 and Transform mixed architecture as a backbone network, combining a dynamic weight distribution mechanism, and constructing a multi-task learning model; a complex electromagnetic signal to be detected is input into the multi-task learning model, a modulation type recognition result and a code element sequence prediction result are obtained, collaborative optimization of the two tasks is achieved, and the negative migration effect caused by task conflicts is effectively restrained through a dynamic weight distribution mechanism. In an electromagnetic signal demodulation process, a modulation mode adopted by a signal is firstly judged, then an original code element sequence is predicted, modulation type identification is the basis of code element sequence prediction, and the accuracy of the modulation type identification directly affects a final result of code element sequence prediction. The multi-task learning model improves the accuracy of modulation type identification through collaborative optimization of two tasks, and improves the accuracy of code element sequence generation at the same time.