A signal processing method for power carrier communication based on OFDM modulation

Abstract

The application discloses a signal processing method of power carrier communication based on OFDM modulation, relates to the technical field of power carrier communication, and comprises the following steps: S1, cyclic prefix and guard interval design and insertion; S2, improved synchronization algorithm for frequency offset processing; S3, noise suppression optimization of a front-end signal processing module; S4, parallel processing architecture design and task allocation; S5, channel estimation at a receiving end; S6, frequency domain equalization processing; S7, OFDM demodulation processing; and S8, symbol synchronization and data recovery. Through the steps S1 and S2, the effects of effectively resisting power line channel multipath delay, reducing inter-symbol interference and inter-carrier interference are achieved. Through the steps S3, S5 and S6, the effects of accurately identifying power line impulse noise and Gaussian noise characteristics and dynamically suppressing noise are achieved. Through the steps S4 and S8, the effects of improving signal processing efficiency by using a parallel processing architecture and avoiding processor overload are achieved, and meanwhile, the data error correction capability is enhanced by convolution coding or Turbo coding, so that the real-time performance and reliability of high-speed data transmission are ensured.

Description

Technical Field

[0001] This invention relates to the field of power line carrier communication technology, and in particular to a signal processing method for power line carrier communication based on OFDM modulation. Background Technology

[0002] With the rapid development of smart grid and Internet of Things technologies, power line carrier communication, as a technology that uses existing power line infrastructure to realize data transmission, has been widely used in smart home control, power monitoring, remote meter reading and other fields due to its advantages such as no need to lay additional communication lines, low cost and convenient deployment. Power line carrier communication based on orthogonal frequency division multiplexing technology divides high-speed data streams into multiple low-speed sub-data streams and modulates them onto multiple orthogonal subcarriers for transmission. It can effectively resist multipath fading and improve spectrum utilization, and has become the mainstream technical solution in the field of power line carrier communication.

[0003] However, the power line channel environment is complex and variable, with serious noise interference, multipath effects, and frequency-selective fading, posing numerous challenges to OFDM-based power line carrier communication. On the one hand, interference signals such as impulse noise and Gaussian noise in power lines can severely affect signal transmission quality, and traditional filtering methods struggle to accurately suppress complex noise. On the other hand, the multipath delay characteristics of power line channels easily lead to inter-symbol interference and inter-carrier interference, while conventional cyclic prefix designs and synchronization algorithms are insufficient to meet high-precision synchronization requirements. Furthermore, the signal processing of power line carrier communication involves a large number of complex calculations, and traditional serial processing architectures have bottlenecks in real-time performance and processing efficiency, making them unsuitable for high-speed data transmission scenarios. Therefore, it is necessary to design a signal processing method for power line carrier communication based on OFDM modulation to solve the aforementioned problems. Summary of the Invention

[0004] To achieve the above objectives, the present invention provides a signal processing method for power line carrier communication based on OFDM modulation, comprising the following implementation steps:

[0005] S1. Cyclic Prefix and Guard Interval Design and Insertion: For the OFDM modulation process in power line carrier communication, based on the multipath delay characteristics of the power line channel, the maximum delay spread value Tmax of the power line channel is obtained through channel measurement, and a specific cyclic prefix length Lcp and guard interval Lgi are designed.

[0006] S2. Improved synchronization algorithm for frequency offset processing: An improved synchronization algorithm is adopted, which utilizes the cyclic correlation characteristics based on pilot symbols to estimate the frequency offset of the received signal at the receiving end.

[0007] The pilot symbol sequence in the received signal is cyclically correlated with the locally known pilot symbol sequence. The frequency offset Δf is calculated by finding the position of the correlation peak, and frequency offset compensation is performed based on the calculation result. Where Ts is the sampling period, x(n) is the original signal, and y(n) is the compensated signal;

[0008] S3. Noise suppression optimization of the front-end signal processing module: By collecting noise samples from the power line at different times and under different load conditions, a dataset containing 1000 sets of noise data is constructed.

[0009] The time-frequency distribution features of power line noise are extracted using short-time Fourier transform. The time-frequency distribution feature image is input into a deep learning feature extraction unit, and training is performed using cross-entropy loss function and stochastic gradient descent algorithm. The impulse interference features and Gaussian noise features of power line noise are identified and classification weights are output.

[0010] S4. Parallel processing architecture design and task allocation: A parallel processing architecture is adopted to allocate signal processing tasks to multiple processor units. The parallel processing module contains at least three independent processor units, which respectively perform cyclic prefix insertion, frequency offset compensation and noise suppression functions.

[0011] Furthermore, a dynamic load balancing strategy is used to allocate task priorities to processor units. A preset threshold for task queue length is set to 50. When the task queue length of any processor unit exceeds the preset threshold, a task migration mechanism is triggered to migrate some tasks to processor units with lighter loads, ensuring the efficiency and timeliness of signal processing.

[0012] Preferably, the following implementation steps are also included:

[0013] S5. Receiver Channel Estimation: At the receiver, channel estimation is performed on the signal after frequency offset compensation and noise suppression. The channel frequency response H(k) is estimated using the least squares method with pilot symbols, and the channel response at the data subcarrier is interpolated using a linear interpolation method to obtain the complete channel frequency response estimate.

[0014] S6. Frequency domain equalization processing: Based on the channel frequency response estimate obtained in step S5, the received signal is equalized using the zero-forcing equalization algorithm or the minimum mean square error equalization algorithm.

[0015] Equalizer coefficients , where H*(k) is the conjugate of the channel frequency response, σ² is the noise variance, and N0 is the noise power spectral density;

[0016] S7, OFDM demodulation: The equalized signal is demodulated using OFDM to remove the cyclic prefix and guard interval, converting the signal from the frequency domain to the time domain, and recovering the original modulation symbol through inverse fast Fourier transform.

[0017] S8. Symbol Synchronization and Data Recovery: The demodulated symbols are synchronized and recovered at a specific time. The Gardner algorithm for timing synchronization is used to accurately detect symbol boundaries and recover the original transmitted data sequence.

[0018] Preferably, in step S1, before inserting the cyclic prefix and guard interval, the orthogonal frequency division multiplexing symbols are pre-coded using Gray coding to reduce the probability of bit errors between symbols.

[0019] The cyclic prefix length Lcp is set to ≥ fs is the sampling frequency, Lgi is set to 0.5*Lcp, and the cyclic prefix and guard interval are inserted at the beginning of the orthogonal frequency division multiplexing symbol;

[0020] In step S2, when estimating the frequency offset, a segmented correlation method is used to divide the received signal into multiple subsequences of length L. Cyclic correlation operations are performed on each subsequence, and then the average value of the frequency offset estimates of each subsequence is taken as the final frequency offset ∆f, thereby improving the accuracy of frequency offset estimation.

[0021] Preferably, in step S3, the deep learning feature extraction unit uses data augmentation techniques during training to rotate, translate, and scale the noise time-frequency distribution feature image to expand the training dataset and improve the model's generalization ability.

[0022] The parameter adjustment unit dynamically adjusts the filter coefficients according to the time-frequency distribution characteristics to achieve precise noise suppression; the convolutional neural network layer uses a GPU for accelerated computation, and the parameter adjustment unit consists of a control chip and a digital-to-analog converter.

[0023] Preferably, in step S4, the dynamic load balancing strategy calculates the load value of each processor unit based on multiple dimensions of indicators such as CPU utilization, memory usage, and task queue length, and allocates task priorities in ascending order of load value.

[0024] Preferably, in step S5, when performing channel estimation, a channel estimation method based on compressed sensing is introduced. When the number of pilots is small, the sparse characteristics of the power line channel are utilized to reconstruct the channel frequency response, thereby improving the efficiency and accuracy of channel estimation.

[0025] Preferably, in step S6, after frequency domain equalization, the bit error rate of the equalized signal is detected. When the bit error rate exceeds a preset threshold of 0.001, the equalization algorithm is automatically switched from ZF equalization algorithm to MMSE equalization algorithm, or a fusion strategy of the two algorithms is adopted for secondary equalization.

[0026] Preferably, in step S8, after symbol synchronization and timed recovery, the recovered original data sequence is subjected to error correction coding processing, using convolutional coding or Turbo coding error correction coding methods to further improve the reliability of data transmission.

[0027] In summary, this invention provides a signal processing method for power line carrier communication based on OFDM modulation, which has the following beneficial effects:

[0028] By designing and inserting the S1 cyclic prefix and guard interval, and implementing the S2 improved synchronization algorithm, the frequency offset processing steps are achieved, effectively resisting multipath delay in power line channels and reducing inter-symbol and inter-carrier interference. Simultaneously, the frequency offset is accurately compensated using the cyclic correlation characteristics of pilot symbols, restoring OFDM subcarrier orthogonality and improving signal demodulation accuracy. Furthermore, the S1 cyclic prefix length Lcp is set to ≥ Lgi is set to 0.5*Lcp to isolate OFDM symbol boundaries in the time domain, effectively eliminating symbol overlap caused by multipath reflections; S2 reduces the frequency offset estimation error to ≤0.001*fs through segmented cyclic correlation operations on pilot symbols, combined with... This improves the inter-carrier interference suppression ratio to over 25dB, ensuring that the QAM demodulation bit error rate remains stable below 0.0001 under 16QAM modulation.

[0029] Through noise suppression optimization in the S3 front-end signal processing module, channel estimation at the S5 receiver end, and frequency domain equalization processing steps, the system achieves accurate identification of power line impulse noise and Gaussian noise characteristics and dynamically suppresses noise. Simultaneously, by combining least squares channel estimation and frequency domain equalization algorithms, channel fading and inter-symbol interference are eliminated, improving the signal-to-noise ratio and integrity of signal transmission. S3, based on a 3-layer convolutional neural network trained with 1000 noise samples, improves the accuracy of impulse noise identification and Gaussian noise classification. Dynamically adjusted filter coefficients reduce in-band noise power by 15-20 dB. S5 introduces compressed sensing channel estimation, achieving a channel frequency response reconstruction error ≤0.05 even with a reduced number of pilots. Step S6, using the MMSE equalization algorithm, improves signal integrity and the overall system signal-to-noise ratio when the channel fading depth is ≥10 dB.

[0030] By designing and allocating tasks using the S4 parallel processing architecture and implementing S8 symbol synchronization and data recovery steps, the system achieves the effect of improving signal processing efficiency and avoiding processor overload through parallel processing architecture. At the same time, it enhances data error correction capabilities through convolutional coding or Turbo coding, ensuring the real-time performance and reliability of high-speed data transmission. Attached Figure Description

[0031] Figure 1 This is a schematic diagram of the flow architecture of a signal processing method for power line carrier communication based on OFDM modulation according to the present invention. Detailed Implementation

[0032] The following is in conjunction with the appendix Figure 1 This application will be described in further detail below.

[0033] Example

[0034] Please see Figure 1 This invention provides a technical solution: a signal processing method for power line carrier communication based on OFDM modulation, comprising the following implementation steps:

[0035] S1. Cyclic Prefix and Guard Interval Design and Insertion: For the OFDM modulation process in power line carrier communication, based on the multipath delay characteristics of the power line channel, the maximum delay spread value Tmax of the power line channel is obtained through channel measurement. Specific cyclic prefix length Lcp and guard interval Lgi are designed. By setting the cyclic prefix length and guard interval, the multipath delay of the power line channel is effectively resisted, inter-symbol interference and inter-carrier interference are reduced, and the multipath resistance capability of the OFDM system is improved.

[0036] S2. Improved synchronization algorithm for frequency offset processing: An improved synchronization algorithm is adopted, which utilizes the cyclic correlation characteristics based on pilot symbols to estimate the frequency offset of the received signal at the receiving end. Through the cyclic correlation operation of pilot symbols, the accuracy of frequency offset estimation is improved and the damage of carrier frequency offset to OFDM subcarrier orthogonality is reduced.

[0037] The pilot symbol sequence in the received signal is cyclically correlated with the locally known pilot symbol sequence. The frequency offset Δf is calculated by finding the position of the correlation peak, and frequency offset compensation is performed based on the calculation result. Where Ts is the sampling period, x(n) is the original signal, and y(n) is the compensated signal. By accurately compensating for frequency offset through y(n), subcarrier orthogonality is restored, ICI is reduced, and the accuracy of signal demodulation is improved.

[0038] S3. Noise suppression optimization of the front-end signal processing module: By collecting noise samples of power lines at different times and under different load conditions, a dataset containing 1,000 sets of noise data is constructed. Through the collection of noise samples in multiple scenarios, the complex noise environment of power lines is covered, providing diversified training data for deep learning models.

[0039] The time-frequency distribution features of power line noise are extracted using short-time Fourier transform. The time-frequency distribution feature image is input into a deep learning feature extraction unit, which is trained using cross-entropy loss function and stochastic gradient descent algorithm. The unit identifies the impulse interference features and Gaussian noise features of power line noise and outputs classification weights. The deep learning network automatically extracts the time-frequency features of noise, achieving accurate classification of impulse noise and Gaussian noise, and providing a decision basis for dynamic filtering.

[0040] S4. Parallel Processing Architecture Design and Task Allocation: A parallel processing architecture is adopted, which allocates signal processing tasks to multiple processor units. The parallel processing module contains at least three independent processor units, which respectively perform cyclic prefix insertion, frequency offset compensation and noise suppression functions. By splitting tasks into independent processor units, the parallelization of the signal processing flow is realized, and the overall processing efficiency is improved.

[0041] Furthermore, a dynamic load balancing strategy is used to allocate task priorities to processor units. A preset threshold for task queue length is set to 50. When the task queue length of any processor unit exceeds the preset threshold, a task migration mechanism is triggered to migrate some tasks to processor units with lighter loads, ensuring the efficiency and timeliness of signal processing. Through dynamic load balancing and task migration, overload of a single processor unit is avoided, ensuring the real-time performance of signal processing and system throughput.

[0042] It also includes the following implementation steps:

[0043] S5. Receiver Channel Estimation: At the receiver, channel estimation is performed on the signal after frequency offset compensation and noise suppression. The channel frequency response H(k) is estimated using the least squares method with pilot symbols. The channel response at the data subcarrier is interpolated using the linear interpolation method to obtain a complete channel frequency response estimate. The complete channel frequency response estimate is obtained through the least squares method and linear interpolation, providing accurate channel state information for subsequent equalization processing.

[0044] S6. Frequency domain equalization processing: Based on the channel frequency response estimate obtained in step S5, the received signal is equalized using the zero-forcing equalization algorithm or the minimum mean square error equalization algorithm. The frequency domain equalization algorithm is used to eliminate channel fading and inter-symbol interference and restore the original amplitude and phase of the signal.

[0045] Equalizer coefficients Where H*(k) is the conjugate of the channel frequency response, σ² is the noise variance, and N0 is the noise power spectral density. The equalizer coefficients are calculated through W(k), and combined with the parameters H*(k) and σ², accurate compensation for channel distortion is achieved.

[0046] S7. OFDM Demodulation Processing: OFDM demodulation is performed on the equalized signal to remove the cyclic prefix and guard interval, convert the signal from the frequency domain to the time domain, and quickly recover the original modulation symbols. Through the OFDM demodulation process, the frequency domain signal is converted into time domain symbols, and the inverse transformation recovery of the modulation symbols is completed.

[0047] S8. Symbol Synchronization and Data Recovery: Symbol synchronization and timed recovery are performed on the demodulated symbols. The Gardner algorithm for timed synchronization is used to accurately detect the symbol boundaries and recover the original transmitted data sequence. The Gardner algorithm achieves symbol-level precise synchronization to ensure that the boundaries of the data sequence are correct and to avoid bit errors caused by symbol misalignment.

[0048] In step S1, before inserting the cyclic prefix and guard interval, the orthogonal frequency division multiplexing symbols are pre-coded. Gray coding is used to encode the symbols to reduce the probability of bit errors between symbols. Gray coding reduces the probability of bit errors between adjacent symbols. Even if a single bit error occurs, it only affects adjacent symbols, thus improving the coding fault tolerance.

[0049] The cyclic prefix length Lcp is set to ≥ fs is the sampling frequency, Lgi is set to 0.5*Lcp, and this cyclic prefix and guard interval are inserted at the beginning of the orthogonal frequency division multiplexing symbol. The specific values ​​of Lcp and Lgi are determined by quantization calculation to ensure that the cyclic prefix length covers the maximum multipath delay, and the guard interval further isolates the symbol boundary to improve the reliability of system transmission.

[0050] In step S2, when estimating the frequency offset, a piecewise correlation method is used to divide the received signal into multiple subsequences of length L. Cyclic correlation is performed on each subsequence, and then the average value of the frequency offset estimates of each subsequence is taken as the final frequency offset ∆f. This improves the accuracy of the frequency offset estimation. By taking the average value through piecewise correlation, the influence of random noise on the frequency offset estimation is reduced, thereby improving the stability and accuracy of the estimation results.

[0051] In step S3, the deep learning feature extraction unit uses data augmentation techniques during training to rotate, translate, and scale the noise time-frequency distribution feature image, thereby expanding the training dataset and improving the model's generalization ability. By expanding the training dataset through data augmentation, the deep learning model's generalization ability to different noise patterns is improved, and the model's reliability is enhanced.

[0052] The parameter adjustment unit dynamically adjusts the filter coefficients according to the time-frequency distribution characteristics to achieve precise noise suppression. The convolutional neural network layer uses a GPU for accelerated computation. The parameter adjustment unit consists of a control chip and a digital-to-analog converter. By dynamically adjusting the filter coefficients, it adaptively suppresses different types of noise. Combined with GPU acceleration, it improves computational efficiency and ensures real-time noise reduction processing.

[0053] In step S4, the dynamic load balancing strategy calculates the load value of each processor unit based on multiple dimensions of indicators such as CPU utilization, memory usage, and task queue length. Task priorities are allocated in ascending order of load value. By dynamically evaluating processor load through multiple indicators, the strategy avoids uneven task distribution caused by bias in a single indicator and improves the scientific nature of load balancing.

[0054] In step S5, a channel estimation method based on compressed sensing is introduced when performing channel estimation. When the number of pilots is small, the channel frequency response is reconstructed by utilizing the sparsity characteristics of the power line channel, thereby improving the efficiency and accuracy of channel estimation. The compressed sensing technique reduces the number of pilots required, thereby improving the efficiency of channel estimation, while the channel sparsity is used to improve the estimation accuracy.

[0055] In step S6, after frequency domain equalization, the bit error rate of the equalized signal is detected. When the bit error rate exceeds the preset threshold of 0.001, the equalization algorithm is automatically switched from ZF equalization algorithm to MMSE equalization algorithm, or a fusion strategy of the two algorithms is adopted for secondary equalization. The algorithm adaptive switching is triggered by real-time bit error rate detection, and the equalization performance is dynamically improved when the channel conditions deteriorate, so as to ensure the system transmission quality.

[0056] In step S8, after symbol synchronization and timed recovery, the recovered original data sequence is subjected to error correction coding. Convolutional coding or Turbo coding is used to further improve the reliability of data transmission. Redundant parity bits are introduced through error correction coding to detect and correct erroneous bits during transmission, thereby further improving the reliability of data transmission.

[0057] The embodiments described in this specific implementation are preferred embodiments of this application and are not intended to limit the scope of protection of this application. Identical components are represented by the same reference numerals. Therefore, all equivalent changes made to the structure, shape, and principle of this application should be included within the scope of protection of this application.

Claims

1. A signal processing method for power line carrier communication based on OFDM modulation, characterized in that: The implementation steps include the following: S1. Cyclic Prefix and Guard Interval Design and Insertion: For the OFDM modulation process in power line carrier communication, based on the multipath delay characteristics of the power line channel, the maximum delay spread value Tmax of the power line channel is obtained through channel measurement, and a specific cyclic prefix length Lcp and guard interval Lgi are designed. S2. Improved synchronization algorithm for frequency offset processing: An improved synchronization algorithm is adopted, which utilizes the cyclic correlation characteristics based on pilot symbols to estimate the frequency offset of the received signal at the receiving end. The pilot symbol sequence in the received signal is cyclically correlated with the locally known pilot symbol sequence. The frequency offset Δf is calculated by finding the position of the correlation peak, and frequency offset compensation is performed based on the calculation result. Where Ts is the sampling period, x(n) is the original signal, and y(n) is the compensated signal; S3. Noise suppression optimization of the front-end signal processing module: By collecting noise samples from the power line at different times and under different load conditions, a dataset containing 1000 sets of noise data is constructed. The time-frequency distribution features of power line noise are extracted using short-time Fourier transform. The time-frequency distribution feature image is input into a deep learning feature extraction unit, and training is performed using cross-entropy loss function and stochastic gradient descent algorithm. The impulse interference features and Gaussian noise features of power line noise are identified and classification weights are output. S4. Parallel processing architecture design and task allocation: A parallel processing architecture is adopted to allocate signal processing tasks to multiple processor units. The parallel processing module contains at least three independent processor units, which respectively perform cyclic prefix insertion, frequency offset compensation and noise suppression functions. Furthermore, a dynamic load balancing strategy is used to allocate task priorities to processor units. A preset threshold for task queue length is set to 50. When the task queue length of any processor unit exceeds the preset threshold, a task migration mechanism is triggered to migrate some tasks to processor units with lighter loads, ensuring the efficiency and timeliness of signal processing.

2. The signal processing method for power line carrier communication based on OFDM modulation according to claim 1, characterized in that: It also includes the following implementation steps: S5. Receiver Channel Estimation: At the receiver, channel estimation is performed on the signal after frequency offset compensation and noise suppression. The channel frequency response H(k) is estimated using the least squares method with pilot symbols, and the channel response at the data subcarrier is interpolated using the linear interpolation method to obtain the complete channel frequency response estimate. S6. Frequency domain equalization processing: Based on the channel frequency response estimate obtained in step S5, the received signal is equalized using the zero-forcing equalization algorithm or the minimum mean square error equalization algorithm. Equalizer coefficients σ² is the conjugate of the channel frequency response, σ² is the noise variance, and N0 is the noise power spectral density. S7, OFDM demodulation: The equalized signal is demodulated using OFDM to remove the cyclic prefix and guard interval, converting the signal from the frequency domain to the time domain, and recovering the original modulation symbol through inverse fast Fourier transform. S8. Symbol Synchronization and Data Recovery: The demodulated symbols are synchronized and recovered at a specific time. The Gardner algorithm for timing synchronization is used to accurately detect symbol boundaries and recover the original transmitted data sequence.

3. The signal processing method for power line carrier communication based on OFDM modulation according to claim 1, characterized in that: In step S1, before inserting the cyclic prefix and guard interval, the orthogonal frequency division multiplexing symbols are pre-coded using Gray coding to reduce the probability of bit errors between symbols. The cyclic prefix length Lcp is set to ≥ fs is the sampling frequency, Lgi is set to 0.5*Lcp, and the cyclic prefix and guard interval are inserted at the beginning of the orthogonal frequency division multiplexing symbol; In step S2, when estimating the frequency offset, a segmented correlation method is used to divide the received signal into multiple subsequences of length L. Cyclic correlation operations are performed on each subsequence, and then the average value of the frequency offset estimates of each subsequence is taken as the final frequency offset ∆f, thereby improving the accuracy of frequency offset estimation.

4. The signal processing method for power line carrier communication based on OFDM modulation according to claim 1, characterized in that: In step S3, the deep learning feature extraction unit uses data augmentation techniques during training to rotate, translate, and scale the noise time-frequency distribution feature image to expand the training dataset and improve the model's generalization ability. The parameter adjustment unit dynamically adjusts the filter coefficients according to the time-frequency distribution characteristics to achieve precise noise suppression; the convolutional neural network layer uses a GPU for accelerated computation, and the parameter adjustment unit consists of a control chip and a digital-to-analog converter.

5. The signal processing method for power line carrier communication based on OFDM modulation according to claim 1, characterized in that: In step S4, the dynamic load balancing strategy calculates the load value of each processor unit based on multiple dimensions of indicators such as CPU utilization, memory usage, and task queue length, and allocates task priorities in ascending order of load value.

6. The signal processing method for power line carrier communication based on OFDM modulation according to claim 2, characterized in that: In step S5, a channel estimation method based on compressed sensing is introduced when performing channel estimation. When the number of pilots is small, the channel frequency response is reconstructed by utilizing the sparsity characteristics of the power line channel, thereby improving the efficiency and accuracy of channel estimation.

7. The signal processing method for power line carrier communication based on OFDM modulation according to claim 2, characterized in that: In step S6, after frequency domain equalization, the bit error rate of the equalized signal is detected. When the bit error rate exceeds the preset threshold of 0.001, the equalization algorithm is automatically switched from ZF equalization algorithm to MMSE equalization algorithm, or a fusion strategy of the two algorithms is used for secondary equalization.

8. The signal processing method for power line carrier communication based on OFDM modulation according to claim 2, characterized in that: In step S8, after symbol synchronization and timed recovery, the recovered original data sequence is subjected to error correction coding processing. Convolutional coding or Turbo coding error correction coding methods are used to further improve the reliability of data transmission.

Images