Low voltage power line multicarrier communication system based on lwt-ofdm

Abstract

This invention relates to the field of low-voltage power line communication technology, specifically a low-voltage power line multi-carrier communication system based on LWT-OFDM, comprising: adaptive processing, subband mapping, parameter pre-configuration, waveform synthesis, and channel transceiver modules. The adaptive processing module encapsulates the original communication data in a specific format and dynamically adjusts the data block size based on real-time detected channel noise characteristics. The subband mapping module performs lifting wavelet transform on the data blocks, mapping them to wavelet function subbands of different scales to form initial subband symbols. The parameter pre-configuration module embeds a reference signal pattern and pre-configures the transmission parameters for each subband. The waveform synthesis module performs amplitude and phase adjustment and time-domain shaping window processing to generate physical waveforms. The channel transceiver module implements signal transmission, synchronization, and data reverse recovery. This system can adapt to changes in channel noise, suppress subband crosstalk, and improve the anti-interference capability and transmission stability of low-voltage power line communication.

Description

Technical Field

[0001] This invention relates to the field of low-voltage power line communication technology, and more particularly to a low-voltage power line multi-carrier communication system based on LWT-OFDM. Background Technology

[0002] Conventional low-voltage power line multi-carrier communication systems mostly employ traditional OFDM communication architectures. Communication data is processed using fixed block specifications, subcarrier allocation is achieved through Fourier transform in the subband mapping stage, transmission parameters are configured with preset fixed values, waveform synthesis is implemented using conventional time-domain window functions, and the receiver performs signal synchronization and data parsing according to a fixed demodulation process. Existing systems only use channel noise detection to passively correct transmission parameters, failing to directly correlate noise characteristics with data block specifications, and subband signal generation does not incorporate the multi-scale processing characteristics of wavelet transform.

[0003] Traditional fixed data segmentation cannot adapt to the real-time fluctuations in noise in low-voltage power line channels. The data segmentation has a low match with channel transmission characteristics, easily exacerbating noise interference and causing signal transmission distortion. In low-voltage power line multipath interference environments, Fourier transform subband mapping exhibits weak anti-interference capabilities for subband signals, with significant crosstalk between subbands. Fixed transmission parameter configurations cannot meet the transmission requirements of each subband signal, and time-domain waveform shaping lacks adaptability to subband signals, limiting signal transmission stability and fidelity.

[0004] This invention needs to adapt to the real-time variation characteristics of low-voltage power line channel noise and realize dynamic control of data block size. It needs to replace the traditional Fourier transform subband mapping method and complete the multi-scale subband mapping of data blocks by improving wavelet transform, thereby optimizing the carrying and transmission form of subband signals. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of the existing technology and propose a low-voltage power line multi-carrier communication system based on LWT-OFDM.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: a low-voltage power line multi-carrier communication system based on LWT-OFDM, comprising: An adaptive processing module acquires raw communication data from low-voltage power lines, performs preliminary format encapsulation on the raw communication data, and performs adaptive block processing on the encapsulated data. The block size is dynamically adjusted based on the channel noise characteristics detected in real time. The subband mapping module performs lifting wavelet transform on each data block and maps it to multiple subbands composed of wavelet functions of different scales to form initial subband symbols. The parameter pre-configuration module embeds a known reference signal pattern at a specific position of the initial sub-band symbol, and uses the reference signal pattern to pre-calculate and pre-configure the transmission parameters of each sub-band before transmission; The waveform synthesis module, based on pre-configured parameters, adjusts the amplitude and phase of the sub-band symbols carrying the data, merges the symbols of all sub-bands, and applies a time-domain shaping window function to generate the physical waveform signal. The channel transceiver module feeds the physical waveform signal into the low-voltage power line channel for transmission, captures and synchronizes the physical waveform signal at the receiving end, performs a reverse processing flow on the synchronized signal, separates each sub-band, and recovers the data.

[0007] As a further aspect of the present invention, the step of adaptively segmenting the encapsulated data, wherein the segment size is dynamically adjusted based on the real-time detected channel noise characteristics, specifically includes: During the intervals between data transmissions, a sequence of probe signals is periodically sent to the power line channel; The distortion and attenuation of the probe signal sequence are received and analyzed to quantify the noise intensity and impulse interference statistical characteristics of the current channel. Establish a channel quality assessment matrix, which includes background noise power, average impulse noise occurrence rate, and channel response flatness index. Based on the indicators in the channel quality assessment matrix, a preset mapping rule table is queried. The mapping rule table defines the optimal combination of data block length and guard interval length corresponding to different channel conditions. The data block length and guard interval length obtained by mapping the current channel conditions are applied to segment the encapsulated data to be transmitted, ensuring that each data block has an independent guard interval in the time domain.

[0008] As a further aspect of the present invention, the step of embedding a known reference signal pattern at a specific position of the initial sub-band symbol specifically includes: Design a reference signal sequence with good autocorrelation characteristics, wherein the reference signal sequence has a non-uniform power distribution in each subband of the lifting wavelet transform domain; Based on the channel quality assessment matrix, the set of subbands with high signal-to-noise ratio and the set of subbands with low signal-to-noise ratio under the current channel conditions are identified; The reference signal sequence is inserted at close intervals on the set of subbands with high signal-to-noise ratio; On the set of subbands with low signal-to-noise ratio, the reference signal sequence is inserted at sparse intervals, and the saved symbol resources are used for data transmission. A unique reference signal pattern embedding location map is generated for each data block, and the location map is transmitted to the receiving end along with the header control information of the data block.

[0009] As a further aspect of the present invention, the steps of pre-calculating and pre-configuring the transmission parameters of each sub-band before transmission using the reference signal pattern specifically include: Before signal transmission, offline simulation or rapid calculation is performed based on the channel quality assessment matrix and the predetermined structure of the reference signal pattern; The estimation accuracy to be achieved when performing channel estimation at the receiver using the reference signal pattern is pre-calculated; Based on the estimated accuracy, an initial range of equalizer coefficient values ​​is preset for each sub-band; To address the statistical characteristics of impulse noise, different time-domain protection mechanism trigger thresholds are pre-configured for different sub-bands; The initial range of equalizer coefficient values ​​and the trigger threshold of the time-domain protection mechanism are used as a sub-band parameter configuration table and associated with the corresponding data blocks.

[0010] As a further aspect of the present invention, the step of adjusting the amplitude and phase of the sub-band symbols carrying data according to pre-configured parameters specifically includes: Read the sub-band parameter configuration table associated with the current data block; For each subband symbol carrying valid data, the pre-compensation weight that should be applied to the subband symbol at the transmitter is derived based on the initial equalizer coefficient value range of its subband. The pre-compensation weights include an amplitude scaling factor and a phase rotation factor, which are used to partially offset the distortion introduced by the channel; The pre-compensation weights of all subbands are combined into a diagonal matrix, and the subband symbol vector carrying the data is multiplied by the diagonal matrix to achieve symbol-level pre-equalization. After pre-equalization is completed, the peak power of each sub-band symbol is checked, and individual symbols that exceed the threshold are subjected to amplitude limiting and smoothing.

[0011] As a further aspect of the present invention, the step of applying a time-domain shaping window function to generate a physical waveform signal specifically includes: Choose a time-domain window function with fast edge roll-off characteristics; The time-domain signal after sub-band merging is multiplied by the time-domain window function to smoothly transition the amplitude of the beginning and end portions of the signal frame to zero. Calculate the cyclic prefix length of the windowed signal frame, which is associated with the guard interval length determined in the adaptive block processing step; A waveform segment of equal length to the cyclic prefix is ​​extracted from the end of the signal frame, copied, and added to the head of the signal frame to form the final physical waveform signal to be sent. The physical waveform signal is converted from digital to analog and amplified in power before being coupled to a low-voltage power line.

[0012] As a further aspect of the present invention, the step of capturing and synchronizing the physical waveform signal at the receiving end specifically includes: Amplify and filter the signal coupled from the power line; Using matched filtering techniques, sliding correlation detection is performed on the special structure in the signal composed of the time-domain shaping window function and the cyclic prefix. When the relevant peak value exceeds the set threshold, the starting position of the captured valid signal frame is determined; Based on the position and shape of the relevant peaks, the sampling clock phase of the receiver is finely adjusted to achieve symbol timing synchronization; Based on the known characteristics of the reference signal pattern, the residual carrier frequency deviation is estimated and compensated in the lifting wavelet transform domain to complete carrier synchronization.

[0013] As a further aspect of the present invention, the step of performing reverse processing on the synchronized signal, separating each sub-band, and performing data recovery specifically includes: Remove the cyclic prefix from the header of the received signal frame; Apply the same time-domain shaping window function as the transmitter to the signal frame after removing the cyclic prefix; The windowed signal is subjected to multi-level lifting wavelet transform to decompose the signal into various sub-bands; Based on the reference signal pattern embedding location map, the known reference signal is extracted from each sub-band; Using the extracted reference signal and the preset value range in the sub-band parameter configuration table, the channel response of each sub-band is accurately estimated, and the final equalizer coefficient is calculated.

[0014] As a further aspect of the present invention, the step of performing reverse processing on the synchronized signal, separating each sub-band, and performing data recovery also includes a data sub-band processing procedure: The final equalizer coefficients are used to perform frequency domain equalization on the symbols carrying data in each sub-band to compensate for channel effects. Monitor the energy mutation of each sub-band symbol in the time domain. If the instantaneous energy of a sub-band symbol exceeds the trigger threshold of its corresponding time domain protection mechanism, it is determined that each sub-band symbol is affected by impulse noise interference. For symbols identified as being interfered with, reconstruction and replacement are performed using interpolation algorithms based on the values ​​of their adjacent sub-bands and symbols at adjacent time points. Demap the data symbols after equalization and interference repair to restore them to a binary bit stream; All binary bit streams restored from the same data block are reassembled in sequence to obtain an estimate of the original communication data.

[0015] As a further aspect of the present invention, the dynamic adjustment step of the time-domain protection mechanism trigger threshold specifically includes: At the receiving end, the number of times each sub-band symbol is identified as being subject to pulse interference and its distribution pattern are continuously statistically analyzed. The statistical results are fed back to the transmitter via the reverse control channel; The transmitter updates the statistical characteristics of impulse noise in the channel quality assessment matrix based on historical feedback information; Based on the updated statistical characteristics, the trigger threshold values ​​of the time-domain protection mechanism corresponding to each sub-band in the subsequent data blocks are dynamically adjusted and written into the new sub-band parameter configuration table.

[0016] Compared with the prior art, the advantages and positive effects of the present invention are as follows: By dynamically adjusting the data block size based on the real-time detected noise characteristics of the low-voltage power line channel, the data block specifications can be synchronized and adapted to the real-time noise state of the channel. This eliminates the adaptation deviation between fixed block specifications and channel transmission characteristics, weakens the direct interference of channel noise on communication data, reduces signal distortion caused by block mismatch during data transmission, and ensures that the block processing form of the original communication data conforms to the real-time transmission conditions of the low-voltage power line channel. This reduces the impact of noise and multipath interference superposition on data blocks, makes the data block processing logic correspond to the channel transmission state, ensures the rationality of data block processing, reduces signal processing deviations in subsequent subband mapping stages, and makes the signal form in the initial data processing stage more suitable for the transmission environment of low-voltage power lines.

[0017] Performing lifting wavelet transform on each data block and mapping it to multiple sub-bands composed of wavelet functions of different scales can leverage the multi-scale analysis characteristics of lifting wavelet transform to change the signal processing method of traditional sub-band mapping, reduce the degree of signal crosstalk between sub-bands, enhance the anti-interference capability of sub-band signals in low-voltage power line multipath interference environments, adapt the signal structure of the initial sub-band symbols to the channel transmission characteristics, reduce the loss of sub-band signals during transmission, optimize the signal distribution state of sub-band symbols, provide a stable signal foundation for subsequent transmission parameter pre-configuration, and sub-band symbol amplitude and phase adjustment, improve the stability of the data carried by each sub-band signal, weaken the signal distortion problem caused by traditional transform methods, and make the transmission characteristics of sub-band signals consistent with the transmission law of low-voltage power line channels, thereby improving the integrity and stability of signal transmission. Attached Figure Description

[0018] Figure 1 This is a timing diagram of the low-voltage power line multi-carrier communication system based on LWT-OFDM described in this invention; Figure 2 A flowchart for embedding a known reference signal pattern; Figure 3 For the waveform synthesis module: comparison of symbol amplitude before and after pre-equalization; Figure 4 This describes the iterative convergence process of the LWT-OFDM equalizer coefficients. Figure 5 The heatmap is dynamically updated to reflect the threshold of the time-domain protection mechanism. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0020] In the description of this invention, it should be understood that the terms "length," "width," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicating orientation or positional relationships, are based on the orientation or positional relationships shown in the accompanying drawings and are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Furthermore, in the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0021] See Figure 1 This invention provides a low-voltage power line multi-carrier communication system based on LWT-OFDM, specifically comprising: The adaptive processing module first acquires the raw communication data from the low-voltage power line, which may originate from an upper-layer application. This module performs preliminary format encapsulation on the raw communication data, such as adding necessary frame headers or protocol information. The module then performs adaptive block processing on the encapsulated data; the block size is not fixed but dynamically adjusted based on real-time detected channel noise characteristics. The segmented data enters the sub-band mapping module. This module performs a lifting wavelet transform on each data block, mapping it to multiple sub-bands composed of wavelet functions of different scales, forming an initial set of sub-band symbols. The parameter pre-configuration module embeds known reference signal patterns at specific positions on the formed initial sub-band symbols. Using these pre-defined reference signal patterns, the module pre-calculates and pre-configures the transmission parameters of each sub-band before signal transmission. The waveform synthesis module adjusts the amplitude and phase of the symbols carrying data on each sub-band according to the pre-configured parameters. Finally, the module merges the symbols from all sub-bands into a time-domain signal and applies a time-domain shaping window function to generate the physical waveform signal to be transmitted. The channel transceiver module is responsible for feeding the generated physical waveform signal into the low-voltage power line channel for transmission. At the receiving end, this module captures and synchronizes the physical waveform signal received from the channel. For the synchronized signal, the module performs a reverse processing flow, separating each sub-band and recovering the data, finally outputting an estimate of the original communication data.

[0022] In one embodiment of the present invention, the adaptive processing module performs adaptive block processing on the encapsulated data. Specifically, during data transmission intervals, the system periodically sends a probe signal sequence to the power line channel. The module receives and analyzes the distortion and attenuation of the probe signal sequence after transmission in the channel, thereby quantifying the noise intensity and impulse interference statistical characteristics of the current channel. Based on this, the module establishes a channel quality assessment matrix, which includes background noise power, average impulse noise occurrence rate, and channel response flatness index. Subsequently, the module queries a preset mapping rule table based on the various indices in the channel quality assessment matrix. This mapping rule table defines the optimal combination of data block length and guard interval length corresponding to different channel conditions. The system applies the data block length and guard interval length obtained from the mapping rule table under the current channel conditions to segment the encapsulated data to be transmitted subsequently. This operation ensures that each data block has an independent guard interval in the time domain.

[0023] In practical implementation, the adaptive processing module of the low-voltage power line multi-carrier communication system performs adaptive block processing on the encapsulated data. During the data transmission interval, the system periodically sends a predefined probe signal sequence to the power line channel. The probe signal sequence consists of a set of pseudo-random symbols with wide-spectrum characteristics. The module at the receiving end captures the probe signal sequence after it has been transmitted through the channel, receives and analyzes the distortion and attenuation of the probe signal sequence. The analysis process includes calculating the mean square error between the received signal and the original probe signal sequence, as well as the abrupt change components in the detected signal waveform. Based on the calculation results, the noise intensity and impulse interference statistical characteristics of the current channel are quantified. The noise intensity is expressed as the equivalent signal-to-noise ratio, and the impulse interference statistical characteristics are expressed as the number of impulses with amplitudes exceeding a threshold per unit time.

[0024] In some embodiments, the adaptive processing module establishes a channel quality assessment matrix, which is a data structure containing multiple quantified indicators. The channel quality assessment matrix records background noise power, average impulse noise occurrence rate, and channel response flatness index. Background noise power is obtained through spectral analysis of the silent segment of the probe signal; the average impulse noise occurrence rate is obtained by statistically analyzing high-intensity interference events within a preset time window; and the channel response flatness index is characterized by calculating the variance of the gain of each sub-band of the probe signal. Based on the indicators in the channel quality assessment matrix, the module queries a preset mapping rule table stored in local memory. The mapping rule table defines the optimal combination of data block length and guard interval length corresponding to different channel conditions, using background noise power, average impulse noise occurrence rate, and channel response flatness index as joint input indices. When constructing the channel quality assessment matrix, the system employs a sliding window statistical model to continuously update background noise power, average impulse noise occurrence rate, and channel response flatness. This model uses a fixed time window as the statistical unit, performing point-by-point calculations on the probe signal sampling points within the current window. It does not rely on preset empirical values ​​or simulation fitting data; all indicators are obtained through real-time sampling and actual channel response calculations, ensuring that the channel state description is consistent with the physical channel. When querying the preset mapping rule table, the rule table uses the measured channel indicators as the sole input, and the output results are only used to determine the data block length and guard interval length, without introducing any additional manually set parameters. The block cutting operation is strictly executed according to the mapping results, ensuring that the temporal structure of each data block matches the real-time transmission capability of the channel.

[0025] Optionally, the mapping rule table is obtained in advance through offline simulation and channel measurement experiments. The core mapping relationship of the mapping rule table is expressed by the following function:

[0026] Where: function character Indicates the calculated data block length, function character Represents the background noise power in the channel quality assessment matrix, function character. The function character represents the average impulse noise occurrence rate in the channel quality assessment matrix. The function represents the channel response flatness index in the channel quality assessment matrix. This represents a non-linear mapping from the three input parameters to the block length. The system uses the data block length obtained by querying the mapping rule table based on the current channel conditions, along with the associated guard interval length, to perform segmentation operations on the encapsulated data to be transmitted subsequently.

[0027] It can be understood that the segmentation operation extracts data blocks from a continuous encapsulated data stream, using the data block length as the unit. Each extracted data block is assigned an independent guard interval in the time domain, and the length of the guard interval, along with the data block length, is given by a mapping rule table. In some embodiments, when the channel quality assessment matrix indicates an increase in the average impulse noise rate, the mapping rule table outputs a combination of a shorter data block length and a longer guard interval length. In a specific implementation, after the adaptive processing module completes the segmentation of the current data frame, it outputs the block data with the guard interval length attached to the subsequent sub-band mapping module, and simultaneously writes the currently used data block length and guard interval length information into the block header control field for synchronous use by the receiving end.

[0028] In one embodiment of the present invention, after the sub-band mapping module completes the mapping of the initial sub-band symbols, the parameter pre-matching module embeds a known reference signal pattern at a specific position of the initial sub-band symbols. This process includes, see [reference] Figure 2This paper designs a reference signal sequence with good autocorrelation characteristics, exhibiting a non-uniform power distribution across subbands in the lifting wavelet transform domain. The lifting wavelet transform employs a multi-scale progressive decomposition algorithm, mapping data blocks sequentially to different scale subbands according to a fixed number of decomposition levels. Each decomposition level uses a linear lifting structure, without using custom nonlinear transformations or fictitious filter coefficients. The wavelet basis function is a commonly used bioorthogonal wavelet basis in the communication field. Subband division is generated progressively based on scale coefficients, maintaining strict orthogonality between subbands and effectively reducing crosstalk between them. This decomposition process is a publicly available and implementable digital signal processing flow, fully reproducible on a general-purpose digital signal processor. The module identifies the sets of subbands with high and low signal-to-noise ratios (SNR) under the current channel conditions based on the channel quality assessment matrix. The system inserts the reference signal sequence at closer intervals into the identified high-SNR subbands and at sparser intervals into the identified low-SNR subbands, using the saved symbol resources to carry effective data transmission. The module generates a unique reference signal pattern embedding location map for each data block, which is transmitted to the receiver along with the header control information of the data block. Using the embedded reference signal pattern, the transmission parameters of each sub-band are pre-calculated and pre-configured before transmission. Specifically, before signal transmission, the system performs offline simulation or rapid calculation based on the existing channel quality assessment matrix and the predetermined structure of the reference signal pattern. The module pre-calculates the estimation accuracy achievable at the receiver when using the reference signal pattern for channel estimation. Based on this estimation accuracy, the module presets an initial equalizer coefficient value range for each sub-band. Simultaneously, considering the statistical characteristics of impulse noise, the module pre-configures different time-domain protection mechanism trigger thresholds for different sub-bands. Finally, the module integrates the initial equalizer coefficient value range and the time-domain protection mechanism trigger thresholds into a sub-band parameter configuration table and associates it with the corresponding data block.

[0029] In practical implementation, after the subband mapping module completes the mapping of the initial subband symbols, the parameter pre-matching module embeds a known reference signal pattern at specific positions of the initial subband symbols. This process begins by designing a reference signal sequence with good autocorrelation characteristics. The reference signal sequence adopts a Gray complement sequence, and the reference signal sequence has a non-uniform power distribution across each subband in the lifting wavelet transform domain. The power distribution is designed based on the reciprocal of the subband center frequency. Based on the channel quality assessment matrix obtained from the adaptive processing module, which contains subband-level signal-to-noise ratio (SNR) estimates, the module identifies the sets of subbands with high SNR and low SNR under the current channel conditions. The identification rule is that subbands with SNR above a threshold are assigned to the high SNR subband set, and the rest are assigned to the low SNR subband set.

[0030] In some embodiments, on the identified high SNR subband set, the system inserts reference signal sequences at dense intervals, with one reference signal inserted every four data symbols. On the identified low SNR subband set, the system inserts the same reference signal sequences at sparse intervals, with one reference signal inserted every sixteen data symbols, and uses the saved symbol resources to carry effective data transmission. The module generates a unique reference signal pattern embedding location map for each data block. The reference signal pattern embedding location map is a binary mask matrix, where the rows of the matrix correspond to the subband index, the columns correspond to the symbol time index, and the position with a value of "1" indicates the insertion of a reference signal. The reference signal pattern embedding location map is encoded along with the header control information of the data block and transmitted to the receiving end.

[0031] Optionally, the transmission parameters of each subband can be pre-calculated and pre-configured before transmission using an embedded reference signal pattern. Before signal transmission, the system performs rapid calculations based on the existing channel quality assessment matrix and the predetermined structure of the reference signal pattern. This rapid calculation is accomplished by looking up pre-stored channel response templates and noise covariance matrices. The module pre-calculates the achievable estimation accuracy at the receiver using the reference signal pattern, with the estimation accuracy represented by the lower bound of the mean square error. Based on this estimation accuracy, the module presets an initial equalizer coefficient value range for each subband. This range is a complex interval, with the center point set based on the reciprocal of the ideal channel response, and the interval radius proportional to the square root of the estimation accuracy.

[0032] In one embodiment of the present invention, the waveform synthesis module adjusts the amplitude and phase of the sub-band symbols carrying data according to pre-configured parameters. Specifically, the module reads the sub-band parameter configuration table associated with the current data block. For each sub-band symbol carrying valid data, the module derives the pre-compensation weight to be applied at the transmitter based on the preset initial equalizer coefficient range for its sub-band in the configuration table. The pre-compensation weight includes an amplitude scaling factor and a phase rotation factor, used to partially offset distortions that may be introduced by the channel. The module forms a diagonal matrix from the pre-compensation weights of all sub-bands, and multiplies the sub-band symbol vector carrying data with this diagonal matrix to achieve symbol-level pre-equalization. After pre-equalization, the module checks the peak power of each sub-band symbol and performs amplitude limiting and smoothing on individual symbols exceeding a preset threshold. Subsequently, the waveform synthesis module applies a time-domain shaping window function to generate the physical waveform signal. This step includes selecting a time-domain window function with fast edge roll-off characteristics. The module performs a dot product operation between the subband-merged time-domain signal and the time-domain window function, ensuring a smooth transition of amplitude to zero at the beginning and end of the signal frame. The module calculates the required cyclic prefix length for the windowed signal frame, which is correlated with the guard interval length determined in the adaptive block processing step. Then, the module extracts a waveform segment equal to the cyclic prefix length from the end of the signal frame, copies it, and adds it to the beginning of the signal frame, forming the final physical waveform signal to be transmitted. Finally, the module performs digital-to-analog conversion and power amplification on this physical waveform signal before coupling it to the low-voltage power line.

[0033] In practice, the waveform synthesis module adjusts the amplitude and phase of the subband symbols carrying the data according to pre-configured parameters. The module reads the subband parameter configuration table associated with the current data block, which contains the initial equalizer coefficient range for each subband. For each subband symbol carrying valid data, the module derives the pre-compensation weights to be applied at the transmitter based on the initial equalizer coefficient range of the subband to which the subband symbol belongs. These pre-compensation weights include an amplitude scaling factor and a phase rotation factor. The amplitude scaling factor is used to pre-compensate for amplitude fading that may be introduced by the channel, and the phase rotation factor is used to pre-correct phase offset that may be introduced by the channel. The module combines the pre-compensation weights of all subbands into a diagonal matrix, multiplies the subband symbol vector carrying the data by this diagonal matrix, and performs symbol-level pre-equalization. The multiplication operation is completed in the lifting wavelet transform domain. After pre-equalization is completed, the waveform synthesis module checks the peak power of each sub-band symbol and performs amplitude limiting and smoothing processing on individual symbols that exceed the preset threshold. The amplitude limiting and smoothing processing uses a soft compression function to smoothly compress the amplitude of the symbols that exceed the threshold to below the threshold value.

[0034] In some embodiments, the specific steps of applying a time-domain shaping window function to generate a physical waveform signal begin with window function selection. The waveform synthesis module selects a time-domain window function with a fast edge roll-off characteristic from a set of candidate functions. The fast edge roll-off characteristic refers to the window function's amplitude dropping from a stable value to zero in a very short time at the time-domain edge. Referring to Table 1, several candidate time-domain window functions and their corresponding roll-off characteristics are shown.

[0035] Table 1: Domain Window Functions and Corresponding Roll-off Characteristics

[0036] The waveform synthesis module performs a dot product operation on the time-domain signal after sub-band combining and a selected time-domain window function. This dot product operation smoothly transitions the amplitude of the beginning and end portions of the signal frame to zero, thereby reducing out-of-band spectral leakage. The waveform synthesis module calculates the cyclic prefix length of the windowed signal frame. This cyclic prefix length is related to the guard interval length determined in the adaptive block processing step, and the correlation is defined by the following formula:

[0037] Where: characters This represents the final calculated length of the cyclic prefix, in characters. This represents the length of the guard interval determined in adaptive block processing, character. This represents the estimated maximum multipath delay spread of the channel recorded in the channel quality assessment matrix, in characters. Represents the system sampling rate, symbol This indicates a rounding down operation. The waveform synthesis module extracts a segment of length [length missing] from the end of the windowed signal frame. The waveform segment is copied and added to the header of the signal frame to form the final physical waveform signal to be sent.

[0038] Optionally, the addition of the cyclic prefix is ​​performed in the time domain. The waveform synthesis module performs digital-to-analog conversion and power amplification on the final physical waveform signal. The sampling rate of the digital-to-analog conversion is strictly matched with the symbol rate of the lifting wavelet transform, and the power amplification gain is set according to the power line channel drive requirements. The power-amplified analog signal is fed into the low-voltage power line through a coupling circuit.

[0039] It is understandable that, in practical implementation, the dot product operation between the window function and the signal, as well as the truncation and addition of the cyclic prefix, constitute the core of physical waveform signal shaping. Before performing the windowing operation, the waveform synthesis module must ensure that the length of the time-domain signal is completely consistent with the length of the window function; if they are inconsistent, the signal is padded with zeros or truncated. In some embodiments, the window function is pre-stored in read-only memory, and the waveform synthesis module retrieves the window function coefficients by looking up a table.

[0040] See Figure 3 In a low-voltage power line multi-carrier communication system based on LWT-OFDM, the symbol pre-equalization and amplitude limiting effects of the waveform synthesis module are visually presented in this figure. Specifically, the waveform synthesis module reads the sub-band parameter configuration table associated with the current data block. Based on the preset initial equalizer coefficient range for each sub-band, it derives the pre-compensation weights composed of the amplitude scaling factor and the phase rotation factor. All sub-band pre-compensation weights are then combined into a diagonal matrix and multiplied with the sub-band symbol vector carrying the data in the lifting wavelet transform domain to complete the symbol-level pre-equalization operation. The dotted curves in the figure represent the amplitude before equalization, reflecting the original amplitude distribution of each sub-band symbol before the pre-compensation operation. It can be seen that the amplitude of some symbols is significantly higher than the preset amplitude limiting threshold (dashed line, threshold value is 2.0), posing a risk of exceeding peak power limits. The boxed curves represent the amplitude after equalization (including amplitude limiting), reflecting the symbol amplitude state after pre-equalization and amplitude limiting smoothing processing: After the pre-equalization operation completes amplitude pre-compensation, the module uses a soft compression function to smooth and limit the amplitude of symbols exceeding the 2.0 amplitude limiting threshold, uniformly constraining the amplitude of symbols exceeding the threshold to below the threshold value, while retaining the amplitude characteristics of symbols below the threshold. Ultimately, this achieves the normalization of all symbol amplitudes, effectively suppressing peak power and ensuring the power compliance and transmission stability of the signal transmission. The figure visually verifies the effectiveness of the pre-equalization and amplitude limiting processing in the waveform synthesis module: pre-compensation offsets potential channel distortion, and amplitude limiting controls peak power, providing amplitude-compliant input signals for subsequent waveform shaping operations such as time-domain windowing and cyclic prefix addition, ensuring the transmission performance and spectral efficiency of multi-carrier communication under low-voltage power line channels.

[0041] In one embodiment of the present invention, the channel transceiver module captures and synchronizes physical waveform signals at the receiving end. This process specifically includes amplification and filtering preprocessing of the signal coupled from the power line. The module utilizes matched filtering technology to perform sliding correlation detection on the special structure in the signal formed by the time-domain shaping window function and cyclic prefix applied by the transmitting end. When the correlation peak exceeds a set threshold, the module determines the starting position of the captured valid signal frame. Based on the position and shape of the correlation peak, the module finely adjusts the sampling clock phase of the receiver to achieve symbol timing synchronization. Based on the known characteristics of the reference signal pattern, the module estimates and compensates for the residual carrier frequency deviation in the lifting wavelet transform domain, completing carrier synchronization. After synchronization, the module performs a reverse processing procedure on the signal to separate each sub-band and recover the data. Part of this reverse process involves removing the cyclic prefix from the header of the received signal frame. The same time-domain shaping window function as that applied at the transmitting end is applied to the signal frame after removing the cyclic prefix. The module performs multi-level lifting wavelet transform on the windowed signal to decompose the signal into each sub-band. The module extracts known reference signals from each sub-band based on the reference signal pattern embedding location map obtained from the data block header information. Using the extracted reference signals and the preset value range in the sub-band parameter configuration table, the module accurately estimates the channel response of each sub-band and calculates the final equalizer coefficients.

[0042] In practical implementation, the channel transceiver module at the receiving end performs acquisition and synchronization of the physical waveform signal, amplifies and filters the signal coupled from the power line, with amplification performed by a programmable gain amplifier and filtering by a bandpass filter to suppress out-of-band noise. The channel transceiver module utilizes matched filtering technology to perform sliding correlation detection on the special structure of the signal, composed of a time-domain shaped window function and a cyclic prefix applied at the transmitting end. The reference template for matched filtering is a locally stored, cyclically expanded time-domain shaped window function waveform. When the correlation peak value output by the sliding correlation calculation exceeds a set threshold, the channel transceiver module determines the starting position of the captured valid signal frame, based on the monotonicity of the correlation peak value across three consecutive sampling points. The channel transceiver module finely adjusts the sampling clock phase of the receiver based on the position and shape of the correlation peak to achieve symbol timing synchronization. This adjustment is accomplished by controlling the phase error input of the digital phase-locked loop. Based on the known characteristics of the reference signal pattern, the channel transceiver module estimates and compensates for the residual carrier frequency deviation in the lifting wavelet transform domain to achieve carrier synchronization. The estimation method is based on the phase difference of the reference signal between adjacent symbols.

[0043] In some embodiments, a reverse processing procedure is performed on the synchronized signal to separate each subband and recover the data. The first step of the reverse processing procedure is to remove the cyclic prefix from the header of the received signal frame. The channel transceiver module removes sampling points of the corresponding header length from the received signal based on the frame start position determined during the synchronization phase and the known cyclic prefix length field. The same time-domain shaping window function as the transmitting end is applied to the signal frame after removing the cyclic prefix; the application process involves multiplying the window function by its conjugate to prevent amplitude distortion. The channel transceiver module performs a multi-level lifting wavelet transform on the windowed signal. This multi-level lifting wavelet transform decomposes the signal into each subband, with the number of decomposition levels matching the number used by the transmitting subband mapping module. Refer to Table 2, which lists the possible combinations of lifting wavelet filters used for signal decomposition at the receiving end and their corresponding computational complexities.

[0044] Table 2: Wavelet Filter Combinations and Corresponding Computational Complexity

[0045] The channel transceiver module extracts known reference signals from each sub-band based on the reference signal pattern embedding location map decoded from the data block header information. The extraction operation directly reads the symbol value at the corresponding position according to the index in the location map. Using the extracted reference signals and the preset value range in the sub-band parameter configuration table, the channel transceiver module accurately estimates the channel response of each sub-band. The channel response estimation process is constrained by a preset equalizer coefficient value range; the estimated values ​​are limited to this complex interval. Finally, the equalizer coefficients are calculated based on the minimum mean square error criterion and solved using the following constraint relationships:

[0046] Where: characters This represents the estimated value of the final equalizer coefficient for the k-th subband obtained from the solution, character... This represents the reference signal vector extracted by the receiver from the k-th sub-band, character... Represents the locally known original reference signal sequence, character Represents the equalizer coefficient variable to be optimized, with the symbol... This represents the complex set of values ​​for the initial equalizer coefficients preset for the k-th subband in the subband parameter configuration table, denoted by the symbol. Let L be the squared norm of a vector. The solution process is performed on a predefined set of complex numbers. The process is carried out internally, using numerical search methods to find the value that minimizes the objective function. Value as .

[0047] Optionally, channel response estimation and equalizer coefficient calculation are iteratively performed in the digital signal processor. The calculated final equalizer coefficients are stored in a buffer for subsequent frequency domain equalization operations on all data symbols within the same data block. It is understood that removing the cyclic prefix and applying a time-domain window are necessary pre-processing steps for the reverse engineering, and their accuracy directly affects the quality of the subsequent lifting wavelet transform decomposition. In specific implementations, the coefficients of the window function must be strictly consistent with those at the transmitter and pre-stored in the receiver's memory. In some embodiments, carrier frequency offset compensation is performed after the lifting wavelet transform, and the compensation algorithm is based on the maximum likelihood criterion, utilizing the phase information of the reference signal pattern.

[0048] See Figure 4 In a low-voltage power line multi-carrier communication system based on LWT-OFDM, the realization of the equalizer coefficient iterative convergence process relies on receiver channel estimation and iterative optimization techniques under the minimum mean square error (MMSE) criterion. Specifically, after signal acquisition, synchronization, and multi-layer lifting wavelet transform, the receiver extracts known reference signals from each sub-band based on the reference signal pattern embedding location map. Combined with the preset initial equalizer coefficient value range in the sub-band parameter configuration table, it accurately estimates the channel response of each sub-band and iteratively updates the equalizer coefficients with the goal of minimizing the mean square error (MSE). In the figure, the vertical axis represents the mean square error (MSE), characterizing the deviation between the estimated equalizer coefficient value and the ideal optimal value, while the horizontal axis represents the number of iterations, reflecting the optimization iterative process of the equalizer coefficients. In the initial iteration phase (0-10 iterations), the equalizer coefficients deviate significantly from the optimal value, and the MSE drops rapidly from approximately 0.7. This phase represents the coarse convergence process of the coefficients. The initial coefficient values ​​are quickly corrected through channel estimation of the reference signal to offset major distortions such as multipath fading and noise interference in the power line channel. In the middle iteration phase (10-30 iterations), the rate of MSE decrease slows down, and the fine-tuning phase begins. Under the constraints of the subband parameter configuration table, the system continuously optimizes within a preset complex range using numerical search methods, gradually approximating the true inverse matrix of the channel response. In the later iteration phase (after 30 iterations), the MSE converges to a stable level close to 0, with only minor fluctuations. This indicates that the equalizer coefficients have converged to the optimal solution, effectively compensating for the amplitude and phase distortions of each subband symbol caused by the channel, and providing reliable coefficient support for subsequent frequency domain equalization and data demapping. The convergence characteristics verify the effectiveness of the MMSE iterative equalization algorithm adopted in this system, which is constrained by the sub-band parameter configuration table. In the complex channel environment of low-voltage power lines, it can converge quickly through a finite number of iterations, taking into account both equalization accuracy and algorithm real-time performance, and ensuring the transmission reliability of the multi-carrier communication system.

[0049] In one embodiment of the present invention, a reverse processing flow is performed on the synchronized signal to separate each sub-band and recover the data. This process also includes a data sub-band processing flow. This flow involves using the finally calculated equalizer coefficients to perform frequency domain equalization on the symbols carrying data in each sub-band to compensate for channel interference. The module monitors the energy mutations of each sub-band symbol in the time domain. If the instantaneous energy of a sub-band symbol exceeds its corresponding time domain protection mechanism trigger threshold, the sub-band symbol is determined to be subject to impulse noise interference. For the symbols determined to be interfered with, the module uses the values ​​of its adjacent sub-bands and symbols at adjacent times to reconstruct and replace them using an interpolation algorithm. The module demaps the data symbols after equalization and interference repair, restoring them to binary bit streams. The module reassembles all the binary bit streams restored from the same data block in their original order to obtain an estimate of the original communication data. The time domain protection mechanism trigger threshold can be dynamically adjusted. This adjustment step involves continuously counting the number of times each sub-band symbol is determined to be subject to impulse interference and its distribution pattern at the receiving end. The module feeds back the statistical results to the transmitting end through a reverse control channel. Based on historical feedback information, the transmitter updates the statistical characteristics of impulse noise in the channel quality assessment matrix. Using these updated statistical characteristics, the transmitter dynamically adjusts the trigger threshold values ​​of the time-domain protection mechanism for each sub-band in subsequent data blocks and writes them into a new sub-band parameter configuration table.

[0050] In practical implementation, the synchronized signal undergoes a reverse processing flow to separate each sub-band and recover the data. This process includes the data sub-band processing flow. The channel transceiver module uses the calculated final equalizer coefficients to perform frequency domain equalization on the symbols carrying data in each sub-band. The frequency domain equalization operation multiplies the received data symbols in each sub-band by the corresponding final equalizer coefficient. The channel transceiver module monitors the energy mutations of each sub-band symbol in the time domain. The monitoring method includes calculating the ratio of the instantaneous power of each symbol to the local moving average power. If the instantaneous energy of a sub-band symbol exceeds the trigger threshold of the time domain protection mechanism corresponding to the sub-band, the channel transceiver module determines that the sub-band symbol is subject to impulse noise interference. For the symbols determined to be interfered with, the channel transceiver module uses the values ​​of its adjacent sub-bands and adjacent time symbols to reconstruct and replace them using an interpolation algorithm. The interpolation algorithm adopts a two-dimensional linear Lagrange interpolation method. The channel transceiver module demaps the data symbols after frequency domain equalization and interference repair. The demapping process restores the complex symbols to a binary bit stream according to the modulation constellation diagram agreed upon by the transmitter. The channel transceiver module reassembles all the binary bit streams restored from the same data block according to the original transmission order. The reassembly operation is completed based on the sequence number in the block header information, and finally obtains the estimated value of the original communication data.

[0051] In some embodiments, the trigger threshold of the time-domain protection mechanism can be dynamically adjusted based on feedback information. At the receiving end, the channel transceiver module continuously counts the number and distribution patterns of symbols in each sub-band that are determined to be subject to impulse interference. The statistics are recorded in data frames, noting the index of the symbol determined to be interfered with and its corresponding instantaneous power value within each sub-band. The channel transceiver module feeds back the statistical results, including interference statistics for each sub-band, to the transmitting end through a reverse control channel. The reverse control channel uses an independent logical channel or utilizes the guard interval of the data frames for in-band transmission.

[0052] Optionally, the transmitter updates the statistical characteristics of impulse noise in the channel quality assessment matrix based on the received historical feedback information. The update operation uses an exponentially weighted moving average algorithm to smoothly estimate the average impulse noise occurrence rate and the average power of interfering impulses. Based on the updated impulse noise statistical characteristics, the transmitter dynamically adjusts the time-domain protection mechanism trigger threshold values ​​corresponding to each sub-band in subsequent data blocks, following the following update rules:

[0053] Where: characters This represents the new time-domain protection mechanism trigger threshold calculated for the k-th sub-band, with power as its dimension. This represents the average power estimate of the interfered symbol in the k-th sub-band, obtained from the feedback information, with the dimension being power. (Character) This represents the original time-domain protection mechanism trigger threshold value for the k-th sub-band, with the dimension being power. (Character) and It is a positive weighting coefficient and satisfies To ensure that the dimensions of the left and right sides of the formula are consistent and represent power, the transmitter will write the calculated new time-domain protection mechanism trigger threshold value into the new sub-band parameter configuration table.

[0054] It is understandable that the dynamic adjustment process allows the trigger threshold of the time-domain protection mechanism to adapt to the energy level of the interference pulse. In specific implementation, the weighting coefficients... and The value can be preset according to the system's tracking speed requirements for channel changes. In some embodiments, the average power estimate is... The calculations exclude obviously anomalous maxima to improve statistical robustness. After applying the new subband parameter configuration table, the channel transceiver module updates its threshold for monitoring energy mutations, thereby altering its sensitivity to impulse noise interference.

[0055] See Figure 5In a low-voltage power line multi-carrier communication system based on LWT-OFDM, the dynamic update process of the time-domain protection mechanism trigger threshold is visually presented through a heatmap. Specifically, the vertical axis distinguishes between the original threshold and the new threshold states, while the horizontal axis corresponds to the eight independent sub-bands obtained from the lifting wavelet transform decomposition. Color codes represent the power values ​​of the time-domain protection threshold for each sub-band, with units of power. In the original threshold state, all sub-bands adopt a unified initial threshold configuration, corresponding to a uniform blue area in the heatmap. The threshold power is uniformly on the order of 2.0, and this initial threshold is pre-configured by the transmitter based on the impulse noise statistical characteristics of the channel quality assessment matrix. In the new threshold state, the transmitter updates the impulse noise statistical characteristics in the channel quality assessment matrix using an exponentially weighted moving average algorithm based on the impulse interference statistical results of each sub-band fed back by the receiver through the reverse control channel. Furthermore, according to preset threshold update rules, the time-domain protection threshold for each sub-band is dynamically adjusted in a differentiated manner. As can be seen from the heatmap, the threshold power of subbands 1-4 remains at a relatively low level of 1.9-2.1, corresponding to subbands with good channel conditions and low impulse interference rate, maintaining high interference decision sensitivity. The threshold power of subbands 5-8 gradually increases with the subband number, reaching a maximum of 2.6, corresponding to subbands with poor channel conditions and high impulse interference energy. Raising the threshold avoids misjudgment while ensuring accurate identification of effective interference. This dynamic update mechanism achieves adaptive matching between the threshold and the subband channel state, maximizing system transmission efficiency and reliability while suppressing impulse noise interference.

[0056] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments that can be applied to other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A low-voltage power line multi-carrier communication system based on LWT-OFDM, characterized in that, include: An adaptive processing module acquires raw communication data from low-voltage power lines, performs preliminary format encapsulation on the raw communication data, and performs adaptive block processing on the encapsulated data. The block size is dynamically adjusted based on the channel noise characteristics detected in real time. The subband mapping module performs lifting wavelet transform on each data block and maps it to multiple subbands composed of wavelet functions of different scales to form initial subband symbols. The parameter pre-configuration module embeds a known reference signal pattern at a specific position of the initial sub-band symbol, and uses the reference signal pattern to pre-calculate and pre-configure the transmission parameters of each sub-band before transmission; The waveform synthesis module, based on pre-configured parameters, adjusts the amplitude and phase of the sub-band symbols carrying the data, merges the symbols of all sub-bands, and applies a time-domain shaping window function to generate the physical waveform signal. The channel transceiver module feeds the physical waveform signal into the low-voltage power line channel for transmission, captures and synchronizes the physical waveform signal at the receiving end, performs a reverse processing flow on the synchronized signal, separates each sub-band, and recovers the data.

2. The low-voltage power line multi-carrier communication system based on LWT-OFDM according to claim 1, characterized in that, The step of adaptively segmenting the encapsulated data, with the segment size dynamically adjusted based on real-time detected channel noise characteristics, specifically includes: During the intervals between data transmissions, a sequence of probe signals is periodically sent to the power line channel; The distortion and attenuation of the probe signal sequence are received and analyzed to quantify the noise intensity and impulse interference statistical characteristics of the current channel. Establish a channel quality assessment matrix, which includes background noise power, average impulse noise occurrence rate, and channel response flatness index. Based on the indicators in the channel quality assessment matrix, a preset mapping rule table is queried. The mapping rule table defines the optimal combination of data block length and guard interval length corresponding to different channel conditions. The data block length and guard interval length obtained by mapping the current channel conditions are applied to segment the encapsulated data to be transmitted, ensuring that each data block has an independent guard interval in the time domain.

3. The low-voltage power line multi-carrier communication system based on LWT-OFDM according to claim 2, characterized in that, The step of embedding a known reference signal pattern at a specific position of the initial sub-band symbol specifically includes: Design a reference signal sequence with good autocorrelation characteristics, wherein the reference signal sequence has a non-uniform power distribution in each subband of the lifting wavelet transform domain; Based on the channel quality assessment matrix, the set of subbands with high signal-to-noise ratio and the set of subbands with low signal-to-noise ratio under the current channel conditions are identified; The reference signal sequence is inserted at close intervals on the set of subbands with high signal-to-noise ratio; On the set of subbands with low signal-to-noise ratio, the reference signal sequence is inserted at sparse intervals, and the saved symbol resources are used for data transmission. A unique reference signal pattern embedding location map is generated for each data block, and the location map is transmitted to the receiving end along with the header control information of the data block.

4. The low-voltage power line multi-carrier communication system based on LWT-OFDM according to claim 3, characterized in that, Using the reference signal pattern, the steps for pre-calculating and pre-configuring the transmission parameters of each sub-band before transmission specifically include: Before signal transmission, offline simulation or rapid calculation is performed based on the channel quality assessment matrix and the predetermined structure of the reference signal pattern; The estimation accuracy to be achieved when performing channel estimation at the receiver using the reference signal pattern is pre-calculated; Based on the estimated accuracy, an initial range of equalizer coefficient values ​​is preset for each sub-band; To address the statistical characteristics of impulse noise, different time-domain protection mechanism trigger thresholds are pre-configured for different sub-bands; The initial range of equalizer coefficient values ​​and the trigger threshold of the time-domain protection mechanism are used as a sub-band parameter configuration table and associated with the corresponding data blocks.

5. The low-voltage power line multi-carrier communication system based on LWT-OFDM according to claim 4, characterized in that, The step of adjusting the amplitude and phase of the sub-band symbols carrying data according to the pre-configured parameters specifically includes: Read the sub-band parameter configuration table associated with the current data block; For each subband symbol carrying valid data, the pre-compensation weight that should be applied to the subband symbol at the transmitter is derived based on the initial equalizer coefficient value range of its subband. The pre-compensation weights include an amplitude scaling factor and a phase rotation factor, which are used to partially offset the distortion introduced by the channel; The pre-compensation weights of all subbands are combined into a diagonal matrix, and the subband symbol vector carrying the data is multiplied by the diagonal matrix to achieve symbol-level pre-equalization. After pre-equalization is completed, the peak power of each sub-band symbol is checked, and individual symbols that exceed the threshold are subjected to amplitude limiting and smoothing.

6. The low-voltage power line multi-carrier communication system based on LWT-OFDM according to claim 5, characterized in that, The step of applying a time-domain shaping window function to generate a physical waveform signal specifically includes: Choose a time-domain window function with fast edge roll-off characteristics; The time-domain signal after sub-band merging is multiplied by the time-domain window function to smoothly transition the amplitude of the beginning and end portions of the signal frame to zero. Calculate the cyclic prefix length of the windowed signal frame, which is associated with the guard interval length determined in the adaptive block processing step; A waveform segment of equal length to the cyclic prefix is ​​extracted from the end of the signal frame, copied, and added to the head of the signal frame to form the final physical waveform signal to be sent. The physical waveform signal is converted from digital to analog and amplified in power before being coupled to a low-voltage power line.

7. The low-voltage power line multi-carrier communication system based on LWT-OFDM according to claim 6, characterized in that, The step of capturing and synchronizing the physical waveform signal at the receiving end specifically includes: Amplify and filter the signal coupled from the power line; Using matched filtering techniques, sliding correlation detection is performed on the special structure in the signal composed of the time-domain shaping window function and the cyclic prefix. When the relevant peak value exceeds the set threshold, the starting position of the captured valid signal frame is determined; Based on the position and shape of the relevant peaks, the sampling clock phase of the receiver is finely adjusted to achieve symbol timing synchronization; Based on the known characteristics of the reference signal pattern, the residual carrier frequency deviation is estimated and compensated in the lifting wavelet transform domain to complete carrier synchronization.

8. The low-voltage power line multi-carrier communication system based on LWT-OFDM according to claim 7, characterized in that, The steps of performing reverse processing on the synchronized signal, separating each sub-band, and performing data recovery specifically include: Remove the cyclic prefix from the header of the received signal frame; Apply the same time-domain shaping window function as the transmitter to the signal frame after removing the cyclic prefix; The windowed signal is subjected to multi-level lifting wavelet transform to decompose the signal into various sub-bands; Based on the reference signal pattern embedding location map, the known reference signal is extracted from each sub-band; Using the extracted reference signal and the preset value range in the sub-band parameter configuration table, the channel response of each sub-band is accurately estimated, and the final equalizer coefficients are calculated. The calculation of the final equalizer coefficients is based on the minimum mean square error criterion and is solved using the following constraints: ； in, This represents the estimated value of the final equalizer coefficient for the k-th sub-band obtained from the solution. This represents the reference signal vector extracted by the receiver from the k-th sub-band. This represents the locally known original reference signal sequence. This represents the equalizer coefficient variable to be optimized. This represents the complex set of values ​​for the initial equalizer coefficients preset for the k-th subband in the subband parameter configuration table. The square of the second norm of a vector is calculated within a predefined set of complex numbers. The process is carried out internally, using numerical search methods to find the value that minimizes the objective function. Value as .

9. The low-voltage power line multi-carrier communication system based on LWT-OFDM according to claim 8, characterized in that, The step of performing reverse processing on the synchronized signal, separating each sub-band, and performing data recovery also includes a data sub-band processing procedure: The final equalizer coefficients are used to perform frequency domain equalization on the symbols carrying data in each sub-band to compensate for channel effects. Monitor the energy mutation of each sub-band symbol in the time domain. If the instantaneous energy of a sub-band symbol exceeds the trigger threshold of its corresponding time domain protection mechanism, it is determined that each sub-band symbol is affected by impulse noise interference. For symbols identified as being interfered with, reconstruction and replacement are performed using interpolation algorithms based on the values ​​of their adjacent sub-bands and symbols at adjacent time points. Demap the data symbols after equalization and interference repair to restore them to a binary bit stream; All binary bit streams restored from the same data block are reassembled in sequence to obtain an estimate of the original communication data.

10. The low-voltage power line multi-carrier communication system based on LWT-OFDM according to claim 9, characterized in that, The dynamic adjustment steps for the trigger threshold of the time-domain protection mechanism specifically include: At the receiving end, the number of times each sub-band symbol is identified as being subject to pulse interference and its distribution pattern are continuously statistically analyzed. The statistical results are fed back to the transmitter via the reverse control channel; The transmitter updates the statistical characteristics of impulse noise in the channel quality assessment matrix based on historical feedback information; Based on the updated statistical characteristics, the trigger threshold values ​​of the time-domain protection mechanism corresponding to each sub-band in the subsequent data blocks are dynamically adjusted and written into the new sub-band parameter configuration table.

Images