Method for suppressing impulse noise in a power line carrier communication system

Abstract

The present application relates to a kind of power line carrier communication system pulse noise suppression method, belong to power line data communication field.The method includes: S1: parameter adaptation: in receiving end, input parameter adaptation module to receiving signal, through the way of sliding window, pulse noise detection module and pulse noise suppression module are adapted condition parameters, i.e.detection parameter and noise suppression parameter;S2: pulse noise detection: received data is input pulse noise detection module according to block, obtain the peak of detection data block, and the peak is compared with the detection parameter of adaptation, judge whether there is pulse noise;S3: pulse noise suppression: through the way of iteration, constantly find out the maximum value of each section of received data and its position, and the maximum value and the data in its adjacent interval are zero, finally, the data higher than a certain threshold is limited to reach the effect of inhibiting pulse noise.The present application method is simple, and does not need to rely on experience to select amplitude threshold.

Description

Technical Field

[0001] This invention belongs to the field of power line data communication and relates to a method for pulse interference detection and suppression in power line carrier communication systems (PLC). Background Technology

[0002] In today's environment where multiple data communication systems coexist, mutual interference frequently occurs. Compared to other communication systems, power line carrier (PLC) systems experience impulse noise more frequently, which has a greater impact on the useful data on each subcarrier within the band.

[0003] From a testing standard perspective, so-called impulse interference refers to intermittent Gaussian noise. Within the time interval of interference, it has continuous characteristics and can overwhelm the signal. The portion of the signal interfered with by Gaussian noise appears in the time domain as a spike or a dense cluster of large signal amplitudes.

[0004] Currently, most impulse noise suppression algorithms employ nonlinear methods in the time domain. The basic idea is to utilize the characteristic that the instantaneous amplitude of impulse noise is much larger than that of normal signals, treating data exceeding a certain amplitude threshold as containing impulse noise, and then setting the data at the impulse noise location to zero or limiting its amplitude. However, the selection of the amplitude threshold is mostly based on experience and rarely can be automatically set and adjusted.

[0005] In addition, there are some statistical model-based methods that use mathematical tools such as steady-state Markov processes or linear prediction models to describe signals or impulse noise. These methods are highly complex to implement and have certain requirements on signal characteristics, and are generally not suitable for engineering applications.

[0006] Therefore, there is an urgent need for a simple method to suppress impulse noise in power line carrier communication systems that does not require empirical selection of amplitude thresholds. Summary of the Invention

[0007] In view of this, the purpose of the present invention is to provide a method for suppressing impulse noise in a power line carrier communication system, which aims to suppress the influence of impulse noise on the received data before demodulation at the receiving end. This method is simple to implement and does not require empirical selection of amplitude thresholds.

[0008] To achieve the above objectives, the present invention provides the following technical solution:

[0009] A method for suppressing impulse noise in a power line carrier communication system specifically includes the following steps:

[0010] S1: Parameter adaptation: The received signal is input into the parameter adaptation module at the receiving end. The condition parameters are adapted to the impulse noise detection module and the impulse noise suppression module through a sliding window, namely the detection parameter (1) and the noise suppression parameter (2).

[0011] S2: Pulse noise detection: Input the received data in blocks into the pulse noise detection module to obtain the peak value of the detected data block, and compare this peak value with the adapted detection parameters to determine whether there is pulse noise;

[0012] S3: Pulse noise suppression: Continuously find the maximum value and its position in each segment of the received data through an iterative method, set the data within the maximum value and its adjacent intervals to zero, and finally limit the data higher than a certain threshold (the clipping threshold PeakInWin in the noise suppression parameters) to achieve the effect of suppressing pulse noise.

[0013] Further, in step S1, parameter adaptation specifically includes the following steps:

[0014] S11: Assume that Len (Len < 1024, the recommended value is 100) is the sliding window length, and step (the recommended value is 60) is the sliding step;

[0015] S12: Divide the received time-domain signal y(n) into blocks according to the FFT points of the OFDM signal, and input the first block of data into the parameter adaptation module;

[0016] S13: First calculate the mean value M(1) of the first window data starting from the first bit, and then calculate the mean value M(2) of the second window data starting from the (step + 1)th bit;

[0017] S14: Compare M(1) and M(2), and take the smaller mean value as the temporary mean value MeanTemp, and record the data of the window corresponding to MeanTemp as TempData;

[0018] S15: Calculate M(n) of the current window data by sliding step bits backward in turn; if M(n) < MeanTemp, update the value of MeanTemp to M(n), and update TempData corresponding to MeanTemp;

[0019] S16: After the last comparison is completed, calculate the peak value PeakInWin1, pseudo-peak value PPeak, and pseudo mean square ratio TempMeanSqu of TempData;

[0020] S17: Input the next two data blocks into the parameter adaptation module to obtain peak values PeakInWin2 and PeakInWin3; and find the maximum value PeakInWin among PeakInWin1, PeakInWin2, and PeakInWin3;

[0021] S18: Input PPeak to the impulse noise detection module, and input TempMeanSqu and PeakInWin to the impulse noise suppression module.

[0022] Furthermore, in step S16, the formula for calculating the pseudo peak PPeak is: PPeak = PeakInWin + MeanTemp.

[0023] Furthermore, in step S16, the formula for calculating the pseudo-mean square ratio TempMeanSqu is:

[0024]

[0025]

[0026]

[0027] Where, σ 2 Let (L+R) be the variance of TempData, and (L+R) be the length of the zeroing interval for each impulse noise suppression module. It is recommended that L=6 and R=25.

[0028] Furthermore, step S2 specifically includes the following steps:

[0029] S21: Input the received signals into the impulse noise detection module in data blocks sequentially;

[0030] S22: Compare the pseudo peak value PPeak with the peak value BPeak of the current data block to determine whether there is impulse noise in the data block;

[0031] S23: If impulse noise is detected, the data block is passed to the impulse noise suppression module; if no impulse noise is detected, the data block is passed directly to the subsequent signal processing module.

[0032] Furthermore, step S3 specifically includes the following steps:

[0033] S31: First, find the peak value of each received data block, BPeak, and the position where the peak value appears, BPeak_index;

[0034] S32: Set the peak value and the data in its adjacent interval [BPeak_index-L,BPeak_index+R] to zero;

[0035] S33: Calculate the mean square ratio of the remaining data (MeanSqu);

[0036] S34: If MeanSqu < TempMeanSqu, it indicates that the data affected by impulse noise in this data block has not been completely suppressed. Steps S31 - S33 should be repeated until MeanSqu > TempMeanSqu, at which point the iteration stops.

[0037] S35: If the iteration does not exceed three times, directly pass the data block after iterative processing into the subsequent processing module of the system. If the number of iterations exceeds three times, the data block after iterative processing needs to be restored to the data before the last three iterations, and after reducing the zero - setting interval of the last three times, re - perform the noise zero - setting process for the last three times.

[0038] S36: Use PeakInWin as the threshold to perform amplitude limiting on the data block after re - zero - setting, and obtain the data after impulse noise suppression. And pass it into the subsequent processing module of the system.

[0039] The beneficial effects of the present invention are as follows: The present invention mainly conducts processing in the time domain, adaptively adjusts the condition parameters of the impulse noise detection module and the impulse noise suppression module through the sliding window method, and adopts the iterative method of zero - setting the noise interval to suppress impulse noise, making the method of the present invention insensitive to the waveform characteristics of impulse noise and simple to implement.

[0040] Other advantages, objectives, and features of the present invention will be described in part in the subsequent specification, and to some extent, will be obvious to those skilled in the art based on the study of the following text, or can be taught from the practice of the present invention. The objectives and other advantages of the present invention can be achieved and obtained through the following specification. Brief Description of the Drawings

[0041] In order to make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be described in detail preferably with reference to the accompanying drawings, where:

[0042] Figure 1 is the principle block diagram of the impulse noise suppression method of the present invention;

[0043] Figure 2 is the flowchart of parameter adaptation;

[0044] Figure 3 is the flowchart of impulse noise suppression;

[0045] Figure 4 is the time - domain diagram containing impulse interference in the input of the simulation example according to the method of the present invention;

[0046] Figure 5 is the time - domain diagram after impulse noise suppression in the simulation example according to the method of the present invention. Detailed Embodiments

[0047] The embodiments of the present invention will be described below through specific examples. Those skilled in the art can easily understand the other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments. Various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the diagrams provided in the following embodiments only illustrate the basic concept of the present invention in a schematic manner. Without conflict, the following embodiments and the features in the embodiments can be combined with each other.

[0048] Please refer to Figures 1-5 , Figure 1 The principle block diagram of a method for suppressing impulse noise in a power line carrier communication system provided by the present invention is shown. This method mainly includes parameter adaptation, impulse noise detection, and impulse noise suppression processing processes, and specifically includes the following steps

[0049] Step 1: Parameter adaptation. The schematic diagram of the parameter adaptation algorithm is as shown in Figure 2 .

[0050] 1) Set Len (Len < 1024, the recommended value is 100) as the sliding window length and step (the recommended value is 六十) as the sliding step.

[0051] 2) Divide the received time-domain signal y(n) into blocks according to the FFT points of the OFDM signal, and input the first block of data into the parameter adaptation module.

[0052] 3) First calculate the mean value (M(1)) of the first window data starting from the first bit, and then calculate the mean value (M(2)) of the second window data starting from the (step + 1)th bit.

[0053] 4) Compare M(1) and M(2) and take the smaller mean value as the temporary mean value (MeanTemp), and record the window data (TempData) corresponding to MeanTemp at the same time.​​​​​​​​​

[0057]

[0058] in,

[0059]

[0060]

[0061] Where, σ 2 Let (L+R) be the variance of TempData, and (L+R) be the length of the zeroing interval for each impulse noise suppression module. It is recommended that L=6 and R=25.

[0062] 7) Pass the next two data blocks to the parameter adaptation module to obtain PeakInWin2 and PeakInWin3. Find the maximum value (PeakInWin) among (PeakInWin1, PeakInWin2, PeakInWin3).

[0063] 8) Pass PPeak to the impulse noise detection module, and TempMeanSqu and PeakInWin to the impulse noise suppression module.

[0064] Step 2: Impulse noise detection.

[0065] 1) Input the raw data into the impulse noise detection module in blocks.

[0066] 2) Compare the PPeak with the peak value of the current data block to determine whether there is impulse noise in the data block.

[0067]

[0068] If if_pulse = 1, it means that impulse noise has been detected, and the data block is passed to the impulse noise suppression module; if if_pulse = 0, it means that impulse noise has not been detected, and the data block is directly passed to the subsequent processing module of the system.

[0069] Step 3: Impulse noise suppression. A schematic diagram of the impulse noise suppression algorithm is shown below. Figure 2 As shown.

[0070] 1) First, find the peak value (BPeak) and the position (BPeak_index) of each received data block.

[0071] 2) Set the peak value and the data in the adjacent interval [BPeak_index-L,BPeak_index+R] to zero.

[0072] 3) Calculate the mean square ratio (MeanSqu) of the remaining data.

[0073] 4) If MeanSqu < TempMeanSqu, it indicates that the data affected by impulse noise in this data block has not been completely suppressed. The peak should be continuously searched for and the data in its adjacent intervals should be set to zero. By this iterative method, the impulse noise is continuously suppressed until MeanSqu > TempMeanSqu and the iteration stops.

[0074] 5) If the iteration does not exceed three times, the data block after iterative processing is directly passed into the subsequent processing module of the system. If the number of iterations exceeds three times, the data block after iterative processing needs to be restored to the data before the last three iterations (temp_y(n)), and the peak positions L_index(1), L_index(2), L_index(3) during the last three iterations should be found.

[0075] Among them, the zeroing interval during the last three iterations is reduced, and then the data in temp_y(n) is zeroed as follows: temp_y(L_index(k)-L2:L_index(k)+R2) = zero(1,L2+R2), k = 1, 2, 3, where L2 is slightly less than L, and R2 is slightly less than R. It is recommended that L2 = 5 and R2 = 20.

[0076] 6) Use PeakInWin as the threshold to perform clipping processing on the data block after re-zeroing, and obtain the data after suppressing impulse noise and pass it into the subsequent processing module of the system.

[0077] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not restrictive. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the purpose and scope of the present technical solution, and they should all be covered within the scope of the claims of the present invention.

Claims

1. A method for suppressing impulse noise in a power line carrier communication system, characterized in that, The method specifically includes the following steps: S1: Parameter adaptation: Input the received signal into the parameter adaptation module at the receiving end, and adapt the conditional parameters, namely the detection parameter and the noise suppression parameter, for the impulse noise detection module and the impulse noise suppression module in the way of a sliding window; The parameter adaptation specifically includes the following steps: S11: Assume that Len is the sliding window length and step is the sliding step size; S12: Block the received time-domain signal y(n) according to the FFT points of the OFDM signal, and input the first block of data into the parameter adaptation module; S13: First calculate the mean value M(1) of the first window data starting from the first position, and then calculate the mean value M(2) of the second window data starting from the (step + 1)-th position; S14: Compare M(1) and M(2), and take the smaller mean value as the temporary mean value MeanTemp, and record the data TempData corresponding to the window of MeanTemp at the same time; S%15: Calculate M(n) of the current window data by sliding step positions backward in turn; if M(n) < MeanTemp, update the value of MeanTemp to M(n), and update the TempData corresponding to MeanTemp; S16: After the last comparison ends, calculate the peak value PeakInWin1, the pseudo-peak value PPeak and the pseudo mean square ratio TempMeanSqu of TempData; S17: Input the subsequent two data blocks into the parameter adaptation module to obtain the peak values PeakInWin2 and PeakInWin3; and find the maximum value PeakInWin among PeakInWin1, PeakInWin2 and PeakInWin3; S18: Input PPeak into the impulse noise detection module, and input TempMeanSqu and PeakInWin into the impulse noise suppression module; S2: Impulse noise detection: Input the received data into the impulse noise detection module block by block, obtain the peak value of the detected data block, and compare the peak value with the adapted detection parameter to judge whether there is impulse noise; 2. The impulse noise suppression method according to claim 1, characterized in that, In step S16, the formula for calculating the pseudo-peak value PPeak is: .

3. The impulse noise suppression method according to claim 1, characterized in that, S%13: Impulse noise suppression: Continuously find the maximum value and its position of each segment of received data through iteration, set the data within the maximum value and its adjacent interval to zero, and finally limit the data higher than the clipping threshold PeakInWin in the noise suppression parameter to achieve the effect of suppressing impulse noise. in, Let (L+R) be the variance of TempData, and (L+R) be the length of the zeroing interval for each impulse noise suppression module.

4. The impulse noise suppression method according to claim 1, characterized in that, In step S16, the calculation formula of the pseudo mean square ratio TempMeanSqu is: Step S2 specifically includes the following steps: [[ID=%15]]S21: Input the received signal into the impulse noise detection module block by block in turn; S22: Compare the size of the pseudo-peak value PPeak and the peak value BPeak of the current data block, and judge whether there is impulse noise in the data block based on this; 5. The impulse noise suppression method according to claim 4, characterized in that, S23: If impulse noise is detected, input the data block into the impulse noise suppression module; if no impulse noise is detected, directly input the data block into the subsequent signal processing module. Step S3 specifically includes the following steps: S31: First, find the peak value BPeak of each received data block and the position BPeak_index where the peak appears; S32: Set the data of the peak value and its adjacent interval [BPeak_index - L, BPeak_index + R] to zero; S33: Calculate the mean square ratio MeanSqu of the remaining data; S34: If MeanSqu < TempMeanSqu, it indicates that the data affected by impulse noise in this data block has not been completely suppressed, and steps S31 - S33 should be repeated until MeanSqu > TempMeanSqu to stop the iteration; S35: If the iteration does not exceed three times, directly pass the iteratively processed data block to the subsequent processing module of the system; if the number of iterations exceeds three times, the iteratively processed data block needs to be restored to the data before the last three iterations, and the noise zeroing process of the last three times should be redone after reducing the zeroing interval of the last three times; S36: Using PeakInWin as the threshold, the data block after being reset to zero is subjected to amplitude limiting to obtain the data after impulse noise suppression. It is then passed to the system's subsequent processing module.

Images