A method and apparatus for impulse noise processing

Abstract

The application discloses a pulse noise processing method and device. Different threshold values are obtained based on an input signal. The input signal is processed based on the threshold values to obtain a processing result of at least one threshold value, and the processing result includes a processed signal and a processed signal weight. A de-noised pulse signal is determined based on the processing results of at least two threshold values. The application adopts a plurality of threshold value determination methods, processes the input signal based on the threshold values, obtains the processed signal and the weight corresponding to the threshold value, and performs weighted average on the processed signal to obtain an output signal. The application does not need to find an optimal threshold value, can be applied to a scene where pulse noise appears densely and in a large number, only needs simple multiple comparisons, can realize the de-noising function, greatly reduces the complexity, and reduces the power consumption.

Description

Technical Field

[0001] This disclosure belongs to the field of signal processing technology, specifically relating to a method and apparatus for processing impulse noise. Background Technology

[0002] Existing impulse noise processing techniques can be broadly categorized into two types: noiseless parametric denoising and nonlinear impulse denoising. Noiseless parametric denoising is only applicable to sparsely occurring impulse noise. If the impulse noise is dense and numerous, it cannot be reconstructed and therefore cannot be suppressed, making it unsuitable for practical engineering applications. Nonlinear impulse denoising requires finding a threshold value T. The choice of this threshold value determines the effectiveness of the impulse processing. If T is too small, useful signals may be misinterpreted as impulses, resulting in the loss of useful information. If T is too large, too much impulse noise is retained, failing to achieve the goal of removing impulse noise. Therefore, it is necessary to find an optimal threshold value T. opt Optimal threshold value T opt The estimation method involves a large amount of computation, especially when calculating the input signal r. k The expectation of multi-moment estimation requires the input signal r k Finding the sum of squares, the sum of cubes, and then the average requires a large number of multipliers and adders. If r k Let the length be M. This method requires M cubes, M squares, and 3M-3 additions. Therefore, it is necessary to solve the problem of large computational load and high complexity of impulse noise processing technology. Summary of the Invention

[0003] To address the aforementioned technical issues, this disclosure employs a method for determining multiple threshold values, which can be applied to scenarios where impulse noise has a high density and large number of occurrences. It also reduces the dependence of nonlinear denoising on the optimal threshold value, while simultaneously reducing computational load and complexity.

[0004] In a first aspect, this disclosure provides an impulse noise processing method that obtains multiple different threshold values ​​based on the input signal;

[0005] The input signal is denoised based on multiple threshold values ​​to obtain a processing result for at least one threshold value. The processing result includes the processed signal and the processed signal weight.

[0006] The output signal is determined based on the processing results of at least two of the threshold values.

[0007] Secondly, this disclosure provides an impulse noise processing apparatus, comprising:

[0008] The threshold value determination module obtains multiple different threshold values ​​based on the input signal;

[0009] The noise reduction processing module performs noise reduction processing on the input signal based on multiple threshold values ​​to obtain a processing result for at least one threshold value. The processing result includes the processed signal and the processed signal weight.

[0010] The signal output module determines the output signal based on the processing results of at least two of the threshold values.

[0011] Thirdly, this disclosure provides a readable storage medium having executable instructions thereon, which, when executed, cause a computer to perform the steps of the above-described impulse noise processing method.

[0012] Fourthly, this disclosure provides an electronic device, the device including a processor and a memory, the memory storing computer program instructions suitable for execution by the processor, the computer program instructions being executed by the processor to perform the steps of the above-described impulse noise processing method.

[0013] This disclosure employs a method of determining multiple threshold values. The input signal is denoised based on these multiple threshold values ​​to obtain the processed signal and its weight corresponding to each threshold value. The output signal is then determined based on the processing results of at least two of the threshold values. This disclosure eliminates the need to find the optimal threshold value, making it applicable to scenarios with high levels and large numbers of impulse noises. Furthermore, it achieves denoising functionality through simple multiple comparisons, significantly reducing complexity and power consumption. Attached Figure Description

[0014] The accompanying drawings illustrate exemplary embodiments of the present disclosure and, together with the description thereof, serve to explain the principles of the present disclosure. These drawings are included to provide a further understanding of the present disclosure and are incorporated in and constitute a part of this specification.

[0015] Figure 1 This is a structural block diagram of the signal processing device in this disclosure;

[0016] Figure 2 This is a schematic diagram of the impulse noise processing method disclosed herein;

[0017] Figure 3 This is a schematic diagram of the algorithm for the impulse noise processing method in this disclosure;

[0018] Figure 4 This is a structural block diagram of the impulse noise processing device in this disclosure;

[0019] Figure 5 This is a block diagram of the readable storage medium disclosed herein;

[0020] Figure 6 This is a structural block diagram of the electronic device disclosed herein. Detailed Implementation

[0021] The present disclosure will now be described in further detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the scope of the disclosure. Furthermore, it should be noted that, for ease of description, only the parts relevant to the present disclosure are shown in the accompanying drawings.

[0022] It should be noted that, where there is no conflict, the embodiments and features described in this disclosure can be combined with each other. This disclosure will now be described in detail with reference to the accompanying drawings and embodiments.

[0023] See Figure 2 As shown, this embodiment of the present disclosure provides a signal processing device, which includes a transceiver 101, a noise reduction device 102, and a central processing unit 103. The transceiver 101 can transmit and receive signals via a power line 201. The transceiver 101 communicates with the noise reduction device 102, and the noise reduction device 102 communicates with the central processing unit 103.

[0024] The transceiver 101 described above can receive various pulse signals or digital signals. These pulse signals or digital signals can be signals transmitted via electronic line carrier technology, signals transmitted via data lines, or signals received by an antenna, but are not limited to these.

[0025] The aforementioned noise reduction device 102 can perform noise reduction processing on pulse signals or digital signals, and then transmit them to the central processing unit for processing.

[0026] The above-mentioned noise reduction methods are diverse. They can adopt existing noise-free parameter denoising methods or nonlinear pulse denoising methods, or they can directly adopt nonlinear zeroing and / or amplitude limiting methods.

[0027] In one example, after receiving a pulse signal, the noise reduction device 102 can denoise the signal using a nonlinear pulse denoising process; the nonlinear pulse denoising process is performed when the transceiver 101 receives the input signal r. k Then, select an appropriate threshold value T and input signal r. k The amplitude comparison is performed on the input signal r. k For sample points exceeding the threshold value T, nonlinear zeroing / limiting processing is applied to the input signal r. k Sample points less than or equal to the threshold value T remain unchanged, and the output signal y after noise removal is output. k .

[0028] However, nonlinear pulse processing methods, whether zeroing or limiting, require finding a threshold value T. The choice of this threshold value determines the effectiveness of pulse processing. If the value of T is too small, useful signals may be misidentified as pulses, resulting in the loss of useful information. If the value of T is too large, too much pulse noise will be retained, failing to achieve the goal of removing pulse noise. Therefore, it is necessary to find an optimal threshold value T. opt Current practices regarding T opt The selection method is to first calculate the received signal r k The multi-moment estimation expectation is used to estimate the probability of impulse noise and the noise variance based on the obtained multi-moment expectation value, and then the optimal threshold value T is calculated. opt It involves a large amount of computation and high complexity, requiring a large number of multipliers and adders.

[0029] Based on this, the present disclosure provides an impulse noise processing method that eliminates the need to find the optimal threshold value. This method can be applied to scenarios where impulse noise has a high density and a large number of occurrences, and can achieve noise reduction by simply making multiple comparisons. This greatly reduces complexity and power consumption.

[0030] Figure 2 , Figure 3 This embodiment of the present disclosure illustrates an impulse noise processing method, which includes:

[0031] S101, Based on input signal r k Obtain multiple different threshold values ​​T i Input signal r k It can be a pulse signal or a digital signal; the above threshold value T i It can be estimated based on experience, or it can be based on the input signal r. k Determine the minimum and maximum threshold values ​​T1 and T2. N At the minimum value T1 and the maximum value T N The threshold values ​​are selected based on the number of threshold values. The number of threshold values ​​N can be preset, or it can be based on the minimum threshold value T1 and the maximum threshold value T. N Preset the threshold values ​​to the minimum value T1 and the maximum value T. N If the difference is small, the number of threshold values ​​N can be reduced. When the minimum threshold value T1 and the maximum threshold value T1 are close, the threshold value N can be reduced. N If the difference is large, the number of threshold values ​​N can be increased. Different threshold values ​​T can be set in an arithmetic progression or in a non-arithmic progression.

[0032] S102, Based on multiple threshold values ​​T i For the input signal r k Noise reduction processing is performed to obtain at least one threshold value T. iThe processing result includes the processed signal r_b i and the processed signal weight α i Noise reduction processing can be applied to the input signal r k Medium greater than the threshold value T i The sample points are set to zero and / or subjected to amplitude limiting.

[0033] Wherein, weight α i With input signal r k Medium greater than the threshold value T i The number of sample points B_N i Related to, when greater than the threshold value T i The number of sample points B_N i The more, the higher the weight α i The smaller the value, the greater the threshold value T. i The number of sample points B_N i The smaller the value, the higher the weight α. i The larger.

[0034] S103, based on at least two of the threshold values ​​T i The processing result determines the output signal y. kk The processed signal r_b can be fully or partially processed. i Perform a weighted average to obtain the output signal y k .

[0035] By adopting multiple threshold values ​​T i The determination method is based on the threshold value T. i For the input signal r k Noise reduction processing is performed to obtain the threshold value T. i The corresponding processed signal r_b i and weight α i The processed signal r_b i Perform a weighted average to obtain the output signal y k There is no need to find the optimal threshold value T. opt It can be applied to scenarios with high density and large number of impulse noises, and can achieve noise reduction function with only simple multiple comparisons, which greatly reduces complexity and power consumption.

[0036] In one embodiment, the above is based on the input signal α i Obtain multiple different threshold values ​​T i ,include:

[0037] First based on the input signal r k Determine the threshold value T i The range, that is, determining the threshold value T. i The minimum value T1 and the maximum value T NThe specific approach is to find the input signal r. k The maximum value r_max and the minimum value T1 are the minimum values ​​of the average of all sample points that are less than k times the maximum value r_max, where k = 1 / 5 - 1 / 2; the maximum value T_max is the minimum value T1 of the threshold. N It is n times the minimum value T1, where n = 2-5; in this embodiment, the exemplary recorded input signal r k For all sample points less than r_max / 4, the mean of these sample points is the minimum threshold value T1. For example, the multiplier is set to 4, and the maximum value is T. N Then it can be determined to be T. N = 4 * T1, and the multiple can also be set to other numbers.

[0038] When two adjacent threshold values ​​T i When the differences are equal, the i-th threshold value can be calculated:

[0039] T i =T1+(i-1)ΔT, i=1,2…N

[0040] Among them, T i Let ΔT be the i-th threshold value, and let ΔT be the minimum difference between the two threshold values, i = 1 to N.

[0041] When two adjacent threshold values ​​T i When the differences are not equal, threshold values ​​such as T1, T2, T3, T5, and T7 can be selected as examples.

[0042] Two threshold values ​​T i The minimum difference satisfies: ΔT=(T N -T1) / (N-1),

[0043] Where ΔT is the minimum difference between the two threshold values, and N is the number of threshold values.

[0044] The number of exemplary threshold values ​​N = 10 to 20, such as 10, 15, 20 or other numbers.

[0045] In another embodiment, based on multiple threshold values ​​T i For the input signal r k Noise reduction processing is performed, including: reducing the noise level of the input signal r. k Medium greater than the threshold value T i The sample points are nonlinearly zeroed out and / or limited to eliminate the identified noise pulses. Zeroing out is preferred due to its simplicity. For example, for r... k Medium greater than the threshold value T i The sample points are nonlinearly zeroed out, and the input signal r is subjected to this process. k Less than or equal to the threshold value Ti The sample points remain unchanged, and the output is the noise-removed signal r_b i For example, for the input signal r k Medium greater than the threshold value T i Amplitude limiting is applied to the sample points of the input signal r. k Less than or equal to the threshold value T i The sample points remain unchanged, and the output is the noise-removed signal r_b i For example, for the input signal r k Medium greater than the threshold value T i The sample points are partially amplitude-limited and partially zeroed out, for the input signal r k Less than or equal to the threshold value T i The sample points remain unchanged, and the output is the noise-removed signal r_b i

[0046] In yet another embodiment, the weight α i Based on the input signal r k Medium greater than the threshold value T i The number of sample points B_N i Determined. In this embodiment, the weight α i The method for determining it is:

[0047] Use the i-th threshold value T i For the input signal r k Make a judgment; if it is greater than the threshold value T i Set the sample points to zero and record the number of sample points B_N. i The signal after being set to zero is denoted as r_b. i Input signal r k Let the total length be M, then the weight is calculated as: α i =1-B_N i /

[0048] In a preferred embodiment, the above is based on a plurality of said threshold values ​​T i For the input signal r k Noise reduction processing is performed to obtain at least one threshold value T. i The processing results include:

[0049] When the input signal r k There exists a value greater than the currently stated threshold T. i The sample points are used to perform noise reduction processing, and the processed signal weight α is determined based on the number of sample points. i ;

[0050] When the (i+1)th threshold value T i+1 The corresponding processed signal weight αi+1 With the i-th threshold value T i The corresponding processed signal weight α i If they are equal, obtain the processing results of the first threshold value to the i-th threshold value, where i is an integer greater than or equal to 1 and less than or equal to N-1.

[0051] With the number of threshold values ​​N being 20, the current calculation is up to the 15th threshold value T. 15 For example, if the current threshold value T 15 weight α 15 Equal to the previous threshold value T 14 The corresponding weight α 14 Discard the current threshold value T 15 The corresponding processed signal r_b 15 and weight α 15 And abandon the calculation of T 15 The processed signal r_b corresponding to the subsequent threshold value i and weight α i This step; and the subsequent processed signal r_b i The weighted average is calculated only up to α1*_b1. 15 *_b 15 The weighted average is used to obtain the output signal y. k .

[0052] When calculating other threshold values ​​T i The corresponding weight α i Equal to the previous threshold value T i-1 The corresponding weight α i-1 Discarding the current threshold value T i The corresponding processed signal r_b i and weight α i The process of obtaining the processed signal r_b corresponding to the threshold value ends. i and weight α i The steps.

[0053] In another preferred embodiment, the above is based on at least two threshold values ​​T. i The processing result determines the output signal y. k ,include:

[0054] According to the threshold value T i The processed signal weight α i The order from largest to smallest is based on at least two threshold values ​​T. i The threshold value T of the preset ratio i The weighted average result is determined based on the processing results.

[0055] The aforementioned preset ratio can be between 50% and 100%; taking a preset ratio of 60% as an example, based on all the threshold values ​​T that need to be calculated. i Obtain the corresponding signal r_b i and weight α i After completion, according to all weights α i Sort by size from largest to smallest, and then perform a weighted average on the top 60% of the weights. This can reduce the amount of computation while still eliminating noise impulses.

[0056] Figure 4 The example provided in this exemplary embodiment of the present disclosure illustrates an impulse noise processing apparatus capable of implementing the above-described impulse noise processing method. The apparatus includes:

[0057] Threshold determination module 1 can determine the threshold value based on the external input signal r. k Obtain multiple different threshold values ​​T i Threshold value T i Passed to noise reduction processing module 2;

[0058] Noise reduction processing module 2, based on multiple threshold values ​​T i For the input signal r k Noise reduction processing is performed to obtain at least one threshold value T. i The processing result includes the processed signal r_b i and the processed signal weight α i Noise reduction processing is applied to the input signal r k Medium greater than the threshold value T i The sample points are nonlinearly zeroed and / or limited.

[0059] Signal output module 3, based on at least two of the threshold values ​​T i The processing result determines the output signal y. k The processed signal r_b can be fully or partially processed. i Perform a weighted average to obtain the output signal y kk .

[0060] The threshold determination module 1, noise reduction processing module 2, and signal output module 3 can all be calculated by the processor, and the threshold determination module 1, noise reduction processing module 2, and signal output module 3 can be connected in sequence to transmit signals.

[0061] In PLCs, pulse noise has a short duration and high energy, which greatly affects the signal. Therefore, it is necessary to adjust the signal after the digital signal r is processed by the ADC (Analog-to-Digital Converter). k After pulse noise processing, the signal enters the synchronization module for synchronization processing.

[0062] The device disclosed herein employs a method of determining multiple threshold values, zeroing and / or limiting sample points that exceed the threshold values, and using the number of zeroed and / or limited sample points to obtain weights for weighted averaging. This impulse noise processing device retains the impulse processing effect while requiring only simple multiple comparisons, eliminating the need for a large number of multipliers, thus greatly reducing complexity and power consumption.

[0063] Figure 5 The exemplary embodiment of this disclosure provides a computer-readable storage medium 4, on which executable instructions 5 are stored. When the executable instructions 5 are executed, they cause a computer to perform the steps of the impulse noise processing method described above. The computer-readable storage medium 4 may be an electronic medium, magnetic medium, optical medium, electromagnetic medium, infrared medium, or semiconductor system or propagation medium. The computer-readable storage medium 4 may also include semiconductor or solid-state memory, magnetic tape, removable computer disk, random access memory (RAM), read-only memory (ROM), hard disk, and optical disk. Optical disks may include optical disc-read-only memory (CD-ROM), optical disc-read / write (CD-RW), and DVD.

[0064] See Figure 6An exemplary embodiment of the present disclosure provides an electronic device comprising a processor 6 and a memory 7. The memory stores computer program instructions suitable for execution by the processor, which, when run by the processor, perform the steps of the impulse noise processing method described above. The processor 6 may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it may also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. For example, the processor 6 may employ a multi-core digital signal processor 6713, with multiple DSP cores having a clock frequency of 500MHz, and may use interrupts to control time precision. The memory 7 may include random access memory (RAM) and may also include non-volatile memory, such as at least one disk storage device. The memory 7 can also be an internal memory of the Random Access Memory (RAM) type. The processor 6 and memory 7 can be integrated into one or more independent circuits or hardware, such as Application Specific Integrated Circuits (ASICs). It should be noted that the computer program in the aforementioned memory 7 can be implemented as a software functional unit and, when sold or used as an independent product, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, electronic device, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of the present invention.

[0065] In the description of this specification, the references to terms such as "one embodiment / mode," "some embodiments / modes," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment / mode or example is included in at least one embodiment / mode or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment / mode or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments / modes or examples. Furthermore, without contradiction, those skilled in the art can combine and integrate the different embodiments / modes or examples described in this specification, as well as the features of different embodiments / modes or examples.

[0066] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "multiple" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0067] Those skilled in the art should understand that the above embodiments are merely for illustrating the present disclosure and are not intended to limit the scope of the disclosure. Those skilled in the art can make other changes or modifications based on the above disclosure, and these changes or modifications still fall within the scope of the present disclosure.

Claims

1. A method for processing impulse noise, characterized in that, include: Multiple different threshold values ​​are obtained based on the input signal; The input signal is denoised based on multiple threshold values ​​to obtain a processing result for at least one threshold value. The processing result includes the processed signal and the processed signal weight. The output signal is determined based on the processing results of at least two of the aforementioned threshold values; The minimum value among the plurality of threshold values ​​is the average value of all sample points that is less than k times the maximum value of the input signal, where k = 1 / 5 to 1 / 2; The maximum value among the multiple threshold values ​​is n times the minimum value, where n = 2 - 5; Each of the aforementioned threshold values ​​satisfies: T i =T1+(i-1)ΔT Among them, T i Let ΔT be the minimum difference between two threshold values, i = 1 to N, and T1 be the minimum threshold value. N This represents the maximum value of the threshold.

2. The impulse noise processing method according to claim 1, characterized in that, The difference between two adjacent threshold values ​​may be equal or unequal; or, The minimum difference between two threshold values ​​satisfies: ΔT = (T N -T1) / (N-1), T1 is the minimum value of the threshold, T N ΔT is the maximum value of the threshold, ΔT is the minimum difference between the two threshold values, and N is the number of threshold values.

3. The impulse noise processing method according to any one of claims 1 to 2, characterized in that: The step of performing noise reduction processing on the input signal based on multiple threshold values ​​to obtain a processing result of at least one threshold value includes: When there are sample points in the input signal that are greater than the current threshold value, noise reduction processing is performed on the sample points, and the weight of the processed signal is determined based on the number of sample points; When the processed signal weight corresponding to the (i+1)th threshold value is equal to the processed signal weight corresponding to the ith threshold value, the processing results of the 1st threshold value to the ith threshold value are obtained, where i is an integer greater than or equal to 1 and less than or equal to N-1, and N is the number of threshold values.

4. The impulse noise processing method according to claim 2, characterized in that: The processed signal weights satisfy: a i =1-B_N i / M Where, α i B_N is the processed signal weight for the i-th threshold value. i M represents the number of sample points in the input signal that are greater than the current threshold value, where M is the total number of sample points over the total length of the input signal.

5. The impulse noise processing method according to claim 1, characterized in that, The output signal is a weighted average of the processed signals of at least two threshold values; or, The process of determining the output signal based on the results of at least two threshold values ​​includes: Based on the processed signal weights of the threshold values ​​in descending order, and the processing results of at least two threshold values ​​representing a preset proportion of the threshold values, a weighted average result is determined.

6. A pulse noise processing device, characterized in that, include: The threshold value determination module obtains multiple different threshold values ​​based on the input signal; The noise reduction processing module performs noise reduction processing on the input signal based on multiple threshold values ​​to obtain a processing result for at least one threshold value. The processing result includes the processed signal and the processed signal weight. The minimum value among the multiple threshold values ​​is the average value of all sample points that is less than k times the maximum value of the input signal, where k = 1 / 5 to 1 / 2. The maximum value among the multiple threshold values ​​is n times the minimum value, where n = 2-5. The signal output module determines the output signal based on the processing results of at least two of the threshold values; Each of the aforementioned threshold values ​​satisfies: T i =T1+(i-1)ΔT Among them, T i Let ΔT be the minimum difference between two threshold values, i = 1 to N, and T1 be the minimum threshold value. N This represents the maximum value of the threshold.

7. A readable storage medium, characterized in that, It has executable instructions that, when executed, cause a computer to perform the steps of the impulse noise processing method as described in any one of claims 1-5.

8. An electronic device, characterized in that, The device includes a processor and a memory, the memory storing computer program instructions suitable for execution by the processor, the computer program instructions being executed by the processor to perform the steps of the impulse noise processing method as described in any one of claims 1-5.

Images