阅读说明：本技术 一种针对宽带电力线载波通信的频段检测方法 (Frequency band detection method for broadband power line carrier communication ) 是由 宋征卫 王世钊 俞建江 于 2021-06-23 设计创作，主要内容包括：本发明公开了一种针对宽带电力线载波通信的频段检测方法,具体包括如下步骤：101)信号预处理步骤、102)信号到达检测步骤、103)信号处理步骤、104)判决处理步骤；本发明提供了实现快速自动检测识别的一种应用于无线跳频网络的快速接入方法。(The invention discloses a frequency band detection method for broadband power line carrier communication, which specifically comprises the following steps: 101) signal preprocessing step, 102) signal arrival detection step, 103) signal processing step, and 104) decision processing step; the invention provides a quick access method applied to a wireless frequency hopping network, which realizes quick automatic detection and identification.)

1. A frequency band detection method for broadband power line carrier communication is characterized by comprising the following steps:

101) signal preprocessing step: adjusting the sampling rate of the output signal of the front-end ADC, so that the output signal rate is 1 time of the symbol rate;

102) signal arrival detection: acquiring delayed autocorrelation energy and autocorrelation energy from a preamble signal of broadband power line carrier communication, and judging whether the signal arrives according to the ratio change of the delayed autocorrelation energy and the autocorrelation energy;

103) a signal processing step: after the signal is judged to arrive, a mark of the signal arrival is given, and a receiver is prompted to enter subsequent digital processing; carrying out peak clipping algorithm on the signal to remove narrow-band interference in broadband power line carrier communication;

104) a judgment processing step: carrying out binary quantization spectrum processing on the signal without the narrow-band interference, wherein the binary quantization spectrum converts the information of a specific frequency point into a frequency domain energy binary sequence consisting of 0 and 1;

carrying out XOR processing on the frequency domain energy binary sequence of the signal without the narrow-band interference and the frequency domain energy binary sequence of four frequency bands defined by the standard, and calculating the accumulated sum; the obtained four calculation results respectively correspond to the similarity between the received signal and the four standard frequency bands, the higher the similarity is, the smaller the accumulation sum is, and the frequency band corresponding to the minimum accumulation sum is the working frequency band of the received signal.

2. The method according to claim 1, wherein the method comprises: the formula for obtaining the time delay autocorrelation energy and the autocorrelation energy in step 102) is as follows:

wherein, P_Dcorr(t) is the delayed cross-correlation energy, P_Acorr(t) is the autocorrelation energy;

R_corr(t)＝P_Dcorr(t)/P_Acorr(t) (equation 3)

R_corr(t) is the ratio of the delayed autocorrelation energy to the autocorrelation energy; when R is_corr(t) is greater than 0.6, indicating that a signal arrival is detected.

3. The method according to claim 2, wherein the method comprises: the time delay autocorrelation energy and the autocorrelation energy are realized in a sliding window mode, the window length is the length of a repetition period of a preamble signal of broadband power line carrier communication, and the window sliding stepping is 1 time of a symbol rate;

the window sliding sample moves in a first-in first-out mode, a group of delay autocorrelation and autocorrelation energy values are output once per step, and a ratio of the delay autocorrelation energy values to the autocorrelation energy values is obtained.

4. The method according to claim 1, wherein the method comprises: step 103) after the signal is detected to arrive, assuming that the time t arrives, selecting N sampling points { x ] from the beginning of the time t₀,x₁,…,x_N-1Acquiring signal frequency domain information y₀,y_1,…,y_N-1The frequency domain information acquisition formula is as follows:

n elements of the frequency domain information represent the amplitude and phase information of the preamble signal on N frequency points; obtaining amplitude information p for each element in the frequency domain information₀,p₁,…,p_N-1Obtaining the signal energy intensity power (y) on each frequency point_n) Instead of obtaining amplitude p by energy intensity_n＝power(y_n)；

When the difference value of the energy intensity of the target frequency point of the detected signal and the energy intensity of the adjacent frequency points before and after exceeds 4 times, the target frequency point is considered to have narrow-band interference, and the target frequency point with the narrow-band interference is replaced by the weighted average value of the energy intensity of the adjacent frequency points before and after, and the specific formula is as follows:

5. the method according to claim 1, wherein the method comprises: the binary quantization spectrum processing in step 104) is specifically as follows:

the average energy statistics of 25 frequency points in total is taken as the noise energy intensity P by selecting the frequency points 15-25 and 498-511_noiseNamely:

decision threshold condition TH of binary quantization spectrum_bDefined as the noise energy intensity P_noise2 times of the frequency domain information, by dividing the frequency domain energy p of each frequency point in the frequency domain information₀,p₁,…,p_N-1H and a decision threshold condition TH one by one_bComparing to obtain binary quantization spectrums { b) of all frequency points₀,b₁,…,b_N-1And fixedly filling 0 in other unused frequency points, namely:

Technical Field

The invention relates to the field of communication transmission, in particular to a frequency band detection method for broadband power line carrier communication.

Background

Power line broadband carrier (HPLC) communication is a communication technology implemented based on power lines, and related technical requirements are made by national grid companies. The HPLC communication system defines 4 different working communication frequency bands, which are: 1.953MHz-11.96MHz, 2.441MHz-5.615MHz, 0.781MHz-2.930MHz, 1.758MHz-2.930MHz, different frequency bands can not be communicated with each other.

The existing realization mode realizes the selection of four working frequency bands by a mode of presetting parameters, which causes the problems of non-intercommunicating mixed equipment loading, large workload of field installation and implementation and the like. Therefore, the automatic detection and identification of the four working frequency bands are realized by a signal processing method, the use complexity of HPLC products can be greatly simplified, and the method has strong practical significance.

Disclosure of Invention

The invention overcomes the defects of the prior art, and provides a quick access method applied to a wireless frequency hopping network for realizing quick automatic detection and identification by fully utilizing the frequency composition mode of four frequency bands of a broadband power line carrier (HPLC) technology, the repetition characteristic of a preamble signal and the processing characteristic of Fourier transform.

In order to solve the technical problems, the technical scheme of the invention is as follows:

a frequency band detection method for broadband power line carrier communication specifically comprises the following steps:

101) signal preprocessing step: adjusting the sampling rate of the output signal of the front-end ADC, so that the output signal rate is 1 time of the symbol rate;

102) signal arrival detection: acquiring delayed autocorrelation energy and autocorrelation energy from a preamble signal of broadband power line carrier communication, and judging whether the signal arrives according to the ratio change of the delayed autocorrelation energy and the autocorrelation energy;

103) a signal processing step: after the signal is judged to arrive, a mark of the signal arrival is given, and a receiver is prompted to enter subsequent digital processing; carrying out peak clipping algorithm on the signal to remove narrow-band interference in broadband power line carrier communication;

104) a judgment processing step: carrying out binary quantization spectrum processing on the signal without the narrow-band interference, wherein the binary quantization spectrum converts the information of a specific frequency point into a frequency domain energy binary sequence consisting of 0 and 1;

carrying out XOR processing on the frequency domain energy binary sequence of the signal without the narrow-band interference and the frequency domain energy binary sequence of four frequency bands defined by the standard, and calculating the accumulated sum; the obtained four calculation results respectively correspond to the similarity between the received signal and the four standard frequency bands, the higher the similarity is, the smaller the accumulation sum is, and the frequency band corresponding to the minimum accumulation sum is the working frequency band of the received signal.

Further, the formula for obtaining the delay autocorrelation energy and the autocorrelation energy in step 102) is as follows:

wherein, P_Dcorr(t) is the delayed cross-correlation energy, P_Acorr(t) is the autocorrelation energy;

R_corr(t)＝P_Dorr(t)/P_Acorr(t) (equation 3)

R_corr(t) is the ratio of the delayed autocorrelation energy to the autocorrelation energy; when R is_corr(t) is greater than 0.6, indicating that a signal arrival is detected.

Furthermore, the time delay autocorrelation energy and the autocorrelation energy are realized in a sliding window mode, the window length is the length of a repetition period of a preamble signal of broadband power line carrier communication, and the window sliding stepping is 1 time of the symbol rate;

the window sliding sample moves in a first-in first-out mode, a group of delay autocorrelation and autocorrelation energy values are output once per step, and a ratio of the delay autocorrelation energy values to the autocorrelation energy values is obtained.

Further, after the signal arrival is detected in step 103), assuming that the time t arrives, selecting N sampling points { x ] from the beginning of the time t₀,x₁,…,x_N-1Acquiring signal frequency domain information y₀,y_1,…,y_N-1The frequency domain information acquisition formula is as follows:

n elements of the frequency domain information represent the amplitude and phase information of the preamble signal on N frequency points; obtaining amplitude information p for each element in the frequency domain information₀,p₁,…,p_N-1Obtaining the signal energy intensity power (y) on each frequency point_n) Instead of obtaining amplitude p by energy intensity_n＝power(y_n)；

When the difference value of the energy intensity of the target frequency point of the detected signal and the energy intensity of the adjacent frequency points before and after exceeds 4 times, the target frequency point is considered to have narrow-band interference, and the target frequency point with the narrow-band interference is replaced by the weighted average value of the energy intensity of the adjacent frequency points before and after, and the specific formula is as follows:

further, the binary quantization spectrum processing in step 104) is specifically as follows:

the average energy statistics of 25 frequency points in total is taken as the noise energy intensity P by selecting the frequency points 15-25 and 498-511_noiseNamely:

decision threshold condition TH of binary quantization spectrum_bDefined as the noise energy intensity P_noise2 times of the frequency domain information, by dividing the frequency domain energy p of each frequency point in the frequency domain information₀,p₁,…,p_N-1H and a decision threshold condition TH one by one_bComparing to obtain binary quantization spectrums { b) of all frequency points₀,b₁,…,b_N-1And fixedly filling 0 in other unused frequency points, namely:

compared with the prior art, the invention has the advantages that:

the invention fully utilizes the characteristics of the preamble information defined by the broadband power line carrier (HPLC) standard, including frequency domain composition characteristics, time domain repeatability and cyclic shift characteristics of Fourier transform, provides a working frequency band detection method, replaces the original parameter setting mode, and greatly improves the convenience of the use of an HPLC system. Meanwhile, the method has no strict requirements on information such as sampling rate, sampling point positioning, frequency offset and the like, and can run in a robust mode; and key operation processing resources of the method, including correlators and Fourier transform, belong to the necessary components of broadband power line carrier HPLC digital processing, and a large amount of operation resources do not need to be additionally increased.

Drawings

FIG. 1 is a flow chart of the present invention;

FIG. 2 is a schematic diagram of a sliding window based process according to the present invention;

FIG. 3 is a schematic diagram of the energy spectrum of the leading signal of four working bands of HPLC;

fig. 4 is a diagram of binary quantization spectra of four working band pilot signals of HPLC.

Detailed Description

The invention is further described with reference to the following figures and detailed description.

As shown in fig. 1 to 4, a frequency band detection method for broadband power line carrier communication specifically includes the following steps:

101) signal preprocessing step: the sampling rate of the output signal of the front-end ADC is adjusted, so that the output signal rate is 1 time of the symbol rate.

I.e. the signal received by the receiver is the sampled output of the front-end ADC, the sampling rate of the ADC is generally higher than the signal bandwidth, i.e. the symbol rate, and usually has an integer multiple relationship. For this scheme, the sampling rate of the sampling output needs to be adjusted to 1 times of the symbol rate, i.e. equivalent to the signal bandwidth.

The adjustment to 1 time symbol rate can be realized by means of signal interpolation, filtering, sampling and the like, wherein the filtering is to filter digital spectrum repetition introduced by interpolation, and if the original sampling rate and the target sampling rate are in integral multiple relation, interpolation is not needed, and a filtering process is also not needed; the specific digital signal interpolation, filtering and sampling method is implemented, and the method is not described in detail, and can be implemented by those skilled in the art in a conventional manner.

102) Signal arrival detection: and acquiring delay autocorrelation energy and autocorrelation energy from a preamble signal of broadband power line carrier communication, and judging whether the signal arrives according to the change of the ratio of the delay autocorrelation energy to the autocorrelation energy. The specific implementation mode is as follows:

as shown in fig. 2, the delayed autocorrelation energy and the autocorrelation energy are implemented by sliding a window, where the window length is the length of the repetition period of the preamble signal of the broadband power line carrier communication, and the window sliding step is 1 times the symbol rate.

The window sliding sample moves in a first-in first-out mode, a group of delay autocorrelation and autocorrelation energy values are output once per step, and a ratio of the delay autocorrelation energy values to the autocorrelation energy values is obtained.

Specifically, in continuous sampling points, two sampling point sequences with the length of N and formed by 2N sampling points are selected in a sliding window mode, and time delay cross-correlation energy P is respectively calculated_Dcorr(t) and autocorrelation energy P_Acorr(t), the calculation formulas of the two are respectively as follows:

simultaneously calculating the correlation energy, and synchronously calculating the ratio R of the correlation energy_corr(t)：

R_corr(t)＝P_Dcorr(t)/P_Acorr(t) (equation 3)

Because the leading signal of the broadband power line carrier communication HPLC is sent repeatedly with the length N as a period, when the signal completely arrives and is in the leading area, theoretically R_corr(t) has a value of 1; while in the absence of signal, R is not repeated in sequence 1 and sequence 2 of the sliding window_corrThe value of (t) is close to 0. Therefore, it can be said that when R is_corr(t) when the signal arrival is detected when the signal arrival is more than 0.6, wherein 0.6 is a signal arrival threshold, the higher the threshold is set, the higher the reliability of the signal arrival detection is, but the probability of missed detection is synchronously increased; in a broadband power line carrier communication HPLC receiver, the preamble repetition period N is generally 1024, and the longer length brings strong increaseAnd the threshold is set to be 0.6, so that the arrival of the signal can be accurately judged.

103) A signal processing step: after the signal is judged to arrive, a mark of the signal arrival is given, and the receiver is prompted to enter the subsequent digital processing. I.e., the core purpose of channel arrival detection, is to give an indication of signal arrival and prompt the receiver for subsequent digital processing in a system employing a burst communication mechanism. And subsequently, carrying out peak clipping algorithm on the signal to remove narrow-band interference in the broadband power line carrier communication.

That is, after the arrival of the signal is detected, assuming that the time t arrives, N sampling points { x ] starting at the time t are selected₀,x₁,…,x_N-1Acquiring frequency domain information y of the signal from the frequency domain information₀,y_1,…,y_N-1The frequency domain information acquisition formula is as follows:

the N elements of the frequency domain information represent the amplitude and phase information of the preamble signal at the N frequency points. Obtaining amplitude information p for each element in the frequency domain information₀,p₁,…,p_N-1Obtaining the signal energy intensity power (y) on each frequency point_n) Instead of obtaining amplitude p by energy intensity_n＝power(y_n)。

After the energy intensity information in the frequency domain information is obtained, the narrow-band interference in a real HPLC system is resisted through a peak clipping algorithm by combining the characteristics that the energy intensities of all frequency points in an HPLC preamble signal frequency domain are the same and the selective fading of the frequency domain is not mutated when an HPLC transmission channel is introduced. Namely, when the difference value between the target frequency point of the detected signal and the energy intensity of the adjacent frequency points in the front and the back exceeds 4 times, the corresponding value is 6 dB. Then, the target frequency point with the narrow-band interference is considered to exist on the target frequency point, and the target frequency point with the narrow-band interference is replaced by the weighted average value of the energy intensity of the adjacent frequency points in the front and the back, and the specific formula is as follows:

104) a judgment processing step: and performing binary quantization spectrum processing on the signal without the narrow-band interference, wherein the binary quantization spectrum converts the information of the specific frequency point into a frequency domain energy binary sequence consisting of 0 and 1. That is, a binary quantization spectrum is represented to indicate whether each frequency point on the frequency domain has a spectrum of a modulation signal, generally 0 indicates that there is no modulation signal but only noise, and 1 indicates that there is a modulation signal. Whether a modulation signal exists on each frequency point can be judged through the relation between the frequency point energy and the noise energy. The noise energy intensity can be obtained by performing statistical averaging on the energy intensity of the idle frequency points defined by the HPLC standard.

As shown in FIG. 3, an ideal energy spectrum diagram of the leading signals of four working bands of HPLC is shown, wherein the data will not be modulated at the frequency points 0-31, 491-_noiseNamely:

decision threshold condition TH of binary quantization spectrum_bDefined as the noise energy intensity P_noise2 times higher, i.e. the signal is required to be 3dB greater than the noise. By comparing the frequency domain energy p of each frequency point in the frequency domain information₀,p₁,…,p_N-1H and a decision threshold condition TH one by one_bComparing to obtain binary quantization spectrums { b) of all frequency points₀,b₁,…,b_N-1And fixedly filling 0 in other unused frequency points, namely:

in the same principle, as shown in fig. 4, the ideal binary quantization spectra of the four working band preambles of HPLC can be directly calculated, which are respectively { b1 }₀,b1₁,…,b1_N-1}，{b2₀,b2₁,…,b2_N-1}，{b3₀,b3₁,…,b3_N-1}，{b4₀,b4₁,…,b4_N-1}。

After binary quantization spectrum sequences of the received signal and the standard signal are obtained respectively, the correlation degree CCoef between the received signal and the standard signal can be measured_nThe calculation is carried out, and for the binary quantization signal, the logical exclusive-or and accumulation operation is directly executed according to elements, namely:

wherein n is 1, 2, 3, 4, corresponding to four standard working frequency bands respectively. Of the four correlation coefficients, the value with the minimum value corresponds to the working frequency band of the currently received signal, and the ideal minimum value may be zero.

Namely, the frequency domain energy binary sequence of the signal without the narrow-band interference is subjected to exclusive or processing with the frequency domain energy binary sequence of four frequency bands defined by the standard, and the accumulated sum is calculated. The obtained four calculation results respectively correspond to the similarity between the received signal and the four standard frequency bands, the higher the similarity is, the smaller the accumulation sum is, and the frequency band corresponding to the minimum accumulation sum is the working frequency band of the received signal.

In summary, the scheme is not affected by phase information, and in the characteristic of Fourier transform, the input signal x_kCyclic shift of y_nIs only reflected in the phase and does not change the amplitude, so that the selected starting sample point x is not required₀Corresponding to the first point of the actual preamble sequence, any sampling point after the arrival of the signal can be selected as the initial sampling point x₀。

If the reliability of the final result needs to be further improved, the step 103) and the step 104) may be repeatedly executed by selecting different sampling data sets for multiple times after the signal arrives, and the correlation information obtained by multiple processing may be subjected to statistical processing.

The foregoing is only a preferred embodiment of the present invention, and it should be noted that, for those skilled in the art, several modifications and decorations can be made without departing from the spirit of the present invention, and these modifications and decorations should also be regarded as being within the scope of the present invention.