Method for measuring time delay through multi-frequency variance weighted fitting

Abstract

The invention discloses a method for measuring time delay through multi-frequency variance weighted fitting, which utilizes a conventional time delay estimation and curve fitting method to obtain high-precision time delay estimation in a frequency sub-band, and utilizes a change rule of noise signal fluctuation along with variance to solve the problem of how to estimate the time delay with high precision under the condition of broadband noise.

Description

Technical Field

[0001] The present invention relates to the technical field of signal processing, and particularly relates to a method for measuring time delay by multi-frequency variance weighted fitting. Background Art

[0002] Research on the influence of ocean environmental noise shows that at low signal-to-noise ratios, the phase differences tend to be uniformly distributed with the largest variance; when the signal-to-noise ratio increases, the phase differences tend to be exponentially distributed and the variance decreases sharply.

[0003] The time delay difference is a basic observable for target azimuth and distance measurement, and high-precision time delay difference measurement is the key to direction finding and ranging. When the signal received by a passive sonar is the continuous noise radiated by a target, it is crucial to solve the problem of the accuracy of time delay estimation under broadband radiated noise conditions. Summary of the Invention

[0004] The present invention provides a method for measuring time delay by multi-frequency variance weighted fitting to solve the problem of the accuracy of time delay measurement for broadband noise signals.

[0005] To solve the above technical problems, the present invention adopts the following technical solutions:

[0006] A method for measuring time delay by multi-frequency variance weighted fitting includes: obtaining broadband noise and dividing the broadband noise into multiple frequency sub-bands; using least squares fitting to find the time delay of each frequency sub-band; calculating the mean time delay for the time delays of multiple frequency sub-bands; calculating the time delay variance of each frequency sub-band at multiple moments according to the mean time delay and the time delay of each frequency sub-band; calculating the sum of the time delay variances of multiple frequency sub-bands to obtain the total variance; calculating the proportion of the time delay variance of each frequency sub-band in the total variance; and fitting the time delays of each frequency sub-band according to the proportion to obtain the time delay of the broadband noise.

[0007] Preferably, using least squares fitting to find the time delay of each frequency sub-band is specifically performed according to the following formula;

[0008] In the formula, f i represents the frequency of the i-th frequency point, θ i represents the phase angle of the i-th frequency point, i represents the serial number of the frequency sub-band, k represents the number of frequency sub-bands, and τ n represents the time delay of the n-th frequency sub-band.

[0009] Preferably, calculating the mean time delay for the time delays of multiple frequency sub-bands is specifically performed according to the following formula;

[0010] In the formula, n = 1,..., N, N represents the number of frequency sub-bands, represents the mean time delay.

[0011] Preferably, the time delay variance is specifically executed according to the following formula; In the formula, σ n represents the time delay variance, m represents the number of moments, and τ n，i represents the time delay of the nth frequency sub-band at the i-th moment.

[0012] Preferably, the total variance is specifically executed according to the following formula; In the formula, σ represents the total variance.

[0013] Preferably, according to the ratio, the time delays of each frequency sub-band are fitted to obtain the time delay of the broadband noise, which is specifically executed according to the following formula; In the formula, τ represents the time delay of the broadband noise, and W n represents the ratio of the time delay variance of each frequency sub-band to the total variance.

[0014] Compared with the prior art, the method for measuring time delay by multi-frequency variance weighted fitting provided by the embodiments of the present invention has the following remarkable advantages:

[0015] Using multi-frequency variance weighted fitting to measure time delay can better approximate the true value and improve the accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 is a schematic diagram of the working process of the method for measuring time delay by multi-frequency variance weighted fitting provided by the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0017] The following combines the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0018] As Figure 1 shown, the embodiments of the present invention provide a method for measuring time delay by multi-frequency variance weighted fitting to solve the problem of time delay measurement accuracy of broadband noise signals. The present invention divides the broadband noise into N sub-bands, and performs high-precision time delay estimation on each sub-band. The time delay accuracy is limited by the signal bandwidth. To improve the estimation accuracy, the phase difference is used to obtain the time delay difference. The effect of directly averaging the time delay difference is not the best. Combining the characteristics of the noise, using multi-frequency variance weighted fitting to measure time delay can better approximate the true value.

[0019] The multi-frequency variance weighted fitting method for measuring time delay provided by the present invention is based on realizing the time delay of each frequency sub-band according to variance weighting. The time delay design of the frequency sub-band is to obtain the high-precision time delay within the frequency sub-band by least squares fitting of the phase, and then perform variance weighting on the time delay of the frequency sub-band to obtain the time delay, so as to realize the high-precision estimation of the time delay within the wide frequency band.

[0020] The specific steps are as follows:

[0021] Step 1: Use least squares fitting to calculate the time delay for each frequency sub-band. The least squares fitting formula is: The sub-band time delay τ n is τ n = ζ n / 2π; f i represents the frequency of the i-th frequency point, θ i represents the phase angle of the i-th frequency point, k represents the number of frequency points, i represents the serial number of the frequency point, and n represents the serial number of the sub-band.

[0022] Step 2: Calculate the mean value of the N frequency sub-band time delays τ n obtained in Step 1

[0023] Step 3: Calculate the variance of the τ n，t obtained in Step 1 for m measurements.

[0024] Step 4: Sum the variances σ n obtained in Step 3, that is

[0025] Step 5: Calculate the percentage of σ n in the total variance σ, that is

[0026] Step 6: Fit multiple time delay values according to the weights and output the time delay, and calculate

[0027] The beneficial effects of the present invention are as follows:

[0028] By using the present invention, high-precision time delay estimation within the frequency sub-band is obtained by using conventional time delay estimation and curve fitting methods, and the variation law of the noise signal fluctuation with variance is utilized, solving the problem of how to estimate the time delay with high precision under the condition of wideband noise.

[0029] In the description of this specification, specific features, structures, materials or characteristics may be combined in a suitable manner in any one or more embodiments or examples.

[0030] Of course, the present invention may also have many other embodiments. Without departing from the spirit and essence of the present invention, those skilled in the art can make various corresponding changes and modifications according to the present invention. However, these changes and modifications should all fall within the protection scope of the claims of the present invention.

Claims

1. A method for measuring time delay by multi-frequency variance weighted fitting, characterized in that: include: Obtaining broadband noise, and dividing the broadband noise into a plurality of frequency sub-bands; Use least squares fitting to find the delay of each frequency subband; For delays of multiple frequency sub-bands, calculate the average delay; Calculate the delay variance of each frequency sub-band at multiple times based on the delay mean and the delay of each frequency sub-band; Calculate the sum of the delay variances of multiple frequency sub-bands to obtain the total variance; Calculate the ratio of the delay variance of each frequency sub-band to the total variance; According to the ratio, the time delay of each frequency sub-band is fitted to obtain the time delay of broadband noise.

2. The method for measuring time delay by multi-frequency variance weighted fitting according to claim 1, characterized in that: The least squares fitting is used to calculate the delay of each frequency sub-band, which is specifically performed according to the following formula: In the formula, f i represents the frequency of the i-th frequency point, θ i represents the phase angle of the i-th frequency point, i represents the sequence number of the frequency subband, k represents the number of frequency subbands, τ n represents the delay of the nth frequency subband.

3. The method for measuring time delay by multi-frequency variance weighted fitting according to claim 2, characterized in that: The delay of the multiple frequency sub-bands is calculated by calculating the delay average, which is specifically performed according to the following formula; Where n=1, ..., N, N represents the number of frequency subbands, Indicates the average delay.

4. The method for measuring time delay by multi-frequency variance weighted fitting according to claim 3, characterized in that: The delay variance is specifically implemented according to the following formula: In the formula, σ n represents the delay variance, m represents the number of moments, τ n,i represents the delay of the nth frequency subband at time i.

5. The method for measuring time delay by multi-frequency variance weighted fitting according to claim 4, characterized in that: The total variance is specifically implemented according to the following formula: In the formula, σ represents the total variance.

6. The method for measuring delay by multi-frequency variance weighted fitting according to claim 5, characterized in that: The time delay of each frequency sub-band is fitted according to the ratio to obtain the time delay of broadband noise, which is specifically performed according to the following formula; Where τ represents the delay of broadband noise, W n It indicates the ratio of the delay variance of each frequency sub-band to the total variance.

Images