A Time Delay Measurement Method Based on Noise Correlation
Through the differential covariance method based on noise correlation, delay information is directly extracted from the frequency signal, solving the complexity of delay measurement in time-frequency simultaneous transmission technology, and achieving high-precision delay measurement.
Patent Information
- Application Number
- CN202310013996.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2023-01-05
- Publication Date
- 2025-07-08
- Estimated Expiration
- 2043-01-05
AI Technical Summary
In the prior art, the transmission delay measurement of time-frequency signals is complex and difficult to be efficiently implemented, especially in the time-frequency simultaneous transmission technology, the delay correction of time signals requires complex system measurement.
By using the differential covariance method based on the noise characteristics of the frequency signal, the noise information in the frequency signal is extracted, the transmission delay of the signal is measured, and the delay measurement is performed using the noise correlation method.
A simple, effective and robust delay measurement method is realized, with the accuracy better than sampling accuracy, and simplifying the acquisition process of delay information.
Smart Images

Figure CN116094960B_ABST
Abstract
Description
Technical Field
[0001] The present invention is mainly applied to the field of generation and processing of time-frequency signals, and mainly solves the related problems of local and wide-area time delay measurement and time signal synchronization. Measuring the transmission time delay of signals is the most important work in the time-frequency field, and it has very important applications in fields such as navigation, communication, and radar. Background Art
[0002] Time-frequency signals have extensive and important applications in modern high-tech fields. The transmission of high-precision frequency signals and the synchronization of time signals play very important roles in them. Currently, the world's common time benchmarks - International Atomic Time and Coordinated Universal Time are achieved through satellite time transfer, and higher-precision time-frequency signal transfer is achieved through fiber optic networks. In these technologies, frequency signals and time signals are often transmitted separately. The frequency signal is a high-performance sine-cosine wave signal, while the time signal is a second pulse or a coded signal.
[0003] Since the frequency signal is continuous and the time signal is pulse-modulated, generally the performance index of the frequency signal is better than that of the time signal. Many fields require both frequency signals and time signals, so time-frequency co-transmission technologies have also been developed. Their principle is to add both frequency signals and time signals in the same channel to achieve the transmission of the two signals. Due to signal transmission delay, the time signal must correct this delay. Therefore, the measurement of delay is an important part of time signal transmission and requires complex system measurement. Summary of the Invention
[0004] Based on the uniqueness and non-repeatability of the noise of the frequency signal, the present invention measures the time delay of the frequency signal by scanning the relevant time, and realizes directly extracting both the frequency signal and the relative time delay information from the frequency signal.
[0005] The present invention is implemented based on the noise processing method of "differential covariance". It is the combination of differential deviation and covariance. The differential variance is a common data processing method in the time-frequency field. The first-order difference is called Allan Deviation (ADEV), and the second-order difference is called Hadamard Deviation (HDEV). We express it in the form of covariance, which is called Allan covariance (ACOV) and Hadamard covariance (HCOV), and is expressed as:
[0006]
[0007] where τ is the integration time, a j (τ) is the average value of a in the time period from jτ to (j + 1)τ, b j(τ, Δt) is the average value of b in the time period from jτ + Δt to (j + 1)τ + Δt. {a j (τ)} and {b j (τ)} are two noise sequences, both having M elements. j is the sequence number. Δt is the relative time delay between the two noise sequences. If the same noise y 1,j is included in the two sequences, denoted as a j = (y1 - y2) j , b j = (y1 - y3) j , then the relative time delay of y1 in the two noise comparison experiments can be measured by measuring the variation of the Allan correlation coefficient (AR) or the Hadamard correlation coefficient (HR) with the relative time delay Δt. We take AR as an example to illustrate:
[0008]
[0009] The maximum value of AR corresponds to the situation where the middle noise y1 of the two noise sequences is aligned, and the corresponding time relationship is t = -Δt. In addition, there are two minimum values, corresponding to t = -Δt ± τ. In this way, the transmission time delay of the signal can be obtained according to the extreme positions in the AR curve. The results of theoretical analysis are as Figure 2 shown, and the results of the experiment are as Figure 3 shown.
[0010] In implementation, first, the local and remote error information of the frequency signal is collected by comparison, and then the collected information is processed to obtain the curve of AR varying with the relative time delay, and the transmission time delay is obtained according to the position of the maximum value. Description of the Drawings
[0011] Figure 1 is the device diagram of the time delay measurement method based on noise correlation.
[0012] Figure 2 is the data processing simulation diagram of the time delay measurement method based on noise correlation. The relative time delay is taken as 0, and the integration time is taken as 30. It can be seen that the results of numerical simulation show that there is a maximum value at the relative time delay of 0, and there are two minimum values at the relative time delays of ±30, which is also expected by the theory.
[0013] Figure 3 is the experimental result of the time delay measurement method based on noise correlation. We take three groups of data with the integration time of 1 second, 2 seconds, and 4 seconds, and find that the time delay between the two signals is 1514.3 seconds, and the measurement accuracy is better than the sampling accuracy (1 second). Detailed Implementation Manner
[0014] The present invention will be further described below in conjunction with the embodiments and the drawings, but the protection scope of the present invention should not be limited thereby.
[0015] See Figure 1 , Figure 1 which is a structural diagram of a time delay measurement method based on noise correlation of the present invention. It includes a frequency source (1), a signal transmission line (2), a local frequency reference (3), a local phase comparator (4), a remote frequency reference (5), a remote phase comparator (6), a local data recorder (7), a remote data recorder (8), and a data processor (9).
[0016] It is actually two sets of frequency comparison systems. The frequency source (1) and the local frequency reference (3) input signals into the local phase comparator (4) to achieve local frequency comparison. The local data recorder (7) records the data to obtain a noise data sequence {y 1,i}}, forming the first set of comparison systems. The frequency source (1) transmits the frequency signal through the signal transmission line (2). At the remote end, the remote frequency reference (5) is connected to the remote phase comparator (6). The comparison result is recorded by the remote data recorder (8) to obtain a noise data sequence {y 2,+l}}, forming the second set of comparison systems. The noise data sequences {y 1,i} and {y 2,+l} are processed according to the method of differential covariance to obtain an AR curve, and the transmission time delay is obtained by locating the maximum value.
[0017] Since the present invention uses noise as the carrier of information for data analysis and extracts very useful time information, it realizes the measurement of time delay information without adding any technical means, and is a simple, effective, and robust time delay measurement method.
Claims
1. A method for measuring the transmission delay of a time-frequency signal based on noise correlation, characterized in that, Comprising the following steps: Providing a frequency source (1), a signal transmission line (2), a local frequency reference (3), a local phase comparator (4), a remote frequency reference (5), a remote phase comparator (6), local data recording (7), remote data recording (8) and data processing (9); The frequency source (1) and the local frequency reference (3) input signals into the local phase comparator (4) to perform local frequency comparison, and record the comparison results in the local data recording (7) to obtain local noise data; Meanwhile, the frequency source (1) transmits frequency signals through the signal transmission line (2), and at the remote end, the remote frequency reference (5) is connected to the remote phase comparator (6) for remote frequency comparison, and the comparison results are recorded in the remote data recording (8) to obtain remote noise data; Processing the data in the local data recording (7) and the remote data recording (8), solving the variation of their differential covariance with the relative time delay, and when the differential covariance is the maximum value, the corresponding relative time delay is the transmission time delay from the frequency source (1) to the remote end.
2. A method for measuring the time delay of time-frequency signal transmission based on noise correlation measurement according to claim 1, characterized in that: ① Using noise as a signal and measuring the time delay by extracting the time information of the noise; ② Measuring the correlation between noises by means of differential covariance; ③ Using the method of relative translation of noise sequences to obtain the variation of the correlation of noise sequences with the relative time delay; ④ Obtaining the transmission time delay by locating the maximum value of the differential covariance vs. relative time delay curve.
Citation Information
Patent Citations
Method and system for remotely calibrating frequency standards
CN101692163A
Noise sensitivity coefficient measurement method based on noise correlation
CN110133383A