Spatial resolution improving method based on incoherent optical frequency domain reflection time delay measurement

The high side lobes are suppressed through the adaptive iteration method, and the optical delay measurement resolution of the incoherent optical frequency domain reflection method is improved, and the problems of insufficient resolution and dispersion effects in traditional methods are solved, achieving higher measurement accuracy and better measurement Low hardware requirements.

CN119027315AInactive Publication Date: 2024-11-26NANJING GUANGYUE TECHNOLOGY CO LTD +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410952697.6
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-16
Publication Date
2024-11-26
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

The existing multi-channel optical delay measurement methods lack spatial resolution, especially the incoherent optical frequency domain reflection method introduces high side lobes when restoring time domain response, which limits the time delay measurement resolution and is easily averse to long-distance measurements. Dispersion effect.

Method used

Adaptive iteration method is adopted to acquire microwave signals and photocurrent signals through a dual-channel amplitude phase receiver, and an error covariance matrix and a weighted least squares cost function are constructed to suppress high side lobes, thereby improving the resolution of optical delay measurement.

Benefits of technology

Without the need to expand the scanning frequency range, the time-delay measurement resolution is improved, hardware requirements are reduced, and the impact of dispersion effect on measurements over long distances is avoided.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119027315A_ABST
    Figure CN119027315A_ABST
Patent Text Reader

Abstract

The invention discloses a spatial resolution improving method based on incoherent optical frequency domain reflection time delay measurement. The method mainly comprises the following steps of extracting amplitude and phase differences of sequentially set frequencies in a microwave signal and a light current signal through an amplitude and phase receiver; setting an optical time delay measurement range and a measurement minimum division value; constructing an error covariance matrix for the time domain response corresponding to each set optical time delay value, and constructing a cost function for minimizing the error covariance matrix based on a weighted least square criterion; a multi-channel time delay estimation result enabling the cost function to be minimum is solved through a self-adaptive iteration method; and finally extracting an optical time delay value corresponding to a maximum value in the estimated time domain response. According to the invention, the hardware condition of the incoherent frequency domain reflection system is reduced, and the influence of the dispersion effect on measurement under long-distance measurement is avoided.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to an optical delay measurement method, and in particular to a method for improving the spatial resolution of delay measurement based on incoherent optical frequency domain reflection. Background Art

[0002] Optical delay is a basic parameter in optical systems, which determines the key performance of the system. At present, electronic information systems are developing towards large-scale and large arrays, and more and more scenarios require multi-path optical delay measurement and control. Taking the multi-channel beam scanning system with optical true delay control as an example (Ren C, Pan B, Chen Y, et al. Orbital Angular Momentum Mode-Group Beamforming System Based on An Integrated Optical TrueTime Delay Line Chip[C] / / 2021IEEE MTT-S International Wireless Symposium(IWS).IEEE, 2021:1-3.), each of its four channels has 128 delay control states, so multi-channel delay measurement technology is required. In order to distinguish the delay states of each channel, multi-channel delay measurement technology must have a high spatial resolution, so improving the spatial resolution of optical delay measurement is of great value and significance.

[0003] At present, the commonly used multi-channel optical delay measurement methods mainly include optical time domain reflectometry, coherent optical frequency domain reflectometry, and incoherent optical frequency domain reflectometry. In the optical time domain reflectometry method, each link to be measured will reflect pulse light during measurement. The system realizes multi-path optical delay measurement by resolving pulse light in the time domain. However, the delay measurement resolution of this method is limited by the pulse width and it is difficult to break through the nanosecond level. The coherent optical frequency domain reflectometry method uses a continuous frequency sweep laser to mix the reflected detection light and the local oscillator light, and the delay information of each link to be measured will be mapped to the frequency of the mixed signal. The delay measurement resolution of this scheme is limited by the bandwidth of the laser frequency scan. At the same time, due to the influence of the laser's own coherent noise, its measurement resolution decreases rapidly as the measurement distance increases. In addition, the coherent frequency domain reflectometry method is not suitable for the measurement of multimode links. Multimode interference will introduce intensity jitter in the coherent measurement system, which seriously affects the measurement results. The incoherent optical frequency domain reflectometry method performs a series of intensity modulations on a low-coherent laser source, measures the amplitude and phase responses of the superposition of multipath links for each modulation frequency, and then performs frequency-time transformation on the amplitude and phase responses to restore the time domain response, thereby obtaining the optical delay information of each link to be measured. This scheme has low requirements for measurement devices, and because the light source is incoherent, this scheme can achieve the measurement of multimode optical links. The delay measurement resolution of this scheme is limited by the frequency bandwidth of the intensity modulation. The limited bandwidth will introduce high side lobes in the frequency-time transformation, thereby limiting the delay measurement resolution; however, if the frequency bandwidth is further expanded, the measurement system is susceptible to dispersion effects. Summary of the invention

[0004] Purpose of the invention: In view of the shortcomings of the prior art, the present invention proposes a method for improving the spatial resolution of time delay measurement based on incoherent optical frequency domain reflection, which solves the problem that the traditional frequency-time transformation method introduces high side lobes when restoring the time domain response, thereby limiting the time domain resolution. The adaptive iterative method is used to suppress the high side lobes, thereby improving the resolution of optical delay measurement.

[0005] Technical solution: The method for improving the optical delay resolution based on incoherent optical frequency domain reflection described in the present invention comprises the following steps:

[0006] Step 1: Use a dual-channel amplitude-phase receiver to collect local oscillator microwave signals and beat frequency photocurrent signals of various frequencies, and extract the amplitude-phase difference of microwave signals of various frequencies in the two channels.

[0007] Step 2: Set the optical delay measurement range and the minimum measurement division value.

[0008] Step 3: construct an error covariance matrix for the time domain response corresponding to each optical delay value set in step 2, and construct a cost function that minimizes the error covariance matrix based on the weighted least squares criterion.

[0009] Step 4: Solve the time domain response through an adaptive iterative method to obtain the multi-channel delay estimation result that minimizes the cost function in step 3.

[0010] Step 5: Extract the optical delay value corresponding to the maximum value in the estimated time domain response.

[0011] Furthermore, the implementation process of step 1 is as follows:

[0012] The microwave source generates M frequencies f1, f2, ..., f in sequence. M The signal is used as the local oscillator microwave signal, which is modulated in turn onto a wide-spectrum optical carrier. The modulated optical signal is converted into photoelectricity after passing through the optical link to be measured to generate a photocurrent signal carrying the frequency. The amplitude and phase receiver extracts the amplitude and phase differences of the M frequency components of the local oscillator microwave signal and the photocurrent signal between the two channels in turn: i(f1), i(f2), …, i(f M ).

[0013] Furthermore, the implementation process of step 2 is as follows:

[0014] Assume that the optical delay range to be measured is τ1, τ2, ..., τ k ,…,τ K , the minimum division value of optical delay measurement is Δτ=τ k -τ k-1 .

[0015] Furthermore, the implementation process of step 3 is as follows:

[0016] Assume that the time domain response to be obtained is ψ=[ψ1,ψ2,…,ψ k ,…,ψ K ]. The frequency response of the link can be expressed as:

[0017] i=F(τ)ψ+e (1)

[0018] where i=[i(f1),i(f2),…,i(f M )] is the amplitude difference of the two signals received by the amplitude-phase receiver; F(τ) is an M×K matrix, defined as F(τ)=[f(τ1),f(τ2),…,f(τ k ),…,f(τ K )], and f(τ k )=[exp(-2πf1τ k ),exp(-2πf2τ k ),…,exp(-2πf M τ k )] T ; e is the additive noise existing in each frequency component received by each amplitude and phase receiver.

[0019] From formula (1), we know that the optical delay value is τ k The error covariance matrix Q(τ k ) is constructed as follows:

[0020] Q(τ k )=F(τ)PF H (τ)-|ψ k | 2 f(τ k )f H (τ k ) (2)

[0021] where P = diag([|ψ1| 2 ,|ψ2| 2 ,…,|ψ K | 2 ]), diag(…) is a diagonal matrix, and P k =|ψ k | 2 .

[0022] According to the weighted least squares criterion, τ k The time domain response ψ k The cost function C(ψ k ):

[0023] C(ψ k )=(if(τ k )ψ k ) H Q -1 (τ k )(if(τ k )ψ k ) (3)

[0024] Furthermore, the implementation process of step 4 is as follows:

[0025] Note k n is the τ obtained in the nth iteration k The time domain response of k 0 is the initial solution; P n =diag([|ψ1 n | 2 ,|ψ2 n | 2 ,…,|ψ K n | 2 ]) is the square of the time domain response amplitude obtained in the nth iteration, P0 is the initial solution; Q n (τ k ) is the corresponding delay τ in the nth iteration kThe resulting error covariance matrix, Q0(τ k ) is the initial solution. The initial solution is obtained by the following formula:

[0026]

[0027] The adaptive iterative process is:

[0028]

[0029] Formula (6) takes n = 1, 2, ..., N. The iteration termination condition is:

[0030] ||P n -P n-1 ||2<P th (6)

[0031] Among them, ||…||2 is the 2-norm of the matrix, P th It is the preset iteration termination threshold condition.

[0032] Furthermore, the implementation process of step 5 is as follows:

[0033] In order to find the delay of each channel, construct the time domain response function ψ(τ k )=ψ k N , the τ corresponding to each maximum point of the function k This is the optical delay to be measured.

[0034] Beneficial effect: Compared with the prior art, the present invention suppresses the high side lobes introduced in the time domain by the traditional frequency-time transform through an adaptive iterative algorithm, and can improve the delay measurement resolution without increasing the frequency scanning range, thereby reducing the hardware conditions of the incoherent frequency domain reflection system, and also avoids the influence of the dispersion effect on the measurement under long-distance measurement. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 This is a flow chart of the method for improving spatial resolution based on incoherent optical frequency domain reflection delay measurement of the present invention.

[0036] Figure 2 This is the structural diagram of a typical incoherent frequency domain reflection system.

[0037] Figure 3 Time domain response recovered by traditional incoherent optical frequency domain reflectometry method for multipath simulation.

[0038] Figure 4 It is the time domain response restored by the method of the present invention in multipath simulation. DETAILED DESCRIPTION

[0039] In view of the limitation of the resolution of the existing incoherent optical frequency domain reflection method for measuring optical delay, the solution of the present invention is to suppress the high side lobes introduced by the traditional frequency-time transformation method when restoring the time domain response through an adaptive iteration method, thereby improving the delay measurement resolution. Compared with the traditional incoherent optical frequency domain reflection technology, which requires the expansion of the microwave frequency scanning range to improve the resolution, the solution of the present invention greatly reduces the hardware requirements, and avoids the influence of the dispersion effect on the measurement at long distances while improving the resolution.

[0040] The present invention will be further described below in conjunction with the accompanying drawings.

[0041] The present invention proposes a method for improving optical delay resolution based on incoherent optical frequency domain reflection, such as Figure 1 As shown, the specific steps include:

[0042] Step 1: Use a dual-channel amplitude-phase receiver to collect local oscillator microwave signals and beat frequency photocurrent signals of various frequencies, and extract the amplitude-phase difference of microwave signals of various frequencies in the two channels.

[0043] Traditional incoherent optical frequency domain reflectometry technology uses a series of swept microwave signals to modulate a low-coherence carrier. Through frequency scanning and frequency-time conversion algorithms, the time domain response of the optical link to be tested can be obtained. The time corresponding to each maximum value in the time domain response is the delay value in the link to be tested. The typical system structure diagram is shown below. Figure 2 As shown, suppose the microwave source generates M microwave frequencies f1, f2, ..., f M Loaded onto the intensity modulator, the amplitude and phase receiver sequentially extracts the amplitude and phase differences of M frequency components i(f1), i(f2), ..., i(f M ).

[0044] Step 2: Set the optical delay measurement range and the minimum measurement division value.

[0045] Assume that the optical delay range to be measured is τ1, τ2, ..., τ k ,…,τ K , the minimum division value of optical delay measurement is Δτ=τ k -τ k-1 .

[0046] Step 3: construct an error covariance matrix for the time domain response corresponding to each optical delay value set in step 2, and construct a cost function that minimizes the error covariance matrix based on the weighted least squares criterion.

[0047] To solve the time domain response of the link to be tested, assume that the time domain response to be obtained is ψ = [ψ1, ψ2, …, ψ k ,…,ψ K ]. If the link to be tested has a delay value of τ k The link, then the time domain response ψ isk There is a maximum value near. The frequency response of the link can be expressed as:

[0048] i=F(τ)ψ+e (1)

[0049] where i=[i(f1),i(f2),…,i(f M )] is the amplitude difference of the two signals received by the amplitude-phase receiver; F(τ) is an M×K matrix, defined as F(τ)=[f(τ1),f(τ2),…,f(τ k ),…,f(τ K )], and f(τ k )=[exp(-2πf1τ k ),exp(-2πf2τ k ),…,exp(-2πf M τ k )] T ; e is the additive noise existing in each frequency component received by each amplitude and phase receiver.

[0050] From formula (1), we know that the optical delay value is τ k The error covariance matrix Q(τ k ) is constructed as follows:

[0051] Q(τ k )=F(τ)PF H (τ)-|ψ k | 2 f(τ k )f H (τ k ) (2)

[0052] where P = diag([|ψ1| 2 ,|ψ2| 2 ,…,|ψ K | 2 ]), diag(…) is a diagonal matrix, and P k =|ψ k | 2 .

[0053] According to the weighted least squares criterion, τ k The time domain response ψ k The cost function C(ψ k ):

[0054] C(ψ k )=(if(τ k )ψ k ) H Q -1 (τ k )(if(τk )ψ k ) (3)

[0055] ψ that minimizes the cost function of equation (3) k The optimal estimate has an analytical expression:

[0056]

[0057] Step 4: Solve the time domain response through an adaptive iterative method to obtain the multi-channel delay estimation result that minimizes the cost function in step 3.

[0058] remember is the τ obtained in the nth iteration k The time domain response of is the initial solution; P n =diag([|ψ1 n | 2 ,|ψ2 n | 2 ,…,|ψ K n | 2 ]) is the square of the time domain response amplitude obtained in the nth iteration, P0 is the initial solution; Q n (τ k ) is the corresponding delay τ in the nth iteration k The resulting error covariance matrix, Q0(τ k ) is the initial solution. The initial solution is obtained by the following formula:

[0059]

[0060] The adaptive iterative process is:

[0061]

[0062] Formula (6) takes n=1, 2, ..., N. The iteration termination condition is:

[0063] ||P n -P n-1 ||2<P th (7)

[0064] Among them, ||…||2 is the 2-norm of the matrix, P th It is the preset iteration termination threshold condition.

[0065] Step 5: Extract the optical delay value corresponding to the maximum value in the estimated time domain response. Construct the time domain response function ψ(τ k )=ψ k N , the τ corresponding to each maximum point of the function kThis is the optical delay to be measured.

[0066] The following simulation experiment is used to verify the method for improving the spatial resolution of incoherent optical frequency domain reflection delay measurement proposed in the present invention, and the experimental results also fully demonstrate the effectiveness of the method of the present invention.

[0067] In the simulation, the non-coherent frequency domain reflection measurement system measures three links simultaneously, and the delay values ​​are 400.4ps, 460.7ps, and 1200.2ps respectively. The frequency scanning range is 1GHz to 9GHz, and the scanning frequency interval is 0.5GHz. The delay measurement range is set to 0ns to 2ns, and the minimum delay measurement division value is 0.1ps. The time domain response recovered using the traditional frequency-time conversion algorithm is as follows Figure 3 As shown, the time domain response restored using the adaptive iterative method proposed in the present invention is as follows Figure 4 As shown in the figure, the result is obtained after 30 iterations. According to the results, it can be seen that the traditional method cannot distinguish the two links of 400.4ps and 460.7ps when the frequency scanning range is 1GHz to 9GHz, and the resolution of measuring the delay is greatly limited. The method of the present invention greatly improves the delay measurement resolution of the incoherent frequency domain reflection method, verifying the effectiveness of this method.

Claims

1. A method for improving spatial resolution of time delay measurement based on incoherent optical frequency domain reflection, characterized in that: The following steps are involved: Step 1: Use a dual-channel amplitude-phase receiver to collect local oscillator microwave signals and beat frequency photocurrent signals of various frequencies, and extract the amplitude-phase difference of microwave signals of various frequencies in the two channels; Step 2: Set the optical delay measurement range and the minimum measurement division value; Step 3: construct an error covariance matrix for the time domain response corresponding to each optical delay value set in step 2, and construct a cost function that minimizes the error covariance matrix based on the weighted least squares criterion; Step 4: Solve the time domain response by adaptive iterative method to obtain the multi-channel delay estimation result that minimizes the cost function in step 3; Step 5: Extract the optical delay value corresponding to the maximum value in the estimated time domain response.

2. The method for improving spatial resolution of time delay measurement based on incoherent optical frequency domain reflection according to claim 1 is characterized in that: The implementation process of step 1 is as follows: the microwave source generates M frequencies of f1, f2, ..., f M The signal is used as the local oscillator microwave signal, which is modulated in turn onto a wide-spectrum optical carrier. The modulated optical signal is converted into photoelectricity after passing through the optical link to be measured to generate a photocurrent signal carrying the frequency. The amplitude and phase receiver extracts the amplitude and phase differences of the M frequency components of the local oscillator microwave signal and the photocurrent signal between the two channels in turn: i(f1), i(f2), …, i(f M ).

3. The method for improving spatial resolution of time delay measurement based on incoherent optical frequency domain reflection according to claim 1 is characterized in that: The implementation process of step 2 is as follows: Assume that the optical delay range to be measured is τ1, τ2, ..., τ k ,…,τ K , the minimum division value of optical delay measurement is Δτ=τ k -τ k-1 .

4. The method for improving spatial resolution of time delay measurement based on incoherent optical frequency domain reflection according to claim 1 is characterized in that: The implementation process of step 3 is as follows: Assume that the time domain response to be obtained is ψ=[ψ1,ψ2,…,ψ k ,…,ψ K ], the frequency response of the link can be expressed as: i=F(τ)ψ+e (1) where i=[i(f1),i(f2),…,i(f M )] is the amplitude difference of the two signals received by the amplitude-phase receiver; F(τ) is an M×K matrix, defined as F(τ)=[f(τ1),f(τ2),…,f(τ k ),…,f(τ K )], and f(τ k )=[exp(-2πf1τ k ),exp(-2πf2τ k ),…,exp(-2πf M τ k )] T ; e is the additive noise existing in each frequency component received by each amplitude and phase receiver; From formula (1), we know that the optical delay value is τ k The error covariance matrix Q(τ k ) is constructed as follows: Q(t k )=F(τ)PF H (t)-|ψ k | 2 f(t k )f H (t k ) (2) where P = diag([|ψ1| 2 ,|ψ2| 2 ,…,|ψ K | 2 ]), diag(…) is a diagonal matrix, and P k =|ψ k | 2 ; According to the weighted least squares criterion, k The time domain response ψ k The cost function C(ψ k ): C(ψ k )=(if(τ k )ψ k ) H Q -1 (t k (if(t) k )ψ k ) (3).

5. The method for improving spatial resolution of time delay measurement based on incoherent optical frequency domain reflection according to claim 1 is characterized in that: The implementation process of step 4 is as follows: Note k n is the τ obtained in the nth iteration k The time domain response of k 0 is the initial solution; P n =diag([|ψ1 n | 2 ,|ψ2 n | 2 ,…,|ψ K n | 2 ]) is the square of the time domain response amplitude obtained in the nth iteration, P0 is the initial solution; Q n (τ k ) is the corresponding delay τ in the nth iteration k The resulting error covariance matrix, Q0(τ k ) is the initial solution, and the initial solution is obtained by the following formula: The adaptive iterative process is: Formula (6) takes n = 1, 2, ..., N, and the iteration termination condition is: ||P n -P n-1 ||2<P th (6) where ||…||2 is the 2-norm of the matrix, P th It is the preset iteration termination threshold condition.

6. The method for improving spatial resolution of time delay measurement based on incoherent optical frequency domain reflection according to claim 1 is characterized in that: The implementation process of step 5 is as follows: In order to find the delay of each channel, construct the time domain response function ψ(τ k )=ψ k N , the τ corresponding to each maximum point of the function k This is the optical delay to be measured.