The invention relates to the field of optical performance monitoring technology, and more particularly, to a method for enhancing optical signal-to-noise ratio measuring precision by correcting spectral resolution.
Optical Signal to Noise Ratio (OSNR) is a ratio of signal power in a channel to noise power in a specific bandwidth range. OSNR indicates the level of noise in the signal, and is one of important technical indicators for optical performance monitoring.
Generally, OSNR is measured by out-of-band monitoring method based on an optical spectrum analyzer (OSA), which is recommended by ITU-T G.697. This method is characterized in that OSNR is calculated by estimating noise in channels with interchannel noise, and is widely used since it is easy and does not affect service.
This OSNR measuring method based on spectrum analysis has the following disadvantage: due to the factors such as mechanical adjustment in OSA or external environment change and so on, its setting resolution may be different from its actual resolution, which results in low precision for measuring OSNR based on the spectrum method.
In view of the above-mentioned problems in the prior art, the present invention provides a method for enhancing optical signal-to-noise ratio measuring precision by correcting spectral resolution, which aims to replace the setting resolution of OSA with corrected resolution, so as to obtain OSNR, thereby solving the problem of large error resulted from the difference between setting resolution and actual resolution when OSNR is measured by using OSA.
In order to achieve the objective of the invention, a method for enhancing optical signal-to-noise ratio measuring precision by correcting spectral resolution is provided, including the following steps of:
(1) obtaining actual power PAi of a series of broad spectrum signals which have a central wavelength of λ, a bandwidth of Δλ1, a power interval of ΔPz, and a power range of (Pmin˜Pmax),
wherein, i=1, 2, . . . , n; z=1; Pmin is the minimal power of broad spectrum signals; Pmax is the maximum power of broad spectrum signals; the power range of (Pmin˜Pmax) of the broad spectrum signals could cover the dynamic range of power of signals under test;
(2) measuring a spectrum of each signal for the above series of broad spectrum signals by using OSA, adding up the power of sampling points which fall into the spectrum range having a central wavelength of λ and a bandwidth of Δλ1, to obtain a sum of power of the sampling points, thereby to obtain a series of sums PBi for the above series of broad spectrum signals; in which, i=1, 2, . . . , n,
wherein, a display central wavelength of the OSA is set to be λ, a display wavelength range of the OSA is set to be Δλ1, a wavelength interval between adjacent sampling points is Δλ2; the number of sampling points is n; the value of setting resolution is Resset; and n˜Resset≥Δλ1;
(3) obtaining corrected resolutions Resi=PBi·Δλ2/PAi, according to the actual power PAi of the series of broad spectrum signals, the sums PBi of power of the sampling points, and the wavelength interval Δλ2;
(4) fitting the corrected resolutions Resi and the sums PBi of power of the sampling points by the least squares method, to obtain a fitting expression of (Res−PB);
(5) substituting the above series of sums PBi of power of sampling points in the step (2) into the fitting expression of (Res−PB), to obtain a series of corrected resolutions RES′i after being fitted;
obtaining measured power of the series of broad spectrum signals P′Ai=PBi·Δλ2/RES′i, according to the series of corrected resolutions RES′i after being fitted, the sums PBi of power of the sampling points, and the wavelength interval Δλ2;
(6) according to the actual power PAi and the measured power P′Ai of broad spectrum signals, calculating relative errors
and determining whether the following formula
is satisfied,
if yes, it is indicated that the corrected resolution can replace the actual resolution;
performing step (7);
if no, it is indicated that there is a large error between the corrected resolution and the actual resolution; then, making ΔPΔ=ΔPz/2, z=z+1, and repeating steps (1)-(5),
wherein, 0<ε<1;
(7) measuring a spectrum of a signal under test under the same settings as in the step (2) by using OSA, to obtain corrected resolutions corresponding to sampling points required by calculating OSNR in the spectrum; replacing the setting resolutions with the corrected resolutions to obtain OSNR.
Preferably, for the above method for enhancing optical signal-to-noise ratio measuring precision by correcting spectral resolution, step (7) includes the following sub-steps of:
(7.1) measuring the spectrum of the signal under test which has a central wavelength of λ and a bandwidth of Δλ3, under the same settings as in the step (2) by using OSA;
substituting PBj=Paj·Δλ1/Δλ2 into the fitting expression of (Res−PB), to obtain the corrected resolutions Resaj corresponding to the sampling points required by calculating OSNR in the spectrum,
wherein, Paj are the power of the sampling points required by calculating OSNR in the spectrum, j=1, 2, . . . , m, and Δλ3≤Δλ1;
(7.2) obtaining a total power of the signal under test
wherein, Pbk are the power of sampling points which fall into the spectrum range having a central wavelength of λ and a bandwidth of Δλ3, k=1, 2, . . . , l; Resbk are the corrected resolutions corresponding to the power of sampling points Pbk;
(7.3) obtaining a total power of noise
wherein, ƒ is an undetermined function, and its specific expression is determined according to a used OSNR measurement method; Pct and Pdp are the power of sampling points in the spectrum range required by calculating noise, t=1, 2, . . . . , s, p=1, 2, . . . . , q; Resct and Resdp are corrected resolutions corresponding to the power Pct and Pdp of sampling points, respectively;
(7.4) obtaining optical signal to noise ratio
wherein, λr is a reference bandwidth, and is set to be 0.1 nm.
In general, comparing to the prior art, the above technical solution in the present invention can achieve the following advantageous effects:
(1) the method for enhancing optical signal-to-noise ratio measuring precision by correcting spectral resolution provided in the invention can obtain the signal power and the noise power in the signals under test more accurately, so as to enhance the OSNR measuring precision, by measuring actual power of the broad spectrum signals in a certain bandwidth, determining the sums of the power of the sampling points in the certain bandwidth for the broad spectrum signals by using an optical spectrum analyzer, obtaining the corrected resolution of optical spectrum analyzer, and replacing the setting resolution with the corrected resolution;
(2) the method is applicable to correct resolution for all optical spectrum analyzers, and also applicable to enhance measuring precision for all OSNR measuring methods based on spectrum analysis, and has the advantages of easiness to handle and implement.
For clear understanding of the objectives, features and advantages of the invention, detailed description of the invention will be given below in conjunction with accompanying drawings and specific embodiments. It should be noted that the embodiments are only meant to explain the invention, and not to limit the scope of the invention.
The method for enhancing optical signal-to-noise ratio measuring precision by correcting spectral resolution provided in the invention aims to enhance OSNR measuring precision by correcting resolution and reducing the difference between the setting resolution and the actual resolution.
Embodiment 1 provides a method for enhancing the OSNR measuring precision by correcting the resolution of volume grating OSA, including the following steps of:
(1) obtaining actual power PA of a broad spectrum signal which has a central wavelength of λ and a bandwidth of Δλ1 by using an optical power meter,
wherein, PA makes the power of broad spectrum signal to be in the dynamic range of power of signal under test; the broad spectrum signal is generated from an erbium doped fiber amplifier (EDFA), the central wavelength λ and bandwidth Δλ1 of the broad spectrum signal can be obtained by adjusting a tunable filter, and its output power can be controlled by adjusting an optical attenuator;
(2) measuring a spectrum of the above broad spectrum signal by using the volume grating OSA, adding up the power of sampling points which fall into the spectrum range having a central wavelength of λ and a bandwidth of Δλ1, to obtain a sum PB of power of the above sampling points,
wherein, a display central wavelength of the volume grating OSA is set to be λ, a display wavelength range is set to be Δλ1, a wavelength interval between adjacent sampling points is Δλ2; the number of sampling points is n; the value of setting resolution is Resset; and n·Resset≥Δλ1;
(3) obtaining a corrected resolution Res=PB·Δλ2/PA, according to the actual power PA of the broad spectrum signal, the sum PB of power of the sampling points, and the wavelength interval Δλ2;
(4) measuring a spectrum of a signal under test under the same settings as in the step (2) by using the volume grating OSA, replacing a setting resolution with a corrected resolution to obtain OSNR, which specifically includes the following sub-steps of:
(4.1) measuring the spectrum of the signal under test which has a central wavelength of λ and a bandwidth of Δλ3, under the same settings as in the step (2) by using the volume grating OSA, wherein Δλ3≤Δλ1;
(4.2) obtaining a total power of the signal under test
wherein, Pk are the power of sampling points which fall into the spectrum range having a central wavelength of λ and a bandwidth of Δλ3, k=1, 2, . . . . , l;
(4.3) obtaining a total power of noise: in the case that OSNR is measured by an out-of-band monitoring method, the total power of noise
wherein, P(λ−Δλ) and P(λ+Δλ) are the power of sampling points at wavelengths of λ−Δλ and λ+Δλ, respectively, wherein, λ indicates central wavelength;
(4.4) obtaining optical signal to noise ratio
wherein, λr is a reference bandwidth, and is set to be 0.1 nm in embodiment 1.
In the present invention, the signal power and the noise power in the signal under test can be obtained more accurately, so as to enhance the OSNR measuring precision, by obtaining a corrected resolution of volume grating OSA, and replacing a setting resolution with a corrected resolution.
Embodiment 2 provides a method for enhancing the OSNR measuring precision by correcting the resolution of OSA based on SBS effect, including the following steps of:
(1) obtaining actual power PAi of a series of broad spectrum signals which have a central wavelength of λ, a bandwidth of Δλ1, a power interval of ΔPz, and a power range of (Pmin˜Pmax) by using an optical power meter,
wherein, i=1, 2, . . . , n; z=1; Pmin is the minimal power of broad spectrum signals; Pmax is the maximum power of broad spectrum signals; the power range of (Pmin˜Pmax) of the broad spectrum signals could cover the dynamic range of power of signals under test; the broad spectrum signals are generated from EDFA, the central wavelength λ and bandwidth Δλ1 of the broad spectrum signals can be obtained by adjusting a tunable filter; and output power can be controlled by adjusting an optical attenuator;
(2) measuring a spectrum of each signal for the series of broad spectrum signals by using OSA based on SBS effect, adding up the power of sampling points which fall into the spectrum range having a central wavelength of λ and a bandwidth of Δλ1, to obtain a sum of power of the sampling points, thereby to obtain a series of sums PBi for the above series of broad spectrum signals, in which, i=1, 2, . . . , n,
wherein, a display central wavelength of OSA based on SBS effect is set to be λ, a display wavelength range is set to be Δλ1, a wavelength interval between adjacent sampling points is Δλ2; the number of sampling points is n; the value of setting resolution is Resset; and n·Resset≥Δλ1;
(3) obtaining corrected resolutions Resi=PBi·Δλ2/PAi, according to the actual power PAi of the series of broad spectrum signals, the sums PBi of power of the sampling points, and the wavelength interval Δλ2;
(4) fitting corrected resolutions Resi and the sums PBi of power of the sampling points by the least squares method, to obtain a fitting expression of (Res−PB);
(5) substituting the above series of sums PBi of power of sampling points in the step (2) into the fitting expression of (Res−PB), to obtain a series of corrected resolutions RES′i after being fitted;
obtaining the measured power of the series of broad spectrum signals P′Ai=PBi·ΔλZ/RES′i, according to the series of corrected resolutions RES′i after being fitted, the sums PBi of power of the sampling points, and the wavelength interval Δλ2;
(6) according to the actual power PAi and the measured power P′Ai of broad spectrum signals, calculating relative errors
and determining whether the following formula
is satisfied,
if yes, it is indicated that the corrected resolution can replace the actual resolution; performing step (7);
if no, it is indicated that there is a large error between the corrected resolution and the actual resolution; then, reducing ΔPz to make ΔPz=ΔPz/2, z=z+1, repeating steps (1)-(5),
wherein, 0<ε<1;
(7) measuring a spectrum of a signal under test under the same settings as in the step (2) by using OSA based on SBS effect, to obtain corrected resolutions corresponding to the sampling points required by calculating OSNR in the spectrum, replacing the setting resolutions with the corrected resolutions to obtain OSNR, which specifically includes the following sub-steps of:
(7.1) measuring the spectrum of the signal under test which has a central wavelength of λ and a bandwidth of Δλ3, under the same settings as in the step (2) by using OSA based on SBS effect;
substituting PBj=Paj·Δλ1/Δλ2 into the fitting expression of (Res−PB), to obtain the corrected resolutions Resaj corresponding to the sampling points required by calculating OSNR in the spectrum,
wherein, Paj are power of the sampling points required by calculating OSNR in the spectrum, j=1, 2, . . . , m; Δλ3≤Δλ1;
(7.2) obtaining a total power of the signal under test
wherein, Pbk are power of sampling points which fall into the spectrum range having a central wavelength of λ and a bandwidth of Δλ3, k=1, 2, . . . , l; Resbk are the corrected resolutions corresponding to the power Pbk of sampling points;
(7.3) obtaining a total power of noise: in the case that OSNR is measured by an out-of-band monitoring method, the total power of noise
wherein, P(λ−Δλ) and P(λ+Δλ) are the power of sampling points at wavelengths of λ−Δλ and λ+Δλ, respectively, λ indicates central wavelength; Res(λ−Δλ) and Res(λ+Δλ) are corrected resolutions corresponding to the power P(λ−Δλ) and P(λ+Δλ) of sampling points, respectively;
(7.4) obtaining optical signal to noise ratio
wherein, λr is a reference bandwidth, and is set to be 0.1 nm.
It should be appreciated that, for correcting resolution of other kinds of OSA, and for enhancing measuring precision of other OSNR measuring methods based on spectrum analysis, all these methods are included in inventive concept of the present invention.
While preferred embodiments of the invention have been described above, the invention is not limited to disclosure in these embodiments and the accompanying drawings. Any changes or modifications without departing from the spirit of the invention fall within the scope of the invention.
Number | Date | Country | Kind |
---|---|---|---|
2016 1 0815635 | Sep 2016 | CN | national |
Filing Document | Filing Date | Country | Kind |
---|---|---|---|
PCT/CN2016/099413 | 9/20/2016 | WO | 00 |
Publishing Document | Publishing Date | Country | Kind |
---|---|---|---|
WO2018/045603 | 3/15/2018 | WO | A |
Number | Name | Date | Kind |
---|---|---|---|
5986782 | Alexander | Nov 1999 | A |
6493088 | Hui | Dec 2002 | B1 |
7340127 | Chang | Mar 2008 | B1 |
9621263 | Wei | Apr 2017 | B2 |
9876566 | Fludger | Jan 2018 | B1 |
20020044322 | Blumenthal | Apr 2002 | A1 |
20030030859 | Youn | Feb 2003 | A1 |
20040212428 | Ode | Oct 2004 | A1 |
20060153498 | Shen | Jul 2006 | A1 |
20080124076 | Rudolph et al. | May 2008 | A1 |
20090257534 | Wu | Oct 2009 | A1 |
20130336647 | Le Taillandier De Gabory | Dec 2013 | A1 |
20150010299 | Wei | Jan 2015 | A1 |
20160056891 | Chen | Feb 2016 | A1 |
20160308611 | Oda | Oct 2016 | A1 |
Number | Date | Country |
---|---|---|
101119174 | Feb 2008 | CN |
103968943 | Aug 2014 | CN |
104052544 | Sep 2014 | CN |
104079347 | Oct 2014 | CN |
1376900 | Jan 2004 | EP |
Entry |
---|
Wang et al; Enhancing the signal -to-noise ratio of FTIR spectrometers by a digital J-Stop; Aug. 2017; Optical Society of America. |
Number | Date | Country | |
---|---|---|---|
20180183515 A1 | Jun 2018 | US |