CN1481084A - A method and device for measuring the performance of an optical transmission system - Google Patents

Abstract

The invention discloses a method and device for measuring the performance of an optical transmission system. The method includes the following steps: 1) selecting a reference level, using the reference level as a judgment level to recover a code stream, and obtaining a reference code stream; Compare the reference code stream with the recovered code stream under the changed decision level, so as to obtain the first "U"-shaped relationship curve between the bit error rate and the decision level, and calculate the "U"-shaped relationship curve The bit error rate at the lowest point of the lowest point; constantly change the reference level, get the bit error rate at the lowest point of multiple "U"-shaped relationship curves, and find the lowest bit error rate, from the lowest bit error rate Calculate the value of the Q factor. Since the reference level recovery data is used as the reference code stream, there is no need to prepare a standard signal as the reference code stream in advance, which greatly facilitates the test and solves the problem of on-site monitoring.

Description

A method and device for measuring the performance of an optical transmission system

Technical field:

The invention relates to a method and device for measuring the performance of an optical transmission system.

Background technique:

Bit Error Rate (BER) and Optical Signal-to-Noise Ratio (OSNR) are important parameters for monitoring optical fiber transmission systems. The bit error rate is not a parameter of the physical layer, and when the bit error rate is small, it takes a considerable time to measure the bit error. It is an alternative method to monitor the transmission characteristics of the system by measuring OSNR. This method is suitable for systems with large dispersion tolerance and noise-limited systems, such as 2.5Gbit/s systems. For higher rate transmission systems, OSNR can no longer describe the actual transmission performance well. A better way is to use Q factor to describe the transmission performance of the system.

The key to measuring the Q factor is to find the best decision level. The bit error rate BER corresponding to the best decision level has a simple relationship with the Q factor. $\mathrm{BER}\≈\frac{\exp ({-Q}^2/2)}{Q\sqrt{2\mathrm{\π}}},$ The Q factor can be obtained by detecting the bit error rate at the optimum level. But usually the bit error rate at the best decision level is small, and it is inconvenient to measure the bit error rate. Bergano proposed a simple method suitable for measuring the Q factor under the condition of Gaussian distributed noise. The principle of this method is given in Fig. 1, by continuously adjusting the amplitude of the decision level, a corresponding series of bit error rate values are obtained. If the decision level is set on the x-axis and the bit error rate is set on the y-axis, a "U"-shaped curve can be obtained. Bergano proposed a method to indirectly determine the minimum bit error rate point. The usual practice is to measure the bit error rate (generally between 10 ^-4 and 10 ^-10 ) where the decision level deviates from the optimal decision level. Through data fitting The method fits the entire "U"-shaped bit error rate versus decision level curve, so as to obtain the lowest bit error rate and the best decision level. In this way, the Q factor can be calculated.

But one problem with this measurement method is that a reference bit stream is needed to compare the actual bit error when measuring bit errors in the line. Generally, the reference bit stream in measurement is a standard signal, so this method is very inconvenient to use, especially not Suitable for on-site monitoring.

Invention content:

The purpose of the present invention is to solve the above problems and provide a method and device for measuring the performance of an optical transmission system, which does not require a reference code stream and is convenient for on-site use.

To achieve the above object, the present invention proposes a method and device for measuring the performance of an optical transmission system.

The method comprises the following steps: 1) Selecting a first reference level, using the reference level as the decision level to restore the code stream to obtain a reference code stream; Compare the code streams to obtain the first "U"-shaped relationship curve between the bit error rate and the decision level, and calculate the bit error rate at the lowest point of the "U"-shaped relationship curve; 2) Select the second Reference level, repeat the above-mentioned process, obtain the BER at the lowest point of another "U" font relationship curve; 3) constantly change the reference level, obtain the BER at the lowest point of multiple "U" font relationship curve Bit error rate, and find out the lowest bit error rate, the reference level at this time is the optimal reference level; 4) Use the above-mentioned optimal reference level as the best decision level, and calculate from the lowest bit error rate Get the value of the Q factor.

The device for measuring the performance of the optical transmission system includes two data recovery modules, which are respectively connected to the data to be measured; the decision level of one of the data recovery modules adopts a reference level, and the decision level of the other data recovery module adopts an optional It also includes two D flip-flops and an OR gate, the input terminals of the D flip-flop are respectively connected to two restored data streams, and the output terminals are connected to the OR gate for comparison; it also includes an error counter, and its input terminal It is connected with the output terminal of the OR gate to record the final different results; the data recovery module of the reference level also includes a clock recovery circuit, and the generated clock enters the above-mentioned D flip-flop and A clock counter, the D flip-flop synchronizes the compared code stream, the clock counter records the number of data generated, and it is input to the bit error rate (BER) calculation device together with the output of the error counter to calculate the bit error rate (BER), and Then use the fitting method to obtain the "U"-shaped relationship curve between the bit error rate (BER) and the decision level (Uth).

Due to the adoption of the above scheme, the reference level recovery data is used as the reference code stream to realize bit error testing. This reference code stream is recovered from the measured data stream, so there is no need to prepare a standard signal as a reference code in advance. The flow greatly facilitates the test and solves the problem of on-site monitoring.

Description of drawings:

Fig. 1 is a schematic diagram of the principle of simplified measurement of Q factor in the prior art.

Fig. 2 is a schematic diagram of the measurement principle of the present invention.

Fig. 3 is a schematic circuit diagram of a specific embodiment of the present invention.

Detailed ways:

The present invention will be described in further detail below through specific embodiments and in conjunction with the accompanying drawings. The method of this patent is shown in Fig. 2, and it is based on the fact that when the reference level is changed, the obtained bit error rate and decision level relationship curve is always U-shaped. Assume a reference level, such as point A in Figure 2. Take the code stream recovered at this reference level (that is, the code stream recovered with this reference level as the decision level) as the reference code stream, and compare it with the code stream recovered at the changed decision level, so as to obtain the bit error rate The "U"-shaped relationship curve between BER and decision level Uth, and calculate (by fitting method) related parameters, so as to calculate the lowest bit error rate, that is, the bit error rate at the lowest point of the curve. The method of estimation can use data fitting method, for example, the formula can be used: $\mathrm{BER}(V,V_R)=\frac{1}{2}\mathrm{erfc}\left(\frac{|u_1-V|}{{\mathrm{\σ}}_1}\right)*(1-\frac{1}{2}\mathrm{erfc}\left(\frac{|u_1-V_R|}{\mathrm{\σ}1}\right))$ $+\frac{1}{2}\mathrm{erfc}\left(\frac{|V-u_0|}{{\mathrm{\σ}}_0}\right)*(1-\frac{1}{2}\mathrm{erfc}\left(\frac{|V_R-u_0|}{\mathrm{\σ}0}\right))$

Where u ₁ and u ₀ are the electrical averages of "0" and "1". σ ₁ and σ ₀ are the variances of the "0" and "1" noise. V and _VR are decision level and reference level. This step is similar to the method of indirectly determining the lowest bit error rate point proposed by Bergano, except that the reference code stream is not a special standard code stream, but a code stream recovered from the measured data stream at the reference level. Obviously, the latter is easier to obtain, but because it is not a real standard code stream, the measured minimum bit error rate is not the real minimum, and the following steps are required for further measurement.

Considering another reference point C, repeat the process described in the previous step to get another lowest BER.

Change the reference level until the lowest bit error rate among all the lowest bit error rates is found (point B in the figure), or when the change of the calculated lowest bit error rate is less than 10 ^-15 , stop changing the reference level. The reference level at this time is the optimal reference level, and this level is also the optimal decision level. Accordingly, the relationship between the bit error rate BER and the Q factor can be considered as $\mathrm{BER}\≈\frac{\exp (-Q^2/2)}{Q\sqrt{2\mathrm{\π}}}$ Established, so that the final Q value can be calculated from the bit error rate at this time.

The selection principle of the reference level and the decision level is to make the measured bit error rate between 10 ^-4 and 10 ^-10 . At this time, the decision level deviates from the optimal decision level, and the bit error rate is higher than the minimum bit error rate. Results can be measured faster.

Figure 3 is a schematic diagram of the implementation circuit for measuring the Q factor. The circuit includes two data recovery modules, which are respectively connected to the data to be measured; the judgment level of one of the data recovery modules adopts a reference level, and the judgment level of the other data recovery module adopts an adjustable judgment level; it also includes Two D flip-flops and an OR gate, the input terminals of the D flip-flop are respectively connected to two recovered data streams, and the output terminals are connected to the OR gate for comparison; an error counter is also included, and the input terminal thereof is connected to the output terminal of the OR gate , used to record the final different results; the data recovery module of the reference level also includes a clock recovery circuit, and the generated clock enters the above-mentioned D flip-flop and a clock counter respectively after passing through the delay circuit, and the D flip-flop Synchronize the compared code stream, the clock counter records the number of generated data, and it is input to the bit error rate (BER) calculation device together with the output of the error counter - CPU to perform bit error rate (BER) calculation, and then use the fitting method A "U"-shaped relationship curve between the bit error rate (BER) and the decision level (Uth) is obtained. The CPU is also connected to a reference decision level generation circuit of one of the data recovery circuits to control the size of the reference level.

The measured data flow is divided into two paths and enters two data recovery modules respectively. The decision level of one data recovery module adopts a reference level (the size of the reference level is controlled by the CPU), and the decision level of the other data recovery module adopts an adjustable level. After the recovered data stream passes through two D flip-flops, it enters an OR gate for comparison. The last difference result is recorded by the error counter. The data recovery module of the reference level also generates a clock, which enters the D flip-flop and the clock counter respectively after a proper delay. Among them, the purpose of the D flip-flop is to synchronize the code streams compared. The clock counter records the number of generated data, combined with the error counter, the bit error rate can be calculated. The calculation of the bit error rate is completed by the CPU, and finally after processing by the CPU, the final result is output.

Claims (8)

1, a kind of method of measuring light transmission system performance is characterized in that comprising the steps:

1) choosing first reference level, is that decision level is recovered code stream with this reference level, obtains with reference to code stream; The code stream that recovers under this decision level with reference to code stream and variation is compared, thereby obtain " U " font relation curve of article one error rate (BER) and decision level (Uth), and extrapolate the error rate at the curve minimum point place of " U " font relation curve;

2) choose second reference level, repeat said process, obtain the error rate that another " U " font relation curve minimum point place (B) locates;

3) constantly change reference level, obtain the error rate at the minimum point place of a plurality of " U " font relation curve, and find out lowest bit error rate wherein, the reference level of this moment is optimum reference level;

4) with above-mentioned optimum reference level as optimum decision level, extrapolate the value of the Q factor from this lowest bit error rate.

2, the method for a kind of measuring light transmission system performance as claimed in claim 1 is characterized in that: the selection principle of reference level and decision level is to make the error rate that records 10 ^-4To 10 ^-10Between.

3, the method for a kind of measuring light transmission system performance as claimed in claim 1 or 2 is characterized in that: when the lowest bit error rate of calculating changes less than set point, just stop to change reference level.

4, the method for a kind of measuring light transmission system performance as claimed in claim 1 or 2 is characterized in that: the error rate of optimum decision level place correspondence and the relation of the Q factor are: $\mathrm{BER}\≈\frac{\exp (-Q^2/2)}{Q\sqrt{2\mathrm{\π}}}.$

5, the method for a kind of measuring light transmission system performance as claimed in claim 1 or 2 is characterized in that in step 1) to 3) in, the method that obtains " U " font relation curve is: measured data flow is divided into two-way, enters two data recovery module respectively; The decision level of a data recovery module adopts reference level, and the decision level of another data recovery module adopts adjustable decision level; The data recovered flow point is through behind two d type flip flops, enters one or compare; Last Different Results is noted down by error counter; Wherein the data recovery module of reference level also produces a clock, this clock is after the process time-delay, enter into d type flip flop and clock counter respectively, d type flip flop makes the code stream of comparison synchronous, the clock counter record produces the data number, calculate the error rate (BER) in conjunction with error counter, and and then utilize fitting process to obtain " U " font relation curve of the error rate (BER) and decision level (Uth).

6, the method for a kind of measuring light transmission system performance as claimed in claim 3, it is characterized in that: the calculating of the error rate is finished by CPU, and by the size that CPU controls reference level, after handling through CPU at last, exports last result.

7, a kind of device of measuring light transmission system performance is characterized in that: comprise two data recovery module, link to each other respectively with measured data; The decision level of one of them data recovery module adopts reference level, and the decision level of another data recovery module adopts adjustable decision level; Also comprise two d type flip flops and one or, described d type flip flop input connects two data recovered streams respectively, output termination or door compare; Also comprise error counter, its input with or the door output link to each other, be used to write down last Different Results; Wherein the data recovery module of reference level also comprises a clock recovery circuitry, the clock that is produced is behind the process delay circuit, enter into above-mentioned d type flip flop and a clock counter respectively, d type flip flop makes the code stream of comparison synchronous, the clock counter record produces the data number, the output of it and error counter together is input to be carried out the error rate (BER) and calculates on the error rate (BER) calculation element, and and then utilizes fitting process to obtain " U " font relation curve of the error rate (BER) and decision level (Uth).

8, the device of a kind of measuring light transmission system performance as claimed in claim 7, it is characterized in that: the described error rate (BER) calculation element is CPU (CPU), it also produces circuit with the reference decision level of one of them data recovery circuit and links to each other the size of control reference level.