WO2004073244A1

WO2004073244A1 - Bit error rate monitoring method and device

Info

Publication number: WO2004073244A1
Application number: PCT/AU2004/000184
Authority: WO
Inventors: Elaine Wong; Thas A. Nirmalathas; Christina Min Ee Lim
Original assignee: The University Of Melbourne
Priority date: 2003-02-17
Filing date: 2004-02-17
Publication date: 2004-08-26
Also published as: AU2003900689A0; CA2519392A1; EP1602189A1

Abstract

The present application discloses a method of bit error rate (BER) monitoring a signal and an apparatus for implementing the method. The method includes the steps of asynchronously sampling the amplitude of the signal to be BER monitored to obtain a plurality of sampled values, and determining at least one of, a proportion of the sample values falling above an upper amplitude threshold, and a proportion of the sample values falling below a lower amplitude threshold, to derive a BER for the monitored signal.

Description

BIG ERROR RATE MONITORING METHOD AND DEVICE

Field of the invention

The present invention relates broadly to a method of bit error rate (BER) monitoring of a digital signal, and to a device for BER monitoring of a digital signal. Background of the invention

Example embodiments of present invention will be described herein in the context of BER monitoring an optical signal, however the present invention should not be construed as being limited only to use with optical signals. It should be understood that embodiments of the present invention will be applicable to BER monitoring of digital signals transmitted over any type of transmission channel including, but not limited to, electrical, microwave and radio transmission channels.

An optical network relies on specific performance monitoring techniques to assess the quality of the offered services and the integrity of signals traversing in the network. A method of monitoring signal integrity is to evaluate the BER of the transmitted signal, i.e. the monitoring of the quality of an optical link at the bit level. However, the ability to accurately and quickly evaluate the BER is a major task with traditional BER measurement techniques being off-line, and often requires the use of an expensive BER tester at every monitoring location.

A feasible solution for fast assessment of the BER is to exploit the statistics of the transmitted signal. In intensity modulated direct detected (IM/DD) systems where Gaussian noise (e.g. relative intensity noise from source lasers, spontaneous emission noise from EDFAs, and thermal noise at the photodiodes) dominates, the most widely used method for fast BER estimation is the Q-factor method which uses Gaussian statistics of the probability density function (PDF) of the transmitted marks and spaces. By obtaining the synchronously sampled amplitude histogram 10 of the transmitted signal as shown in Fig. 1, the Q-factor and hence BER can be estimated from the measured means and standard deviation of the marks and spaces. The Q-factor and BER are calculated using equations (1) and (2) respectively. μ^ -μ δ (i) σ- + σ₀

The parameters μ_\ and μ₀ are the mean values 12, 14, and σi and σ₀ are the standard deviations of the histogram at mark and space levels. To carry out this measurement, data signals need to be detected with an appropriate clock recovery scheme so that a precise sampling system can be used to obtain the amplitude histogram and thus Q and BΕR measurements. This method of measurement is usually referred to as synchronous sampled histogram measurement.

Recently, methods to obtain asynchronously sampled histograms have been proposed, as they do not require the clock recovery of the transmitted signals. The signal is sampled randomly at a rate often lower than the transmission bit rate throughout the bit stream, thus offering the possibility of bit-rate and data format transparency. Fig. 2 depicts the asynchronously sampling histograms 20 of the mark and space levels in which samples are obtained at random times irrespective to the optimum decision time. As can be observed, a significant amount of extraneous and unwanted samples are collected between the mean values

22, 24 of the mark and space levels. The asynchronous Q factor is estimated by post processing these samples, specifically by fitting two Gaussian distributions to the histograms of the mark and space levels. As such, the respective mean and standard deviation values can therefore be evaluated. Often times, the asynchronous Q factor is related back to the actual Q factor which is derived from the measured BΕR, using a bit-rate independent proportionality.

The synchronous and asynchronous sampling histogram methods for Q-factor evaluation described above entail a substantial number of samples to be acquired in order to accurately form signal distributions of both the mark and space levels having the requisite Gaussian characteristics. Hence, the time required to detect degradation of the transmitted signal from the onset of a signal perturbation may be large.

In at least preferred embodiments, the present invention seeks to provide a BΕR monitoring method and device enabling a reduced degradation detection time, whilst maintaining the benefit of asynchronous sampling, i.e. bit rate independence. Summary of the invention

In a first aspect the present invention provides a method of bit error rate (BER) monitoring a digital signal. The method includes the steps of:

- asynchronously sampling the amplitude of a signal to be BER monitored to obtain a plurality of sampled values;

- determining at least one of, a proportion of the sample values falling above an upper amplitude threshold, and a proportion of the sample values falling below a lower amplitude threshold, to derive a BER for the monitored signal.

The method can further include determining from the amplitude of the signal to be BER monitored at least one of the upper amplitude threshold and the lower amplitude threshold.

The upper amplitude threshold can be determined by, determining a cross-over amplitude defining a transition between a sample value being detected as an upper binary value and a lower binary value; determining a mean amplitude value for the signal to be BER monitored when representing the upper binary value; and using the determined mean amplitude and the cross-over amplitude to arrive at the upper amplitude threshold.

Preferably the upper amplitude threshold is set at a level substantially twice the mean amplitude value for the sampled values representing the upper binary value above the cross-over amplitude.

The lower amplitude threshold can be determined by, determining a cross-over amplitude defining a transition between a sample value being detected as an upper binary value and a lower binary value; determining a mean amplitude value for the signal to be BER monitored when representing the lower binary value; and using the determined mean amplitude and the cross-over amplitude to arrive at the lower amplitude threshold.

Preferably the lower amplitude threshold is set at a level substantially twice, the mean amplitude value for the sampled values representing the lower binary value, below the crossover amplitude.

Preferably the signal to be BER monitored is an optical signal. It is also preferable that the amplitude measurement of the signal is performed in the electrical domain. The signal to be BER monitored is preferably converted to an electrical signal with the amplitude of the optical signal being represented by a voltage level of the electrical signal.

Preferably the method includes, DC-blocking the converted electrical signal prior to measuring the amplitude of the signal to set the cross-over amplitude of the converted signal at zero volts.

The method can further includes splitting the signal to be BER monitored into a plurality of signal components, and conducting the amplitude measurement on a first signal component.

A second signal component can be used for determining the proportion of sampled values that have an amplitude above the upper amplitude threshold. A third signal component can be used for determining the proportion of sampled values that have an amplitude below the lower amplitude threshold.

Preferably the signal to be BER monitored is sampled at a rate lower than a minimum data rate of the signal.

In a second aspect the present invention provides a bit error rate (BER) monitoring device, including a threshold generating stage configured to generate at least one of an upper amplitude threshold and a lower amplitude threshold on the basis of the amplitude of the signal to be BER monitored; a sample collection stage configured to asynchronously sample the signal to be BER monitored to generate a plurality of sample values; a comparator stage configured to compare the sampled values to at least one of the upper amplitude threshold and the lower amplitude threshold; and a control stage configured to determine at least one of, a proportion of the sample values falling above the upper amplitude threshold, and a proportion of the sample values falling below the lower amplitude threshold and to derive a BER for the signal being monitored.

Preferably the threshold generating stage includes amplitude measuring means configured to measure the amplitude of the signal to be BER monitored and to determine at least one of, a mean amplitude for the signal to be BER monitored when representing the upper binary value, and a mean amplitude for the signal to be BER monitored when representing the lower binary value.

The threshold generating stage can be further configured to determine the upper amplitude threshold on the basis of the determined mean amplitude for the signal to be BER monitored when representing the upper binary value and a cross-over amplitude defining a transition between a sample value being detected as an upper binary value and a lower binary value.

The threshold generating stage can be further configured to determine the lower amplitude threshold on the basis of the determined mean amplitude for the signal to be BER monitored when representing the lower binary value and a cross-over amplitude defining a transition between a sample value being detected as an upper binary value and a lower binary value.

Preferably the signal to be BER monitored is an optical signal. The signal to be BER monitored is processed in the electrical domain as an electrical signal having a voltage level representative of the amplitude of the signal to be BER monitored .

The monitor can further include a pre-processing stage configured to convert an optical signal to be BER monitored into an electrical signal. Preferably the pre-processing stage incudes a DC blocking circuit configured to block a D.C. component of the electrical signal. The BER monitoring device can further include a plurality of signal paths wherein at least one signal path bypasses the threshold generating stage. Preferably the device includes a first signal path for communicating a first signal component to the threshold generating stage for use in determining at least one of the upper amplitude threshold and the lower amplitude threshold; and at least one second signal path for communicating a respective second signal component to the sample collection stage.

The comparator stage can includes a first comparator configured to compare the plurality of sample values to the upper amplitude threshold and a second comparator configured to compare the plurality of sample value to the lower amplitude threshold.

The device can include at least one counter configured to count the number of samples with amplitudes above the upper amplitude threshold. The device may additionally, or alternatively, include at least one counter configured to count the number of samples with amplitudes below the lower amplitude threshold.

The control stage can be further configured to send a control signal if the BER of a signal being monitored is above a predetermined BER threshold. Preferably the sample rate of the sample collection stage is less than a minimum data rate of the signal being BER monitored.

The signal to be BER monitored is can be digital signal transmitted via any one of the following types of transmission signal: a microwave signal, a radio signal, an electrical signal. In a third aspect the present invention provides a method of bit error rate (BER) monitoring an optical signal including the steps of:

- converting the optical signal into an electrical signal having a voltage corresponding to an amplitude of the optical signal;

- deriving from the electrical signal an upper voltage threshold and a lower voltage threshold;

- asynchronously sampling the voltage of the electrical signal to obtain a plurality of sampled values; and

- determining a proportion of the sampled values falling above the upper voltage threshold and a proportion of the sample values falling below the lower voltage threshold, to derive a BER for the monitored signal.

In a fourth aspect the present invention provides a device for bit error rate (BER) monitoring of an optical signal, including; a signal reception stage configured to receive an optical signal and convert it into an electrical signal having a voltage representative of the amplitude of the optical signal; a threshold generating stage configured to generate an upper amplitude threshold and a lower amplitude threshold on the basis of the voltage of the electrical signal; a sample collection stage configured to asynchronously sample the signal to generate a plurality of sampled values; a comparator stage configured to compare the sampled values to the upper amplitude threshold and the lower amplitude threshold; and a control stage configured to determine a proportion of the sample values falling above the upper amplitude threshold and a proportion of the sample values falling below the lower amplitude threshold, and to derive a BER for the signal being monitored. Brief description of the drawings

Preferred embodiments of the present invention will now be described, by way of example only, with reference to the accompanying drawings.

Figure 1 is a diagram illustrating an estimation of BER using prior art asynchronous amplitude histograms.

Figure 2 is a diagram illustrating an estimation of BER using prior art asynchronous amplitude histograms.

Figure 3 is a diagram illustrating estimation of BER utilising asynchronous amplitude histograms embodying principles underlying the method of the present invention. Figure 4 is a schematic diagram illustrating an asynchronous BER monitoring circuit embodying the present invention.

Figures 5 a and 5b are schematic diagrams illustrating an experimental setup for testing a BER monitoring circuit embodying the present invention.

Figure 6 shows a graph illustrating a comparison of directly measured BER and measure of BER obtained utilising a monitoring circuit embodying the present invention and simulations thereof.

Detailed description of the embodiments

In order to reduce the degradation detection time whilst benefiting from the asynchronous sampling method, the preferred embodiment described provides a novel method of calculating BER by correlating the probability of samples falling outside a predefined window with the actual BER measurement.

Referring now to Fig. 3, an asynchronously sampled amplitude histogram 30 of a DC- blocked signal is shown, whereby the optimum cross-over threshold 32 is situated at zero. The area under the "one" bit component 34 of the histogram 30 from minus infinity to zero (indicated as region 33 on the left of the cross-over threshold 32) is therefore the error probability of transmitting a "one" bit but detecting it as a "zero" bit. However, as described above with reference to Figure 2, it is not possible to count samples falling in this region in the asynchronously sampled amplitude histogram 30. However, it has been recognised by the applicant that due to the symmetrical property of the sample distribution 34, there is an area that is equivalent to region 33 as indicated by region 35 on the right of the one bit component 34 which is "accessible" and therefore which can be used to determine the probability of error. As will be recognised same principle applies to the determining probability of transmitting a "zero" bit but it being received as a "one" bit. That is, the area as indicated by region 37 on the left of the "zero" bit component 39 is equivalent to the area under the "one" bit component 39 of the histogram 30 from plus infinity to zero, indicated as region 31 on the right of the zero cut-off threshold 32.

Thus it has been found by the present inventors that by determining either one or both of, the proportion of the sample values falling above an upper amplitude threshold and/or the proportion of the sample values falling below a lower amplitude threshold, a BER rate for the monitored signal can be derived.

It will be noted the current example of BER monitoring an optical signal, has noise sources that cause a Gaussian distribution of sample amplitudes, however it should be appreciated that certain embodiments of the present invention could advantageously be applied to the BER monitoring of signals having an amplitude distribution that is symmetrical but not

Gaussian in shape.

In the example embodiment a window P_w is defined by defining the upper and lower voltage window threshold values 36, 38. hi the present example, the noise affecting the "ones" is the same as the noise affecting the "zeros" and therefore the window defined by the upper and lower thresholds is symmetrical around the zero voltage threshold 32. In certain embodiments the means μi and μ can lie at different distances from the zero voltage threshold and thus the window defined by the upper and lower thresholds will not be symmetrical. However in both cases the window will have a width of twice the distance between the means μi and μ₂. As described above, a measure of BER is obtained through a specified proportionality between the probability of samples falling outside this window, P_w, and the BER.

Fig. 4 illustrates a schematic diagram of an asynchronous BER monitoring circuit 40 embodying the present invention. A fraction of the optical power of a transmitted signal is tapped off using an optical coupler (not shown). The asynchronous BER monitoring circuit 40 comprises four main stages: a transimpedance front-end 42; a threshold generating stage 44; a combined asynchronous sampling and comparator stage 46; and a controller stage 48. The front-end 42, which comprises a PIN photodiode 50 of bandwidth similar to the transmitted signal, transimpedance amplifier 52 and a DC block 54, detects and amplifies the incoming optical signal and converts it into an electrical signal which has no DC offset.

The resulting electrical signal is then split into three paths, the lowest path 56 is sent to the threshold generating stage 44 and the upper and middle paths 58, 60 bypass this stage and are sent to the asynchronous sampling and comparator stage 46. At the threshold generating stage 44, the amplitude of the DC blocked signal is measured, and the upper and lower thresholds of the window are calculated with programmable logic device 59, which is part of a processing unit 45 of the circuit 40. For example the programmable logic device can be a Complex programmable logic device (CPLD) or preferably a Field Programmable Gate Array (FPGA) which can be externally programmed to evaluate the amplitude and required thresholds. The upper threshold is calculated on the basis of the mean value of the signal amplitude when representing a "one" value and the zero cross-over value (in this case ON as the signal is D.C. blocked), and the lower threshold is calculated on the basis of the mean value of the signal amplitude when representing a "zero" value and the zero cross-over value. In the present example the upper threshold is set at twice the mean "one" amplitude above the zero cross-over, and the lower threshold is set at twice the mean "zero" value below the zero cross-over.

The signals on the upper and middle paths 58, 60 are fed into upper and lower comparators 62, 64 of the asynchronous sampling stage 46 respectively. The decision thresholds of these comparators are the calculated upper and lower thresholds from the threshold generating stage 44. The signals are asynchronously clocked at a rate determined by an independent pulse source 66, with a rate lower than the transmission bit rate.

A controller 61 which is also clocked by the independent pulse source 66 maintains counters of the number of bits falling outside the window, i.e. above the upper threshold or below the lower threshold, B, and the number of pulses, P. The controller 61 is also part of the processing unit 45 of the circuit 40.

The probability of bits falling outside the window, P_w, is calculated by dividing B by P.

In use if P_w increases above a specified level, e.g. corresponding to BER = 10^"9, a digital output signal is transmitted by the control stage for use in signal control and restoration purposes, hi order to minimise the time needed to detect signal degradation, the counters may be reset periodically. Figure 5 shows an experimental setup for testing the direct correspondence of Pw and BER in an intensity modulated direct detection fibre submission experiment. A continuous wave (CW) laser 102 (1550nm) is externally modulated with a 2²³-l pseudo random bit sequence, at a 1 Gb/s. The optical output of the Mach Zehnder modulator circuit 104 is combined with amplified spontaneous emission (ASE) noise from an erbium doped fibre amplifier (EDFA) 106, which is used as a Gaussian noise source. The combined noisy optical signal is then launched into 10km of single mode fibre (SMF) 108.

An attenuator 109 is provided after the 10km SMF 108 and prior to a PIN photodetector 110. The PIN photo detector 110 with a 1.2 GHz bandwidth directly detects the optical signal. The resulting electrical signal is then fed into a conventional bit error rate testset (BERT) 112 to obtain the true BER measurement of the transmitted signal. A DC block 111 is utilised to DC- block the converted electrical signal prior to the BER (or P ₅ see below) measurements.

During the experiment, the noise from the EDFA 106 broadens the histogram characteristics of the modulated signal travelling in the SMF 108. At the optical to electrical conversion point, the ratio between the modulated, broadened optical signal and an intrinsic noise of the PIN photodetector 110 can be controlled through the attenuator 109, which in turn results in variations in the BER as a result of the varying signal-to-noise ratio.

To test the correspondence of Pw and BER, the BERT 112 (Figure 5 a) is replaced by a detection circuit embodying the present invention, implemented utilising a communications signal analyser (CSA) 114, in conjunction with a 125 Mb/s clock source 116, as shown in Figure 5b. In this example implementation, the upper and lower threshold values are "manually" determined utilising the CSA, and relevant comparator settings are manually adjusted for counting the number of samples falling outside of the voltage window, the signal being asynchronously sampled utilising the 125 Mb/s clock 116.

Figure 6 shows the directly measured BER (Figure 5a) as a function of power applied to the Gaussian noise source 106 for back-to-back transmission and with fibre measured in experiments and simulations (curve 120). For comparison, the P_w measurements obtained utilising the detection circuit 114 (Figure 5b) embodying the present invention are also shown in Figure 6 (curve 122). As can be seen, it is clear that the P_w generally track the BER thus confirming that the BER can be effectively evaluated for optical signal performance by correlating Pw to the actual BER measurement.

Returning now to Figure 5b, in the example embodiment which utilises a photodiode 110 for conversion of the optical into an electrical system, it is expected that the bandwidth of the photodiode typically needs to be at a minimum of the order of a maximum incoming bit-rate of the optical data signal being tested, in order to recover the complete behaviour at the output of the first-stage optical electrical conversion. However, it is noted that further investigation may well point to a possibility of relaxing the bandwidth requirement in such photodiode embodiments. It will be appreciated by the person skilled in the art that numerous modifications and/or variations may be made to the present invention as shown in the specific embodiments without departing from the spirit or scope of the invention as broadly described. The present embodiments are, therefore, to be considered in all respects to be illustrative and not restrictive.

For example, it will be appreciated that the present invention can be implemented by counting only sample values falling below a lower threshold value, or alternatively only sample values that fall above an upper threshold, as opposed to counting both occurrences as described in the preferred embodiment. For those implementations, it will be appreciated that the correlation between the probability of sample values falling either below the lower threshold or above the upper threshold and the BER remains valid. Furthermore, it will be appreciated that monitoring time for signal degradation is dependent on the asynchronous sampling rate. Intuitively, there is a direct relationship between the sampling rate and the time needed to acquire a specified number of samples outside the window. For example, the higher the sampling rate, the faster it is from the onset of a perturbation to achieve a certain probability which indicates signal degradation. On the other hand, it may be desirable to not having to sample at high rates but rather benefit from the intuitively less critical and lower-cost under-sampling implementations. Accordingly, the present invention is not limited to a specific sampling rate, but rather, that rate can be chosen to suit requirements such as availability of pulse source, costs, and signal degradation detection time requirements. hi the claims that follow and in the summary of the invention, except where the context requires otherwise due to express language or necessary implication the word "comprising" is used in the sense of "including", i.e. the features specified may be associated with further features in various embodiments of the invention.

Claims

1. A method of bit error rate (BER) monitoring a signal including the steps of:

- asynchronously sampling the amplitude of the signal to be BER monitored to obtain a plurality of sampled values; - determining at least one of, a proportion of the sample values falling above an upper amplitude threshold, and a proportion of the sample values falling below a lower amplitude threshold, to derive a BER for the monitored signal.

2. The method of claim 1 including, deriving from the amplitude of the signal to be BER monitored at least one of the upper amplitude threshold and the lower amplitude threshold.

3. A method as claimed in either one of claims 1 or 2 wherein the upper amplitude threshold is derived by: determining a cross-over amplitude defining a transition between a sample value being detected as an upper binary value and a lower binary value; determining a mean amplitude value of the signal to be BER monitored when representing the upper binary value; and using the determined mean amplitude value and the cross-over amplitude to arrive at the upper amplitude threshold.

4. A method as claimed in claim 3 wherein the upper amplitude threshold is set at a level substantially twice the mean amplitude value for the sampled values representing the upper binary value above the cross-over amplitude.

5. A method as claimed in any one of claims 1 to 4 wherein the lower amplitude threshold is determined by: determining a cross-over amplitude defining a transition between a sample value being detected as an upper binary value and a lower binary value; determining a mean amplitude value for the signal to be BER monitored when representing the lower binary value; and using the determined mean amplitude value and the cross-over amplitude to arrive at the lower amplitude threshold.

6. A method as claimed in claim 5 wherein the lower amplitude threshold is set at a level substantially twice the mean amplitude value for the sampled values representing the lower binary value, below the cross-over amplitude.

7. A method as claimed in any one of the preceding claims wherein the signal to be BER monitored is an optical signal.

8. A method as claimed in any one of the preceding claims wherein the amplitude measurement is performed in the electrical domain.

9 A method as claimed in claim 8 wherein the signal to be BER monitored is converted to an electrical signal with the amplitude of the optical signal being represented by a voltage level of the electrical signal.

10. A method as claimed in claim 9, further including, DC-blocking the converted electrical signal prior to measuring the amplitude of the signal to set the cross-over amplitude of the converted signal at zero volts.

11. A method as claimed in any one of the preceding claims, wherein the method includes splitting the signal to be BER monitored into a plurality of signal components, and conducting the amplitude measurement on a first signal component.

12. A method as claimed in claim 11 that further includes, using a second signal component for determining the proportion of sampled values that have an amplitude above the upper amplitude threshold.

13. A method as claimed in either of claims 11 or 12 that further includes using a third signal component for determining the proportion of sampled values that have an amplitude below the lower amplitude threshold.

14. A method as claimed in any one of the preceding claims, wherein the signal to be BER monitored is sampled at a rate lower than a minimum data rate of the signal.

15. A bit error rate (BER) monitoring device, including a threshold generating stage configured to generate at least one of an upper amplitude threshold and a lower amplitude threshold on the basis of the amplitude of the signal to be BER monitored; a sample collection stage configured to asynchronously sample the signal to be BER monitored to generate a plurality of sampled values; a comparator stage configured to compare the sampled values to at least one of the upper amplitude threshold and the lower amplitude threshold; and a control stage configured to determine at least one of, a proportion of the sample values falling above the upper amplitude threshold, and a proportion of the sample values falling below the lower amplitude threshold and to derive a BER for the signal being monitored.

16. A bit error rate monitor as claimed in claim 15 wherein threshold generating stage includes amplitude measuring means configured to measure the amplitude of the signal to be BER monitored and to determine at least one of, a mean amplitude for the signal to be BER monitored when representing the upper binary value, and a mean amplitude for the signal to be BER monitored when representing the lower binary value.

17. A bit error rate monitor as claimed in either of claims 15 and 16 wherein the threshold generating stage is further configured to determine the upper amplitude threshold on the basis of the determined mean amplitude for the signal to be BER monitored when representing the upper binary value and a cross-over amplitude defining a fransition between a sample value being detected as an upper binary value and a lower binary value.

18. A bit error rate monitor as claimed in any one of claims 15 to 17 wherein the threshold generating stage is further configured to determine the lower amplitude threshold on the basis of the determined mean amplitude for the signal to be BER monitored when representing the lower binary value and a cross-over amplitude defining a transition between a sample value being detected as an upper binary value and a lower binary value.

19. A bit error rate monitor as claimed in any one of claims 15 to 18 wherein the signal to be BER monitored is an optical signal.

20. A bit error rate monitor as claimed in any one of claims 15 to 19 wherein the signal to be BER monitored is processed in the electrical domain as an electrical signal having a voltage level representative of the amplitude of the signal to be BER monitored.

21. A bit error rate monitor as claimed in any one of claims 15 to 20 wherein the monitor further includes a pre-processing stage configured to convert an optical signal to be BER monitored into an electrical signal.

22. A bit error rate monitor as claimed in claim 21 wherein the pre-processing stage includes a DC blocking circuit configured to block a DC component of the electrical signal.

23. A bit error rate monitor as claimed in any one of claims 15 to 23 which further includes a plurality of signal paths wherein at least one signal path bypasses the threshold generating stage.

24. A bit error rate monitor as claimed in claim 23 wherein the device includes a first signal path for communicating a first signal component to the threshold generating stage for use in determining at least one of the upper amplitude threshold and the lower amplitude threshold; and at least one second signal path for communicating a respective second signal component to the sample collection stage.

25. A bit error rate monitor as claimed in any one of claims 15 to 24 wherein the comparator stage includes a first comparator configured to compare the plurality of sampled values with the upper amplitude threshold and a second comparator configured to compare the plurality of sampled value with the lower amplitude threshold.

26. A bit error rate monitor as claimed in any one of claims 15 to 25 which further includes at least one counter configured to count the number of samples with amplitudes above the upper amplitude threshold.

27. A bit error rate monitor as claimed in any one of claims 15 to 26 which further includes at least one counter configured to count the number of samples with amplitudes below the lower amplitude threshold.

28. A bit error rate monitor as claimed in any one of claims 15 to 27 wherein the control stage is further configured to send a control signal if the BER of a signal being monitored is above a predetermined BER threshold.

29. A bit error rate monitor as claimed in any one of claims 15 to 28 in which the sample rate of the sample collection stage is less than a minimum data rate of the signal being

BER monitored.

30. A method of bit error rate (BER) monitoring an optical signal including the steps of:

- converting the optical signal into an electrical signal having a voltage corresponding to an amplitude of the optical signal; - deriving from the electrical signal an upper voltage threshold and a lower voltage threshold;

31. A method as claimed in claims 30 wherein the upper amplitude threshold is derived by:

DC blocking the electrical signal so that a cross-over amplitude between the upper binary value and lower binary value corresponds to a zero voltage level in the electrical signal; determining a mean positive voltage of the signal to be BER monitored when representing the upper binary value; and setting the upper voltage threshold to double the mean positive voltage.

32. A method as claimed in claims 31 which further includes: determining a mean negative voltage of the signal to be BER monitored when representing the lower binary value; and setting the lower voltage threshold to double the mean negative voltage.

33. A device for bit error rate (BER) monitoring of an optical signal, including; a signal reception stage configured to receive an optical signal and convert it into an electrical signal having a voltage representative of the amplitude of the optical signal; a threshold generating stage configured to generate an upper amplitude threshold and a lower amplitude threshold on the basis of the voltage of the electrical signal; a sample collection stage configured to asynchronously sample the signal to generate a plurality of sampled values; a comparator stage configured to compare the sampled values to the upper amplitude threshold and the lower amplitude threshold; and a control stage configured to determine a proportion of the sample values falling above the upper amplitude threshold and a proportion of the sample values falling below the lower amplitude threshold, and to derive a BER for the signal being monitored.

34. A device for bit error rate (BER) monitoring of an optical signal as claimed in claim 33 which further includes a DC blocking circuit configured to block a DC component of the electrical signal.

35. A device for bit error rate (BER) monitoring of an optical signal as claimed in either of claims 33 or 34 which includes a first signal path for communicating a first signal component to the threshold generating stage for use in determining the upper and lower amplitude thresholds; a second signal path for communicating a second signal component to a first comparator of the comparator stage that is configured to compare the plurality of sampled values with the upper amplitude threshold; and a third signal path for communicating a third signal component a second comparator of the comparator stage that is configured to compare the plurality of sampled values with the lower amplitude threshold.

36. A method as claimed in any one of claims 1 to 6 and 8 to 14 wherein the signal to be BER monitored is a digital signal transmitted via any one of the following types of transmission signal: a microwave signal, a radio signal, an electrical signal.

37. A method substantially as hereinbefore described with reference to the illustrative embodiments described herein.

38. A method substantially as hereinbefore described with reference to figure 4 of the accompanying drawings.