GB1581165A - Measuring noise/error characteristic of a modulated telecommunications signal transmission system - Google Patents

Description

(54) IMPROVEMENTS IN MEASURING NOISE/ERROR CHARACTERISTIC OF A MODULATED TELECOMMUNICATIONS SIGNAL TRANSMISSION SYSTEM (71) We, THE POST OFFICE, a British corporation established by Statute, of 23 Howland Street, London, W1P 6HQ, do hereby declare the invention, for which we pray that a patent may be granted to us, and the method by which it is to be performed, to be particularly described in and by the following statement: This invention relates to methods and apparatus for measuring the relationship between carrier to noise (C/N) ratio and error rate in a modulated telecommunications signal transmission system, the invention being particularly, but not exclusively, applicable to phase shift keyed systems.

According to one aspect of the invention, said apparatus comprises an attenuator for controllably attenuating the modulated signal before it is received at a receiver, said attenuator being driven by one output of drive means having a second output providing one input to an indicator whereby said one input is related to the magnitude of the attenuated modulated signal, and an error detector associated with the receiver for detecting the incidence of errors in the received signal, an output from the detector providing a second input to said indicator, said indicator being adapted to indicate the relationship between the carrier to noise ratio and the error rate as attenuation of said modulated signal is altered by said drive means.

Said drive means can be a voltage ramp generator whereby said C/N ratio-error rate relationship is indicated automatically during a sweep of the ramp generator.

Said attenuator is preferably a PIN diode attenuator.

Said indicator can be an X-Y plotter, the C / N ratio being preferably converted to logarithmic form by passing the second output of the drive means through a logarithmic amplifier and the output of said error detector being fed through a digital to analogue converter.

According to another aspect of the invention there is provided a method of measuring the relationship between carrier to noise ratio and error rate in a modulated telecommunications signal transmission system, said method comprising operating drive means both to controllably attenuate the modulated signal before it is received at a receiver and to provide an input to an indicator which input is thereby related to the magnitude of the attenuated signal and, following detection of the incidence of errors in the received signal, generating a second input to the indicator related to the rate of occurrence of errors, said indicator being adapted to indicate the relationship between the carrier to noise ratio and the error rate as attenuation .of said modulated signal is altered.

An embodiment of the invention will now be described by way of example with reference to the accompanying drawings, in which: Figure 1 is a block schematic diagram of a transmission system showing details of apparatus according to the invention Figures 2A, 2B, 2C and 3 are graphical illustrations linking time, C/N ratio and error rate, and Figure 4 is a block schematic diagram corresponding generally to Figure 1 but showing further practical details.

The apparatus shown in Figure l of the accompanying drawings makes it possible to make measurements of C/N ratio against error rate automatically, by a sweep type method, over a range sufficient for checking of system performance. This method represents a considerable time saving in C/N ratio against error rate measurements which are frequently made during digital transmission system development and may be neces sary as part of maintenance procedures on working systems.

The method involves changing the received C/N ratio by causing the loss between a transmitter 1 and a receiver 2 to vary steadily while the error rate is repeatedly measured over a short period.

The errors measured in the period are converted into a level proportional to the error rate which is then used to drive one axis on an X-Y plotter 3 while the other axis is driven by a signal derived from that controlling the change of signal level, so producing a C/N ratio against error rate characteristic.

The two axes of the plotter are calibrated in C/N ratio and error rate respectively.

It is possible to measure a range of error rate between 1 in 101 ratio and 1 in 102 with error samples taken over 0.1 sec periods for bit rates in the region of 120 Mbit/S the whole characteristics being plotted in about 20 sec. Background error rate is, however, to be measured separately over a longer period to allow sufficient errors to be measured.

The items which are added to a standard measurement system to provide an automatic system for measuring C/N ratio against error rate are shown in Figure 1 within the broken outline 4. A PIN diode attenuator 5 is inserted in the transmission path (this could be at i.f. or r.f.) and in conjunction with the precision r.f. variable attenuator 6 is used to provide the required range of C/N ratio variation. The output of an error detector 7 is fed to a digital to analogue converter where the errors in constant sampling periods are converted to a level proportion to error rate which is then used to drive one axis of the X-Y plotter 3. A pattern generator 11 is used to supply an input signal to the transmitter 1.

The PIN attenuator 5 is driven from a ramp generator 9 which also provides another output which is passed through a logarithmic amplifier 10 and drives the other axis of the X-Y plotter. Providing the PIN attenuator has a linear drive characteristic, then the logarithmic amplifier output will be proportional to C/N ratio in dB.

A typical system C/N ratio against error rate characteristic lies between C/N ratios of about 10dB and 22dB for error rates of 1 in 107 respectively. If these error rates occur for higher C/N ratios and only part of the characteristic is measured, this will indicate an unsatisfactory performance. If required, in this situation, the range of C/N variation could be adjusted so that the whole characteristic is measured.

The period over which errors are counted should ideally be fairly long so that sufficient errors are counted to give an accurate measurement of error rate, but the whole swept measurement needs to be completed in a fairly short period so that this method has a significant advantage over point by point measurements. A reasonable compromise might be about 0.1 sec periods for error rate sampling and about 20 sec for the complete measurement. This represents a rate of change of C/N ratio of in the region of ldB/sec which would give a change of about 0.1 dB in C/N ratio during the error sampling period. This would typically represent an error rate change of about 10% which can be neglected.

A practical test of a C/N ratio against error rate measuring system of the type described was made using a digital to analogue converter derived from error rate monitoring equipment used on field tests of 25 M bit/s and 60 M bit/s 11 Ghz lineof-sight systems. This D/A converter consisted of a chain of decade counters the outputs of which operate relays which switch resistors into circuit such that the current drawn from a constant voltage supply is proportional to the errors counted in the sampling period. At the end of the sampling period the count is stopped and for a similar period the relays are energised to give an output from the converter indicating the count reached by the decade counters. This circuit distinguishes changes of error rate of X2 only but was sufficiently accurate for this swept type of measurement.The circuit was originally operated with a sampling period of 60 sec but it was found that the reed relays used in this circuit operate satisfactorily at 10 times per sec.

Typical outputs of the main stages of the measurement system are shown diagrammatically in Figures 2A to 2C. One of the first characteristics plotted using the system is shown in Figure 3. This characteristic was plotted without the logarithmic amplifier 10 between the ramp generator 9 and the X-Y plotter 3 and thus does not yet have the familiar C/N ratio against error rate characteristic shape. The characteristic shown indicates the capabilities of the measurement system. In Figure 3, the C/N ratio against error rate characteristic is shown as plotted by the X-Y plotter.

Figure 2A shows the C/N ratio of the system plotted against time (that is, it shows the effect of the attenuator 5). Figure 2B shows the output of the digital / analogue converter 8 (that is, error rate) plotted against time. Each level in the curve of Figure 2B consists of smoothed 0.1 sec samples from the D/A converter 8. Each step indicates where a relay has operated, switching in another resistor in consequence of an increase in error rate X 2 (that is, doubling of error rate). Figure 2c shows error rate plotted against C/N ratio. The measurement of C/N ratio against error rate characteristics may also be required as part of system maintenance procedures after repairs or changes have been made or at intervals to detect drift in system performance.

The system described provides a useful development and maintenance aid.

Further practical details will now be described with reference to Figure 4, Figure 4 corresponding generally to blocks of Figure 1 but showing more detail. Blocks of Figure 4 corresponding to blocks of Figure 1 are given corresponding but primed reference numerals.

Referring to Figure 4, a pseudo-random error pattern generator 111 is connected to a four-phase coherent phase shift keying (PSK) transmitter 11 operating at 11.115 gigahertz and 120 megabits/second. The output of the transmitter 1' is connected to the input of a calibrated precision attenuator 61, the output of which is connected to the input of a PIN-diode attentuator 5'. The output of the attenuator 51 is connected to a receiver 21 which has its automatic gain control (AGC) output connected to an AGC interface circuit 40. The signal output of the receiver 21 is connected to an error detector 71 having its output connected to an error counter 41 in the form of a ripple counter.

The error counter 41 has its output connected to a multiplexor 42 which is provided to determine the state of fill of the ripple counter 41. A control circuit 43 to provide the functions of error count gating, error count reset and error count calibration receives a 120 megahertz clock signal and is connected to supply outputs to the multiplexor 42 and to an error display and latches circuit 44. The multiplexor 42 has its output connected to the input of the error display and latches circuit 44 which has its output connected to a digital / analogue converter 8'. The output of the digital/analogue converter 81 is connected to the X input of an X-Y plotter 31 The Y input to the X-Y plotter is taken from the AGC interface circuit 40.

A ramp generator 91 has one output connected to control the attenuator 51 and a second output connected to control the lifting of the pen of the X-Y plotter 31 The operation of the arrangement shown in Figure 4 is generally the same as that of Figure 1 except that instead of the ramp generator directly providing the Y input to the X-Y plotter it provides the said input indirectly through the action of the AGC in the receiver 21. The level of the AGC signal in the receiver 21 varies with signal strength and can thus be used as a measure of C/N ratio. The AGC interface circuit converts the AGC output of the receiver 21 into a logarithmic input acceptable to the X-Y plotter 31.

An exemplary realisation of some of the blocks of Figure 4 is given below.

A Hewlett Packard microwave switch No.

33102A is used as the PIN diode attenuator 51 The error counter 41 takes the form of eight integrated circuit decade counters connected as a resettable ripple counter. A gate is provided to feed error pulses to the ripple counter for a predetermined time period so that the state of fill of the counter is a measure of the error rate. The multiplexor 42 takes the form of a number of integrated circuit multiplexors arranged to produce an output indicating the state of fill of the ripple counter in exponent and mantissa form.

The error display and latches circuit 44 include an integrated latching device and a respective semiconductor, numerical display device for the exponent and mantissa.

The digital/analogue convertor 81 includes a linear converter for the exponent and a logarithmic converter for mantissa, a summing amplifier being used to add the converted exponent and mantissa. Integrated circuit switching circuits connected to a plurality of resistors of different "weightings" are used in the conversions.

The ramp generator 91 includes a monostable device triggerable to start the ramp and resetting after a predetermined time. A capacitor charged, under control of the monostable device, by a constant current generator is used to produce the ramp. A buffer amplifier is connected to receive the capacitor voltage as input and has its output connected to a level-setting amplifier which adjusts the level of the start and finish of the ramp to the correct values for attenuator 51. A circuit is provided to cause the X-Y plotter pen to be lifted when the ramp is not being generated.

It is possible to use a motor driven precision attenuator in place of the described PIN diode attenuator.

WHAT WE CLAIM IS:- 1. Apparatus for measuring the relationship between carrier to noise (C/N) ratio and error rate in a modulated telecommunications signal transmission system, the apparatus comprising an attenuator for controllably attenuating the modulated signal before it is received at a receiver, drive means having a first output to drive the attenuator and a second output arranged to provide one input to an indicator such that said one input is related to the magnitude of the attenuated modulated signal, and an error detector for association with the receiver for detecting the incidence of errors in the received signal, the error detector having an output arranged to provide a second input to the indicator, the indicator being arranged to indicate the relationship

**WARNING** end of DESC field may overlap start of CLMS **.

Claims (12)

**WARNING** start of CLMS field may overlap end of DESC **. of system maintenance procedures after repairs or changes have been made or at intervals to detect drift in system performance. The system described provides a useful development and maintenance aid. Further practical details will now be described with reference to Figure 4, Figure 4 corresponding generally to blocks of Figure 1 but showing more detail. Blocks of Figure 4 corresponding to blocks of Figure 1 are given corresponding but primed reference numerals. Referring to Figure 4, a pseudo-random error pattern generator 111 is connected to a four-phase coherent phase shift keying (PSK) transmitter 11 operating at 11.115 gigahertz and 120 megabits/second. The output of the transmitter 1' is connected to the input of a calibrated precision attenuator 61, the output of which is connected to the input of a PIN-diode attentuator 5'. The output of the attenuator 51 is connected to a receiver 21 which has its automatic gain control (AGC) output connected to an AGC interface circuit 40. The signal output of the receiver 21 is connected to an error detector 71 having its output connected to an error counter 41 in the form of a ripple counter. The error counter 41 has its output connected to a multiplexor 42 which is provided to determine the state of fill of the ripple counter 41. A control circuit 43 to provide the functions of error count gating, error count reset and error count calibration receives a 120 megahertz clock signal and is connected to supply outputs to the multiplexor 42 and to an error display and latches circuit 44. The multiplexor 42 has its output connected to the input of the error display and latches circuit 44 which has its output connected to a digital / analogue converter 8'. The output of the digital/analogue converter 81 is connected to the X input of an X-Y plotter 31 The Y input to the X-Y plotter is taken from the AGC interface circuit 40. A ramp generator 91 has one output connected to control the attenuator 51 and a second output connected to control the lifting of the pen of the X-Y plotter 31 The operation of the arrangement shown in Figure 4 is generally the same as that of Figure 1 except that instead of the ramp generator directly providing the Y input to the X-Y plotter it provides the said input indirectly through the action of the AGC in the receiver 21. The level of the AGC signal in the receiver 21 varies with signal strength and can thus be used as a measure of C/N ratio. The AGC interface circuit converts the AGC output of the receiver 21 into a logarithmic input acceptable to the X-Y plotter 31. An exemplary realisation of some of the blocks of Figure 4 is given below. A Hewlett Packard microwave switch No. 33102A is used as the PIN diode attenuator 51 The error counter 41 takes the form of eight integrated circuit decade counters connected as a resettable ripple counter. A gate is provided to feed error pulses to the ripple counter for a predetermined time period so that the state of fill of the counter is a measure of the error rate. The multiplexor 42 takes the form of a number of integrated circuit multiplexors arranged to produce an output indicating the state of fill of the ripple counter in exponent and mantissa form. The error display and latches circuit 44 include an integrated latching device and a respective semiconductor, numerical display device for the exponent and mantissa. The digital/analogue convertor 81 includes a linear converter for the exponent and a logarithmic converter for mantissa, a summing amplifier being used to add the converted exponent and mantissa. Integrated circuit switching circuits connected to a plurality of resistors of different "weightings" are used in the conversions. The ramp generator 91 includes a monostable device triggerable to start the ramp and resetting after a predetermined time. A capacitor charged, under control of the monostable device, by a constant current generator is used to produce the ramp. A buffer amplifier is connected to receive the capacitor voltage as input and has its output connected to a level-setting amplifier which adjusts the level of the start and finish of the ramp to the correct values for attenuator 51. A circuit is provided to cause the X-Y plotter pen to be lifted when the ramp is not being generated. It is possible to use a motor driven precision attenuator in place of the described PIN diode attenuator. WHAT WE CLAIM IS:-

1. Apparatus for measuring the relationship between carrier to noise (C/N) ratio and error rate in a modulated telecommunications signal transmission system, the apparatus comprising an attenuator for controllably attenuating the modulated signal before it is received at a receiver, drive means having a first output to drive the attenuator and a second output arranged to provide one input to an indicator such that said one input is related to the magnitude of the attenuated modulated signal, and an error detector for association with the receiver for detecting the incidence of errors in the received signal, the error detector having an output arranged to provide a second input to the indicator, the indicator being arranged to indicate the relationship

between the carrier to noise ratio and the error rate as attenuation of the modulated signal is altered by the drive means.

2. Apparatus as claimed in claim 1, wherein said drive means is a voltage ramp generator arranged so that said C/N ratioerror rate relationship is indicated automatically during a sweep of the ramp generator.

3. Apparatus as claimed in claim 2, wherein the drive means includes an AGC interface circuit to derive said second output from an AGC signal of the said receiver.

4. Apparatus as claimed in any preceding claim, wherein the error detector includes a ripple counter to count detected errors and a multiplexor circuit to determine the state of fill of the ripple counter.

5. Apparatus as claimed in claim 4, wherein the multiplexor circuit is arranged to produce an output in exponent and mantissa form.

6. Apparatus as claimed in any preceding claim, wherein the said attenuator is a PIN diode attenuator.

7. Apparatus as claimed in any preceding claim, wherein the said indicator is an X-Y plotter.

8. Apparatus as claimed in any preceding claim, wherein both of the outputs provided as inputs to the indicator are arranged to be in logarithmic form.

9. Apparatus for measuring the relationship between carrier to noise (C/N) ratio in a modulated telecommunications signal, the apparatus being substantially as herein described with reference to and as illustrated by Figure 1 of the accompanying drawings.

10. Apparatus for measuring the relationship between carrier to noise (C/N) ratio in a modulated telecommunications signal, the apparatus being substantially as herein described with reference to and as illustrated by Figure 4 of the accompanying drawings.

11. A method of measuring the relationship between carrier to noise ratio and error rate in a modulated telecommunications signal transmission system, said method comprising generating drive means both to controllably attenuate the modulated signal before it is received at a receiver and to provide an input to an indicator, which input is related to the magnitude of the attenuated signal, and, following detection of the incidence of errors in the received signal, generating a second input to the indicator related to the rate of occurrence of errors, said indicator being adapted to indicate the relationship between the carrier to noise ratio and the error rate as attenuation of said modulated signal is altered.

12. A method as claimed in claim 11, wherein the input related to the carrier to noise ratio is derived from an AGC signal of the receiver.

A T TT O~ T1k A'T] A wr