DE10223964B3 - Method for detecting a burst in a transmitted data stream and circuit arrangement for carrying out the method - Google Patents

Abstract

The invention relates to a method for detecting a burst (100a) in a transmitted data stream (102), the transmitted data stream being received in a receiving unit (101), and an electrical signal (102a) corresponding to the transmitted data stream (102) from the receiving unit ( 101) is output, the electrical signal (102a) output by the receiving unit (101) is converted into an amplifier output signal (102b) by means of a transimpedance amplifier (103), a first mean value (M1) of the amplifier output signal (102b) with a first averaging time constant (T1) is formed in a first averaging unit (119-1), a second average (M2) of the amplifier output signal (102b) is formed with a second averaging time constant (T2) in a second averaging unit (119-2), a comparison of the first average value (M1) with the second average value (M2) of the amplifier output signal (102b) by means of a comparison unit (120) rd, and at least one amplifier time constant of the transimpedance amplifier (103) is switched as a function of the comparison carried out by means of a switching signal (114).

Description

Burst detection method in a transferred Data stream and circuit arrangement for performing the method

The present invention relates to a method for recognizing signals with different time profiles, in particular coded waveforms, and particularly relates to a method of detecting a burst in a transferred Data stream and a circuit arrangement for carrying out the Process.

Burst mode receivers are used to receive Burst mode receiver signals used, these receiver signals or the transferred streams are preferably composed of individual bursts, which in carry information in a coded manner. During a burst, one must following circuit arrangement with short amplifier time constants work to do the fast signal changes within a burst dissolve to be able while a longer time between individual bursts, i.e. a transition period can pass in which a large amplifier time constant, i.a. to suppression of noise, is required. There is a fundamental problem with burst mode receivers in that at least two amplifier time constants were used with these must and that, if necessary, between the two amplifier time constants must be switched flexibly.

Burst mode receivers are preferred in passive optical networks (PON = Passive Optical Network), where data rates for data transmission via Ethernet of currently up to 1.25 Gbit / s. Between two successive ones Data packets in the transmitted Data stream can be a time gap (also called "gap"), depending on the respective transmission method, vary widely.

There will also be a transition period (also as "Settling Time ") specified that passes to start a new data packet to be able to output correct data. Typical transition times are in the range of 50 nanoseconds (ns) to 1000 ns. This transition period is determined by an amplifier time constant - for example T = R⋅C - determined those in the burst mode receiver can be specified in terms of circuitry.

Conventional data stream receivers have for this reason, always an AC-coupled (AC-coupled) Offset and gain control for one continuous operation. For one In burst mode operation, these receivers are not suitable because of transition times can only be provided in the range of 500 ns to 1000 ns. A shorter Control time constant and / or a shorter coupling time constant would the received Change data flow through a control and / or coupling.

A well known method to do this Overcoming problem and fast transition times ("Settling times") exists therein, a DC voltage coupling (DC coupling) from transimpedance amplifier and postamplifier to realize.

That way, a quick one Settling can also be provided below 100 ns. On major disadvantage of this conventional circuit arrangement is that DC coupling is essential higher switching effort in the entire burst mode receiver arrangement requires.

Here, a functionality of AC coupling (AC) through an offset adjustment via digital-to-analog converter simulated. In addition to additional Components such as digital-to-analog converters, special logic, connection units, etc. the conventional DC-coupled burst mode receiver arrangement a higher one Dynamic range as an AC-coupled burst mode receiver arrangement exhibit.

In the publication “A SiGe BiCMOS Burst-Mode 155 Mb / s Receiver for PON "by S. Brigati et al, Italtel, IFX, BMR for APON @ ESSCIRC 2001, is such a conventional one Burst mode receiver for passive Optical networks (PON = Passive Optical Network) described.

The disadvantage of this receiver arrangement is that provided a complex and costly circuit arrangement must be, which is inappropriately extremely prone to failure. Furthermore, an amplitude or gain control in the transimpedance amplifier used by a nonlinear Characteristic curve of the transimpedance amplifier replaced, which leads to the disadvantage that low supply voltages are no longer used can.

Another traditional approach to deployment short transition times consists of the levels of logic 0 signals and 1 signals with analog Sample and hold circuits in real time within one bit measure up. This can a correct offset within the first few bits of the data stream transmission as well as a correct gain setting be determined.

A major disadvantage of conventional circuit arrangements is that a switchover signal for switching over the transimpedance amplifier between at least two different amplifier time constants must take place via a separate switchover signal which is fed to a switchover connection. This switchover signal must be, for example, by external logic are specified, which enables switching between, for example, a fast and a superfast mode or switching between a large amplifier time constant and a small amplifier time constant.

Another disadvantage of such external switching is that the transimpedance amplifier with an additional, accessible from outside Switching connection must be provided. Furthermore, it is inappropriate that external circuit units must be provided, which the timing and size of a Set switch signal.

The EP 0 580 567 A1 describes a receiver for an optical transmission network which has a variable amplifier time constant which, despite its simple construction, does not require long synchronization signals. An AC voltage coupling stage of the receiver has a timing element with a variable time constant, as a result of which a timing element can be controlled by a control circuit. In a disadvantageous manner, a first, small time constant and then a second, large time constant are initially assigned to this timer without a synchronization signal or a burst being recognizable. So it is inappropriately with that in the EP 0 580 567 A1 described circuit arrangement also not possible to provide an automatic switchover between different time constants of the timing element. In a disadvantageous manner, the switchover signal must be processed externally.

The US 5,412,498 discloses a receiver with a multiple RC time constant, the switching thereof being provided by a control unit by means of a control signal. Also in the in the US 5,412,498 Disclosed circuit arrangement has the disadvantage that a burst is not recognized and thus an automatic switchover cannot be provided in this way. In particular, it is disadvantageous that a clamp signal generator and a mode controller must be provided in order to provide such a switching of the time constants by means of resistors which can be connected in parallel. Instead, first and second switches are inappropriately provided, which connect the connections of respective coupling capacitors to ground during the first and second clamping intervals.

It is therefore an object of the present invention to provide a method and circuit arrangement that enables a simple detection of bursts and an automatic Switchover provides without externally accessible switchover connections provided Need to become and without having to process a changeover signal externally.

This object is achieved by a solved method specified in claim 1.

Furthermore, the task by a Circuit arrangement with the features of claim 7 solved.

Further refinements of the invention result from the subclaims.

An essential idea of the invention is an output signal of one used in a burst mode receiver Transimpedance amplifier to prepare a switchover signal.

According to the properties of the Amplifier output analyzed by a circuit arrangement according to claim 7.

The result is advantageous the analysis returned directly to the transimpedance amplifier, depending on switching from at least one of predetermined limit values Amplifier time constant in the transimpedance amplifier allows becomes. Another advantage of the circuit arrangement according to the invention and the inventive method is that no additional Switching connection can be provided on the impedance amplifier got to.

Furthermore, it is very useful that automatic detection of bursts in burst mode receiver signals allows is, so the entire circuitry of the burst mode receiver continues can be simplified and automated.

The inventive method for recognizing a Bursts in a broadcast Data flow essentially has the following steps:

a) detecting the transmitted data stream, the has a coded signal curve in a receiving unit;

b) Outputting the transmitted Data stream corresponding electrical signal from the receiving unit;

c) amplifying the output from the receiving unit electrical signal into an amplifier output signal by means of a transimpedance amplifier;

d) forming a first mean of the amplifier output signal with a first averaging time constant in a first averaging unit;

e) forming a second mean of the amplifier output signal with a second averaging time constant in a second averaging unit;

f) performing a comparison of the first mean with the second mean value of the amplifier output signal by means of a comparison unit; and

g) switching at least one amplifier time constant of the transimpedance amplifier dependent on of the comparison carried out in step f) by means of a Switching signal.

Furthermore, the circuit arrangement according to the invention to detect a burst in a transmitted data stream, the a transimpedance amplifier for reinforcement into an amplifier output signal supplied is essentially on:

a) A first averaging unit for Formation of a first average value of the data stream transmitted Amplifier output signal, with a first averaging time constant;

b) a second averaging unit to form a second mean value corresponding to the transmitted data stream Amplifier output signal, with a second averaging time constant;

c) a comparison unit for carrying out a Comparison of the first mean with the second mean of the Amplifier output signal; and

d) a difference value determination device to determine a difference between the first mean and the second mean.

There are advantageous ones in the subclaims Developments and improvements to the subject of Invention.

According to a preferred development The present invention uses a burst in a transmitted Data stream or in a burst mode receiver signal, the one encoded signal curve, by the comparison according to step f) identified in claim 1 such that the first mean of deviates from the second mean by a predetermined difference value. This way it’s easy to do bursts, i.e. signaling events with a short time constant, to be recognized in a transmitted data stream.

According to another preferred A further development of the present invention is the first averaging time constant for forming the first mean greater than the second averaging time constant for Form the second mean.

According to yet another preferred development In the present invention, the comparison unit forms the difference value by summing the first mean with the inverted second Mean or vice versa. In this way it becomes a difference value get which is a current voltage-related distance of the two mean values equivalent.

According to yet another preferred development of the present invention then becomes a switch signal to the transimpedance amplifier output when the difference value exceeds a first reference value.

According to yet another preferred development of the present invention then becomes a switch signal to the transimpedance amplifier output when the difference value falls below a second reference value. That way the difference value advantageously between an upper Limit and a lower limit.

According to yet another preferred development of the present invention has the difference value determining means at least one first comparator unit for determining an upper one Limit on. In this case, a switchover signal can advantageously are output if the difference value exceeds the first reference value.

According to yet another preferred development of the present invention has the difference value determination means at least one second comparator unit for determining a lower one Limit of the difference value. This can be done in an advantageous manner a switch signal to the transimpedance amplifier are output if the difference value falls below a second reference value.

According to yet another preferred development The present invention is a logic unit for evaluation of the signals output by the differential value determination device provided.

According to yet another preferred development The present invention is a flip-flop unit for providing of a changeover signal depending provided by a signal output by the logic unit.

Embodiments of the invention are shown in the drawings and in the description below explained in more detail.

The drawings show:

1 a block diagram of an overall circuit arrangement of a burst mode receiver with a receiving unit, a transimpedance amplifier and a post-amplifier;

2 a typical coded signal curve in a transmitted data stream, two bursts being illustrated by way of example; and

3 a block diagram of a circuit arrangement for detecting a burst in a transmitted data stream according to a preferred embodiment of the present invention.

In the figures denote the same The same or functionally identical components or steps.

1 shows a circuit arrangement of a burst mode receiver in a block diagram.

In the in 1 Circuit arrangement shown for burst mode receivers is a transmitted data stream 102 a receiving unit 101 fed which is a receiving element 113 having. At high data transmission rates (for example 1.25 Gbit / s), this receiving element is designed as a light-sensitive or optical receiving element. In this way, data streams in different passive optical networks can be converted into an electrical signal 102 be implemented.

The electrical signal 102 will eventually a transimpedance amplifier 103 supplied which has at least one controllable amplifier unit 107 having. The amplifier unit is preferably, but not exclusively 107 designed as an operational amplifier. In the feedback branch of the amplifier unit 107 two parallel resistor elements are formed, one of which is a resistance control unit 106 is varied. In this way, the entire gain of the transimpedance amplifier 103 varied. The output signal of the amplifier unit 107 becomes a converter unit 108 supplied, which at a second input terminal a bias voltage via a biasing unit 109 is fed. The converter unit converts “single-ended” signals from the amplifier unit 107 into differential signals.

In this way it is possible to offset an amplifier output signal 102b variable. The output signal of the converter unit 108 is tapped and the gain control unit 106 fed. The gain control unit 106 consists of a gain control unit 106a and a level detection unit 106b , The level detection unit 106b detects that at the output of the converter unit 108 applied level and acts via the gain control unit 106a on the feedback branch of the operational amplifier unit 107 , so that a gain control loop is provided.

According to the invention, a connection is made to the transimpedance amplifier 103 following amplifier 104 by means of an alternating voltage coupling (AC coupling). This is in 1 represented by the two capacitors C. In a switchover unit according to the invention 105 switching between different amplifier time constants is provided. In the in 1 shown, preferred embodiment, a switch between two different amplifier time constants is shown. However, it should be pointed out that a switchover between more than two different amplifier time constants is possible. The switchover presets an amplifier time constant adaptively. The amplifier time constant results approximately from the product of the resistance value of the (connected in parallel) resistance elements and the capacitance value of the capacitors C.

The respective amplifier time constant is calculated as: T = R⋅C.

In 1 it can be seen that when the switch S is flipped, the amplifier time constant is reduced. In this exemplary embodiment, the flipping of a switch is carried out by an external switching signal 114 provided that a switch port 115 of the transimpedance amplifier 103 is fed.

Furthermore, in 1 shown that a switching unit 105 also in the gain control unit 106a the gain control unit 106 is provided. When the AC output coupling is switched over, the gain control is switched over at a constant level of the amplifier output signal 102b to maintain. This amplifier output signal 102b eventually becomes the post amplifier 104 fed. The post-amplifier is constructed in a conventional manner and will therefore only be described briefly. In the exemplary embodiment shown, the post-amplifier consists of at least one differential amplifier unit 110 , wherein the differential amplifier units connected in series 110 via an offset control unit 112 to be controlled. The offset control unit 112 consists of an offset detection unit 112b which have an offset at the output of the differential amplifier units connected in series 110 detected, the detected offset, the offset control unit 102 is supplied to offset control of the first of the differential amplifier units connected in series 110 provide.

Another offset control stage consists of two differential amplifier units 110 and a threshold detector 111 , wherein the output signal of the threshold detector 111 the second offset detection unit 112b which in the second offset control unit 112 is integrated, is fed. The second offset control unit also contains 112 also an offset control unit 112a which on the first differential amplifier unit 110 the second amplifier stage of the post-amplifier 104 acts.

In 2 is a typical coded waveform 100 one of a transmitted data stream 102 corresponding amplifier output signal 102b shown. It can be clearly seen that between two successive bursts 100a and 100b there is a transition period Ü.

In 2 is shown as an example, like averaged signals of the coded waveform 100 run with different averaging time constants. The dashed line shows a signal curve averaged with a large averaging time constant T1 as a mean value M1, while the solid line shows a signal curve averaged with a small averaging time constant T2 as a mean value M2. Due to the different averaging time constants T1 and T2, the mean values M1 and M2 differ from one another increasingly over the course of time from the time of averaging.

It is thus clearly recognizable that a difference value D, which is also variable over time, occurs between the two mean values M1 and M2. According to the invention, this mean value D is now used to determine whether a burst is present in a burst mode receiver signal or not. Switching between different amplifier time constants of the transimpedance ver stärkers 103 , please refer 1 is only necessary if the difference value D between the first average M1 and the second average M2 exceeds the limits specified in terms of amount.

It's over 2 Clearly recognizable that in the exemplary embodiment shown, the first mean value M1 follows the second mean value M2, since the averaging time constant T1 is greater than the averaging time constant T2. For different bursts 100a . 100b , ..., 100n in the burst mode receiver signal, the second mean M2 now moves around the first mean M1. During a burst 100a it is not necessary to switch the transimpedance amplifier, which is operated with a small amplifier time constant at this time. If the burst were to last for a longer time, the first average M1 and the second average M2 would approximate or match. Without the presence of a burst, the same applies, ie the two mean values M1 and M2 would match within predefinable limits.

If a second burst occurs suddenly 100b after the end of a first burst 100a however, a range arises in which the two mean values M1, M2 differ, since the averaging is carried out with different averaging time constants T1, T2 and thus a different reaction to a change in the coded signal curve 100 he follows.

In the exemplary embodiment according to the invention, the transimpedance amplifier is now used 103 a switching signal 114 always provided when the second mean value M2 deviates from the first mean value M1 by a predeterminable limit, ie when the difference value D falls below a predeterminable reference value (the difference value D is in the region (II) in 2 ), or if the difference value D exceeds a first reference value (the difference value is in a range (I) in 2 ).

3 shows a preferred embodiment of a circuit arrangement for performing the method according to the invention.

As in 3 is shown, the amplifier output signal 102b of the transimpedance amplifier 103 two averaging units simultaneously, ie a first averaging unit 119-1 and a second averaging unit 119-2 fed. The first averaging unit 119-1 by means of the signal supplied to it with a first averaging time constant T1, while the second averaging unit 119-2 averages the signal supplied to it with a second averaging time constant T2. That is, the amplifier output signal 102b is averaged with two different averaging time constants T1 and T2, so that different first and second average values M1 and M2 result.

To determine the difference value D, see 2 , serves the comparison unit, which is designed, for example, as a summation unit, one of the two signals, in this case the output signal of the second averaging unit 119-2 is inverted.

The difference value D becomes a difference value determination device 116 supplied, in which the position and size of the difference value D can be detected. The difference value determination device 116 contains a first comparator unit in the preferred embodiment 123 and a second comparator unit 124 , The output signal of the comparison unit fed to the difference value determination device 120 is the first inputs of the first comparator unit 123 and the second comparator unit 124 fed. The second input of the first comparator unit 123 with a first reference value 121 applied (upper reference value), while the second input of the second comparator unit 124 with a second reference value (lower reference value).

The first and second comparator units 123 respectively. 124 switch when the signal fed to their first input terminals switches the first reference value 121 or the second reference value 122 exceeds. This means referring to 2 that the difference value D exceeded an upper limit or fell below a lower limit.

It should be noted that the difference value D, as referring to FIG 2 is defined, can assume positive and negative values since the comparison unit sums the first and second mean values M1 and M2, the mean value M2 being inverted, so that the second mean value M2 can be below or above the first mean value M1 in terms of voltage. One of the difference value determination device 116 downstream logic unit determines whether one of the two comparator units 123 respectively. 124 the difference value determination device 116 an overshoot or undershoot of the first or second reference value 121 respectively. 122 Show. The difference value has the through the first reference value 121 and the second reference value 122 predeterminable upper or lower limits exceeded, so the logic unit 117 a control signal to the flip-flop unit 118 off, which toggles into an ON state for a predeterminable time and a switchover signal 114 outputs which of the switch port 115 of the transimpedance amplifier 103 is fed.

This switchover signal switches over to the amplifier time constant of the transimpedance amplifier in accordance with the detected burst. Furthermore, the output signal of the flip-flop unit 118 to the logic unit 117 returned to the end of a switching process by the flip-flop unit 118 is specified, the logic unit 117 provide.

By the inventive method for recognizing a burst in a transmitted Data stream and the circuit arrangement for performing the method is thus possible; without an external control signal for switching and without an external switching connection of the transimpedance amplifier to enable switching of the amplifier time constant of the transimpedance amplifier at low cost.

Inexpensive post amplifiers can also be used which operate in a "slow" mode of operation can.

Advantageously, in a standard TD housing integrated transimpedance amplifier used be, the TD standard housing four Have connection pins, with two connection pins for the supply are provided and two pins for a differential output are provided.

Can advantageously by the circuit arrangement according to the invention be avoided that an additional Connection pin or an additional one Switching connection must be provided as an automatic Burst detection is provided.

100

coded waveform

100a,

bursts

100b,

100n

101

receiver unit

102

transferred data stream

102

electrical signal

102b

Amplifier output

103

Transimpedance amplifier

104

postamplifier

105

switching

106

Gain control unit

106a

Gain control unit

106b

Level detection unit

107

amplifier unit

108

converter unit

109

biasing unit

110

Differential amplifier unit

111

threshold

112

Offset control unit

112a

Offset control unit

112b

Offset acquisition unit

113

receiving element

114

switching

115

switching terminal

116

Difference value determining device

117

logic unit

118

Kippstufeneinheit

119-1

First Averaging unit

119-2

Second Averaging unit

120

comparing unit

121

first reference value

122

second reference value

123

First comparator

124

Second comparator

Claims (11)

Burst Detection Method ( 100a ) in a transmitted data stream ( 102 ), with the following steps: a) acquisition of the transmitted data stream ( 102 ), which has a coded waveform ( 100 ) in a receiving unit ( 101 ); b) output of a data stream transmitted ( 102 ) corresponding electrical signal ( 102 ) from the receiving unit ( 101 ); c) Implementation of the from the receiving unit ( 101 ) output electrical signal ( 102 ) into an amplifier output signal ( 102b ) using a transimpedance amplifier ( 103 ); d) forming a first mean value (M1) with a first averaging time constant (T1) in a first averaging unit ( 119-1 ); e) forming a second mean value (M2) with a second averaging time constant (T2) in a second averaging unit ( 119-2 ); f) performing a comparison of the first means value (M1) with the second mean value (M2) of the amplifier output signal ( 102b ) by means of a comparison unit ( 120 ); and g) switching at least one amplifier time constant of the transimpedance amplifier ( 103 ) as a function of the comparison carried out in step f) by means of a changeover signal ( 114 ). A method according to claim 1, characterized in that a burst ( 100a . 100b ) in a transmitted data stream ( 102 ), which has a coded waveform ( 100 ) is identified by the comparison in such a way that the first mean (M1) deviates from the second mean (M2) by a predetermined difference value (D). Method according to Claim 1 or Claim 2, characterized in that that the first averaging time constant (T1) to form the first Mean (M1) is greater than the second averaging time constant (T2) for forming the second mean (M2). A method according to claim 1, characterized in that the comparison unit ( 120 ) forms the difference value (D) by summing the first mean value (M1) with the inverted second mean value (M2). A method according to claim 1 and 2, characterized in that to the transimpedance amplifier ( 103 ) the switching signal ( 114 ) is output when the difference value (D) has a first reference value ( 121 ) exceeds. Method according to claims 2 and 5, characterized in that to the transimpedance amplifier ( 103 ) the switching signal ( 114 ) is output when the difference value (D) has a second reference value ( 122 ) falls below. Circuit arrangement for detecting a burst ( 100a ) in a transmitted data stream ( 102 ) connected to a transimpedance amplifier ( 103 ) for amplification into an amplifier output signal ( 102b ) can be supplied, comprising: a) a first averaging unit ( 119-1 ) to form a first mean value (M1) of the transmitted data stream ( 102 ) corresponding signal, with a first averaging time constant (T1); b) a second averaging unit ( 119-2 ) to form a second mean value (M2) of the transmitted data stream ( 102 ) corresponding signal, with a second averaging time constant (T2); c) a comparison unit ( 120 ) to carry out a comparison of the first mean (M1) with the second mean (M2) of the signal; and d) a difference value determination device ( 116 ) for determining a difference value (D) between the first mean value (M1) and the second mean value (M2). Circuit arrangement according to claim 7, characterized in that the differential value determination device ( 116 ) at least one first comparator unit ( 123 ) to determine an upper limit of the difference value (D). Circuit arrangement according to claim 7, characterized in that the differential value determination device ( 116 ) at least one second comparator unit ( 124 ) to determine a lower limit of the difference value (D). Circuit arrangement according to claim 7, characterized in that a logic unit ( 117 ) for the evaluation of the difference value determination device ( 116 ) output signals is provided. Circuit arrangement according to claim 7 and 10, characterized in that a flip-flop unit ( 118 ) to provide a switching signal ( 114 ) depending on one of the logic unit ( 117 ) output signal is provided.