Circuit for compensating a degraded signal and associated method

1. Field of the Invention

The present invention relates to a circuit for receiving a transmission signal and associated method, and more particularly, to a circuit for dynamically compensating a baseline wander of the transmission signal and associated method.

2. Description of the Prior Art

Please refer to FIG. 1 , which is a schematic diagram of typical user terminals 10 A and 10 B connected to each other by a network transmission line 18 . The user terminals 10 A and 10 B can be network servers, routers, clients, or other kinds of network terminals. The transmission line 18 is a network transmission line, such as an Unshielded Twisted Pair Category 5 (UTP-Cat.5) Ethernet cable. The user terminal 10 A and the user terminal 10 B utilize a signal circuit 12 A and a signal circuit 12 B respectively to transmit signals via the transmission line 18 . The signal circuits 12 A, 12 B include transformers 16 A, 16 B and resistors R 0 a , R 0 b respectively that match the impedance of the transmission line 18 . A transmitter 14 A of the signal circuit 12 A utilizes a differential transmission signal. That is, the transmission signal includes both positive and negative transmission signals transmitted by a pair of conductive lines.

The differential transmission signal is sent through two output terminals of the transmitter 14 A to nodes P 0 A, P 1 A, and is then transformed onto the transmission line 18 by the transformer 16 A. The differential transmission signal is received by the signal circuit 12 B of the user terminal 10 B after the differential transmission signal travels through two wires of the transmission line 18 . The differential transmission signal is transformed by the transformer 16 B and is sent to nodes P 0 B, P 1 B of the signal circuit 12 B. The differential transmission signal is then sent to two differential input terminals of a receiver 14 B, so that the user terminal 10 B receives the signal from the user terminal 10 A via the transmission line 18 .

Several problems can occur during the signal transmission process. For instance, each signal circuit must match the electrical characteristic of the transmission line 18 in order for signal transmission to be effective. However, a high pass characteristic of the transformer in the signal circuit undesirably affects the signal level of the transmission signal.

The Ethernet network system suffers from this high pass problem. In Ethernet, an MLT-3 coded transmission signal is sent from the transmitter, transformer, to the transmission line. The coded transmission signal has a fixed baseline that is a longtime average of a level-changeable digital signal. The electrical characteristic of this signal is that the baseline of the signal relates to a low frequency (LF) component of the signal, which relates to the transmission data. When the coded transmission signal passes through the transformer and onto the transmission line, the direct current (DC) component of the LF is filtered out due to the high pass characteristic of the transformer. After the transmission signal passes through the transmission line to the signal circuit of another user terminal, the original baseline of the signal is degraded and baseline wander occurs.

For the purposes of explaining the baseline wandering phenomenon, please refer to FIG. 2 showing a prior art waveform-time diagram of the transmission signal during the signal transmission process in the network system shown in FIG. 1 . The horizontal axis in FIG. 2 is time, and the vertical axis is signal amplitude. In FIG. 2 , because the transmission signal is a differential signal, a waveform of a positive transmission signal shown is representative. At the user terminal 10 A in FIG. 1 , the transmission signal is generated as illustrated by a waveform 20 . The waveform 20 is composed of three different waveform levels: a high level, a zero level, and a low level representing the different digital signals according to the MLT-3 code. A dotted line 20 A and a dotted line 20 B perfectly represent the overall signal profile of the waveform 20 .

After the transmission signal passes through the transformer 16 A, the transmission line 18 and the transformer 16 B of the user terminal 10 B, the LF component is filtered out due to the high pass characteristic of the transformer. The transmission line affects the original baseline and baseline wander occurs. A waveform 22 of the transmission signal affected by baseline wander is received at the node P 0 B of the user terminal 10 B. The level of the waveform 22 has different deviations at different times, so that enveloping signal levels 22 A and 22 B of the waveform 22 appear to wander between a high level and low level. Therefore, the waveform 22 does not represent the original digital signal of the waveform 20 correctly because of the baseline wandering phenomenon.

Signal transmission errors caused by the baseline wandering phenomenon will now be briefly explained with reference to FIG. 1 and FIG. 2 . A signal with a level higher than a fixed high reference level 24 A is determined as a high level digital signal. A signal with a level lower than a fixed low reference level 24 B is determined as a low level digital signal. The waveform 20 is originally generated by the transmitter 14 A with respect to the high reference level 24 A and low reference level 24 B. The waveform 20 is then affected by the baseline wandering phenomenon during its transmission as previously described, and is finally received at the receiver 14 B. The receiver 14 B then interprets the waveform 22 with reference to the high reference level 24 A and low reference level 24 B and generates a received signal waveform 24 .

The waveform 24 contains errors in time periods T 1 and T 2 . In time period T 1 the waveform 22 wanders low enough so that the original high signal in period T 1 falls below the high reference level 24 A. The original high signal during period T 1 is interpreted as zero level as shown by waveform 24 , incorrectly. A plurality of low level pulses error in a similar way during time period T 2 . The baseline wandering phenomenon thus causes the digital signal of the network transmission to not be received and decoded correctly.

In the prior art signal circuit, a fixed DC bias is added to the received transmission signal at the receiving terminal to compensate for baseline wandering. This solution is not adequate as the deviation of the baseline changes over time, and a fixed compensation cannot eliminate the signal deviation phenomenon entirely.

It is therefore a primary objective of the claimed invention to provide a signal compensation circuit and associated method for compensating signals by adjusting signal levels dynamically to solve the above-mentioned problem, and to ensure the quality of network signal transmissions.

According to the claimed invention, the signal compensation circuit comprises a detection circuit for detecting an enveloping signal level of a transmission signal transmitted via a transmission line, and a correction circuit for dynamically compensating the transmission signal according to a comparison result generated by comparing the transmission signal with a reference level.

It is an advantage that the claimed invention uses a dynamic detection and compensation method to adjust the signal levels so as to correct the deviations of the signal levels caused by the baseline wandering phenomenon. This ensures that the receiver receives the digital signal correctly, thus improving the signal receiving quality of network communication.

These and other objectives of the claimed invention will no doubt become obvious to those of ordinary skill in the art after reading the following detailed description of the preferred embodiment that is illustrated in the various figures and drawings.

FIG. 1 is a schematic diagram of a prior art computer network.

FIG. 2 shows waveforms at different nodes in the computer network shown in FIG. 1 .

FIG. 3 is a schematic diagram of a signal circuit according to the present invention.

FIG. 4 shows waveforms by the signal circuit shown in FIG. 3 .

Please refer to FIG. 3 , which is a schematic diagram of a signal circuit 30 used in a network system according to the present invention. A user terminal 10 A and a user terminal 50 are connected with each other via a network transmission line 18 . The user terminals 10 A and 50 can be network servers, routers, clients, or other network terminals. The transmission line 18 can be an Ethernet Unshielded Twisted Pair Category 5 (UTP-Cat.5) cable, and a transmission signal can be an MLT-3 coded, or 100Base-T coded signal. The user terminal 10 A and the user terminal 50 use a signal circuit 12 A and a signal circuit 30 respectively to process transmission signals. The signal circuit 12 A includes a transformer 16 A and a resistor R 0 a that is matched to the impedance of the transformer 16 A. The signal circuit 30 includes a transformer 32 and a resistor R 0 that is matched to the impedance of the transformer 32 . A transmitter 14 A of the signal circuit 12 A generates a transmission signal. A receiver 36 of the signal circuit 30 receives the transmission signal via the transmission line 18 .

A detection circuit 38 and a correction circuit 40 are provided in the signal circuit 30 for the purpose of compensating baseline wander of the transmission signal to be received by the receiver 36 . The detection circuit 38 detects a signal level of the transmission signal at a node N 3 and a node N 4 , and compares the signal level with a reference level. If the signal level is larger than the reference level, a corresponding first control signal 38 A and a corresponding second control signal 38 B are generated by the detection circuit 38 . The correction circuit 40 compensates for baseline wander of the transmission signal according to the results generated by comparing the transmission signal with the reference level by the detection circuit 38 . After the differential transmission signal passes through the transformer 32 of the signal circuit 30 , the positive and negative transmission signals are sent to a node N 1 and a node N 2 respectively. A second resistor R 2 and a current source 34 A are disposed on a signal transmitting path from the node N 1 to the node N 3 in the correction circuit 40 . A second resistor R 2 and a current source 34 B are disposed on another signal transmitting path from the node N 2 to the node N 4 in the correction circuit 40 . On the two signal paths for transmitting the differential signal, the node N 3 and the node N 4 are connected to a common-mode power supply by two first resistors R 1 . The common-mode power supply is used for providing the differential signal with a common-mode voltage V CM at a node N 5 . The common-mode voltage V CM for an MLT-3 coded, or 100 Base-T coded, signal is 1.8 volts.

The current sources 34 A and 34 B of the correction circuit 40 include controllable current sources 46 A, 48 A and 46 B, 48 B respectively. Bias circuits 41 A, 42 A and 41 B, 42 B provide appropriate bias voltages to the current sources 34 A and 34 B so that the current sources 34 A and 34 B can operate normally. The current intensities of the controllable current sources 48 A and 48 B are controlled by the first control signal 38 A generated by the detection circuit 38 . The current intensities of the controllable current sources 46 A and 46 B are controlled by the second control signal 38 B also generated by the detection circuit 38 . The controllable current sources 46 A and 48 A of the current source 34 A together provide a compensation current Ic 1 at the node N 3 . The controllable current sources 46 B and 48 B of the current source 34 B together provide a compensation current Ic 2 at the node N 4 . A compensation voltage Vc 1 is generated across the resistor R 1 , between nodes N 3 and N 5 , by the compensation current Ic 1 . Likewise, a compensation voltage Vc 2 is generated across the resistor R 1 , between nodes N 4 and NS, by the compensation current Ic 2 . After the differential transmission signal passes through the node N 1 and the node N 2 , the positive and negative transmission signals pass through the resistors R 2 to the node N 3 and the node N 4 , respectively. The positive and negative transmission signals refer to the common-mode voltage V CM as a center voltage level and utilize the resistors R 2 as a loading. The compensation voltages Vc 1 and Vc 2 are added to the positive and negative transmission signals, respectively. In the preferred embodiment of the present invention, the intensities of the compensation currents Ic 1 and Ic 2 are substantially equal. The two first resistors R 1 , having the same resistance, are provided so that the compensation voltages Vc 1 and Vc 2 are also substantially equal. After compensation, the transmission signal is received by the receiver 36 at the node N 3 and the node N 4 in a differential manner so that the transmission signal is transferred from the user terminal 10 A to the user terminal 50 .

The operation for correcting baseline wander according to this invention is described as follows. Please refer to FIG. 4 , which is a waveform of related signals during baseline wander correction of the transmission signal. The horizontal axis in FIG. 4 represents time, and the vertical axis represents signal magnitude. A waveform 54 shown in FIG. 4 is the waveform of the positive component of the differential transmission signal, at the node N 3 . An enveloping signal level 56 represents the positive signal profile of the waveform 54 . As mentioned previously, the enveloping signal level 56 of the waveform 54 deviates with time due to baseline wander, so that the enveloping signal level 56 of the waveform 54 of the transmission signal does not form an ideal straight horizontal line. The detection circuit 38 in the present invention detects a difference between the enveloping signal level 56 and a fixed reference level 52 at a time t1. As illustrated in FIG. 4 , the enveloping signal level 56 is larger than the reference level 52 . Accordingly, the detection circuit 38 generates the first control signal 38 A and the second control signal 38 B to control the current source 34 A and the current source 34 B respectively to produce the negative compensation current Ic 1 and the negative compensation current Ic 2 (the flowing directions of the compensation currents Ic 1 and Ic 2 can be seen in FIG. 3).To accomplish this, the second control signal 38 B turns off the controllable current sources 46 A and 46 B, and the first control signal 38 A turns on the controllable current sources 48 A and 48 B. The negative compensation currents Ic 1 and Ic 2 cause the compensation voltages Vc 1 and Vc 2 to be negative as well (the polarization directions of the compensation voltages Vc 1 and Vc 2 are shown in FIG. 3 ). The resulting negative compensation voltage Vc 1 is added to the positive transmission signal at the node N 3 , and the enveloping signal level 56 , which is larger than the reference level 52 , is reduced and thus corrected.

After the positive transmission signal at the node N 3 has been corrected, the detection circuit 38 detects the enveloping signal level 56 and the reference level 52 every predetermined period. At a time t2, the detection circuit 38 detects that the enveloping signal level 56 has been corrected but is still larger than the reference level 52 . The detection circuit 38 generates the first control signal 38 A and second control signal 38 B to control the current source 34 A and the current source 34 B to further negatively compensate the enveloping signal level 56 via the compensation voltage Vc 1 . Finally, after this correction procedure is repeated several times, shown as times t2 through t6 in FIG. 4 , the enveloping signal level 56 will be eventually corrected.

The signal level may also wander lower than a predetermined level. Suppose that the detection circuit 38 detects an enveloping signal level 56 that is lower than the reference level 52 . The detection circuit 38 controls the current sources 34 A and 34 B to generate a positive compensation current Ic 1 resulting in a positive compensation voltage Vc 1 . The positive compensation voltage Vc 1 then adjusts the enveloping signal level 56 positively.

Each control current source can be implemented by a plurality of unit current sources, with each unit current source providing a fixed current. For example, one or more unit current sources can be activated in the controllable current source 46 A to increase the positive compensation current Ic 1 . Similarly, one or more unit current sources can be activated in the controllable current source 48 A to generate the negative compensation current Ic 1 .

In summary, the above-mentioned process of comparing the enveloping signal level 56 with the reference level 52 using the detection circuit 38 , and then generating the compensation current from the current source, and finally compensating the transmission signal will be repeated to dynamically compensate the signal level deviations due to baseline wander. Although the above description uses the positive transmission signal at the node N 3 as an example, the negative transmission signal at the node N 4 is compensated for at the same time. The node N 3 and the node N 4 are symmetrically provided with respect to the common-mode power supply of the node N 5 . Consequently, the controllable current sources 48 A and 48 B controlled by the first control signal 38 A and the controllable current sources 46 A and 46 B controlled by the second control signal 36 A cause the compensation current Ic 1 to be substantially equal to the compensation current Ic 2 and the compensation voltages Vc 1 and Vc 2 to be substantially thus equal. The polarizations of the compensation voltages correspond with the positive and negative signals of the differential transmission signal to respectively compensate the positive and negative signals.

The symmetric operation of the correction circuit 40 will now be explained. A waveform 54 n shown in FIG. 4 is a waveform of the negative transmission signal of the differential transmission signal at the node N 4 . A negative signal envelope 56 n represents the extent of the waveform 54 n . A horizontal signal level 60 represents a stable DC level provided by the common-mode voltage V CM at the node N 5 . The waveform 54 n of the negative transmission signal is thus the negative mirror image of the waveform 54 of the positive transmission signal, with reference to the common-mode voltage V CM acting as a voltage reference. If the enveloping signal level 56 n of the negative transmission signal deviates from a reference level 52 n because of the baseline wandering phenomenon, the detection circuit 38 shown in FIG. 3 detects the difference between the enveloping signal level 56 n and the fixed reference level 52 n . In fact, the nature of the differential signal prescribes mirrored deviations of the positive and negative transmission signals. From the circuit diagram shown in FIG. 3 of the present invention, the voltage at the node N 3 is equal to V CM +Ic 1 *R 1 , and the voltage at the node N 4 is equal to V CM −Ic 2 *R 1 . Due to the symmetrical nature of the positive and negative differential signals and the construction of the correction circuit 40 , the compensation current Ic 1 is equal to the compensation current Ic 2 . The mirror image property of the correction circuit 40 compensates the positive and negative transmission signals equally in magnitude but opposite in direction. As shown in FIG. 4 , the waveform 54 of the positive transmission signal at time t2 is corrected downward as the waveform 54 n of the negative transmission signal is corrected upward. Therefore, the positive and negative components of the differential transmission signal are corrected simultaneously.

Compared with the prior art, which uses a fixed DC bias voltage, the present invention dynamically detects and compensates the deviations of the signal level caused by baseline wandering. The present invention thus ensures that the receiver receives the digital signal correctly, and significantly improves the receiving quality of network communication.

Those skilled in the art will readily observe that numerous modifications and alterations may be made while retaining the teachings of the invention. Accordingly, the above disclosure should be construed as limited only by the metes and bounds of the appended claims.