CN113098563B

CN113098563B - Interference canceller and interference cancellation method

Info

Publication number: CN113098563B
Application number: CN201911335779.1A
Authority: CN
Inventors: 黄亮维; 杨锦记
Original assignee: Realtek Semiconductor Corp
Current assignee: Realtek Semiconductor Corp
Priority date: 2019-12-23
Filing date: 2019-12-23
Publication date: 2022-11-11
Anticipated expiration: 2039-12-23
Also published as: CN113098563A

Abstract

An interference canceller and an interference cancellation method are capable of generating a cancellation signal for reducing cross-port heterogeneous near-end crosstalk (XPAN) of a received signal of a receiver of a network device. The interference canceller includes: an oversampling circuit for oversampling an XPAN transmission signal from a transmitter of the network device to generate an oversampled signal having a frequency higher than a reception frequency of the receiver; a frequency difference calculating circuit for calculating a frequency difference between the transmitting frequency and the receiving frequency of the transmitter; a sampling rate converter for processing the over-sampled signal according to the frequency difference to generate a converted signal having a frequency lower than the frequency of the over-sampled signal and equal to the frequency of the receiving frequency; and an XPAN canceller, generating the cancellation signal according to the converted signal, and determining coefficients of the XPAN canceller according to an error signal, the error signal being dependent on a difference between the XPAN of the received signal and the cancellation signal.

Description

Interference canceller and interference cancellation method

Technical Field

The present invention relates to an interference canceller and an interference cancellation method, and more particularly, to an interference canceller and an interference cancellation method for a network device.

Background

To solve the problem that a transmission signal of a transmitter of a network device may interfere with a reception signal of a receiver of the network device, the applicant proposed the invention of US patent US9252994B 2; however, the above-mentioned patent invention relies on a cancellation signal generating circuit (e.g., a circuit using Least Mean Square (LMS) technique) to track and compensate for phase shift (phase shift) and mixing (aliasing) of the cancellation signal and the interference signal in a certain time, so that the cancellation signal generating circuit may have poor performance when the timing difference between the transmitter and the receiver is large; in addition, during the period when the network device enters the quiescent State (Quiet State) of the Low Power Idle (Low Idle) mode of the Energy-saving Ethernet (EEE), the cancellation signal generating circuit may not track the change of the phase offset, which makes the cancellation signal generating circuit unable to generate an appropriate cancellation signal when leaving the quiescent State, and causes the signal-to-noise ratio of the receiver to decrease.

Disclosure of Invention

An objective of the present invention is to provide an interference canceller and an interference cancellation method for dealing with the interference problem caused by the transmission signal.

The interference canceller of the present invention can generate a cancellation signal for reducing cross-port alien-end crosstalk (XPAN) of a received signal of a receiver of a network device. One embodiment of the interference canceller includes an oversampling circuit, a frequency difference calculating circuit, a sample rate converter, and an XPAN canceller. The over-sampling circuit is used for over-sampling an XPAN transmission signal to generate an over-sampled signal, wherein the XPAN transmission signal is originated from a transmitter of the network device, and the frequency of the over-sampled signal is higher than that of a receiving frequency of the receiver. The frequency difference calculating circuit is used for calculating a frequency difference between a transmitting frequency of the transmitter and the receiving frequency of the receiver. The sampling rate converter is used for processing the over-sampled signal according to the frequency difference to generate a converted signal, wherein the frequency of the converted signal is lower than that of the over-sampled signal, and the frequency of the converted signal is equal to that of the receiving frequency. The XPAN canceller is configured to generate the cancellation signal according to the converted signal, and further configured to determine at least one coefficient of the XPAN canceller according to an error signal, wherein the error signal is dependent on a difference between the XPAN of the received signal and the cancellation signal.

The interference cancellation method of the present invention is performed by a network device, and the method is capable of generating a cancellation signal for reducing cross-port heterogeneous near-end crosstalk (XPAN) of a received signal of a receiver of the network device. One embodiment of the method comprises the steps of: oversampling an XPAN transmission signal to generate an oversampled signal, wherein the XPAN transmission signal originates from a transmitter of the network device, and the frequency of the oversampled signal is higher than that of a receiving frequency of the receiver; calculating a frequency difference between a transmitting frequency and the receiving frequency of the transmitter; processing the over-sampled signal according to the frequency difference to generate a converted signal, wherein the frequency of the converted signal is lower than that of the over-sampled signal, and the frequency of the converted signal is equal to that of the receiving frequency; and using a filter to generate the cancellation signal according to the converted signal, and determining at least one coefficient of the filter according to an error signal, wherein the error signal depends on a difference between the XPAN of the received signal and the cancellation signal.

The features, operation and efficacy of the present invention will be described in detail below with reference to the accompanying drawings.

Drawings

Fig. 1 shows an embodiment of a network device comprising the interference canceller of the present invention;

fig. 2 shows an embodiment of the interference canceller of the present invention;

FIG. 3 shows an embodiment of the sample rate converter of FIG. 2;

FIG. 4 shows an example of the relationship between the over-sampled signal and the converted signal of FIG. 3;

FIG. 5 shows another example of the relationship between the over-sampled signal and the converted signal of FIG. 3;

FIG. 6 shows an embodiment of the receiver of FIG. 2;

fig. 7 shows another embodiment of the interference canceller of the present invention;

FIG. 8a shows an example of the trend of the energy of the converted signal and the XPAN transmission signal of FIG. 7;

FIG. 8b is a diagram showing another example of the trend of the energy of the converted signal and the XPAN transmission signal of FIG. 7;

FIG. 9 shows one embodiment of the steps performed by the calculation and statistics circuitry of FIG. 7;

FIG. 10 shows another embodiment of the steps performed by the compute and statistic circuitry of FIG. 7; and

FIG. 11 shows an embodiment of the interference cancellation method of the present invention.

Detailed Description

The invention discloses an interference eliminator and an interference elimination method, which can achieve better interference elimination performance when the time sequence difference between a transmitter and a receiver is larger, and can track the change of the relation between an interference signal and an elimination signal in the period of a Low Power Idle (Low Power Idle) mode of an Energy-saving Ethernet (EEE).

Fig. 1 shows an embodiment of a network device comprising the interference canceller of the present invention. The network device 100 of fig. 1 comprises a digital transmitting circuit 110, an analog transmitting circuit 120 (including a digital-to-analog converter), a mixing circuit 130, an analog receiving circuit 140 (including an analog-to-digital converter), a digital receiving circuit 150, and an interference canceller 160 of the present invention, wherein the digital transmitting circuit 110 and the analog transmitting circuit 120 belong to a transmitter (operating in a first frequency domain) of a first transceiver, and the analog receiving circuit 140 and the digital receiving circuit 150 belong to a receiver (operating in a second frequency domain) of a second transceiver, the first transceiver and the second transceiver being different. The interference canceller 160 outputs a cancellation signal to the digital receiving circuit 150 according to the output signal of the digital transmitting circuit 110 and the feedback signal of the digital receiving circuit 150, so that the digital receiving circuit 150 cancels the interference component in the received signal output by the analog receiving circuit 140.

Fig. 2 shows an embodiment of the interference canceller of the present invention. The interference canceller 200 of fig. 2 is capable of generating a cancellation signal for reducing cross-port alien-end crosstalk (NEXT) of a received signal (e.g., the output of the analog receiving circuit 140 of fig. 1) of a receiver of a network device. The interference canceller 200 includes an oversampling circuit 210, a frequency difference calculating circuit 220, a sampling rate converter 230, and an XPAN canceller 240. The over-sampling circuit 210 is used to over-sample an XPAN transmit signal (TX) _XPAN ) (for example: the output of the digital transmission circuit 110 of fig. 1) to generate an over-sampled signal (TX) _{XPAN_UP} ) (ii) a The XPAN transport signal originates from a transmitter of the network device; the frequency of the over-sampled signal is higher than a receiving frequency (RX) of the receiver _CLK ) In order for the output of the sample rate converter 230 to include mixing (aliasing) for the XPAN canceller 240 to generate a mixing that is similar to the received signal. The frequency difference calculating circuit 220 is used for calculating a transmission frequency (TX) of the transmitter _CLK ) And a frequency difference (CLK) between the received frequencies _DIFF ). The sampling rate converter 230 is used for processing the over-sampled signal according to the frequency difference to generate a converted Signal (SRC) _OUT ) Wherein the frequency of the converted signal is lower than the frequency of the over-sampled signal and equal to the frequency of the received signal, so as to eliminate/reduce the phase offset between the XPAN transmitted signal and the received signal. XPAN canceller 240 is configured to respond to the transitionGenerating the cancellation signal (XPAN) by switching signals _CANCEL ) Is used for receiving an error signal (S) _ERROR ) At least one coefficient of XPAN canceller 240 is determined in an attempt to reduce the error indicated by a subsequent error signal that depends on the difference between the XPAN and the cancellation signal of the received signal (as shown in fig. 6).

Please refer to fig. 2. The over-sampling circuit 210 of fig. 2 can over-sample the XPAN transmitted signal using known techniques (e.g., zero-order hold (zero-order hold) or raised cosine (raised cosine) techniques) or self-developed techniques, wherein the zero-order hold technique can be used to duplicate an input signal (e.g., a signal with an input AB) to generate an output signal (e.g., a signal with an output AABB). In one embodiment, the network device is an Ethernet device conforming to the 2.5G BASE-T standard; the receiver receives signals through one connection end (such as a port) of the Ethernet device, and the transmitter transmits signals through the other connection end (such as a port) of the Ethernet device and causes interference to the signals received by the receiver; the receiving rate of the receiver is 2.5 gbits/sec, the transmission speed of the transmitter is also 2.5 gbits/sec, the receiving frequency (here, equivalent to Baud) is 200MHz, the transmitting frequency (here, equivalent to Baud) is 200MHz plus a deviation, and the oversampling circuit 210 can generate the oversampling signal at twice the transmitting frequency, so that the frequency of the oversampling signal is twice the frequency of the transmitting frequency and higher than the frequency of the receiving frequency. In one embodiment, the receiving rate of the receiver is 2.5 gbits/s, the transmission speed of the transmitter is 1 gbits/s, the receiving frequency is 200MHz, and the transmitting frequency is 125MHz plus a deviation, the oversampling circuit 210 can generate the oversampling signal at twice the transmitting frequency, such that the frequency of the oversampling signal is twice the transmitting frequency and higher than the receiving frequency. The implementation and other embodiments of the over-sampling circuit 210 can be derived from the above description.

Please refer to fig. 2. The frequency difference calculation circuit 220 of fig. 2 may derive the frequency difference by known or self-developed techniques. In one embodiment, the frequency difference calculating circuit 220 compares the frequency number of the receiving frequency with the frequency number of the transmitting frequency in the same time period to obtain a ratio of the two frequency numbers representing the frequency difference. For example, the frequency of the receiving frequency is 200MHz, the frequency of the transmitting frequency is 200MHz plus a deviation, and the frequency difference calculating circuit 220 can calculate the value of x according to the ratio of the two frequency numbers 1; for example, the frequency of the receiving frequency is 200MHz, the frequency of the transmitting frequency is 125MHz plus a deviation, and the frequency difference calculating circuit 220 can calculate the value of y according to the ratio of the two frequency numbers 8 + 5+ y (200 MHz, (125 MHz + deviation)). The implementation of the frequency difference calculating circuit 220 and other embodiments can be derived from the above description.

FIG. 3 shows an embodiment of the sample rate converter 230 of FIG. 2. The sample rate converter 230 of fig. 3 comprises: a sample-selector (sample-selector) 310 for selecting K output values of the over-sampled signal according to the frequency difference to generate a selection result (SEL), wherein K is a positive integer; and an interpolator (interpolator) 320 for receiving the selection result and performing interpolation according to the frequency difference and the over-sampled signal to generate K output values of the converted signal corresponding to the over-sampled signal, wherein the frequency of the converted signal is equal to the frequency of the received frequency.

Please refer to fig. 3. In one embodiment, the frequency of the transmission frequency is 125MHz, the frequency of the over-sampled signal is 250MHz, and the frequency of the reception frequency is 200MHz, so that, in the same period of time, the ratio of the number of output values of the over-sampled signal to the number of output values of the converted signal is 1.25 (which is related to the frequency difference), so for the sampling rate converter 230, the time for receiving five output values of the over-sampled signal should be equal to the time for outputting four output values of the converted signal; as described above, as shown In FIG. 4, the sampling selector 310 selects the first four (In (1), in (2), in (3), in (4)) of the five output values (In (1), in (2), in (3), in (4)) of the over-sampled signal according to the frequency difference (i.e., K output values of the over-sampled signal) without selecting the fifth output signal (In (5)) to generate the selection result, and the interpolator 320 is connected to the interpolator 320Receiving all the output values of the over-sampled signal and the selection result, and performing interpolation with the over-sampled signal according to the frequency difference, thereby generating four output values (Out (1), out (2), out (3), out (4)) of the converted signal respectively corresponding to the four output values (In (1), in (2), in (3), in (4)) of the over-sampled signal; more specifically, as shown in FIG. 4, the interval between any two adjacent output values of the over-sampled signal is T _IN1 The timing of Out (1) is aligned with that of In (1), and the interpolator 320 can know that the difference between the timing of Out (2) and the timing of In (2) is 0.25T according to the frequency difference _IN1 Out (3) timing differs from In (3) timing by 0.5T _IN1 And the timing of Out (4) differs from the timing of In (4) by 0.75T _IN1 Therefore, the interpolator 320 can generate four output values (Out (1), out (2), out (3), out (4)) of the converted signal by using the appropriate output values of the over-sampled signal according to the above-mentioned timing relationship and a known or self-developed interpolation algorithm. The subsequent selection and interpolation of the sample rate converter 230 can be deduced from the above description. It is noted that, for simplicity, the interpolator 320 may also selectively receive/use a portion of the output value of the over-sampled signal to perform interpolation according to the frequency difference and the portion of the output value of the over-sampled signal.

Please refer to fig. 3. In one embodiment, the frequency of the transmission frequency is 200.02MHz, the frequency of the over-sampled signal is 400.04MHz, and the frequency of the reception frequency is 200MHz, so that, in the same period of time, the ratio of the number of output values of the over-sampled signal to the number of output values of the converted signal is 2.0002 (which can be regarded as twice the frequency difference), so that for the sampling rate converter 230, the time for receiving twenty-zero two output values of the over-sampled signal should ideally be equal to the time for outputting ten thousand output values of the converted signal; as described above, as shown In FIG. 5, the sampling selector 310 selects ten thousand (In (1), in (3), in (5), \8230;, in (19996), in (19998), in (20002) of the two ten-thousand output values of the over-sampled signal according to the frequency difference, in which In (20000) and In (20001) are skipped (i.e., K output values of the over-sampled signal) to generate the selection result, and the interpolator 320 receives all the output values of the over-sampled signalOutput values and the selection result to perform interpolation with the over-sampled signal according to the frequency difference, thereby generating ten thousand output values (Out (1), out (2), \8230;, out (9999), out (10000)) of the converted signal respectively corresponding to ten thousand output values (In (1), in (3), in (5), \8230;, in (19996), in (19998), in (20002)) of the over-sampled signal; more specifically, as shown in FIG. 5, the interval between any two adjacent odd/even output values of the over-sampled signal is T _IN2 The timing of Out (1) is aligned with that of In (1), and the interpolator 320 can know that the difference between the timing of Out (2) and the timing of In (3) is 0.0002T _IN2 Out (3) timing differs from In (5) timing by 0.0004T _IN2 And so on, interpolator 320 generates ten thousand output values (Out (1), out (2), \ 8230;, out (9999), out (10000)) of the converted signal according to the timing relationships and a known or self-developed interpolation algorithm.

Please refer to fig. 1-3. In one embodiment, when the network device including the interference canceller 200 of the present invention enters the low power Idle mode of the energy-saving ethernet network, the sample rate converter 230 can generate the transition signal when leaving the Idle State (e.g., when the network device enters a recovery State or an Idle State) according to the frequency difference, a time length of a quiescent State (Quiet State) of the EEE mode, and a relationship between the transition signal and the over-sampled signal before entering the quiescent State (e.g., before entering the quiescent State, the sample rate converter 230 outputs the transition signal Out (5000) of fig. 5 corresponding to the over-sampled signal In (10000)); more specifically, the sampling rate converter 230 estimates a number of output values of the converted signal that should be output when the mobile device leaves the quiescent state according to the frequency difference and the time duration, and generates the converted signal accordingly. Since the relevant specifications of EEEs are common knowledge in the art, the details thereof are omitted here.

Please refer to fig. 2. In one embodiment, XPAN canceller 240 is a known Least Mean Square (LMS) filter or a known Recursive Least Square (RLS) filter. XPAN canceller 240 may adjust the coefficients of XPAN canceller 240 according to the change in the error signal to further reduce the error indicated by the subsequent error signal; the smaller the error, the better the interference cancellation results. In one embodiment, the XPAN canceller 240 is located in the network device, and is able to receive the error signal as soon as possible. In one embodiment, the step size of the XPAN canceller 240 (e.g., the step size of the LMS filter) is set as much as possible to cope with various interference cancellation situations.

Please refer to fig. 1 and fig. 2. One embodiment of the receiver of fig. 2 is shown in fig. 6, and includes the following circuitry: a first circuit 610 (e.g. an adder/subtractor) for receiving the signal (RXD) in dependence on the received signal (RXD) _{_RAW} ) The difference between the cancellation signal and the input signal (RXD) generates an input signal (RXD) _IN ) (ii) a A slicer 620 for generating a series of level values according to the input signal to form an output signal (RXD) _OUT ) (ii) a And a second circuit 630 (e.g., an adder/subtractor) for outputting the error signal to the XPAN canceller 240 according to a difference between the input signal and the output signal. In one embodiment, the error signal indicates a smaller error when the difference between the input signal and the output signal is smaller. Embodiments of the first circuit 610, the cutter 620, and the second circuit 630 are described in U.S. patent No. US9252994B 2.

Fig. 7 shows another embodiment of the interference canceller of the present invention. Compared to the interference canceller 200 of fig. 2, the interference canceller 700 of fig. 7 further comprises a calculating and statistics circuit 710, wherein the path from the output of the sample rate converter 230 to the calculating and statistics circuit 710 is selective, depending on implementation requirements. The calculating and counting circuit 710 is coupled to the sampling rate converter 230 and the XPAN canceller 240, and is used to make the energy variation trend of the converted signal consistent with the energy variation trend of the XPAN transmission signal as shown in fig. 8a, thereby reducing the workload of the XPAN canceller 240. In other words, if the energy variation trend of the converted signal is not consistent with the energy variation trend of the XPAN transmission signal as shown in fig. 8b, which means that the variation of the converted signal is not synchronous with the variation of the XPAN transmission signal, the XPAN canceller 240 needs more time and calculation amount to adjust the coefficients of the XPAN canceller 240, so as to generate an appropriate cancellation signal according to the converted signal to cancel the interference component in the XPAN transmission signal.

Please refer to fig. 7. In one embodiment, the frequency difference between the transmitting frequency and the receiving frequency is composed of a main difference and an offset (e.g., when the frequencies of the transmitting frequency and the receiving frequency are 200.02MHz and 200MHz, respectively, the main difference depends on the difference of the integer part of the two frequencies (| 200MHz-200MHz | = 0), the bias depends on the difference of the fractional part of the two frequencies (| 0.02MHz-0MHz | =0.02 MHz)), the sample rate converter 230 is configured to process the over-sampled signal according to the main difference and the offset to generate the converted signal in an operating mode (e.g., fig. 5), in which the frequency of the converted signal is equal to the frequency of the receiving frequency; the sample rate converter 230 is further configured to selectively process the over-sampled signal according to only the main difference to generate the converted signal in a calibration mode (e.g., modified in fig. 5 to simply output odd/even output values of the over-sampled signal as output values of the converted signal) such that the trend of the energy of the cancellation signal output by the XAPN canceller 240 according to the converted signal approximates the trend of the energy of the XAPN transmit signal, wherein the frequency of the converted signal is equal to the frequency of the receive frequency in the calibration mode. In the above embodiment, the calculation and statistics circuit 710 is used to perform the steps of FIG. 9:

step S910: in the calibration mode, after the calibration mode starts for a predetermined time, the intensity of the cancellation signal is calculated and recorded according to a first time interval to obtain M records, wherein M is an integer greater than two. As the XPAN canceller 240 continuously adjusts the coefficients of the XPAN canceller 240 according to the error signal during the predetermined time, the variation of the cancellation signal converges, which means that the energy variation trend of the cancellation signal approximates the energy variation trend of the XPAN transmission signal. For example, when the transmitting frequency and the receiving frequency are 200.02MHz and 200MHz, respectively, the time length of ten thousand zero output values of the XPAN transmitting signal should be equivalent to the time length of ten thousand output values of the cancellation signal; accordingly, the calculating and counting circuit 710 can take the time length of five hundred output values as the first time interval to calculate and record the strength of the cancellation signal, and obtain at least 20 records to complete sampling of at least ten thousand output values of the cancellation signal for one output value of ten thousand corresponding to one output value of ten thousand of XPAN transmission signals, wherein the 20 records include a maximum value. It is noted that the calculation of the signal strength is common knowledge in the art, and the details thereof are omitted here.

Step S920: obtaining the energy variation trend of the elimination signal according to the M records, and outputting the energy variation trend (PW) of the elimination signal _TREND ) (or calculation and statistical results) to the sample rate converter 230, so that the sample rate converter 230 generates the converted signal according to the energy variation trend of the cancellation signal when the operation mode is started, and the energy variation trend of the converted signal is close to the energy variation trend of the cancellation signal (approximate to the energy variation trend of the XPAN transmission signal) in the operation mode. For example, the frequencies of the transmitting frequency and the receiving frequency are 200.02MHz and 200MHz, the time length associated with 20 records of the calculating and counting circuit 710 is equal to one period of the energy variation trend of the cancellation signal, the 6 th record of the 20 records is known as a maximum value by the sampling rate converter 230 according to the output of the calculating and counting circuit 710, and then the sampling rate converter 230 estimates the time point when the energy of the cancellation signal reaches the maximum peak next time, according to the main difference and deviation of the frequency difference and according to the position of the maximum value in the period of the energy variation trend

The oversampled signal is interpolated to generate and output the maximum value of the transformed signal (e.g., 10000 × 0.3 output value Out (3000) in fig. 5), so that the energy variation trend of the transformed signal is consistent with the energy variation trend of the XPAN transmission signal, thereby improving the operation performance of the XPAN canceller 240.

Please refer to fig. 7. In another embodiment, the calculation and statistics circuitry 710 is configured to perform the steps of FIG. 10 in the operational mode:

step S1010: and calculating and recording the intensity of the conversion signal and the intensity of the elimination signal according to a second time interval so as to know whether the energy variation trend of the conversion signal deviates from the energy variation trend of the elimination signal. The second time interval is equal to or unequal to the first time interval. The calculation and recording of the signal strength can be seen in the description of the embodiment of fig. 9.

Step S1020: if the energy variation trend of the converted signal deviates from the energy variation trend of the cancellation signal, the sampling rate converter 230 is notified to adjust the generation of the converted signal according to the energy variation trend of the cancellation signal, so that the energy variation trend of the converted signal approaches the energy variation trend of the cancellation signal. The adjustment of the generation of the converted signal can be seen from the description of the embodiment of fig. 9.

Fig. 11 shows an embodiment of the interference cancellation method according to the present invention, which is performed by a network device (e.g., the network device). The embodiment is capable of generating a cancellation signal for reducing cross-port heterogeneous near-end crosstalk (XPAN) of a received signal of a receiver of the network device, comprising the steps of:

step S1110: oversampling an XPAN transmission signal to generate an oversampled signal, wherein the XPAN transmission signal originates from a transmitter of the network device, and a frequency of the oversampled signal is higher than a frequency of a receiving frequency of the receiver. This step may be performed by the over-sampling circuit 210 of fig. 2 or fig. 7.

Step S1120: a frequency difference between a transmitting frequency of the transmitter and the receiving frequency of the receiver is calculated. This step may be performed by the frequency difference calculation circuit 220 of fig. 2 or fig. 7.

Step S1130: the over-sampled signal is processed according to the frequency difference to generate a converted signal, wherein the frequency of the converted signal is lower than that of the over-sampled signal, and the frequency of the converted signal is equal to that of the receiving frequency. This step may be performed by the sample rate converter 230 of fig. 2 or fig. 7.

Step S1140: a filter is used to generate the cancellation signal based on the transformed signal, and at least one coefficient of the filter is determined based on an error signal, wherein the error signal is dependent on a difference between the XPAN of the received signal and the cancellation signal. One embodiment of this filter is XPAN canceller 240 of fig. 2 or fig. 7.

Since the details and variations of the present invention can be understood by those skilled in the art with reference to the disclosure of the present invention, the technical features of the present invention can be reasonably applied to the present invention, and therefore, the repeated and redundant description thereof will be omitted herein.

It is to be noted that, when the implementation is possible, a person skilled in the art can selectively implement some or all of the technical features of any one of the above embodiments, or selectively implement a combination of some or all of the technical features of the above embodiments, thereby increasing the flexibility in implementing the invention. It is noted that the term "A equals B" in this specification is based on engineering considerations and can be interpreted as meaning that A is completely equal to B or that there is a negligible error between A and B; in addition, the term "a approaches B" in this specification may be interpreted as the difference between a and B is less than a predetermined threshold.

In summary, the cancellation signal includes the mixing and phase offset similar to the received signal, so as to achieve better interference cancellation performance when the timing difference between the transmitter and the receiver is large; in addition, the invention can track the change of the relationship between the interference signal and the elimination signal in the period of the low power consumption idle mode of the energy-saving Ethernet.

Although the embodiments of the present invention have been described above, these embodiments are not intended to limit the present invention, and those skilled in the art can make variations on the technical features of the present invention according to the explicit or implicit contents of the present invention, and all such variations may fall within the scope of the patent protection sought by the present invention.

[ description of reference ]

100. Network device

110. Digital transmission circuit

120. Analog transmission circuit

130. Hybrid circuit

140. Analog receiving circuit

150. Digital receiving circuit

160. Interference eliminator

200. Interference eliminator

210. Oversampling circuit

220. Frequency difference calculating circuit

230. Sampling rate converter

240 XPAN eliminator

TX _XPAN XPAN transport signal

TX _{XPAN_UP} Over-sampled signal

RX _CLK Receiving frequency

TX _CLK Frequency of transmission

CLK _DIFF Frequency difference

SRC _OUT Switching signals

XPAN _CANCEL Cancelling signals

S _ERROR Error signal

310. Sampling selector

320. Interposer

SEL selection results

Output values of In (1), in (2), in (3), in (4), in (5) oversampling signals

Out (1), out (2), out (3), out (4) conversion signal output values

T _IN1 Spacing between adjacent output values of oversampled signals

Output values of In (1), in (3), in (5), \ 8230;, in (19996), in (19998) and In (20002) oversampled signals

Out (1), out (2), out 8230, out (9999) and Out (10000) conversion signals

T _IN2 Spacing between adjacent output values of oversampled signals

610. First circuit

620. Cutter

630. Second circuit

RXD _{_RAW} Receiving a signal

RXD _IN Input signal

RXD _OUT Output signal

710. Calculation and statistics circuit

PW _TREND Eliminating energy variation trend of signal

S910 to S920

S1010-S1020 steps

S1110 to S1140.

Claims

1. An interference canceller capable of generating a cancellation signal for reducing cross-port alien near-end crosstalk (XPAN) of a received signal of a receiver of a network device, the interference canceller comprising:

an oversampling circuit for oversampling an XPAN transmission signal to generate an oversampled signal, wherein the XPAN transmission signal originates from a transmitter of the network device, and the frequency of the oversampled signal is higher than that of a receiving frequency of the receiver;

a frequency difference calculating circuit for calculating a frequency difference between a transmitting frequency of the transmitter and the receiving frequency of the receiver;

a sampling rate converter for processing the over-sampled signal according to the frequency difference to generate a converted signal, wherein the frequency of the converted signal is lower than the frequency of the over-sampled signal and equal to the frequency of the receiving frequency; and

an XPAN canceller configured to generate the cancellation signal according to the converted signal, the XPAN canceller further configured to determine at least one coefficient of the XPAN canceller according to an error signal, wherein the error signal is dependent on a difference between the XPAN of the received signal and the cancellation signal.

2. The interference canceller in claim 1, wherein the oversampling circuit employs a zero-order hold or raised cosine technique to generate the oversampling signal.

3. The interference canceller of claim 1, wherein the sample rate converter comprises:

a sampling selector for selecting K output values of the over-sampled signal according to the frequency difference, wherein K is a positive integer; and

an interpolator for performing interpolation according to the frequency difference and the over-sampled signal to generate K output values of the converted signal, the K output values of the converted signal corresponding to the K output values of the over-sampled signal.

4. The interference canceller of claim 1, wherein the XPAN canceller is a least mean square filter or a recursive least square filter.

5. The interference canceller of claim 1, wherein the receiver includes circuitry to provide the error signal:

a first circuit for generating an input signal according to a difference between the received signal and the cancellation signal;

a cutter for generating an output signal according to the input signal; and

a second circuit for outputting the error signal to the XPAN canceller according to the difference between the input signal and the output signal.

6. The interference canceller of claim 1, further comprising:

a calculating and counting circuit coupled to the sample rate converter and the XPAN canceller for performing the following steps:

in a correction mode, after the correction mode starts for a preset time, calculating and recording the intensity of the elimination signal according to a first time interval to obtain M records, wherein M is an integer greater than two; and

and obtaining the energy variation trend of the elimination signal according to the M records, and outputting the energy variation trend of the elimination signal to the sampling rate converter so as to generate the conversion signal according to the energy variation trend of the elimination signal when the sampling rate converter starts an operation mode, so that the energy variation trend of the conversion signal approaches to the energy variation trend of the elimination signal in the operation mode.

7. The interference canceller according to claim 6, wherein said calculating and statistics circuitry is further adapted to perform the following steps in said operational mode:

calculating and recording the intensity of the conversion signal and the intensity of the elimination signal according to a second time interval so as to know whether the energy variation trend of the conversion signal deviates from the energy variation trend of the elimination signal; and

if the energy variation trend of the conversion signal deviates from the energy variation trend of the elimination signal, the sampling rate converter is informed, so that the sampling rate converter adjusts the generation of the conversion signal according to the energy variation trend of the elimination signal, and the energy variation trend of the conversion signal approaches the energy variation trend of the elimination signal.

8. The interference canceller of claim 1, further comprising:

calculating and recording the intensity of the conversion signal and the intensity of the elimination signal according to a time interval so as to know whether the energy change trend of the conversion signal deviates from the energy change trend of the elimination signal; and

if the energy variation trend of the conversion signal deviates from the energy variation trend of the elimination signal, the sampling rate converter is informed to adjust the generation of the conversion signal according to the energy variation trend of the elimination signal, so that the energy variation trend of the conversion signal approaches the energy variation trend of the elimination signal.

9. The interference canceller as claimed in claim 1, wherein the sample rate converter generates the transition signal when leaving the quiescent state according to the frequency difference, a time length of a quiescent state of the low-power idle mode of the EEE, and a relationship between the transition signal and the over-sampled signal before entering the quiescent state after the receiver enters the low-power idle mode of an energy-saving Ethernet EEE.

10. An interference cancellation method performed by a network device, the method capable of generating a cancellation signal for reducing cross-port alien crosstalk (XPAN) of a received signal of a receiver of the network device, the method comprising:

oversampling an XPAN transmission signal to generate an oversampled signal, wherein the XPAN transmission signal originates from a transmitter of the network device, and the frequency of the oversampled signal is higher than that of a receiving frequency of the receiver;

calculating a frequency difference between a transmitting frequency of the transmitter and the receiving frequency of the receiver;

processing the over-sampled signal according to the frequency difference to generate a converted signal, wherein

The frequency of the converted signal is lower than that of the over-sampled signal and equal to that of the receiving frequency; and

a filter is used to generate the cancellation signal based on the transformed signal, and at least one coefficient of the filter is determined based on an error signal, wherein the error signal is dependent on a difference between the XPAN of the received signal and the cancellation signal.