[미국특허]
Far-end crosstalk canceling method and device, and signal processing system
원문보기
IPC분류정보
국가/구분
United States(US) Patent
등록
국제특허분류(IPC7판)
H04B-003/32
H04B-015/00
H04J-001/12
H04L-027/28
출원번호
US-0271903
(2011-10-12)
등록번호
US-8295369
(2012-10-23)
우선권정보
CN-2006 1 0162179 (2006-12-07)
발명자
/ 주소
Fang, Liming
출원인 / 주소
Huawei Technologies Co., Ltd.
대리인 / 주소
Leydig, Voit & Mayer, Ltd.
인용정보
피인용 횟수 :
4인용 특허 :
2
초록▼
A method for remote crosstalk cancellation, includes: filtering, at a first transmitting end which is connected to a first receiving end over a first line, crosstalk source signals according to values of filtering parameters; synthesizing, at the first transmitting end, a first signal to be transmit
A method for remote crosstalk cancellation, includes: filtering, at a first transmitting end which is connected to a first receiving end over a first line, crosstalk source signals according to values of filtering parameters; synthesizing, at the first transmitting end, a first signal to be transmitted over the first line and the filtered crosstalk source signals; transmitting the synthesized signal from the first transmitting end to the first receiving end over the first line; receiving, at the first transmitting end, a noise statistic related value fed back from the first receiving end; and adjusting, at the transmitting end, the values of the filtering parameters according to the noise statistic related value. A corresponding device and a signal processing system for remote crosstalk cancellation are provided.
대표청구항▼
1. A far-end crosstalk canceling method, comprising: filtering, at a first transmitting end which is connected to a first receiving end over a first line, crosstalk source signals according to values of filtering parameters, wherein the crosstalk source signals are related to signals to be transmitt
1. A far-end crosstalk canceling method, comprising: filtering, at a first transmitting end which is connected to a first receiving end over a first line, crosstalk source signals according to values of filtering parameters, wherein the crosstalk source signals are related to signals to be transmitted over second lines, each of the second lines having a second transmitting end connected to a second receiving end;synthesizing, at the first transmitting end, a first signal to be transmitted over the first line with the filtered crosstalk source signals;transmitting the synthesized first signal from the transmitting end to the first receiving end over the first line;receiving, at the transmitting end, a noise statistic related value fed back from the first receiving end, wherein the noise statistic related value reflects a numeric value of a signal-to-noise ratio of the synthesized first signal received by the first receiving end;adjusting, at the transmitting end, the values of the filtering parameters according to the noise statistic related value,wherein the adjusted values of the filtering parameters are used in filtering the crosstalk source signals. 2. The method according to claim 1, wherein the step of filtering the crosstalk source signals comprises: filtering frequency-domain components of the crosstalk source signals per sub-carrier in a frequency domain, wherein the noise statistic related value includes components corresponding to respective frequency bands to which respective sub-carriers belong; wherein the step of synthesizing the first signal with the filtered crosstalk signals comprises: synthesizing frequency-domain components of the first signal with the filtered crosstalk source signals per sub-carrier in the frequency domain. 3. The method according to claim 2, wherein the step of synthesizing frequency-domain components of the first signal and the filtered crosstalk source signals per sub-carrier in the frequency domain comprises: filtering the frequency-domain components of the first signal over the respective sub-carriers according to one of the filtering parameters that can be updated according to a feedback from the first receiving end;synthesizing the filtered frequency-domain components of the filtered first signal with the filtered crosstalk source signals per sub-carrier in the frequency domain. 4. The method according to claim 2, wherein after synthesizing the frequency-domain components of the first signal with the filtered crosstalk source signals per sub-carrier in the frequency domain, the method further comprising: filtering the synthesized frequency-domain components according to one of the filtering parameters per sub-carrier in the filtered frequency domain, wherein the value of the filtering parameter can be adjusted according to a feedback from the first receiving end. 5. The method according to claim 1, wherein the step of filtering crosstalk source signals comprises filtering the crosstalk source signals by performing a time-domain filtering on each of the crosstalk source signals. 6. The method according to claim 5, wherein the first signal is an output of a time-domain filter filtering a signal to be pre-coded using one of the filtering parameters that can be derived from a feedback from the first receiving end. 7. The method according to claim 5, wherein the first signal is an output of a delay filter delaying a signal to be pre-coded. 8. The method according to claim 1, wherein the step of adjusting the values of the filtering parameters according to the noise statistic related value comprises performing the following steps until the values of the filtering parameters converge: iterating current values of the filtering parameters respectively with a plurality of gradient vectors;determining a gradient vector that maximizes a signal-to-noise ratio according to the noise statistic related value corresponding to the iterated sets of the filtering parameters fed back from the first receiving end; andusing the iterated values of the filtering parameters iterated with the gradient vector as updated current values of the filtering parameters. 9. The method according to claim 8, wherein the current values of the filtering parameters are iterated with the gradient vectors in the following formula: w(n+1)=w(n)+λ×sign[e(n)]×u(n)where w(n+1) and w(n) denote values of the filtering parameters after and before the (n+1)th iteration respectively, λ denotes a selected step, sign[e(n)] denotes a sign of an estimated noise error, and u(n) denotes input signals filtered using the values of the filtering parameter w(n). 10. The method according to claim 1, wherein the step of adjusting the values of the filtering parameters according to the noise statistic related value comprises performing the following steps iteratively until the values of the respective filtering parameters converge: fixing values of other sets of filtering parameters except one set of filtering parameters to be adjusted;searching for update values for the set of filtering parameters to be adjusted to reduce the crosstalk according to the noise statistic related value fed back from the first receiving end; andusing the update values as updated current values of the set of filtering parameters to be adjusted, and proceeding with searching for updated values of each of the remaining sets of filtering parameters. 11. The method according to claim 10, wherein searching for the update values for the set of filtering parameters to be adjusted to reduce the crosstalk increase comprises: comparing variations of the noise statistic reflected by the noise statistic related value fed back from the first receiving end if values of the set of filtering parameters to be adjusted are w+λ, w, and w−λ respectively, where w denotes values of the set of the filtering parameters to be adjusted before adjusting, and λ denotes a selected step; andif the values corresponding to the minimum noise statistic are w, reducing the step λ for further comparison until the values corresponding to the minimum noise statistic are other than w, and using the values corresponding to the minimum noise statistic as the updated values of the set of the filtering parameters to be adjusted. 12. The method according to claim 10, wherein searching for the update values of the set of filtering parameters to be adjusted to reduce the crosstalk comprises: searching for the values of the set of the filtering parameters to be adjusted in a certain direction with a selected step λ, and changing the search direction if the numeric value of the noise statistic has an incremental variation;recording the last three search values of the set of filtering parameters to be adjusted if a decremented variation of the numeric value of the noise statistic changes to the incremental variation thereof between the two adjacent searches;fitting a curve of a function relationship between the values of the set of the filtering parameters to be adjusted and the noise statistic according to the last three search values of the set of filtering parameters to be adjusted and the corresponding numeric value of the noise statistic;estimating filtering parameter values corresponding to the minimum noise statistic as the update values of the set of the filtering parameters to be adjusted according to the fitted curve; andshortening the step λ to search for the values of the set of filtering parameters to be adjusted in a next round of cyclic search process. 13. The method according to claim 12, wherein a curve fitting formula to estimate the update values of the set of filtering parameters to be adjusted is: w^=((emid-emin)×(wmax-wmin)emax-emin+wmin+wmid)/2,where ŵ denotes the update values of the set of the filtering parameters to be adjusted resulting from fitting, wmax, wmid and wmin denote the filtering parameter values in the last three search values of the set of filtering parameters to be adjusted corresponding to the maximum, median and minimum numerical value of the noise statistic respectively, and emax, emid and emin denote the numeric value of the noise statistic corresponding to wmax, wmid and wmin respectively. 14. A far-end crosstalk canceling device, which is applied in a first signal transmitting end connected over a first line to a first signal receiving end, comprising a plurality of adaptive filters, a signal synthesis module, a signal transmitting module and a feedback receiving module, wherein: each of the plurality of adaptive filters is configured to filter a crosstalk source signal according to a value of a filtering parameter which can be adjusted according to a noise statistic related value received by the feedback receiving module, wherein the crosstalk source signals are related to signals to be transmitted over second lines, each of the second lines having a second transmitting end connected to a second receiving end;the signal synthesis module is configured to synthesize a first signal to be transmitted over the first line with the filtered crosstalk source signals;the signal transmitting module is configured to transmit the synthesized first signal to the first receiving end; andthe feedback receiving module is configured to receive a noise statistic related value fed back from the first receiving end, wherein the noise statistic related value reflects a numeric value of a signal-to-noise ratio of the synthesized first signal received by the first receiving end. 15. The device according to claim 14, further comprising a delay filter connected to the signal synthesis module, configured to delay the first signal and send the delayed first signal to the signal synthesis module. 16. The device according to claim 14, further comprising an adaptive time-domain filter connected to the signal synthesis module, configured to perform a time-domain filtering on the first signal, which is received from a digital signal input unit, according to values of time-domain filtering parameters, and send the filtered first signal to the signal synthesis module; and adjust the values of the filtering parameters of the adaptive time-domain according to the noise statistic related value received by the feedback receiving module. 17. The device according to claim 14, further comprising: a modulating unit connected, respectively, to the signal synthesis module and to the plurality of adaptive filters, configured to modulate components of the first signal and components of the crosstalk source signals over respective sub-carriers, and transmit components of the modulated first signal to the signal synthesis module and components of the modulated crosstalk source signals to the plurality of the adaptive filters; anda signal output unit connected to the signal synthesis module, configured to receive components over the sub-carriers synthesized by the signal synthesis module, and perform an Inverse Fast Fourier Transformation on received frequency-domain components. 18. The device according to claim 17, further comprising an adaptive frequency-domain filter connected between the modulating unit and the signal synthesis module, configured to perform a frequency-domain filtering on the frequency-domain components of the modulated first signal over the respective sub-carriers and forward frequency-domain components of the filtered signal to the signal synthesis module. 19. The device according to claim 17, further comprising an adaptive frequency-domain filter connected between the signal synthesis module and the signal output unit, configured to: perform a frequency-domain filtering on frequency-domain components output from the signal synthesis module. 20. A signal processing system, comprising a first signal transmitting device and a first signal receiving device connected thereto over a first line, wherein: the first signal transmitting device comprises a plurality of adaptive filters, a signal synthesis module, a signal transmitting module and a feedback receiving module;each of the plurality of adaptive filters is configured to filter a crosstalk source signal according to a value of a filtering parameter, wherein the crosstalk source signals are related to signals to be transmitted over second lines, each of the second lines having a second transmitting end connected to a second receiving end;the signal synthesis module is configured to synthesize a first signal to be transmitted over the first line with the filtered crosstalk source signals;the signal transmitting module is configured to transmit the synthesized first signal to the first receiving end; andthe feedback receiving module is configured to receive a noise statistic related value fed back from the first receiving end, wherein the noise statistic related value reflects a numeric value of a signal-to-noise ratio of the synthesized first signal received by the first receiving end,wherein the values of the filtering parameters is adjusted according to the noise statistic related value, and the adjusted values of the filtering parameters are used in filtering the crosstalk source signals.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.