WO2016101420A1 - 一种用户线远端串扰消除矢量计算方法及装置 - Google Patents

一种用户线远端串扰消除矢量计算方法及装置 Download PDF

Info

Publication number
WO2016101420A1
WO2016101420A1 PCT/CN2015/074496 CN2015074496W WO2016101420A1 WO 2016101420 A1 WO2016101420 A1 WO 2016101420A1 CN 2015074496 W CN2015074496 W CN 2015074496W WO 2016101420 A1 WO2016101420 A1 WO 2016101420A1
Authority
WO
WIPO (PCT)
Prior art keywords
far
user
end crosstalk
user port
port
Prior art date
Application number
PCT/CN2015/074496
Other languages
English (en)
French (fr)
Inventor
沈伦英
徐志兵
王军
Original Assignee
中兴通讯股份有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 中兴通讯股份有限公司 filed Critical 中兴通讯股份有限公司
Publication of WO2016101420A1 publication Critical patent/WO2016101420A1/zh

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04B—TRANSMISSION
    • H04B3/00—Line transmission systems
    • H04B3/02—Details
    • H04B3/32—Reducing cross-talk, e.g. by compensating

Definitions

  • the present invention relates to the field of access network technologies, and in particular, to a second generation very high digital subscriber line far-end crosstalk cancellation vector calculation method and apparatus.
  • the main noise in digital subscriber line DSL systems includes inter-line crosstalk (including far-end crosstalk FEXT, near-end crosstalk NEXT), background noise interference, impulse noise interference, and RFI radio noise.
  • inter-line crosstalk including far-end crosstalk FEXT, near-end crosstalk NEXT
  • background noise interference including impulse noise interference, and RFI radio noise.
  • VDSL2 very high digital subscriber line
  • the VDSL2 system adopts the frequency division multiplexing FDM mode, most of the NEXT and the interfered signals at the receiving end are separated in the frequency band, and the influence can be eliminated or greatly reduced by the filter; however, the interfered signals of the FEXT and the receiving end are generally not available.
  • the frequency band is separated and cannot be eliminated by the filter.
  • the VDSL2 transmission distance is short, which causes the FEXT on the line to be more serious than other DSL technologies. Therefore, FEXT is the main source of crosstalk in the VDSL2 system, which will cause the signal-to-noise ratio to decrease and the line to be reduced.
  • the transmission rate or the bit error rate is increased, and even heavy training is generated to affect the stability of the system.
  • Theoretical analysis shows that the higher the frequency, the larger the FEXT, the shorter the line, and the larger the FEXT.
  • the digital subscriber line access multiplexer DSLAM equipment generally provides the broadband access service to as many users as possible, and is affected by the FEXT of the line pair.
  • the rate of multi-port port construction is much lower.
  • the single-link networking rate of the port and there are some port stability problems, in order to solve these problems, a variety of solutions for FEXT have emerged, and the far-end crosstalk cancellation vector computing technology is one of them.
  • the embodiment of the invention provides a method and a device for calculating a far-end crosstalk cancellation vector of a subscriber line, so as to solve the problem that the long-distance crosstalk cancellation vector operation of a large number of ports in the related art takes too long.
  • an embodiment of the present invention discloses a method for calculating a far-end crosstalk cancellation vector of a subscriber line, including:
  • the far-end crosstalk value of each user port is calculated, and the N user ports are classified into M types according to the far-end crosstalk value of each user port.
  • a user port is extracted from each type of user port, and the far-end crosstalk values of the extracted M user ports are respectively calculated, and the far-end crosstalk values of the M user ports are respectively used as the remaining users of the user ports to which the user port belongs.
  • the far-end crosstalk value of the port
  • the crosstalk calculation matrix of the entire system is formed, and the far-end crosstalk elimination vector calculation is performed;
  • the process of dividing the N user ports into the M class according to the size of the far-end crosstalk value of each user port includes:
  • the user ports whose difference is greater than the set range are divided into different classes.
  • separately calculating a far-end crosstalk value of each user port includes:
  • the pilot sequence is sent to all the user ports except the user port, and the far-end crosstalk value of the user port is calculated according to the error sample message fed back by all the user ports.
  • the foregoing method further includes:
  • the foregoing method further includes:
  • the M-type user port in the entire system is re-divided into M'-type user ports, and the entire system is composed according to the far-end crosstalk value of the re-divided M'-type user port.
  • the crosstalk calculation matrix performs far-end crosstalk cancellation vector calculation.
  • the embodiment of the invention further discloses a user line far-end crosstalk cancellation vector computing device, which comprises:
  • the far-end crosstalk value calculation unit is configured to extract a user port from each type of user port, calculate a far-end crosstalk value of the extracted M user ports, and use the far-end crosstalk values of the M user ports as the user port respectively.
  • the far-end crosstalk cancellation vector calculation unit is configured to form a crosstalk calculation matrix of the entire system according to the far-end crosstalk value of the M-type user port, and perform far-end crosstalk cancellation vector calculation.
  • the classification unit is configured to divide the N user ports into M classes according to the size of the far-end crosstalk value of each user port by:
  • the two user ports whose difference is greater than the set range are divided into different classes.
  • the far-end crosstalk value calculation unit is configured to separately calculate a far-end crosstalk value of each user port by:
  • the pilot sequence is sent to all the user ports except the user port, and the far-end crosstalk value of the user port is calculated according to the error sample message fed back by all the user ports.
  • the foregoing apparatus further includes:
  • Update unit set to have a user port offline and go online again in a type of user port Recalculating the far-end crosstalk value of the user port by using the far-end crosstalk value calculation unit, and re-dividing the user port according to the recalculated far-end crosstalk value of the user port, performing vector calculation separately and completing synchronization.
  • the updating unit is further configured to: when the offline uplink rate of the user port reaches a set threshold, invoke the classification unit to re-divide the M-type user port in the entire system into M′.
  • the class user port forms a crosstalk calculation matrix of the entire system according to the far-end crosstalk value of the re-divided M' class user port, and performs far-end crosstalk cancellation vector calculation.
  • the embodiment of the invention further discloses a computer storage medium, wherein the computer storage medium stores computer executable instructions, and the computer executable instructions are used to execute the above method.
  • the technical solution of the present application has made progress in improving the time for vector calculation of the system, saving the time of vector calculation and improving the user experience.
  • FIG. 1 is a flowchart of a method for calculating a far-end string interference cancellation vector of a subscriber line according to an embodiment of the present invention
  • FIG. 2 is a flowchart of another method for calculating a far-end string interference cancellation vector of a subscriber line according to an embodiment of the present invention
  • FIG. 3 is a flowchart of calculating a far-end crosstalk cancellation vector of a subscriber line in an alternative provided by an embodiment of the present invention
  • FIG. 4 is a schematic diagram of user line classification according to an embodiment of the present invention.
  • FIG. 5 is a schematic diagram of another user line classification according to an embodiment of the present invention.
  • FIG. 6 is a schematic diagram of a device for calculating a far-end crosstalk cancellation vector of a subscriber line according to an embodiment of the present invention.
  • This embodiment provides a method for calculating a far-end crosstalk cancellation vector of a subscriber line, which mainly includes the following operations:
  • the far-end crosstalk value of each user port is calculated, and the N user ports are classified into M types according to the far-end crosstalk value of each user port.
  • a user port is extracted from each type of user port, and the far-end crosstalk values of the extracted M user ports are respectively calculated, and the far-end crosstalk values of the M user ports are respectively used as the remaining users of the user ports to which the user port belongs.
  • the far-end crosstalk value of the port
  • the crosstalk calculation matrix of the entire system is formed, and the far-end crosstalk elimination vector calculation is performed;
  • the process of dividing the N user ports into the M class according to the size of the far-end crosstalk value of each user port includes the following steps:
  • the two user ports whose calculated difference is smaller than the set range are divided into the same class; the two user ports whose calculated difference is larger than the set range are divided into different classes.
  • the process of calculating the far-end crosstalk value of a user port includes: when calculating a far-end crosstalk value of a user port, sending a pilot sequence to all user ports except the user port in the system, according to The error sample message fed back by all the user ports calculates the far-end crosstalk value of the user port.
  • the far-end crosstalk value of the user port can be recalculated, and the user port is reclassified according to the far-end crosstalk value of the recalculated user port. Complete the vector calculation and complete the synchronization.
  • the M-type user port in the entire system can be re-divided into M'-type user ports, and the entire cross-talk value of the re-divided M'-type user port is composed.
  • the crosstalk calculation matrix of the system performs far-end crosstalk cancellation vector calculation.
  • the N user ports are classified into M types according to the size of the far-end crosstalk value of each user port; and one user port is extracted from each type of user port, and the extracted M user ports are respectively calculated.
  • the remote crosstalk value is used as the far-end crosstalk value of the remaining user ports of the user ports to which the user port belongs, and the far-end crosstalk value of the M-type user port is used to form the entire far-end crosstalk value.
  • the crosstalk calculation matrix of the system, and the process of performing far-end crosstalk cancellation vector calculation includes:
  • Step 1 two vector differences in plane coordinates: If Calculate the corresponding chord length of C in plane coordinates: If Calculate the corresponding chord length of C in plane coordinates
  • the N user ports are classified into M types according to the size of the far-end crosstalk value of each user port;
  • Step 3 When the N lines are vector-operated as a whole, the M lines are extracted from the M classes, the pilot sequence is sent, and the es (error sample) collection is completed, and M vector compensation values ⁇ TFExCoef 1 , TFExCoef 2 are calculated. , whil,TFExCoef m ⁇ ;
  • one user port is extracted from each type of user port divided in the above step 2, and the far-end crosstalk values of the extracted M user ports are respectively calculated.
  • step 4 ⁇ TFExCoef 1 , TFExCoef 2 , ..., TFExCoef m ⁇ (the values converted after the operation are ⁇ val 1 , val 2 , ..., val m ⁇ ) are respectively compensated to the corresponding M classes.
  • the entire crosstalk matrix is obtained (assumed as follows) to complete the vector calculation of far-end crosstalk cancellation.
  • the far-end crosstalk values of the M user ports are respectively used as the far-end crosstalk values of the remaining user ports of the user ports to which the user port belongs, and the crosstalk calculation matrix of the entire system is formed according to the far-end crosstalk value of the M-type user ports. Perform far-end crosstalk cancellation vector calculation.
  • the method for calculating the far-end crosstalk cancellation vector of the subscriber line by using the above method is as shown in FIG. 1 and includes the following steps:
  • Step 101 The digital user port (referred to as the user port, the same below) of the second generation very high digital subscriber line is initialized, and the feedback error sample (referred to as es) message is collected through the feedback channel EOC or the service channel.
  • the feedback error sample referred to as es
  • step 102 the user port performs a vector sum operation on the es message.
  • the first user port is initialized, and a pilot sequence is sent to all other ports in the system.
  • the far-end crosstalk vector value TFExCoef 1 Total of far-end xCrosstalk Coef, TFExCoef
  • the two user ports are initialized.
  • the pilot sequence is sent to all other ports in the system, and the far-end crosstalk vector value TFExCoef 2 received is calculated according to the error sample message.
  • Step 103 Starting from calculating the far-end crosstalk vector value of the third user port, comparing the value with the far-end crosstalk vector value of the existing classification to obtain a comparison value.
  • the third user port initiates an initial handshake signal
  • the corresponding es message is collected, and the corresponding crosstalk value TFExCoef 3 is calculated.
  • Step 104 The user port is divided into a plurality of classes according to the comparison value and the preset precision deviation range r, and each class includes at least one user port.
  • Step 105 If there is a new port synchronization, extract one path in each class, add the new port to an existing class or create a new class according to the algorithm of the above steps.
  • Step 106 After all the ports are classified, according to the M categories ⁇ class 1 , class 2 , class 3 , ..., class M ⁇ , a preset control signal is sent, and a user port is selected from each category, again.
  • TFExCoef the far-end crosstalk vector value
  • TFExCoef the far-end crosstalk vector value of the remaining user ports in the same class
  • Step 107 Extract M user ports from M categories, and complete M far-end crosstalk calculations, and obtain M far-end crosstalk values ⁇ TFExCoef 1 , TFExCoef 2 , . . . , TFExCoef m ⁇ , and the converted values are respectively calculated.
  • M far-end crosstalk values ⁇ TFExCoef 1 , TFExCoef 2 , . . . , TFExCoef m ⁇
  • the crosstalk calculation matrix of the entire system N lines
  • Step 108 performing vector calculation on the above crosstalk matrix.
  • the digital user ports in the system are classified, and the ports in the same type are substantially close to each other, and the far-end crosstalk value of a digital user port is used as a representative value, and the remaining ports in the same type are the current round.
  • the es collection and calculation can be performed without using the representative value for the far-end crosstalk compensation, and the vector calculation of the whole system is involved, which indirectly reduces the communication time and calculation time overhead required for the far-end crosstalk estimation process. It also saves the corresponding memory overhead.
  • the following describes a user line far-end crosstalk cancellation vector calculation process, mainly based on the above method, as shown in FIG. 2, the process includes the following steps:
  • Step 201 During the showtime phase of the user port of the entire system, after the system is maintained for a period of time, the next vector calculation cycle comes and the vector calculation of the entire system is performed.
  • the user ports in the entire system are synchronized, if all the user ports in the above M categories are always in a stable state, when the next vector calculation cycle comes, only the M categories are selected, and the M ports are selected again according to the user port selection algorithm.
  • the user port calculates the far-end crosstalk value, and finally forms a crosstalk calculation matrix of the entire system (N-way line) according to the M far-end crosstalk values, and performs vector calculation.
  • Algorithm 1 according to the principle of random scheduling, obtain a user port from the same type. But there are many times If the same port is selected, there is an unfair phenomenon for other ports in the same class.
  • Algorithm 2 according to the polling scheduling principle, sequentially acquires a user port from the same type, so that the defect of the first method can be avoided.
  • Algorithm 3 the combination of specific user priority and polling scheduling, guarantees high-priority user performance requirements, and also gives other users the opportunity to choose.
  • the algorithm of the user port in the same type may adopt any algorithm, which is not limited in this embodiment.
  • Step 202 Extract M user ports from M categories, and complete M far-end crosstalk calculations, and obtain M far-end crosstalk values ⁇ TFExCoef 1 , TFExCoef 2 , . . . , TFExCoef m ⁇ according to the M far-end crosstalks.
  • Step 203 performing vector calculation on the crosstalk matrix.
  • the user line far-end crosstalk value elimination vector calculation process adds a subsequent operation based on the foregoing method, as shown in FIG. 3, including the following steps:
  • Step 301 During the whole system operation, if a user in a category is offline and goes online again, according to the above algorithm, the port is reclassified, the vector calculation is completed separately, and synchronization is completed.
  • step 302 the classification of the system is re-established when the offline rate of the user port in the system reaches a certain threshold.
  • the offline or re-online occurs for a plurality of reasons, which causes the internal balance of the original M classifications to be broken, and finally the overall performance has a large error.
  • multiple users receive crosstalk vectors of other users, and the vectors that are close to each other in a two-dimensional plane are drawn with a vertex of one of the vectors and a circle with a radius of r. That is, the circle shown in 401 in the figure, all users protected in this circle, can be seen as a valid classification.
  • FIG. 5 Another scheme for forming a plurality of effective classifications for all user ports of the subscriber line is as shown in FIG. 5, and includes the following steps:
  • multiple users form a plurality of valid classifications in a two-dimensional plane, which provides a basic condition for the user ports in the entire system to form a far-end crosstalk matrix.
  • This embodiment provides a user line far-end crosstalk cancellation vector computing device. As shown in FIG. 6, at least the following units are included.
  • the far-end crosstalk value calculation unit is configured to extract a user port from each type of user port, calculate a far-end crosstalk value of the extracted M user ports, and use the far-end crosstalk values of the M user ports as the user port respectively.
  • the foregoing classification unit is configured to first calculate a difference between the far-end crosstalk values of each two user ports; and further divide the user ports whose difference is less than the set range into the same class; The range of user ports can be divided into different classes.
  • the far-end crosstalk cancellation vector calculation unit is configured to form a crosstalk calculation matrix of the entire system according to the far-end crosstalk value of the M-type user port, and perform far-end crosstalk cancellation vector calculation.
  • the far-end crosstalk value calculation unit is configured to calculate a far-end crosstalk value of a user port, and send a pilot sequence to all user ports except the user port in the system, and the error sample is fed back according to all user ports.
  • the message calculates the far-end crosstalk value of the user port.
  • the foregoing apparatus may further include: an updating unit, configured to: when a user port is offline and re-online in a type of user port, call the far-end crosstalk value calculation unit to recalculate the far-end crosstalk value of the user port. And re-dividing the user port according to the far-end crosstalk value of the recalculated user port, completing the vector calculation separately and completing the synchronization.
  • an updating unit configured to: when a user port is offline and re-online in a type of user port, call the far-end crosstalk value calculation unit to recalculate the far-end crosstalk value of the user port. And re-dividing the user port according to the far-end crosstalk value of the recalculated user port, completing the vector calculation separately and completing the synchronization.
  • the updating unit is further configured to: when the offline uplink rate of the user port reaches a set threshold, invoke the classification unit to re-divide the M-type user port in the entire system into the M′-type user port, according to the re-division
  • the far-end crosstalk value of the M' class user port constitutes the crosstalk calculation matrix of the entire system, and performs far-end crosstalk cancellation vector calculation.
  • the classification unit is configured to send a pilot sequence to all other ports in the system, calculate the far-end crosstalk value received according to the error sample message, and classify the port into one class.
  • the pilot sequence is sent to all other ports in the system, and the far-end crosstalk value received by the error sample packet is calculated, and the port is classified into another type of user.
  • the classification unit is configured to collect the corresponding es message and calculate the corresponding crosstalk value.
  • the crosstalk value is compared with the crosstalk value of the user port in the previous ports 1 and 2, and the port is attributed to the existing category according to the size of the difference, or a new class including its own port is added.
  • the far-end crosstalk cancellation vector calculation unit is configured to issue a specific control signal, extract a user port from each classification, and calculate the far-end crosstalk value of the port again, and use the crosstalk value as the cross-talk value.
  • the far-end crosstalk value of the remaining user ports in the same class is configured to issue a specific control signal, extract a user port from each classification, and calculate the far-end crosstalk value of the port again, and use the crosstalk value as the cross-talk value.
  • the far-end crosstalk value of the remaining user ports in the same class is configured to issue a specific control signal, extract a user port from each classification, and calculate the far-end crosstalk value of the port again, and use the crosstalk value as the cross-talk value.
  • the far-end crosstalk value of the remaining user ports in the same class is configured to issue a specific control signal, extract a user port from each classification, and calculate the far-end crosstalk value of the port again, and use the crosstalk value as the cross-talk value.
  • the M user ports are extracted by M categories.
  • the far-end crosstalk cancellation vector calculation unit is configured to form a crosstalk calculation matrix of the entire system (N lines) according to the M far-end crosstalk values. Perform vector calculations.
  • the update unit is set to only need to select the algorithm from the above M classification according to the user port. / or the specific user priority principle, extract M user ports again, calculate the far-end crosstalk value, and then form the crosstalk calculation matrix of the whole system (N-channel) according to the M far-end crosstalk values, and perform vector calculation.
  • the update unit is set to classify the port according to the above algorithm, complete the vector calculation separately and complete the synchronization.
  • the update unit is set to re-divide the existing M classifications in the entire system into M′, and finally maintain the far-end crosstalk in the entire system.
  • the accuracy of the vector calculation matrix is set to re-divide the existing M classifications in the entire system into M′, and finally maintain the far-end crosstalk in the entire system.
  • the above technical solution has made progress in improving the time for the system to perform vector calculation, saving the time of vector calculation and improving the user experience.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Cable Transmission Systems, Equalization Of Radio And Reduction Of Echo (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)

Abstract

一种用户线远端串扰消除矢量计算方法及装置，涉及接入网络技术领域。所述方法包括：根据每个用户端口的远端串扰值的大小将N个用户端口划分为M类；从每类用户端口中分别抽取一个用户端口，分别计算抽取的M个用户端口的远端串扰值，将M个用户端口的远端串扰值分别作为该用户端口所属的一类用户端口中其余用户端口的远端串扰值；根据M类用户端口的远端串扰值，组成整个系统的串扰计算矩阵，进行远端串扰消除矢量计算。本发明实施例还公开了一种用户线远端串扰消除矢量计算装置。本申请技术方案在改进系统进行矢量计算的时间方面取得了进步，节省了矢量计算的时间，改善了用户体验效果。

Description

一种用户线远端串扰消除矢量计算方法及装置 技术领域
本发明涉及接入网络技术领域，尤其涉及第二代甚高数字用户线远端串扰消除矢量计算方法及装置。
背景技术
一般情况下，数字用户线DSL系统中主要噪声包括线路间串扰(包括远端串扰FEXT，近端串扰NEXT)、背景噪声干扰、脉冲噪声干扰、RFI无线电噪声等。
第二代甚高数字用户线(VDSL2)和其他DSL技术相比，由于使用更宽的频段，用于更短的传输距离，受到的线路噪声干扰也更为严重；而且高频信号本身的衰减很大，对线路噪声非常敏感，因此噪声已经成为制约VDSL2系统传输性能和稳定性的主要因素。
由于VDSL2系统采用频分复用FDM方式，因此大部分NEXT与接收端的被干扰信号在频段上是分开的，其影响可通过滤波器消除或大大降低；但是FEXT与接收端的被干扰信号一般无法在频段上分开，不能用滤波器消除，同时VDSL2传输距离较短，导致线路上的FEXT较其他DSL技术更为严重，因此FEXT是VDSL2系统中主要的串扰来源，会导致信噪比下降而降低线路传输速率或增大误码率，甚至产生重训练从而影响系统的稳定性。理论分析表明，频率越高，FEXT越大，线路越短，FEXT越大。
在实际工程运用中，数字用户线路接入复用器DSLAM设备一般是整板建链向尽量多的用户提供接宽带入服务，受线对间FEXT影响，多路端口建链时的速率远低于端口的单路建链速率,且存在一些端口稳定性问题，为解决这些问题，出现了多种针对FEXT的解决方案，远端串扰消除矢量计算技术就是其中之一。
当前已有的第二代甚高数字用户线远端串扰消除矢量计算方法，有效地提升用户性能的同时也存在这样的问题：为了计算每个端口所受到的远端串扰，每个端口需要收集其他端口对自身的远端串扰参数，整个过程比较耗费 时间；且随着整个系统接入的用户越多，整个系统做远端串扰消除矢量运算的时间会越长。一旦用户端口同步时间变长，必然会给用户体验带来了负面感受。
发明内容
本发明实施例提供一种用户线远端串扰消除矢量计算方法及装置，以解决相关技术中大量端口进行远端串扰消除矢量运算耗时过长的问题。
为了解决上述技术问题，本发明实施例公开了一种用户线远端串扰消除矢量计算方法，包括：
系统包括有N个用户端口时，分别计算每个用户端口的远端串扰值，根据每个用户端口的远端串扰值的大小将N个用户端口划分为M类；
从每类用户端口中分别抽取一个用户端口，分别计算抽取的M个用户端口的远端串扰值，将M个用户端口的远端串扰值分别作为该用户端口所属的一类用户端口中其余用户端口的远端串扰值；
根据M类用户端口的远端串扰值，组成整个系统的串扰计算矩阵，进行远端串扰消除矢量计算；
其中，N、M均为整数，N>2，M<＝N-2。
可选地，上述方法中，根据每个用户端口的远端串扰值的大小将N个用户端口划分为M类的过程包括：
计算每两个用户端口的远端串扰值之间的差值；
将所述差值小于设定范围的用户端口划分为同一类；
将所述差值大于设定范围的用户端口划分为不同类。
可选地，上述方法中，分别计算每个用户端口的远端串扰值包括：
计算一用户端口的远端串扰值时，向系统中除该用户端口以外的所有用户端口分别发送导频序列，根据所有用户端口反馈的错误样本报文，计算该用户端口的远端串扰值。
可选地，上述方法还包括：
如果一类用户端口中有用户端口离线并再次上线，则重新计算该用户端口的远端串扰值，根据重新计算的用户端口的远端串扰值对该用户端口重新划分类，单独完成矢量计算并完成同步。
可选地，上述方法还包括：
如果用户端口的离线上线率达到设定阀值，则再次对整个系统中的M类用户端口重新划分为M′类用户端口，根据重新划分的M′类用户端口的远端串扰值组成整个系统的串扰计算矩阵，进行远端串扰消除矢量计算。
本发明实施例还公开了一种用户线远端串扰消除矢量计算装置，包括：
分类单元，设置为分别计算系统中每个用户端口的远端串扰值，根据每个用户端口的远端串扰值的大小将系统包含的N个用户端口划分为M类，其中，N、M均为整数，N>2，M<＝N-2；
远端串扰值计算单元，设置为从每类用户端口中分别抽取一个用户端口，分别计算抽取的M个用户端口的远端串扰值，将M个用户端口的远端串扰值分别作为该用户端口所属的一类用户端口中其余用户端口的远端串扰值；
远端串扰消除矢量计算单元，设置为根据M类用户端口的远端串扰值，组成整个系统的串扰计算矩阵，进行远端串扰消除矢量计算。
可选地，上述装置中，所述分类单元，是设置为通过如下方式实现根据每个用户端口的远端串扰值的大小将N个用户端口划分为M类：
计算每两个用户端口的远端串扰值之间的差值；
将所述差值小于设定范围的两个用户端口划分为同一类；
将所述差值大于设定范围的两个用户端口划分为不同类。
可选地，上述装置中，所述远端串扰值计算单元，是设置为通过如下方式实现分别计算每个用户端口的远端串扰值：
计算一用户端口的远端串扰值时，向系统中除该用户端口以外的所有用户端口分别发送导频序列，根据所有用户端口反馈的错误样本报文，计算该用户端口的远端串扰值。
可选地，上述装置还包括：
更新单元，设置为在一类用户端口中有用户端口离线并再次上线，则调 用所述远端串扰值计算单元重新计算该用户端口的远端串扰值，并根据重新计算的用户端口的远端串扰值对该用户端口重新划分类，单独完成矢量计算并完成同步。
可选地，上述装置中，所述更新单元，还设置为在用户端口的离线上线率达到设定阀值时，调用所述分类单元再次对整个系统中的M类用户端口重新划分为M′类用户端口，根据重新划分的M′类用户端口的远端串扰值组成整个系统的串扰计算矩阵，进行远端串扰消除矢量计算。
本发明实施例还公开了一种计算机存储介质，所述计算机存储介质中存储有计算机可执行指令，所述计算机可执行指令用于执行上述方法。
本申请技术方案在改进系统进行矢量计算的时间方面取得了进步，节省了矢量计算的时间，改善了用户体验效果。
附图概述
图1是本发明实施例提供的一种用户线远端串扰扰消除矢量计算方法流程图；
图2是本发明实施例提供的另一种用户线远端串扰扰消除矢量计算方法流程图；
图3是本发明实施例提供的可选方案中用户线远端串扰消除矢量计算流程图；
图4是本发明实施例提供的一种用户线分类示意图；
图5是本发明实施例提供的另一种用户线分类示意图；
图6是本发明实施例提供的一种用户线远端串扰消除矢量计算装置示意图。
本发明的较佳实施方式
下文将结合附图对本发明实施例的技术方案作进一步详细说明。需要说明的是，在不冲突的情况下，本申请的实施例和实施例中的特征可以任意相 互组合。
实施例1
本实施例提供一种用户线远端串扰消除矢量计算方法，主要包括如下操作：
系统包括有N个用户端口时，分别计算每个用户端口的远端串扰值，根据每个用户端口的远端串扰值的大小将N个用户端口划分为M类；
从每类用户端口中分别抽取一个用户端口，分别计算抽取的M个用户端口的远端串扰值，将M个用户端口的远端串扰值分别作为该用户端口所属的一类用户端口中其余用户端口的远端串扰值；
根据M类用户端口的远端串扰值，组成整个系统的串扰计算矩阵，进行远端串扰消除矢量计算；
其中，N、M均为整数，N>2，M<＝N-2。
上述方法中，根据每个用户端口的远端串扰值的大小将N个用户端口划分为M类的过程又包括如下几个操作步骤：
首先，计算每两个用户端口的远端串扰值之间的差值；
然后，将所计算的差值小于设定范围的两个用户端口划分为同一类；将所计算差值大于设定范围的两个用户端口划分为不同类。
可选地，计算一个用户端口的远端串扰值过程包括，计算一用户端口的远端串扰值时，向系统中除该用户端口以外的所有用户端口分别发送pilot sequence(导频序列)，根据所述所有用户端口反馈的error sample(错误样本)报文，计算该用户端口的远端串扰值。
另外，如果一类用户端口中有用户端口离线并再次上线时，还可以重新计算该用户端口的远端串扰值，根据重新计算的用户端口的远端串扰值对该用户端口重新划分类，单独完成矢量计算并完成同步。
如果用户端口的离线上线率达到设定阀值，则可以再次对整个系统中的M类用户端口重新划分为M′类用户端口，根据重新划分的M′类用户端口的远端串扰值组成整个系统的串扰计算矩阵，进行远端串扰消除矢量计算。
可选地，上述方法中，根据每个用户端口的远端串扰值的大小将N个用户端口划分为M类；从每类用户端口中分别抽取一个用户端口，分别计算抽取的M个用户端口的远端串扰值，将M个用户端口的远端串扰值分别作为该用户端口所属的一类用户端口中其余用户端口的远端串扰值；根据M类用户端口的远端串扰值，组成整个系统的串扰计算矩阵，进行远端串扰消除矢量计算的过程包括：
步骤1，平面坐标中两个矢量差：
Figure PCTCN2015074496-appb-000001
若
Figure PCTCN2015074496-appb-000002
计算C在平面坐标中对应的弦长
Figure PCTCN2015074496-appb-000003
根据星座调制方法，以l为半径画圆，以圆心所在的点对应的矢量作为代表值，代表圆内所有其他矢量，即定义l为精度偏差半径r(0<r<＝经验值)。
步骤2，第二代甚高数字用户线(N条数字用户线)系统中，将每条线路受到其余线路总的远端串扰矢量和两两计算差值，并与r比较，最终形成M(M<＝N-2)个分类class＝{class1,class2,......,classm}，且每个class包含x(x<M)个矢量，也即x个线路classx＝{Line1,Line2,......,Linex}；
即根据每个用户端口的远端串扰值的大小将N个用户端口划分为M类；
步骤3，N条线路整体进行矢量运算时，分别从M个class中抽取M路线路，发送pilot sequence，并完成es(错误样本)收集，且计算出M个矢量补偿值{TFExCoef1,TFExCoef2,……,TFExCoefm}；
即从上述步骤2中所划分的每类用户端口中分别抽取一个用户端口，分别计算抽取的M个用户端口的远端串扰值。
步骤4，将{TFExCoef1,TFExCoef2,……,TFExCoefm}(运算时转换后的值分别为{val1，val2，……，valm})分别补偿到对应的M个class中的每个串扰线路上，得出整个串扰矩阵(假设如下)，即可完成远端串扰消除的矢量计算。
Figure PCTCN2015074496-appb-000004
即将M个用户端口的远端串扰值分别作为该用户端口所属的一类用户端口中其余用户端口的远端串扰值，根据M类用户端口的远端串扰值，组成整个系统的串扰计算矩阵，进行远端串扰消除矢量计算。
下面结合应用场景及附图说明上述方法的实现过程。
采用上述方法进行用户线远端串扰消除矢量计算过程如图1所示，包括如下步骤：
步骤101，第二代甚高数字用户线连接的数字用户端口(简称用户端口，下同)初始化阶段，通过反馈通道EOC或业务通道，收集反馈的error sample(简称es)报文。
步骤102，用户端口对es报文做矢量和运算。
第1个用户端口初始化，对系统中所有其他端口发送pilot sequence，根据反馈的error sample报文，计算其受到的远端串扰矢量值TFExCoef1(Total of far-end xCrosstalk Coef，简称TFExCoef)；第2个用户端口初始化，同样，对系统中所有其他端口发送pilot sequence，根据反馈的error sample报文，计算其受到的远端串扰矢量值TFExCoef2。将初始的两个端口作为基准比较类，分为两类用户class1＝{Line1}，class2＝{Line2}；
步骤103，从计算第3个用户端口的远端串扰矢量值开始，将其值与已有分类的远端串扰矢量值进行比较，得到一个比较值。
如当第3个用户端口发起初始化握手信号，同理，收集相应的es报文，并计算相应的串扰值TFExCoef3。将该串扰值与之前的1和2号用户端口的串 扰矢量值TFExCoef1和TFExCoef2进行比较，分别计算矢量差值vector1＝TFExCoef3-TFExCoef1和vector2＝TFExCoef3-TFExCoef2，并分别计算矢量vector1和vector2对应的弦长l1和l2；
步骤104，根据比较值以及预置的精度偏差范围r，将该用户端口分成多个类(class)，每个类至少包含一个用户端口。可选的，归类方法如下：若l1＝l2<＝r且TFExCoef3离TFExCoef1近且TFExCoef3离TFExCoef2远，对应的Line3则放置于类class1{Line1,Line3}，若l1＝l2<＝r，且TFExCoef3离TFExCoef1远且TFExCoef3离TFExCoef2近，放置于类class2＝{Line2,Line3}；
若l1＝l2>r，则单独新建一个类class3＝{Line3}；
若l1<＝r<l2,则Line3放置class1＝{Line1，Line3}；
若l2<＝r<l1,则Line3放置class2＝{Line2，Line3}；
步骤105，若有新端口同步，在每个class中抽取一路，按照上述步骤的算法，将该新端口加入一个已有的class或建立一个新class。
依此类推，将所有用户端口进行分类，最终形成M(M<＝N-2)个类{class1，class2，class3，……，classM}，且每个class由多条远端串扰特性相似的线路组成如class＝{Line1,Linei,……,Linej}；
步骤106，完成所有端口的分类后，根据M个分类{class1，class2，class3，……，classM}，下发预置的控制信号，从每个分类中选取一个用户端口，再次计算该端口的远端串扰矢量值TFExCoef，并以该串扰矢量值作为同一个类中其余用户端口的远端串扰矢量值，如class＝{Line1,Linei,……,Linej},该类中的每个用户端口的远端串扰值为：FEXT＝{TFExCoef,TFExCoef,……,TFExCoef}。
步骤107，从M个分类抽取M个用户端口，完成M次远端串扰计算后，得到M个远端串扰值{TFExCoef1,TFExCoef2,……,TFExCoefm}，运算时转换后的值分别为{val1，val2，……，valm}，依据这M个远端串扰值，最终组成整个系统(N路线路)的串扰计算矩阵，类似如下：
Figure PCTCN2015074496-appb-000005
步骤108，对上述串扰矩阵执行矢量计算。
本发明实例中，将系统中的数字用户端口进行分类，同类中的端口，彼此受到的远端串扰基本接近，以一个数字用户端口的远端串扰值作为代表值，同类中的其余端口本轮矢量计算时，可以不需进行es收集和计算，而直接采用该代表值进行远端串扰补偿，参与整个系统的矢量计算，间接地减少了远端串扰估算过程需要的通信时间和计算时间开销，同时也节省了相应的内存开销。
下面再介绍一种用户线远端串扰消除矢量计算过程，主要在上述方法的基础上，如图2所示，该过程包括如下步骤：
步骤201，整个系统的用户端口showtime阶段，系统维持一段时间后，下一个矢量计算周期来到，执行整个系统的矢量计算。
整个系统中用户端口都同步后，若以上M个分类中所有用户端口一直处于稳定状态，当下一个矢量计算周期来到时，只需从以上M分类中，根据用户端口选中算法，再次抽取M个用户端口，计算其远端串扰值，再根据这M个远端串扰值，最终组成整个系统(N路线路)的串扰计算矩阵，执行矢量计算。
其中，同类中用户端口选择算法，可选的算法有多种，例如：
算法一，根据随机调度原则，从同类中获取一个用户端口。但存在多次 都选中同一个端口的情况，故对同类中其它端口存在不公平的现象。
算法二，根据轮询调度原则，从同类中依次获取一个用户端口，这样就可以避免方法一的缺陷。
算法三，特定用户优先和轮询调度相结合原则，保证高优先级用户性能要求的同时，也给其它用户选择的机会。
要说明的是，同类中用户端口的算法可以采用任意算法，本实施例不作限制。
步骤202，从M个分类抽取M个用户端口，完成M次远端串扰计算后，得到M个远端串扰值{TFExCoef1,TFExCoef2,……,TFExCoefm}，依据这M个远端串扰值，最终组成整个系统(N路线路)的串扰计算矩阵。
步骤203，对上述串扰矩阵执行矢量计算。
本实施例中，系统一旦维持稳定的状态后，再次对整个系统中数字用户线的远端串扰进行重新估算和矢量计算时，只需要从M个分类中抽象M和/或2M(2M<＝N-2)个用户端口进行es报文采集和远端串扰估算，即可组成整个系统的远端串扰矩阵和矢量计算，直接减少了远端串扰估算过程需要的通信时间和计算时间开销。
可选方案中，用户线远端串扰值消除矢量计算过程在上述方法的基础上，增加了后续操作，如图3所示，包括如下步骤：
步骤301，整个系统运行期间，若一个分类中有用户离线并再次上线，则根据以上算法，重新将该端口进行归类，单独完成矢量计算并完成同步。
步骤302，当系统中用户端口离线率达到特定的阀值后，重建系统的分类。
整个系统运行期间，若用户端口的离线上线率达到特定阀值后，再次对整个系统中的已经存在的M个分类进行重新划分为M′，最终维持整个系统中远端串扰矢量计算矩阵的精确性。
本实施例中，针对系统中可能存在个别或部分用户，因多种原因出现离线或再上线的情况，导致原有的M个分类的内部平衡会打破，最终导致整体性能出现较大误差。为了改善该情况，需要重新建立新的M′分类，实现再平 衡而减小误差。
在上述方法的基础上，另一些可选方案提供了一种用户线的用户端口有效分类的方案，该分类过程如图4所示，将多个用户端口，经过矢量运算后，最终构成一个有效的分类。
本实施例中，二维平面内，多个用户收到其他用户的串扰矢量，图示在二维平面内，彼此靠近的矢量，以其中一个矢量的顶点为圆心，以r为半径画圆，即图中401所示的圆，此圆内保护的所有用户，即可看出是一个有效的分类。
另外一种用户线的所有用户端口形成多个有效分类的方案如图5所示，包括如下步骤：
步骤501，用户端口a，经过矢量运算后，单独构成一个有效的分类classa＝{Linea}。
步骤502，用户端口b，经过矢量运算后，单独构成一个有效的新分类classb＝{Lineb}。
步骤503，用户端口c，经过矢量运算后，单独构成一个有效的新分类classc＝{Linec}。
步骤504，用户端口d，经过矢量运算后，单独构成一个有效的新分类classd＝{Lined}。(虚线标示的)用户端口e，参考图1中步骤105，经过矢量运算后，同已有的classc和classd进行比较，决定其最终归属的分类classd＝{Lined。Linee}。
本实施例中，在二维平面内，多个用户构成多个有效的分类，为整个系统中的用户端口构成远端串扰矩阵提供了基础条件。
实施例2
本实施例提供一种用户线远端串扰消除矢量计算装置，如图6所示，至少包括如下单元。
分类单元，设置为分别计算系统中每个用户端口的远端串扰值，根据每个用户端口的远端串扰值的大小将系统包含的N个用户端口划分为M类，其中，N、M均为整数，N>2，M<＝N-2；
远端串扰值计算单元，设置为从每类用户端口中分别抽取一个用户端口，分别计算抽取的M个用户端口的远端串扰值，将M个用户端口的远端串扰值分别作为该用户端口所属的一类用户端口中其余用户端口的远端串扰值；
可选地，上述分类单元，是设置为首先计算每两个用户端口的远端串扰值之间的差值；再将差值小于设定范围的用户端口划分为同一类；将差值大于设定范围的用户端口划分为不同类即可。
远端串扰消除矢量计算单元，设置为根据M类用户端口的远端串扰值，组成整个系统的串扰计算矩阵，进行远端串扰消除矢量计算。
可选地，远端串扰值计算单元，是设置为计算一用户端口的远端串扰值时，向系统中除该用户端口以外的所有用户端口分别发送pilot sequence，根据所有用户端口反馈的error sample报文，计算该用户端口的远端串扰值。
可选地，上述装置还可以包括：更新单元，设置为主要在一类用户端口中有用户端口离线并再次上线，则调用所述远端串扰值计算单元重新计算该用户端口的远端串扰值，并根据重新计算的用户端口的远端串扰值对该用户端口重新划分类，单独完成矢量计算并完成同步。
可选地，上述更新单元，还设置为在用户端口的离线上线率达到设定阀值时，调用分类单元再次对整个系统中的M类用户端口重新划分为M′类用户端口，根据重新划分的M′类用户端口的远端串扰值组成整个系统的串扰计算矩阵，进行远端串扰消除矢量计算。
下面结合实际应用，说明上述装置的工作过程：
首先，第1个用户端口初始化过程中，分类单元设置为对系统中所有其他端口发送pilot sequence，根据反馈的error sample报文，计算其受到的远端串扰值，并将该端口归为一类用户；第2个用户端口初始化过程中，同样，对系统中所有其他端口发送pilot sequence，根据反馈的error sample报文，计算其受到的远端串扰值，并将该端口归为另一类用户；
接着，当第3个用户端口初始化过程中，同理，分类单元设置为收集相应的es报文，并计算相应的串扰值。将该串扰值与之前的1和2号端口中的用户端口的串扰值进行比较，根据差值的大小，将其端口归于已有的类别中，或新增一个包含自身端口的分类。
依次类推，分类单元设置为将所有用户端口进行分类，最终形成M(M<＝N-2)个类。
其次，根据这M个分类，远端串扰消除矢量计算单元设置为下发特定的控制信号，从每个分类中抽取一个用户端口，再次计算该端口的远端串扰值，并以该串扰值作为同一个类中其余用户端口的远端串扰值。
M个分类抽取M个用户端口，完成M次远端串扰计算后，远端串扰消除矢量计算单元设置为依据这M个远端串扰值，最终组成整个系统(N路线路)的串扰计算矩阵，执行矢量计算。
整个系统中用户端口都同步后，若以上M个分类中所有用户端口一直处于稳定状态，当下一个矢量计算周期来到时，更新单元设置为只需从以上M分类中，根据用户端口选中算法和/或特定用户优先原则，再次抽取M个用户端口，计算其远端串扰值，再根据这M个远端串扰值，最终组成整个系统(N路线路)的串扰计算矩阵，执行矢量计算。
整个系统运行期间，若一个分类中有用户离线并再次上线，则更新单元设置为根据以上算法，重新将该端口进行归类，单独完成矢量计算并完成同步。
整个系统运行期间，若用户端口的离线上线率达到特定阀值f后，更新单元设置为再次对整个系统中的已经存在的M个分类进行重新划分为M′，最终维持整个系统中远端串扰矢量计算矩阵的精确性。
以上所述，仅为本发明的较佳实例而已，并非用于限定本发明的保护范围。
本领域普通技术人员可以理解上述方法中的全部或部分步骤可通过程序来指令相关硬件完成，所述程序可以存储于计算机可读存储介质中，如只读 存储器、磁盘或光盘等。可选地，上述实施例的全部或部分步骤也可以使用一个或多个集成电路来实现。相应地，上述实施例中的各模块/单元可以采用硬件的形式实现，也可以采用软件功能模块的形式实现。本申请不限制于任何特定形式的硬件和软件的结合。
工业实用性
上述技术方案在改进系统进行矢量计算的时间方面取得了进步，节省了矢量计算的时间，改善了用户体验效果。

Claims (11)

  1. 一种用户线远端串扰消除矢量计算方法，包括：
    系统包括有N个用户端口时，分别计算每个用户端口的远端串扰值，根据每个用户端口的远端串扰值的大小将N个用户端口划分为M类；
    从每类用户端口中分别抽取一个用户端口，分别计算抽取的M个用户端口的远端串扰值，将M个用户端口的远端串扰值分别作为该用户端口所属的一类用户端口中其余用户端口的远端串扰值；
    根据M类用户端口的远端串扰值，组成整个系统的串扰计算矩阵，进行远端串扰消除矢量计算；
    其中，N、M均为整数，N>2，M<＝N-2。
  2. 如权利要求1所述的方法，其中，根据每个用户端口的远端串扰值的大小将N个用户端口划分为M类的过程包括：
    计算每两个用户端口的远端串扰值之间的差值；
    将所述差值小于设定范围的两个用户端口划分为同一类；
    将所述差值大于设定范围的两个用户端口划分为不同类。
  3. 如权利要求1或2所述的方法，其中，分别计算每个用户端口的远端串扰值包括：
    计算一用户端口的远端串扰值时，向系统中除该用户端口以外的所有用户端口分别发送导频序列，根据所有用户端口反馈的错误样本报文，计算该用户端口的远端串扰值。
  4. 如权利要求3所述的方法，该方法还包括：
    如果一类用户端口中有用户端口离线并再次上线，则重新计算该用户端口的远端串扰值，根据重新计算的用户端口的远端串扰值对该用户端口重新划分类，单独完成矢量计算并完成同步。
  5. 如权利要求4所述的方法，该方法还包括：
    如果用户端口的离线上线率达到设定阀值，则再次对整个系统中的M类 用户端口重新划分为M′类用户端口，根据重新划分的M′类用户端口的远端串扰值组成整个系统的串扰计算矩阵，进行远端串扰消除矢量计算。
  6. 一种用户线远端串扰消除矢量计算装置，包括：
    分类单元，设置为分别计算系统中每个用户端口的远端串扰值，根据每个用户端口的远端串扰值的大小将系统包含的N个用户端口划分为M类，其中，N、M均为整数，N>2，M<＝N-2；
    远端串扰值计算单元，设置为从每类用户端口中分别抽取一个用户端口，分别计算抽取的M个用户端口的远端串扰值，将M个用户端口的远端串扰值分别作为该用户端口所属的一类用户端口中其余用户端口的远端串扰值；
    远端串扰消除矢量计算单元，设置为根据M类用户端口的远端串扰值，组成整个系统的串扰计算矩阵，进行远端串扰消除矢量计算。
  7. 如权利要求6所述的装置，其中，所述分类单元，是设置为通过如下方式实现根据每个用户端口的远端串扰值的大小将N个用户端口划分为M类：
    计算每两个用户端口的远端串扰值之间的差值；
    将所述差值小于设定范围的两个用户端口划分为同一类；
    将所述差值大于设定范围的两个用户端口划分为不同类。
  8. 如权利要求6或7所述的装置，其中，所述远端串扰值计算单元，是设置为通过如下方式实现分别计算每个用户端口的远端串扰值：
    计算一用户端口的远端串扰值时，向系统中除该用户端口以外的所有用户端口分别发送导频序列，根据所有用户端口反馈的错误样本报文，计算该用户端口的远端串扰值。
  9. 如权利要求8所述的装置，还包括：
    更新单元，设置为在一类用户端口中有用户端口离线并再次上线，则调用所述远端串扰值计算单元重新计算该用户端口的远端串扰值，并根据重新计算的用户端口的远端串扰值对该用户端口重新划分类，单独完成矢量计算并完成同步。
  10. 如权利要求9所述的装置，
    所述更新单元，还设置为在用户端口的离线上线率达到设定阀值时，调用所述分类单元再次对整个系统中的M类用户端口重新划分为M′类用户端口，根据重新划分的M′类用户端口的远端串扰值组成整个系统的串扰计算矩阵，进行远端串扰消除矢量计算。
  11. 一种计算机存储介质，所述计算机存储介质中存储有计算机可执行指令，所述计算机可执行指令用于执行权利要求1～5中任一项所述的方法。
PCT/CN2015/074496 2014-12-24 2015-03-18 一种用户线远端串扰消除矢量计算方法及装置 WO2016101420A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201410817086.7A CN105790792B (zh) 2014-12-24 2014-12-24 一种用户线远端串扰消除矢量计算方法及装置
CN201410817086.7 2014-12-24

Publications (1)

Publication Number Publication Date
WO2016101420A1 true WO2016101420A1 (zh) 2016-06-30

Family

ID=56149043

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2015/074496 WO2016101420A1 (zh) 2014-12-24 2015-03-18 一种用户线远端串扰消除矢量计算方法及装置

Country Status (2)

Country Link
CN (1) CN105790792B (zh)
WO (1) WO2016101420A1 (zh)

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107979390B (zh) * 2016-10-25 2021-10-12 中兴通讯股份有限公司 一种矢量化系统识别外部串入端口的方法及装置
CN106844850B (zh) * 2016-12-16 2020-02-18 南京航空航天大学 一种针对远端串扰噪声的消除电路设计方法
CN108259276B (zh) * 2016-12-28 2021-12-14 中兴通讯股份有限公司 一种控制超宽带铜线接入技术g.fast端口的方法及装置
CN109728837B (zh) * 2017-10-30 2020-11-17 桐乡市定邦信息技术有限公司 一种抵消串扰信号的方法、装置及系统
CN109995693A (zh) * 2017-12-31 2019-07-09 中国移动通信集团四川有限公司 抵消网线回传串扰方法、装置、设备及介质

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101471695A (zh) * 2007-12-27 2009-07-01 华为技术有限公司 一种线路管理方法、装置及系统
CN102308486A (zh) * 2011-07-05 2012-01-04 华为技术有限公司 生成扩展信道的方法和装置
CN103067316A (zh) * 2012-12-28 2013-04-24 华为技术有限公司 一种串扰信道估计方法、装置和系统

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101453245B (zh) * 2007-11-29 2012-08-08 华为技术有限公司 一种消除数字用户线串扰的方法、设备和系统
CN101567864B (zh) * 2008-04-23 2012-11-14 瑞昱半导体股份有限公司 具有远端串扰消除功能的网络装置
CN102388588B (zh) * 2011-09-30 2014-04-16 华为技术有限公司 一种串扰信道估计方法、装置及系统

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101471695A (zh) * 2007-12-27 2009-07-01 华为技术有限公司 一种线路管理方法、装置及系统
CN102308486A (zh) * 2011-07-05 2012-01-04 华为技术有限公司 生成扩展信道的方法和装置
CN103067316A (zh) * 2012-12-28 2013-04-24 华为技术有限公司 一种串扰信道估计方法、装置和系统

Also Published As

Publication number Publication date
CN105790792A (zh) 2016-07-20
CN105790792B (zh) 2021-06-15

Similar Documents

Publication Publication Date Title
WO2016101420A1 (zh) 一种用户线远端串扰消除矢量计算方法及装置
RU2616968C1 (ru) Выполнение выравнивания символа восходящего потока в условиях fext
CN109995435B (zh) 一种基于多纤芯光网络的虚拟网络映射方法及系统
JP5186046B2 (ja) デジタル通信システムにおける伝送資源の管理方法及びシステム
CN109992419A (zh) 一种优化的协同边缘计算低延迟任务分配卸载方法
WO2012149791A1 (zh) 一种串扰信道估计方法、装置及系统
US9407561B2 (en) Systems and methods for traffic engineering in software defined networks
US9813112B2 (en) Power control method, device, and system
WO2017097269A1 (zh) 一种干扰估计方法和设备
Tsiaflakis et al. Improved dual decomposition based optimization for DSL dynamic spectrum management
JP5989093B2 (ja) Ｇ．ｖｅｃｔｏｒ初期化のためのシステムおよび方法
WO2014101847A1 (zh) 一种串扰信道估计方法、装置和系统
Perepelkin et al. Improved adaptive routing algorithm in distributed data centers
WO2021098403A1 (zh) 资源分配方法、服务器及存储介质
WO2017045207A1 (zh) 信号处理方法、装置和系统
US9325371B2 (en) Method for managing data transport using crosstalk data
CN109714273A (zh) 一种多核网络设备的报文处理方法及装置
US9525459B2 (en) Method, related device, and system for crosstalk cancellation processing on port line of access device
CN106209164B (zh) 功率控制方法及装置
Moraes et al. Intercarrier interference in DSL networks due to asynchronous DMT transmission
Moraes et al. Dynamic spectrum management in DSL with asynchronous crosstalk
WO2016206225A1 (zh) 一种强滤波抵消方法和装置、计算机存储介质
CN115001975B (zh) 一种5g切片接入的配电网保护优化方法及系统
CN116915709B (zh) 负载均衡的方法及装置、电子设备和存储介质
WO2018120062A1 (zh) 信号处理方法、网络侧设备及客户终端设备

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 15871498

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 15871498

Country of ref document: EP

Kind code of ref document: A1