Abstract: At least one example embodiment discloses a method of initializing a plurality of communication paths in a system. The method includes obtaining a first matrix from a storage medium, the first matrix including at least one channel sub-matrix of the system associated with a subset of the plurality of communication paths, obtaining an indication of which communication paths are to become active and determining, after the obtaining the indication, new compensation coefficients based on the at least one channel sub-matrix and active compensation coefficients, the active compensation coefficients for compensating for crosstalk between the active communication paths prior to the number of active communication paths changing.

Abstract: In one embodiment, the access node includes a vectoring control unit for estimating crosstalk coefficients between a set of subscriber lines based on crosstalk measurements carried out over the set of subscriber lines during a crosstalk measurement phase, and a vectoring processing unit for jointly processing, for crosstalk mitigation, signals to be transmitted over, or received from, the set of subscriber lines based on the estimated crosstalk coefficients. The access node further includes a controller configured to either postpone the execution of power adaptation over a particular line out of the set of subscriber lines after the completion of the crosstalk measurement phase, or to postpone the crosstalk measurement phase after the completion of power adaptation over the particular line.

Abstract: In a multi-tone data transmission system, data are modulated on a set of M tones for being transmitted between a transmitter and a receiver. The data retransmission request device in such system is adapted to request retransmission of a data transmission unit that is hierarchically modulated at a particular hierarchical level and that is modulated on a particular group of N tones, N being a positive integer value greater than one and smaller than or equal to M.

Abstract: In an embodiment the method includes, at a transmit side, grouping carriers into carrier groups, and selecting a group constellation diagram for modulation of the carriers of a given carrier group. The selected group constellation diagram is built as a hierarchical sum of I super-imposed constellations vectors having predefined decreasing powers with the P strongest and the Q weakest constellation vectors being omitted. The carriers of the carrier group are modulated by means of the selected group constellation diagram, and P and/or Q parameters per carrier group and per multi-carrier data symbol are adjusted.

Abstract: A crosstalk cancellation device cancels crosstalk noise of at most M disturbing communication lines in a victim communication line forming part of a vectoring group. The crosstalk cancellation device gradually decreases the cancellation depth of a cancelled disturbing communication line in the victim communication line upon a crosstalk noise variation in the vectoring group requiring cancellation of crosstalk noise of a not yet cancelled disturbing communication line in the victim communication line.

Abstract: A crosstalk emulator for a cable, preferably a xDSL telecommunication cable, represented by several emulation paths each comprising a first segment (L1a/L4a) serially connected to a second segment (L1b/L4b) on a printed circuit board PCB. The first and second segments of a same emulation path form an angle, e.g. of about 90 degrees, at their junction point. All the emulation paths have a same length and preferably run in parallel over the PCB. As a result, each emulation path crosses only once (X21) any other emulation path at a cross-point. Furthermore, the area occupied by the crosstalk emulator on the PCB is reduced with respect to a matrix topology, whereby the present topology can easily be extended to large numbers of couplings, allowing design guidance for a passive coupling emulator with a large amount of coupling elements (CP) at the cross-points.

Abstract: In a network for digital data packet transmission, quality records are stored in a memory. Each of the quality records is indicative for a reception quality being sensed of one or more received data packets. Thereupon, upon overflow of the memory, one or more of the quality records is discarded from the memory. For at least one of the quality records, discarding weights are generated as a function of at least one of the quality records and associated to the quality records. Hereby, upon such overflow, the discarding weights are taken into account in predefined rules and conditions according to which the discarding of the quality records is performed.

Abstract: In one embodiment, the line termination card includes a data output terminal configured to output a data sequence. The card further includes a vectoring entity configured to parse and encode the data sequence into frequency samples according to a carrier loading parameter, configured to scale the frequency samples into scaled frequency samples according to a carrier scaling parameter, and, configured to process the scaled frequency samples for crosstalk compensation. A controller is configured to adjust the carrier loading parameter and the carrier scaling parameter, and a forwarder is coupled to the data output terminal and to the controller. The forwarder is configured to forward the data sequence, the carrier loading parameter and the carrier scaling parameter towards a further line termination card.

Abstract: In a network for digital data packet transmission, reception quality of received data packets is sensed (111) by a remote terminal (101). Thereupon, a bit string indicative for the sensed reception quality is generated (112). This bit string or a compressed version of this bit string is transferred (114) in real-time over a management channel to a first central processing unit (121) in a network node (102) where the bit string is compacted or processed. By processing the bit string, the first central processing unit (121) may determine impulse noise characteristics. Alternatively, the bit string or a compressed or compacted version thereof may be forwarded to a second central processing unit (131) in a network manager (103) that processes the bit string and determines impulse noise characteristics.

Abstract: In a network for digital data packet transmission, quality records are stored in a memory. Each of the quality records is indicative for a reception quality being sensed of one or more received data packets. Thereupon, upon overflow of the memory, one or more of the quality records is discarded from the memory. For at least one of the quality records, discarding weights are generated as a function of at least one of the quality records and associated to the quality records. Hereby, upon such overflow, the discarding weights are taken into account in predefined rules and conditions according to which the discarding of the quality records is performed.

Abstract: In a network for digital data packet transmission, reception quality of received data packets is sensed (111) by a remote terminal (101). Thereupon, a bit string indicative for the sensed reception quality is generated (112). This bit string or a compressed version of this bit string is transferred (114) in real-time over a management channel to a first central processing unit (121) in a network node (102) where the bit string is compacted or processed. By processing the bit string, the first central processing unit (121) may determine impulse noise characteristics. Alternatively, the bit string or a compressed or compacted version thereof may be forwarded to a second central processing unit (131) in a network manager (103) that processes the bit string and determines impulse noise characteristics.