Abstract: A method of detecting faults in three-phase power lines, including sampling current from die three phase power lines using a set of three sensors each located on a respective power line, processing the current samples from each of the three power lines and translating them to a representation of the current in the respective power lines using a processor provided in each of the sensors, wherein one of the sensors serves as a master sensor and two of the sensors serve as slave sensors, transmitting to the master sensor the representation of the current using a transceiver provided in each sensor, calculating, a representation of a three phase current from the representation of the current from the master sensor and the slave sensors, and identifying at the master sensor, a possible fault in the three phase power lines responsive to the calculated representation of the three phase current.

Abstract: A system for harvesting power from AC current in a power line, including a split magnetic core having two halves, wherein the two halves of the magnetic core are configured to enclose around the power line; a coil is wrapped around one or two halves of the magnetic core with a first end and a second end of the coil configured to provide current from the coil; a power harvesting circuit is configured to harvest current from AC current in the power line, the power harvesting circuit including a rectifier and at least one voltage multiplier electrically connected in parallel to the rectifier; and wherein an output of the coil is connected to an input of said power harvesting circuit and an output of the power harvesting circuit is configured to supply voltage and current to load circuits.

Abstract: A system for the detection of discharges over high voltage power-line insulator strings of an electrical grid. A sensor unit is deployed on a non conducting cable of the grid. The sensor feeds an array of band cut-off filters receiving each signals from said at least one sensor unit. An analogue processor is associated with each of the band cut-off filters associated each with a current peak detector. An ADC unit is associated with each one of said analogue processors. In addition a digital processor is provided for calculating data received from each one of the ADCs and one radio transmitter unit sends data to a gateway of a network. A digital processor is adapted to send data processed based on the output of the ADCs, to the radio transmitter for conveying the data to processor associated with a WAN.

Abstract: A method for adaptive-rate communication includes setting a first target signal margin with respect to an actual noise level and a second target signal margin with respect to a predetermined noise level. The actual noise level is measured at a receiver on a communication channel between a transmitter and the receiver. A transmission rate is selected at which to transmit a signal on the channel such that for the selected transmission rate, a first signal-to-noise ratio (SNR) of the signal relative to the measured actual noise level is greater than a baseline SNR level by at least the first target signal margin, and a second SNR of the signal relative to the predetermined noise level is greater than the baseline SNR by at least the second target signal margin.

Abstract: A method for transmitting a stream of data over a channel made up of a plurality of subchannels having respective subchannel rates. The method includes partitioning the data among the subchannels, such that successive words of the data are mapped to the subchannels in alternation responsive to the respective subchannel rates. The words of the data are transmitted over the subchannels to which they are mapped, and are then received and processed to recover the stream of data.

Abstract: A method for transmitting data includes arranging a predetermined quantity of the data together with one or more overhead bytes in an overhead frame and dividing the overhead frame into equal segments, each of the segments including a number of bits that is not an integer multiple of eight. Each of the segments is encoded as a symbol for transmission over a communication channel.

Abstract: An interpolation filter for processing a sequence of input samples provided at an input rate controlled by an input clock, so as to generate interpolants at an output rate controlled by an output clock, which is substantially independent of the input clock. The filter includes a plurality of stages arranged in a succession in which each of the stages, except for a first stage in the succession, is coupled respectively to a preceding one of the stages.

Abstract: A receiver, for receiving a signal transmitted over a channel by a transmitter that includes a Tomlinson-Harashima precoder. The receiver includes a feed-forward equalizer (FFE), adapted to apply a feed-forward equalization function to the signal, so as to generate a sequence of equalized samples. A decision block is coupled to receive the equalized samples from the FFE and to generate a sequence of decision output samples responsive thereto, the decision block comprising an adaptive filter, having coefficients determined adaptively responsive to a characteristic of the channel. A Tomlinson-Harashima decoder is coupled to receive and decode the decision output samples so as to reconstruct the sequence of input symbols.

Abstract: A method for decoding a sequence of symbols encoded by an encoding system with a plurality of possible states and transmitted over a channel, wherein the encoding system output can be represented by a trellis diagram. The method includes receiving a signal corresponding to the sequence of symbols transmitted over the channel and responsive to the signal, finding respective old state metrics for one or more of the possible states of the encoding system that correspond to a given symbol in the sequence. Respective transition metrics are determined for state transitions that link the one or more possible states for which the old state metrics were found to at least some of the possible states of the encoding system that correspond to a succeeding one of the symbols in the sequence.

Abstract: A method for providing Digital Subscriber Line (DSL) service to a group of subscribers over respective subscriber loops in the presence of another, existing digital network service provided on at least some of the subscriber loops. Lower, middle and upper frequency bands are assigned for transmission over the subscriber loops, such that the existing digital network service is provided on the lower band. DSL signals are transmitted downstream on the upper band and are received upstream on the middle band, generally without transmitting or receiving DSL signals on the lower band, over those subscriber loops on which the existing digital network service, is provided. Over those subscriber loops on which the existing digital network service is not provided, the DSL signals are transmitted downstream on the lower and upper bands and are received upstream on the lower and middle bands.

Abstract: Impulsive noise is detected for each discrete multi-tone (DMT) symbol. If impulsive noise is detected, all bytes, which belong to the associated DMT symbol are tagged by “erasure bits”. After interleaving, Reed-Solomon decoding is initially performed without erasures. If the decoding fails, it is performed again, this time with erasures. Reed-Solomon decoders report failure with relatively high certainty, and thus, if the first stage (decoding without erasures) includes failure or errors due to impulsive noise, the second stage of decoding is performed again with erasures.

Abstract: A method for transmitting a succession of input data samples over a channel having a given channel response includes generating a set of candidate signal point sequences responsive to the input data samples, such that the input data samples are recoverable from a final signal derived from the candidate sequences. The succession of input data samples is mapped into a succession of interim signal point sequences, selected from among the candidate signal point sequences according to a signal shaping criterion. The interim signal point sequences are precoded according to the given channel response and further according to a whitening transformation. The precoded sequences are filtered in accordance with the whitening transformation, thus generating the final signal to be transmitted over the channel to a receiver.

Abstract: A method for transmitting a stream of data over a channel made up of a plurality of subchannels having respective subchannel rates. The method includes partitioning the data among the subchannels, such that successive words of the data are mapped to the subchannels in alternation responsive to the respective subchannel rates. The words of the data are transmitted over the subchannels to which they are mapped, and are then received and processed to recover the stream of data.

Abstract: An interpolation filter for processing a sequence of input samples provided at an input rate controlled by an input clock, so as to generate interpolants at an output rate controlled by an output clock, which is substantially independent of the input clock. The filter includes a plurality of stages arranged in a succession in which each of the stages, except for a first stage in the succession, is coupled respectively to a preceding one of the stages.

Abstract: A method is described for mapping digital input data, including a sequence of input samples, into an interim signal point sequence, for use in conjunction with a single carrier modulation transmitter. The method includes generating a set of candidate signal point sequences, for each input sample, selecting an interim signal point sequence, into which to map the sequence of input samples, from among the set of candidate signal point sequences, so as to obtain a low peak power value of a final signal generated from the interim signal point sequence, and using a single carrier modulation transmitter to transmit the final signal to a receiver.

Abstract: A method for adaptive-rate communication includes setting a first target signal margin with respect to an actual noise level and a second target signal margin with respect to a predetermined noise level. The actual noise level is measured at a receiver on a communication channel between a transmitter and the receiver. A transmission rate is selected at which to transmit a signal on the channel such that for the selected transmission rate, a first signal-to-noise ratio (SNR) of the signal relative to the measured actual noise level is greater than a baseline SNR level by at least the first target signal margin, and a second SNR of the signal relative to the predetermined noise level is greater than the baseline SNR by at least the second target signal margin.

Abstract: A receiver, for receiving a signal transmitted over a channel by a transmitter that includes a Tomlinson-Harashima precoder. The receiver includes a feed-forward equalizer (FFE), adapted to apply a feed-forward equalization function to the signal, so as to generate a sequence of equalized samples. A decision block is coupled to receive the equalized samples from the FFE and to generate a sequence of decision output samples responsive thereto, the decision block comprising an adaptive filter, having coefficients determined adaptively responsive to a characteristic of the channel. A Tomlinson-Harashima decoder is coupled to receive and decode the decision output samples so as to reconstruct the sequence of input symbols.

Abstract: A system and method is presented for reducing the power consumed by a line driver driving a signal to be transmitted through a communications media. The preferred method includes the steps of: (1) delaying the signal; (2) determining whether the signal has reached a predetermined threshold value; (3) increasing power supplied to the line driver in response to an indication from the determining step that the threshold has been reached; and (4) delivering the delayed signal to the line driver. The preferred system includes: (1) a digital signal processor; (2) a line driver; (3) a digital delay line disposed between the digital signal processor and the line driver; and (4) a power regulator disposed between the threshold detector and the line driver. The system and method, while applicable to many communications formats, is particularly applicable to communications methods transmitting signals having a relatively large dynamic range, such as xDSL communications techniques.

Abstract: A signal modulation method comprising receiving at least first and second synchronized incoming streams of complex symbols, thereby to define a plurality of incoming vectors each including at least first and second synchronized complex symbols, mapping each complex symbol into a signal component comprising a linear combination of an in-phase signal and a quadrature signal, the quadrature signal comprising a Hilbert transform of said in-phase signal, wherein all of the signal components are substantially mutually orthogonal and wherein the frequency spectrums of all signal components mapped from a single incoming stream are centered around a common frequency location which is unique to the single incoming stream and wherein the frequency spectrums of signal components mapped from different incoming streams having adjacent common frequency locations are partially overlapping and wherein signal components mapped from sequential incoming symbols partially overlap in time and combining all of the signal components