Marvell International Limited

Marvellworld Trade Ltd.

Leland Stanford Junior University

Marvell Asia Pte., Ltd.

Kok-Wui Cheong

Hui-Ling Lou

Shaoan Dai

Zhenzhong Gu

Jin Zhang

Junqing Phil Sun

Randy Tsang

Dimitrios-Alexandros Toumpakaris

Runsheng He

Xing Wu

Jing Fang

Meng Zeng

Xiaolin Che

Yuji Meng

Yicheng Chen

Zining Wu

John Matthew Cioffi

Jamal Riani

Gang Wang

Jagadish Venkataraman

Won-Joon Choi

Johnny Li-Shung Fan

Systems are provided for cable length measurement in a communication system. The systems include a transmitter, a receiver, a signal sampler and a cable length calculation unit. The transmitter is configured to transmit a plurality of data symbols at a first data rate via a wired data communication link, and the receiver is configured to receive a reflection signal. The signal sampler is configured to sample the received reflection signal using a phase shift number of shifting sampling phases to generate reflection samples, and combine the reflection samples with different sampling phases to generate a series of reflection samples corresponding to a second data rate higher than the first data rate. The cable length calculation unit is configured to determine a delay parameter from the series of reflection samples, and generate an estimate of a length of the data communication link.

Systems and methods for high precision cable length measurement in a communication system

Embodiments described herein provide echo cancellation power saving management at a cable transceiver. An echo response signal having a first number of signal components is obtained, via an echo cancellation filter. At a first iteration for calculating a first accumulative echo power, a respective echo tap that corresponds to the first iteration is identified. The first accumulative echo power is calculated for the respective iteration by summing powers of outputs from a last echo tap to the respective echo tap. It is then determined whether the first accumulative echo power, exceeds a pre-determined echo power threshold. If the first accumulative echo power exceeds the pre-determined echo power threshold, a first turn-off indication is sent to the echo cancellation filter to turn off all echo taps including and between the last echo tap to the first echo tap.

Systems and methods for echo or interference cancellation power-saving management in a communication system

A decoder includes a syndrome calculator, a Key Equation Solver (KES) and an error corrector. The syndrome calculator receives an n-symbol code word encoded using a Reed Solomon (RS) code to include (n&#x2212;k) redundancy symbols, calculates for the code word 2t syndromes Si, t=(n&#x2212;k)/2 is a maximal number of correctable erroneous symbols. The KES derives an error locator polynomial {circumflex over (&#x2003;)}(x) whose roots identify locations of erroneous symbols, by applying to the syndromes a number of t iterations. In each iteration the KES calculates two discrepancies between {circumflex over (&#x2003;)}(x) and respective two candidates of {circumflex over (&#x2003;)}(x), and derives from the two candidates an updated candidate of {circumflex over (&#x2003;)}(x). The error corrector recovers the code word by correcting the erroneous symbols using the derived error locator polynomial {circumflex over (&#x2003;)}(x).

Low-latency decoder for Reed Solomon codes

According to various embodiments, a method for demarcating data bursts includes receiving, via a coax network, a first data burst and a second data burst at a network interface device, wherein the first data burst and the second data burst at least partially overlap in the time domain. In various embodiments, the first data burst includes a start marker, first data elements following the start marker, and an end marker following the first data elements. In various embodiments, the start marker is orthogonal to the end marker. The method further includes distinguishing, at the network interface device, the first data burst from the second data burst based on the start marker and the end marker; and transmitting the first data burst and the second data burst via an optical network such that the first data burst and the second data burst do not overlap in time.

Method and device for demarcating data bursts

The systems include a transmitter, a receiver, a signal sampler and a cable length calculation unit. The transmitter is configured to transmit a plurality of data symbols at a first data rate via a wired data communication link, and the receiver is configured to receive a reflection signal. The signal sampler is configured to sample the received reflection signal using a phase shift number of shifting sampling phases to generate reflection samples, and combine the reflection samples with different sampling phases to generate a series of reflection samples corresponding to a second data rate higher than the first data rate. The cable length calculation unit is configured to determine a delay parameter from the series of reflection samples, and generate an estimate of a length of the data communication link.

Systems and techniques relating to channel degradation detection for communication systems are described. A described system includes a processor and an interface to transmit signals and receive signals via a channel that includes a cable. The processor can be configured to perform echo cancellation based on echo tap values to remove portions of the transmitted signals that appear as echoes within the received signals, signal equalization based on equalizer tap values, or both. The processor can be configured to determine a channel quality indicator of the channel based on one or more of the echo tap values, one or more of the equalizer tap values, or both. The processor can be configured to generate a warning indication based on the channel quality indicator indicating a degradation of the cable or the channel.

Channel quality indicator for wireline channel degradation detection

Various aspects of this disclosure describe detecting a signal and measuring power with a multi-band filter. Examples include a signal detection and power measurement module in a receiver capable of detecting a training signal and calculating a reliable power measurement in the presence of narrowband interference. A received signal is filtered by a multi-band filter comprising a plurality of sub-band filters. For instance, sub-band filters may be bandpass filters with non-overlapping pass-bands. A training signal is detected by comparing powers of each of the outputs of the sub-band filters to a plurality of thresholds. For example, each sub-band may be assigned a different threshold value. Responsive to detecting a training signal, a power measurement is determined from at least one sub-band filter output. A gain is set based on the determined power measurement, and applied in the receiver.

Signal detection and power measurement with multi-band filter to mitigate narrowband interference

A transceiver to communicate in a vehicle via a single twisted-pair Ethernet cable includes a transmitter and a receiver, an analog front end, an equalizer, and a controller. The transmitter transmits signals via the single twisted-pair Ethernet cable. The receiver receives signals via the single twisted-pair Ethernet cable. The analog front end receives a first signal received by the transceiver via the single twisted-pair Ethernet cable and outputs a second signal. The equalizer includes a notch filter to cancel electromagnetic interference from the second signal. The controller makes a frequency response of the equalizer independent of the electromagnetic interference by controlling tap values of the notch filter. The controller controls gain of one or more of the analog front end and the equalizer based on the frequency response of the equalizer.

Interference mitigation in single twisted-pair communications

Systems and techniques relating to channel degradation detection for communication systems are described. A described system includes an interface to transmit signals and receive signals via a channel that includes a cable; an echo canceller coupled with the interface, the echo canceller to perform echo cancellation based on echo tap values to remove portions of the transmitted signals that appear as echoes within the received signals; an equalizer coupled with the interface, the equalizer to perform signal equalization based on equalizer tap values, the equalizer tap values being determined based on at least a portion of the received signals to adjust an impulse response of the channel and reduce inter-symbol interference within the received signals; and circuitry configured to determine a return loss channel quality indicator of the channel based on the echo tap values, determine an insertion loss channel quality indicator of the channel based on the equalizer tap values, or both.

A transceiver to communicate in a vehicle via a single twisted-pair Ethernet cable includes a transmitter to transmit signals via the single twisted-pair Ethernet cable and a receiver to receive signals via the single twisted-pair Ethernet cable. The transceiver includes an equalizer, a signal-to-noise ratio estimator, and a controller. The equalizer includes a notch filter and a slicer. The equalizer receives an input signal received by the transceiver via the single twisted-pair Ethernet cable. The notch filter cancels electromagnetic interference from the input signal and to output a filtered signal. The slicer slices the filtered signal. The signal-to-noise ratio estimator estimates a signal-to-noise ratio based on an output of the slicer. The controller controls a rate of adapting the equalizer by controlling a rate of change of tap values of the notch filter based on the signal-to-noise ratio.

Systems and methods to mitigate electro-magnetic interference in single twisted-pair-based communication systems

A transceiver includes an equalizer, an adaptation module, an error detector, and a controller. The equalizer receives an input signal via the single twisted-pair Ethernet cable, including outputs of an analog front end of the receiver and an echo canceller, to cancel electromagnetic interference from the input signal. The adaptation module adapts parameters of one or more of i) the equalizer, ii) the analog front end, and iii) the echo canceller based on an error in an output of the equalizer due to the electromagnetic interference. The error detector detects when the error is greater than or equal to a predetermined threshold and in response sets an error indicator to indicate no error for a predetermined period of time. The controller controls adaptation of the parameters of the one or more of i) the equalizer, ii) the analog front end, and iii) the echo canceller based on the error indicator.

Systems and methods to adaptively mitigate electro-magnetic interference (EMI) in automotive and industrial communication systems

A packet is detected in an input data stream. A stream of fast Fourier transform (FFT) blocks are scanned for a plurality of candidate symbols for a preamble of the packet. The stream of FFT blocks correspond to the input data stream. A symbol boundary is detected between adjacent candidate symbols of the plurality of candidate symbols. The FFT blocks are aligned with the symbol boundary. A transition boundary is detected between different adjacent candidate symbols of the preamble based on the detected symbol boundary. A packet detection indication is provided based on the detected transition boundary.

Growing community of inventors

San Jose, CA, United States of America

Kok-Wui Cheong