Broadcom Corporation

Huawei Technologies Co., Limited

Stmicroelectronics Gmbh

Davide Tonietto

Ali Ghiasi

Henry Wong

Ichiro Fujimori

Afshin Momtaz

Yuchun Lu

Marc-Andre Lacroix

Thomas J Kolze

Bruce Joseph Currivan

Guangming Yin

Mario Caresosa

Euhan Chong

David Chung

Anthony Chan Carusone

David Kyong-Sik Chung

Dustin Tarl Dunwell

John Hogeboom

Semyon Lebedev

Zhonggui Xiang

Petar Ivanov Krotnev

De-serializing and serializing methods are disclosed for data transmission between chips involving an arrangement of an analog frond end and a digital signal processor for the de-serializing and serializing methods. According to the present disclosure, processing in the digital domain may be carried out by a chip with smaller technology-node geometry while processing in the analog domain may be carried out by a chiplet suitable for analog components. Heat generated by the AFE is isolated from the DSP. Furthermore, costs associated with development and manufacturing may be lowered.

Systems, apparatuses, methods, and non-transitory computer-readable storage devices for de-serializing and serializing data transmission

A method includes deactivating transmitters of a first plurality of transceivers that are associated with an endpoint to multi-channel communication fabric. A given transceiver of the first plurality of transceivers includes a receiver. The method includes controlling the given transceiver to cause the given transceiver to couple a reference source of the given transceiver to a first node of the receiver, measure a first value at a second node of the receiver, and determine a gain between the first node and the second node based on the measured first value. The method includes controlling the given receiver to cause the given receiver to isolate the reference source from the first node of the receiver; and measuring, by the given transceiver, a second value at the second node and determining, by the given transceiver, an intrinsic noise based on the measured second value. The method includes activating the deactivated transmitters; measuring, by the given transceiver, a third value at the second node and determining, by the given transceiver, a composite noise based on the measured third value; and determining, by the given transceiver, a crosstalk noise at the first node of the receiver based on the determined gain, the determined intrinsic noise and the determined composite noise.

Methods and systems for in-situ crosstalk measurements in communication systems

A device includes a first inductor positioned on a first substrate. The first inductor has at least one turn in a plane that is perpendicular to a plane of the first substrate. The first inductor is positioned for near field coupling with a second inductor. The second inductor is positioned on a second substrate, with at least one turn that is in a plane perpendicular to a plane of the second substrate. The second inductor is substantially parallel to the first inductor. Such an arrangement may be used for near field coupling, including edge-to-edge coupling, between two integrated circuits.

Inductors for chip to chip near field communication

A system and method for Feed Forward Equalizer (FFE)-Aided Clock Data Recovery (CDR) to calibrate phase offset and enhance gain in baud rate sampling phase detector is provided. In an embodiment, a clock data recovery (CDR) apparatus includes an incremental feed forward equalizer (INC-FFE) in a CDR path and a calibration component in an equalization path, the calibration component connected to the INC-FFE, the calibration component configured to adjust FFE coefficients for the INC-FFE according to a phase code (PC) index in a PC index table and one of a signal-to-noise-ratio (SNR) and a bit error rate (BER) of a sampled signal, wherein the PC index table comprises adjustment values for the FFE coefficients, and wherein the PC index is linearly related to a sampling phase.

FFE-aided CDR to calibrate phase offset and enhance gain in baud rate sampling phase detector

Methods, integrated circuits and computer-readable media for communicating back-channel data over a data link by modulating the phase or frequency of a clock signal of a data signal transmitted over the data link. Slow modulation of the clock signal allows it to be detected and extracted by a receiver without affecting the integrity or bit rate of the data signal. Some embodiments allow the functionality to be implemented without the use of extra hardware in the transmitter or receiver or either.

Frequency/phase-shift-keying for back-channel serdes communication

Methods, systems and devices for dynamically controlling resolution of an analog-to-digital converter (ADC). The ADC receives an analog input signal and outputs digital data. A statistical unit coupled to the ADC obtains samples of the output signal and transmits a control signal to the ADC to adjust the resolution of the ADC. The control signal is generated by the statistical unit based on a comparison of at least one performance indicator with a target performance level. The at least one performance indicator is calculated using the samples.

Dynamic control of ADC resolution

A system and method for Feed Forward. Equalizer (FFE)-Aided Clock Data Recovery (CDR) to calibrate phase offset and enhance gain in baud rate sampling phase detector is provided. In an embodiment, a clock data recovery (CDR) apparatus includes an incremental feed forward equalizer (INC-FFE) in a CDR path and a calibration component in an equalization path, the calibration component connected to the INC-FFE, the calibration component configured to adjust FFE coefficients for the INC-FFE according to a phase code (PC) index in a PC index table and one of a signal-to-noise-ratio (SNR) and a bit error rate (BER) of a sampled signal, wherein the PC index table comprises adjustment values for the FFE coefficients, and wherein the PC index is linearly related to a sampling phase.

The present disclosure provides a system, circuit, and method for correcting clock skew in time-interleaved analog-to-digital converters. At least two clock signals are received along respective channels. A delay of a first channel, carrying a first clock signal, is accounted for by applying one or more first adjustment factors to the channels until an edge of the first clock signal is aligned with a transition point of a reference signal. The first clock signal is swapped to the second channel, and vice-versa. A value of the reference signal as sampled by the first clock signal is compared to values of the reference signal as sampled by the second clock signal to determine a skew of the second channel vis-&#xe0;-vis the first channel, and one or more second adjustment factors are applied to the second channel based on the determined skew of the second channel.

Skew detection and correction in time-interleaved analog-to-digital converters

Apparatus and methods are taught for quickly determining whether a Loss of Signal (LOS) condition has occurred for a receiver which includes an internal reference clock, a LOS circuit and a Clock and Data Recovery (CDR) circuit. The CDR circuit recovers the clock and data of an incoming signal. However, the LOS circuit can determine whether a received incoming signal includes an active signal, independent of said CDR circuit such that it samples said incoming signal utilizing said internal reference clock to determine a loss of signal prior to said CDR recovering the clock of said incoming signal. The LOS circuit includes an analog voltage threshold stage which samples the incoming signal, and produces at least one sample stream indicative of transitions in the incoming signal. The LOS circuit further includes a digital transition stage which counts transitions in order to discriminate between an active signal and noise.

System and method for detecting loss of signal

A high-speed bit stream interface module interfaces a high-speed communication media to a communication Application Specific Integrated Circuit (ASIC) via a Printed Circuit Board (PCB). The high-speed bit stream interface module includes a line side interface, a board side interface, and a signal conditioning circuit. The line side interface includes a media coupler that receives the line side media, such as copper media or optical media. The board side interface couples the high-speed serial bit stream interface module to the PCB. A signal conditioning circuit communicatively couples to the line side interface and to the board side interface. The signal conditioning circuit receives an RX signal from the line side interface, conditions the RX signal, and provides the RX signal to the board side interface. The signal conditioning circuit receives a TX signal from the board side interface, conditions the TX signal, and provides the TX signal to the board side interface.

Conditioning circuit that spectrally shapes a serviced bit stream

A first device transmits data over a first branch of a communications link toward a second device. That second device loops the received data pattern back over a second branch of the communications link. A bit error rate of the looped back data pattern is determined and a pre-emphasis applied to the transmitted data pattern is adjusted in response thereto. The first device further perturbs the data pattern communications signal so as to increase the bit error rate. The pre-emphasis is adjusted so as to reduce the determined bit error rate in the looped back data pattern in the presence of the perturbation. The steps for perturbing the signal and adjusting the pre-emphasis are iteratively performed, with the perturbation of the signal increasing with each iteration and adjustment of the pre-emphasis being refined with each iteration. The signal is perturbing by injecting modulation jitter into the signal (increasing each iteration) and adjusting amplitude of the signal (decreasing each iteration).

SERDES with jitter-based built-in self test (BIST) for adapting FIR filter coefficients

Systems and methods for optimizing operation of a transceiver device are disclosed. The method may include parallel processing an input signal through a first path having a first frequency response and a second path having a second frequency response. The second frequency response may be higher than the first frequency response. Signals from the first and second paths may be combined, creating an output signal having a desired gain and frequency. The parallel processing may adjust a gain of at least one of the first path and the second path. The parallel processing may equalize at least one of the first frequency response and the second frequency response. The input signal may be from a 10 GBit Ethernet channel and/or a Fibre channel.

System and method for programmably adjusting gain and frequency response in a 10-gigabit ethernet/fibre channel system

A high-speed bit stream interface module interfaces a high-speed communication media to a communication Application Specific Integrated Circuit (ASIC) via a Printed Circuit Board (PCB). The high-speed bit stream interface includes a line side interface, a board side interface, and a signal conditioning circuit. The signal conditioning circuit services each of an RX path and a TX path and includes a limiting amplifier and a clock and data recovery circuit. The signal conditioning circuit may also include an equalizer and/or an output pre-emphasis circuit. The limiting amplifier applies respective gains to the RX path and to the TX path that are based upon respective dynamic ranges of the incoming signals.

Bit stream conditioning circuit having adjustable input sensitivity

Various example embodiments are disclosed. According to an example embodiment, an apparatus may include a continuous time filter, a decision feedback equalizer, a clock and data recovery circuit, and an adaptation circuit. The adaptation circuit may be configured to adapt equalization according to at least one dithering algorithm by adjusting a delay adjust signal based on a mean square error of equalized data signals.

Growing community of inventors

Ottawa, Canada

Davide Tonietto