Finisar Corporation

University of Illinois

Naresh Shanbhag

Hyeon Min Bae

Andrew Carl Singer

Jonathan Ashbrook

Jinki Park

Paul Suppiah

Daniel Mahgerefteh

Christopher R Cole

Philip T Krein

The'linh Nguyen

Roy H Campbell

Robert J Drost

Mingu Kang

Heyon Min Bae

Min-Sun Keel

A compute memory system can include a memory array and a controller that generates N-ary weighted (e.g., binary weighted) access pulses for a set of word lines during a single read operation. This multi-row read generates a charge on a bit line representing a word stored in a column of the memory array. The compute memory system further includes an embedded analog signal processor stage through which voltages from bit lines can be processed in the analog domain. Data is written into the memory array in a manner that stores words in columns instead of the traditional row configuration.

Compute memory

A system includes a circuit having a plurality of electronic function blocks interconnected in series, a power source unit coupled to the circuit, for supplying power to the plurality of electronic function blocks, and a control unit coupled to each of the plurality of the electronic function blocks and to the power source unit. The control unit is configured to monitor activity levels of each of the electronic function blocks, and adjusts the activity level of each of the plurality of electronic function blocks. The control unit determines a voltage level suitable for the corresponding adjusted activity level, and adjusts the power supplied to each of the plurality of electronic function blocks in order to achieve the corresponding determined voltage level at each of the plurality of electronic function blocks.

System and method for improving power conversion for advanced electronic circuits

An adaptive Analog to Digital Converter (ADC) that adjusts the representation levels used in the conversion process so as to optimize system performance. By establishing system performance criteria by which to select or adjust the signal value range associated with each digital representation and/or the digital representation, substantially fewer bits may be used in the ADC. The systems and methods described herein enable lower-power, smaller form-factor designs as well as very high-speed operation. In particular, this technology may be beneficial for use in communications systems because it enables ADC's to operate at speeds where traditional ADC designs simply cannot.

Method and system having adjustable analog-to-digital conversion levels

An example method includes receiving a phase correction signal representing a phase difference between a source signal and a reference signal, generating a first control voltage from the phase correction signal using a charge pump circuit, generating a second control voltage from the phase correction signal in response to a digitally filtered version of the phase correction signal, wherein the second control voltage corrects for an offset error present in the first control voltage, calculating a VCO control signal based on a linear combination of the first and the second control voltages; and generating the source signal in response to the VCO control signal.

Baseband phase-locked loop

A system and method is disclosed for controlling signal conditioning parameters and a sampling parameter controlling conversion of a received signal to digital sampled values prior to decoding. The sampled values are decoded according to a comparison with expected values calculated according to a model of a transmission channel. The model is also updated from time to time by comparing the expected values with actual sampled values. Variation of the expected values over time is calculated. One or more of the signal conditioning parameters and the sampling parameter are adjusted according to a numerical minimization method such that the system BER is reduced.

Tuning system and method using a simulated bit error rate for use in an electronic dispersion compensator

A high-speed optical transmitter comprises multiple digital lanes that are provided to a bank of digital-to-analog converters. The analog signals are then used to Phase Shift Keyed (PSK) modulation using a Chirp Managed Laser (CML)-based transmitter, and potentially using dual polarization. A corresponding optical receiver receives the sequence of optical signals at a demodulator. For each polarization, the demodulator includes a corresponding demodulation channel that is configured to demodulate that polarization component of the optical signal into one or more signal components. Each of these signal components is converted into a corresponding digital signal using a corresponding analog-to-digital converter. In the case of higher-order PSK modulation (e.g., 8PSK or higher), for each polarization, the analog converter has a lower sampling rate than for QPSK modulation.

Phase shift keyed modulation of optical signal using chirp managed laser

A peak detector circuit that responds rapidly to power transients, and yet is able to avoid interpreting data fluctuations as power transients by generating dual peak signals from an amplifier's differential output signal, where the dual peak signals have data ripple components that tend to cancel one another. The system and methods permit the peak detectors to be much more responsive to power transients by expanding their bandwidth (shortening the time constants) to the point that low frequency data components affect the individual peak detector signals, but the effects are cancelled out when the individual components are added together. The peak detector described herein may be used in an AGC system to provide ripple-free gain control signals, while rapidly following any power transients in transmitted signals.

Peak detector with active ripple suppression

Methods and systems are provided for an improved phase detector utilizing analog-to-digital converter (ADC) components. In an embodiment, the method includes from an ADC having a sampling clock signal that determines sampling instants, obtaining a first comparison value between an analog signal and a first threshold voltage at a first sampling instant, and obtaining a second comparison value between the analog signal and a second threshold voltage at a second sampling instant. The method further includes, from a supplemental circuit, obtaining a third comparison value between the analog signal and a third threshold voltage at a third sampling instant between the first and second sampling instants. The method further includes processing the first, second, and third comparison values to determine a phase relationship between the analog signal and the sampling clock.

Phase detector utilizing analog-to-digital converter components

A variable gain amplifier and offset cancellation loop circuit and methods for tracking and correcting DC offset errors that may vary in accordance with the gain of the variable gain amplifier. The circuit is designed to provide tracking of rapid changes in the offset error while maintaining a desired overall frequency response of the combined variable gain amplifier and offset loop. The offset loop cancellation circuit has a wide enough bandwidth to allow the offset cancellation loop to track rapid changes in offset errors that result from rapid changes to the amplifier's gain setting. A control circuit is provided to prevent the large offset cancellation loop bandwidth from having a detrimental effect on the amplifier's overall bandwidth when the amplifier is set to high levels of forward gain by adjusting the offset cancellation loop gain as the forward gain of the amplifier is altered.

Variable gain amplifier having variable gain DC offset loop

A phase detector apparatus and method used for clock recovery from a data signal is provided. The phase detector provides phase correction signals to a clock signal generator, where the phase correction signals are only generated if a predetermined data sample pattern is observed. In particular, the predetermined data sample pattern is preferably a transition from one to zero. Thus, transitions from zero to one will not provide a valid phase update output signal, even though a transition has occurred. In other embodiments the predetermined data sample pattern is preferably a one to zero transition preceded by an additional logic one sample.

Pattern-dependent phase detector for clock recovery

An electronic amplifier circuit that provides improved gain control linearity characteristics resulting from having a controllable field effect transistor (FET) acting as a degeneration resistance (degeneration resistance FET) and a controllable load resistance FET. The overall gain function of the amplifier exhibits improved linearity in part due to the presence of the load FET, which tends to cancel the nonlinear behavior emanating from the degeneration FET. The circuit also includes a control circuit for generating non-linear control signals that are responsive to process characteristics of the FETs, such that the degeneration resistance FET and load resistance FETs may be controlled more consistently and independently from process variations.

Variable gain amplifier having dual gain control

A noise tolerant voltage controlled oscillator is described. The voltage controlled oscillator include a varactor element as part of an LC tank circuit. The varactor element is biased by a bias signal and a bias-dependent control signal. The bias-dependent control signal tunes the LC tank circuit. Because the control signal is bias-dependent, noise and other deleterious influences do not cause the varactor element to deviate in capacitance. Instead, the bias-dependent control signal is a tuning signal that is centered around the bias signal, which allows the varactor element to provide a constant capacitance in the event of a varying bias signal.

Growing community of inventors

Champaign, IL, United States of America

Naresh Shanbhag