Abstract: A method of constructing a waveform from N sampled data captured at N successive points in time, includes, in part, applying the N sampled data, K data at a time, to each of M delayed replicas of a filter that includes K taps so to generate N×M interpolated data. The waveform is then constructed from the N sampled data and the N×M interpolated data.

Abstract: A method of constructing a waveform from N sampled data captured at N successive points in time, includes, in part, applying the N sampled data, K data at a time, to each of M delayed replicas of a filter that includes K taps so to generate N×M interpolated data. The waveform is then constructed from the N sampled data and the N×M interpolated data.

Abstract: An adaptation process for automatic gain control is adapted to obtains the statistical characteristics of the received signal to estimate the input signal power level. The estimated input signal power level is used to adjust the gain of a variable gain amplifier for the automatic gain control.

Abstract: A multiple data-rate receiver uses a signal rate detection technique that employs comparators to obtain information on the incoming data rate for enabling the appropriate receiver. Once a data rate is determined, only then is the appropriate receiver activated. Hence, power dissipation is kept to a minimum during the autonegotiation phase. This is a significant improvement over existing art which require two (or more) receivers to be active during the autonegotiation phase, consequently demanding high power dissipation. Because the autonegotiation phase can be lengthy, the present technique is preferable in many cases.

Abstract: A single, 10/100 twisted-pair driver for Ethernet and Fast Ethernet applications designed in a 0.25 &mgr;m, 3.3V, digital CMOS process is described. The combined driver sources either 10 Mbaud Manchester data or 125 Mbaud MLT3/NRZI data. In 10 Mbaud mode the driver incorporates programmable features such as an additional common-mode current for various linearity performance as function of power supply dissipation. In 125 Mbaud mode of operation, the driver can be configured for different duty cycle distortion and for various rise and fall-times as a trade-off in EMI performance and remote receiver cable length performance. As well, rise/fall-time asymmetry performance is programmable as a trade-off in supply power dissipation. The driver can be run optimally during a transition and sub-optimally when sending multiple “0” symbols to save power. These power dissipation numbers can be reduced through programmable modes when more relaxed performance targets are acceptable.

Abstract: A data receiver equalization technique utilizes a receiver clock that has a frequency different that the incoming data frequency (i.e., an “asynchronous clock”) to asynchronously sample the incoming data waveform. The resulting information about the data, typically including the rise and/or fall times, overshoot and/or undershoot, and amplitude, may be used to equalize a data channel. Other adjustments to the receiver, including the gain level and DC offset compensation, may also be made using the resulting information. An illustrative embodiment using clocked comparators, continuous-time comparators, and a statistical analysis circuit is shown.

Abstract: A line-driver receives as an input a Manchester-encoded digital data signal, which is then converted into an analog signal by a wave-shaping circuit. The wave-shaping circuit comprises a bank of current sources with a combined output, each current source being actuated by a switch controlled by a switching signal. The current sources are each scaled by a coefficient, and are actuated in a selected sequence during each input data pulse, thereby generating as a combined output a staircase pulse signal with n steps for each input data pulse.

Abstract: A method for tuning an adaptive equalizer in order to receive digital signals from a transmission medium both coarse and fine tuning methods to adaptively equalize a signal received from the transmission medium. The coarse tuning method adjusts an equalizer such that the post equalized signal starts to resemble a known data pattern, such as an MLT3 data pattern. The coarse tuning method monitors and corrects for several things: illegal transitions, over equalization, statistical data pattern anomalies and saturation conditions. Fine tuning methods operate concurrently with the coarse tuning methods and function from the point at which the coarse tuning methods stop being efficient. Additionally, the fine tuning methods hold the waveform locked in. In addition to coarse tuning and fine tuning of the equalizer, the present invention also adjusts gain of the overall signal such that the post equalized signal is always a certain amplitude.

Abstract: A resistor-string DAC employing resistors of different value and a switching circuit for producing a differential analog voltage output which can be shaped with respect to an arbitrary common-mode voltage.

Abstract: The invention is a receiver front end for a data communications system having adaptive correction for intersymbol interference, DC offset, baseline wander, and flat loss and related method. Each of the compensation circuits is adaptive and is controlled by adaptation logic via a digital feedback loop including a digital integrator for providing perfect or near-perfect integration of the adaptation algorithm feedback error signal. The architecture further utilizes multiple comparators, including continuous-time and clocked comparators, for separately detecting various aspects of the received data signals that are used to determine the signal degradation characteristics needed to generate error signals for the adaptation feedback loops.