TECHNICAL FIELD
This disclosure relates generally to electrical communication.
BACKGROUND
A transmission channel may distort high-frequency (HF) signals communicated through it. The distortion may be a result of frequency-dependent signal attenuation caused, for example, by skin effect or dielectric effect in the transmission channel and may vary according to one or more characteristics of the transmission channel, such as a length or an insulator material of the transmission channel. To compensate for the distortion, a transmitter transmitting signals through the transmission channel may include a pre-emphasis driver. A receiver receiving the signals through the transmission channel may include one or more equalizers.
BRIEF DESCRIPTION OF THE DRAWINGS
FIG. 1 illustrates example pulse responses of an example transmission channel, an example linear equalizer (LE), and an example DFE.
FIG. 2 illustrates an example receiver architecture.
FIG. 3 illustrates example cancellation of intersymbol interference (ISI) by the receiver architecture in FIG. 2.
FIG. 4 illustrates another example receiver architecture.
FIG. 5 illustrates example cancellation of ISI by the receiver architecture in FIG. 4.
FIG. 6 illustrates another example receiver architecture.
FIG. 7 illustrates example cancellation of ISI by the receiver architecture in FIG. 6.
FIG. 8 illustrates an example DFE method.
DESCRIPTION OF EXAMPLE EMBODIMENTS
A transmission channel may distort high-frequency (HF) signals communicated through it. The distortion may be a result of frequency-dependent signal attenuation caused, for example, by skin effect or dielectric effect in the transmission channel and may vary according to one or more characteristics of the transmission channel, such as a length or an insulator material of the transmission channel. To compensate for the distortion of HF signals communicated through a transmission channel, a receiver receiving the signals through the transmission channel may include one or more equalizers to increase the effective length of the transmission channel or to increase communication speed. For example, a receiver may include a DFE, which tend to have greater noise immunity than LEs. A DFE requires parameters specifying the amount of components to apply to the received signals. Adaptive parameters are often desirable because one or more of the characteristics of the transmission channel are often unknown. In particular, it is often desirable to have the DFE apply an amount of compensation that matches the level of attenuation of the received signals from communication across the transmission channel, so that output characteristics of the received signals remain consistently independent of the transmission channel.
The present disclosure incorporates by reference U.S. Patent Application Publication No. 2007/0280383, published 6 Dec. 2007, entitled System and Method for Adjusting Compensation Applied to a Signal.
FIG. 1 illustrates example pulse responses of an example transmission channel, an example LE, and an example DFE. The transmission channel communicates a signal from a transmitter to a receiver that includes the LE and the DFE, which process the received signal. The LE and DFE may communicate the processed signal in any suitable manner to a decision circuit or any other circuit components. After transmission over the channel, the received signal (or pulse) has a long tail due to high-frequency loss in the transmission channel. The long tail causes postcursor ISI because the long tail interferes with successively transmitted symbols. Particular embodiments use an LE to cancel such interference. In particular embodiments, there may be two symbols: such as 0 and 1 or −1 and +1. The present disclosure contemplates any suitable symbols. In FIG. 1, the LE moderately amplifies an attenuated high-frequency component of the received signal to reduce residual ISI occurring after a particular delay. An LE boosts the gain for HF components, but it does not distinguish crosstalk and reflection noise (which are often severe in backplane systems) from the received signal. Crosstalk and reflection are HF noise, and the LE amplifies them together with the received signal. To compensate, the bandwidth of the LE may be reduced to a quarter of the data rate of the received signal. In FIG. 1, the DFE cancels the residual ISI occurring approximately immediately after the pulse. If the DFE is a 1-tap DFE, it may use a feedback loop from a decision circuit in the DFE to cancel residual ISI occurring after a delay of 1.0 UI.
FIG. 2 illustrates an example receiver architecture that includes a continuous-time linear equalizer (CTLE) and 1-tap DFE with direct feedback. The CTLE cancels precursor ISI and boosts low-to-middle frequency components in the received signal. The 1-tap DFE employs direct feedback to cancel ISI on a 1.0-UI-delayed data value and to cancel ISI on 0.5-UI-delayed and 1.5-UI-delayed boundary values, as FIG. 3 illustrates. In FIG. 2, the clock and data recovery (CDR) circuit takes the outputs of the boundary and data decision latches (DLs), shifts the boundary clock (BCLK) to an earlier or later phase to lock the incoming data boundary. The data clock (DCLK) is delayed 0.5 UIs with respect to BCLK. The DCLK samples at an approximate center of the data eye.
The following is a typical problem with the system of FIG. 2. Feedback from the data slicer must settle before the boundary slicer. As a result, the critical path includes the delay associated with the data slicer Tclk-to-q, the delay associated with multiplication with the reference voltage Vref, the delay associated with the summation, and the setup time of the boundary slicer. The sum of all the delay must be less than 0.5 UIs, which is difficult to achieve at high data rates. As an example and not by way of limitation, for a data rate of 10 Gbps, 0.5 UIs is only 50 picoseconds.
FIG. 4 illustrates another example receiver architecture that includes a CTLE and a 1-tap DFE. The system of FIG. 4 uses speculative or loop-unfolding techniques to ease the critical timing path. It precalculates an equalized eye for either prior input-data polarity, latches both results, and chooses a result once the previous bit decision is known. The phase detector takes the data without ISI cancellation on the 0.5 UI and 1.5 UI boundaries. The CTLE cancels precursor ISI and boosts low-to-middle frequency components. The 1-tap DFE cancel ISI on 1.0-UI data, as FIG. 5 illustrates. The CDR circuit takes the outputs of the boundary and data DLs, shifts BCLK to an earlier or later phase to lock the incoming data boundary. DCLK is delayed 0.5 UIs with respect to BCLK.
The following is a typical problem with the system of FIG. 4. ISI from the signal received 0.5 UIs prior to the boundary sample produces approximately x UIs of delay in zero crossing. The ISI contribution from the bit received 1.5 UIs before the boundary sample may be written as ±Δ, where the sign depends on the bit that produces the ISI. If binary 1 and binary 0 appear randomly, the CDR loop averages the contribution (+A) to zero. So the net contribution of the ISI is x-UI delay. Depending on the amount of ISI, the actual delay x may vary from 0.0 to 0.5 UIs. Uncancelled boundary ISI delays the data boundary, and DCLK is not at a center of the data eye, which reduces timing margin.
FIG. 6 illustrates another example receiver architecture. In the system of FIG. 6, DCLK is a delayed version of BCLK. Due to uncancelled boundary ISI, the data boundary is delayed. To make DCLK sample at an approximate center of the data eye, DCLK is delayed less than 0.5 UIs with respect BCLK. In particular embodiments, the minimum possible delay is 0.0 UIs, which completely eliminates the delay circuit. FIG. 7 illustrates example cancellation of ISI by the receiver architecture in FIG. 6. In FIG. 7, 0.3 UIs is an optimal data sampling point. In particular embodiments, such sampling improves timing margin, which is important in high data-rate transmissions. In particular embodiments, compared with the system of FIG. 2, the system of FIG. 6 eases the critical timing path and interleaves the data, which provides a better solution for higher-speed operation. In particular embodiments, compared with the system of FIG. 4, the system of FIG. 6 shifts the data sampling clock to an optimal eye center with little or no design effort. In addition, in particular embodiments, improving timing margin further facilitates higher data-rate transmission.
FIG. 8 illustrates an example DFE method. The method begins at step 800, where a data DL and boundary DL receive an input signal from a receiver. The input signal includes ISI when communicated from the receiver. At step 802, the data DL receives a DCLK signal. At step 804, the boundary DL receives a BCLK signal. At step 806, based on the input signal and the DCLK signal, the data DL recovers data from the input signal to produce a first output signal. At step 808, based on the input signal and the BCLK signal, the boundary DL recovers boundaries between bits in the input signal to produce a second output signal. At step 810, based on the first and second output signals, a CDR circuit produces the DCLK and BCLK signals, with the DCLK signal being delayed with respect to the BCLK signal less than approximately 0.5 unit intervals (UIs) and greater than or equal to approximately zero UIs, at which point the method ends. Particular embodiments may continuously repeat the steps of the method of FIG. 8, according to particular needs. Although the present disclosure describes and illustrates particular steps of the method of FIG. 8 as occurring in a particular order, the present disclosure contemplates any suitable steps of the method of FIG. 8 occurring in any suitable order. Although the present disclosure describes and illustrates particular components carrying out particular steps of the method of FIG. 8, the present disclosure contemplates any suitable components carrying out any suitable steps of the method of FIG. 8.
The present disclosure encompasses all changes, substitutions, variations, alterations, and modifications to the example embodiments herein that a person having ordinary skill in the art would comprehend. Similarly, where appropriate, the appended claims encompass all changes, substitutions, variations, alterations, and modifications to the example embodiments herein that a person having ordinary skill in the art would comprehend.
Patent Grant
8270464
