DFEClkFwd

DFE with Clock Forwarding in the receiver AMI model

Since R2023b

• 

×

Libraries:
SerDes Toolbox / Datapath Blocks

Description

In a parallel link channel, a separate clock (or strobe) signal is sent along with the data signals, which is used to latch each individual data bit. This is known as clock-forwarding. The IBIS-AMI Specification describes a mechanism by which the clock waveform, or recovered clock times, can be passed to the data AMI receiver model to allow the model maker access to the clock. This enables the receiver model to include clock phase noise, clock tree delays, phase interpolators, and so on.

You can use the DFEClkFwd block to simulate and generate the clock forwarding DLLs according to the IBIS-AMI Specification.

Ports

Input

WaveIn — Input baseband signal

scalar | vector

Input baseband signal, can be a sample-by-sample signal specified as a scalar, or an impulse response vector signal.

Data Types: double

Output

WaveOut — Estimated channel output

scalar | vector

Estimated channel output. If the input signal is a sample-by-sample signal specified as a scalar, the output is also scalar. If the input signal is an impulse response vector signal, the output is also a vector.

Data Types: double

Parameters

DFE

Mode — DFE operating mode

Fixed (default) | Off | Adapt

DFE operating mode:

• Off — DFE is bypassed and the input waveform remains unchanged.

• Fixed — DFE applies the input DFE tap weights specified in Initial tap weights (V) to the input waveform.

• Adapt — The DFE block finds the optimum DFE tap values for the best eye height opening for statistical analysis. During time domain simulation, the block uses the adapted values as the starting point and applies them to the input waveform. For more information, see Statistical Analysis in SerDes Systems.

Programmatic Use

• Use get_param(gcb,'Mode') to view the current DFEClkFwd Mode.

• Use set_param(gcb,'Mode',value) to set DFEClkFwd to a specific Mode.

Initial tap weights (V) — Initial DFE tap weights

[0 0 0 0] (default) | row vector

Initial DFE tap weights, specified as a row vector in volts. The length of the vector specifies the number of DFE taps. The vector element value specifies the strength of the tap at that element position. Setting a vector element value to zero only initializes the tap.

You can use a valid MATLAB expression to evaluate the Initial tap weights (V) row vector.

Example: set_param(gcb,'TapWeights',"zeros(1,100)") creates a DFE with 100 taps.

Programmatic Use

• Use get_param(gcb,'TapWeights') to view the current value of DFEClkFwd Initial tap weights (V).

• Use set_param(gcb,'TapWeights',value) to set DFEClkFwd to a specific Initial tap weights (V) vector value.

Data Types: double

Adaptive gain — Controls DFE tap weight update rate

9.6e-5 (default) | positive real scalar

Controls DFE tap weight update rate, specified as a unitless positive real scalar. Increasing the value of Adaptive gain leads to a faster convergence of DFE adaptation at the expense of more noise in DFE tap values.

Programmatic Use

• Use get_param(gcb,'EqualizationGain') to view the current DFEClkFwd Adaptive gain value.

• Use set_param(gcb,'EqualizationGain',value) to set DFEClkFwd to a specific value of Adaptive gain.

Data Types: double

Adaptive step size (V) — DFE adaptive step resolution

1e-06 (default) | nonnegative real scalar | nonnegative real-valued row vector

DFE adaptive step resolution, specified as a nonnegative real scalar or a nonnegative real-valued row vector in volts. Specify as a scalar to apply to all the DFE taps or as a vector that has the same length as the Initial tap weights (V).

Adaptive step size (V) specifies the minimum DFE tap change from one time step to the next to mimic hardware limitations. Setting Adaptive step size (V) to 0 yields DFE tap values without any resolution limitation.

Programmatic Use

• Use get_param(gcb,'EqualizationStep') to view the current DFEClkFwd Adaptive step size (V) value.

• Use set_param(gcb,'EqualizationStep',value) to set DFEClkFwd to a specific value of Adaptive step size (V).

Data Types: double

Minimum DFE tap value (V) — Minimum value of adapted taps

-1 (default) | real scalar | real-valued row vector

Minimum value of the adapted taps, specified as a real scalar or a real-valued row vector in volts. Specify as a scalar to apply to all the DFE taps or as a vector that has the same length as the Initial tap weights (V).

Programmatic Use

• Use get_param(gcb,'MinimumTap') to view the current DFEClkFwd Minimum DFE tap value (V) value.

• Use set_param(gcb,'MinimumTap',value) to set DFEClkFwd to a specific value of Minimum DFE tap value (V).

Data Types: double

Maximum DFE tap value (V) — Maximum value of adapted taps

1 (default) | nonnegative real scalar | nonnegative real-valued row vector

Maximum value of the adapted taps, specified as a nonnegative real scalar or a nonnegative real-valued row vector in volts. Specify as a scalar to apply to all the DFE taps or as a vector that has the same length as the Initial tap weights (V).

Programmatic Use

• Use get_param(gcb,'MaximumTap') to view the current DFEClkFwd Maximum DFE tap value (V) value.

• Use set_param(gcb,'MaximumTap',value) to set DFEClkFwd to a specific value of Maximum DFE tap value (V).

Data Types: double

2x tap weights — Multiply DFE tap weights by a factor of two

on (default) | off

Select to multiply the DFE tap weights by a factor of two.

The output of the slicer in the DFEClkFwd block from the SerDes Toolbox™ is [-0.5 0.5]. But some industry applications require the slicer output to be [-1 1]. 2x tap weights allows you to quickly double the DFE tap weights to change the slicer reference.

CDR

Phase offset (symbol time) — Manual clock phase offset

0 (default) | real scalar in the range [−0.5, 0.5]

Manual clock phase offset to move the recovered clock phase, specified as a real scalar in the range [-0.5, 0.5] in the fraction of symbol time. Phase offset (symbol time) is used to manually shift the clock probability distribution function (PDF) for a better bit error rate (BER).

Programmatic Use

• Use get_param(gcb,'PhaseOffset') to view the current DFEClkFwd Phase offset (symbol time) value.

• Use set_param(gcb,'PhaseOffset',value) to set DFEClkFwd to a specific value of Phase offset (symbol time).

Data Types: double

Reference offset (ppm) — Reference clock offset impairment

0 (default) | real scalar in the range [−300, 300]

Reference clock offset impairment, specified as a real scalar in the range [−300, 300] in parts per million (ppm). Reference offset (ppm) is the deviation between transmitter oscillator frequency and receiver oscillator frequency.

Programmatic Use

• Use get_param(gcb,'ReferenceOffset') to view the current DFEClkFwd Reference offset (ppm) value.

• Use set_param(gcb,'ReferenceOffset',value) to set DFEClkFwd to a specific value of Reference offset (ppm).

Data Types: double

Early/late count threshold — Early or late CDR count threshold to trigger phase update

5 (default) | positive real integer ≥5

Early or late CDR count threshold to trigger a phase update, specified as a unitless positive real integer ≥5. Increasing the value of Early/late count threshold provides a more stable output clock phase at the expense of convergence speed. Because the bit decisions are made at the clock phase output, a more stable clock phase has a better bit error rate (BER).

Early/late count threshold also controls the bandwidth of the CDR.

Programmatic Use

• Use get_param(gcb,'Count') to view the current DFEClkFwd Early/late count threshold value.

• Use set_param(gcb,'Count',value) to set DFEClkFwd to a specific value of Early/late count threshold.

Data Types: double

Step (symbol time) — Clock phase resolution

0.005 (default) | real scalar

Clock phase resolution of the recovered clock, specified as a real scalar in fraction of symbol time. Step (symbol time) is the inverse of the number of phase adjustments in the CDR. If the CDR has 128 steps of phase adjustment, the Step (symbol time) value is 1/128.

Programmatic Use

• Use get_param(gcb,'ClockStep') to view the current DFEClkFwd Step (symbol time) value.

• Use set_param(gcb,'ClockStep',value) to set DFEClkFwd to a specific value of Step (symbol time).

Data Types: double

Sensitivity (V) — Sampling latch metastability voltage

0 (default) | real scalar

Sampling latch metastability voltage, specified as a real scalar in volts. If the data sample voltage lies within the region of (±Sensitivity (V)), there is a 50% probability of bit error.

Programmatic Use

• Use get_param(gcb,'Sensitivity') to view the current DFEClkFwd Sensitivity (V) value.

• Use set_param(gcb,'Sensitivity',value) to set DFEClkFwd to a specific value of Sensitivity (V).

Data Types: double

More About

Specify Content of Clock Times Input

To specify the content of the clock_times input, open the SerDes IBIS-AMI manager dialog box and select the Rx_Use_Clock_Input reserved AMI parameter in the AMI-Rx tab. You can choose between:

• None — No clock is forwarded.

• Times — Forward clock from external clock times data.

• Wave — Forward clock from generated or external waveform data.

The block uses the CDR only when Rx_Use_Clock_Input parameter is set to None.

Note

You must click the Refresh Init button in the Rx Init subsystem after modifying the Rx_Use_Clock_Input parameter.

Clock Forwarding in Stimulus

When you enable clock forwarding, you can use the Clock Forwarding tab in the Stimulus block to generate clock stimulus patterns. For more information, see Stimulus.

Version History

Introduced in R2023b