Jitter Mode Setup (Advanced tab)

Advanced with RJ/RN Separation Method

Click Measure > Configure Jitter Mode Measurements to open this dialog.

Use the Advanced tab of the Jitter Mode Measurements Setup dialog, to configure advanced jitter mode settings. In this picture, click on the Separation Method, Stabilization and RJ/RN Compensation tabs to learn more.

Probability/SER

Use to specify PAM4 jitter analysis to be based on probability or Symbol Error Ratio (SER). TJ, TI, eye width, and eye height measurements in the latest standards are made at probabilities instead of SER.

CRE Jitter Optimization

If JSA is available, the CRE Jitter Optimization field will be displayed. Select Enable CRE Jitter Optimization to apply clock-recovery emulation to Jitter Mode's RJ measurement. This feature is only available if JSA is turned on in Jitter mode and the JSA spectrum is set for Embedded. You can also apply jitter optimization by clicking the CRE Jitter Optimize button that is located on the JSA/CRE toolbar. Click the Setup CRE PLL button to configure the clock-recovery PLL emulation.

Specify Pattern Symbol Sequence

Use this setting to specify the input waveform's pattern symbol sequence (NRZ or PAM4). As shown in the following picture, you can select to have the pattern automatically detected (default), select from a standard Known Pattern, or import a BERT Pattern File (*.ptrn). Available known patterns are listed in the following table. If you select to import a PAM4 Pattern File, you should indicate if the pattern uses Gray Coding. Gray coding, or reflected binary code, is a coding pattern where successive symbols differ by one binary bit. For example in the case of PAM4, binary bit sequences 00, 01,10, and 11 represent levels 0, 1, 2, and 3.

Standard Pattern Selections

Selection	Format		Command Argument	Description
	NRZ	PAM4
PRBS 7 (127 symbols)	♦	♦	PRBS7	27−1 pseudo-random symbol sequence (127 symbols).
PRBS 7 (128 symbols)	♦	♦	EPRBs7	27 pseudo-random symbol sequence (128 symbols).
PRBS 9 (511 symbols)	♦	♦	PRBS9	29−1 pseudo-random symbol sequence (511 symbols).
PRBS 9 (512 symbols)	♦	♦	EPRBs9	29 pseudo-random symbol sequence (512 symbols).
PRBS 11 (2047 symbols)	♦	♦	PRBS11	211−1 pseudo-random symbol sequence (2,047 symbols).
PRBS 11 (2048 symbols)	♦	♦	EPRBs11	211 pseudo-random symbol sequence (2,048 symbols).
PRBS 13 (8191 symbols)	♦	♦	PRBS13	213−1 pseudo-random symbol sequence (8,191 symbols).
PRBS 13 (8192 symbols)	♦	♦	EPRBs13	213 pseudo-random symbol sequence (8,192 symbols).
PRBS 15 (32767 symbols)	♦	♦	PRBS15	215−1 pseudo-random symbol sequence (32,767 symbols).
PRBS 15 (32768 symbols)	♦	♦	EPRBs15	215 pseudo-random symbol sequence (32,768 symbols).
Linearity (160 symbols)		♦	LPAM4
K28.5 (20 symbols)	♦		K28P5
PRBS9Q (511 symbols)		♦	PRBQ9	PRBS9Q (511 symbols).
PRBS13Q (8191 symbols)		♦	PRBQ13	PRBS13Q (8,191 symbols).
CEI SSPR (32762 symbols)	♦		SSPR
SSPRQ (65535 symbols)		♦	QSSPr	SSPRQ (65,535 symbols).

 

Separation Method Selection

This setting determines the method used to decompose RJ and RN. There are three choices: Auto, Spectral, and Tail Fit. The recommended setting is Auto, which automatically selects the spectral or tail fit method to ensure the most accurate RJ and RN measurements. A significant presence of Bounded Uncorrelated Interference (BUI) or Bounded Uncorrelated Jitter (BUJ) on a signal, may adversely affect the RJ and RN measurements, and the Tail Fit method is designed to correct this. However, manually selecting the Tail Fit selection is not recommended.

The following table shows how the Separation Method affects the measurement algorithm as well as the measurements reported in the Jitter and Amplitude results tables. When Auto is selected, both the Spectral and the Tail Fit methods are performed for each edge type: rising and falling. The resulting RJ and RN values are compared between the two methods, with the value reported for an edge type based on the criteria shown in the following table. Depending on the data, the value for different edge types can be calculated using different methods. Regardless of the methods used in Auto, the measurement table always reports results for both BUJ and BUI parameters.

The Spectral separation method uses the aliased spectrum, obtained from the RJ/PJ or RN/PI histograms, to measure RJ / RN. This is the Legacy Jitter Mode algorithm and provides the highest accuracy for signals that have small amounts of Bounded Uncorrelated Jitter/Noise (such as cross talk), which can degrade RJ measurements. Uncertainties are determined from the variance of the variance of the RJ/PJ histogram.

The Tail Fit separation method uses the RJ/PJ and RN/PI histogram data also but instead of using the aliased spectrum to compute RJ and RN this method performs a straight-line fit to the Q-scale tails of the Cumulative Distribution Function (CDF). This method is optimized for data signals with medium to high levels of Bounded Uncorrelated Jitter/Noise (such as cross talk). Uncertainties for the Tail Fit method are determined by the variance of the slope of the straight line fit of the Q-scale tails of the CDF.

Effects of Separation Method Selection on Algorithm and Measurements Reported

Separation Selection	Separation Method Used	Measurement Reported in Tables
		Jitter Table	Amplitude Table
Auto Selected
Definitions: TF = Tail Fit method. SP = Spectral method.
If RJ TF + tail-fit uncertainty ≥ RJ SP – spectral uncertainty	Spectral	BUJ (rms), BUJ (δ - δ)	not applicable
If RJ TF + tail-fit uncertainty < RJ SP − spectral uncertainty	Tail Fit
If RN TF + tail-fit uncertainty ≥ RN SP – spectral uncertainty	Spectral	not applicable	BUI (rms), BUI (δ - δ)
If RN TF + tail-fit uncertainty < RN SP − spectral uncertainty	Tail Fit
Spectral Selected	Spectral	PJ (rms), PJ (δ - δ)	PI (rms), PI (δ - δ)
Tail Fit Selected	Tail Fit	BUJ (rms), BUJ (δ - δ)	BUI (rms), BUI (δ - δ)