IP Library Granted Patent US 10,999,051
Granted Patent B2
US 10,999,051 · App. 16/905,635 · Granted May 4, 2021

Reference noise compensation for single-ended signaling

Inventors: Xi Chen (Milpitas, CA); Nikola Nedovic (San Jose, CA); Carl Thomas Gray (Apex, NC); Stephen G Tell (Chapel Hill, NC)
Assignee: NVIDIA Corp.
H04L7/0079H03K3/037H04B1/12H04B1/16H04L7/0016
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 10,999,051
App. No.
16/905,635
Granted
May 4, 2021
Kind
B2
Abstract

A receiver circuit includes a clock lane propagating a clock signal. A self-sampled clock applies a delayed version of the clock signal to the clock signal and compensation logic controls an amount of delay of the delayed version of the clock, based on a reference voltage offset (difference) between the receiver and a transmitter. The delayed version of the clock is centered on one unit interval of the clock. An offset correction based on a clock duty cycle error is applied to data receiver front ends.

Claims (37)

1. A circuit comprising:

a tunable receiver for a data signal;

a clock signal input;

a sampling circuit configured to generate a sample of a clock signal received at the clock signal input using a one unit interval delayed version of the clock signal; and

a compensation circuit to generate an offset signal from the sample of the clock signal and to apply the offset signal to the tunable receiver for the data signal.

2. The signal receiver circuit of claim 1 , further wherein the offset compensates for a reference voltage error of the clock signal input.

3. The signal receiver circuit of claim 2 , wherein:

the sampling circuit is configured to sample the clock signal at a half-rate; and

the offset generated by the compensation circuit is based on the reference voltage error.

4. The signal receiver circuit of claim 2 , wherein:

the sampling circuit is configured to interpret a sampled “1” as a positive reference voltage error; and

the sampling circuit is configured to interpret a sampled “0” as a negative reference voltage error.

5. The signal receiver circuit of claim 1 , the compensation circuit comprising a configurable step size for an offset compensation loop coupled to the clock signal input.

6. A receiver circuit coupled to receive a clock signal from a transmitter circuit, the receiver circuit comprising:

a clock sampler coupled to a clock lane configured to propagate a clock signal;

a data lane receiver;

the clock sampler configured to apply a delayed version of the clock signal to the clock signal to generate a self-sampled clock signal;

compensation logic to determine a reference voltage error between the receiver circuit and the transmitter circuit based on the self-sampled clock signal; and

logic to apply an offset voltage to the data lane receiver, the offset voltage applied to the data lane receiver based on the reference voltage error.

7. The receiver circuit of claim 6 configured to center the delayed version of the clock signal on one unit interval of the clock signal.

8. The receiver circuit of claim 6 configured to apply an offset voltage to the clock lane, the offset voltage applied to the clock lane based on the reference voltage error.

9. The receiver circuit of claim 8 , wherein the compensation logic comprises a configurable offset voltage adjustment step size.

10. The receiver circuit of claim 6 wherein the compensation logic is configured to apply a trim to the delayed version of the clock signal.

11. The receiver circuit of claim 10 , wherein the trim is based on environmental conditions of the receiver circuit.

12. The receiver circuit of claim 10 , the compensation logic configured to control an amount of the trim based on a number of sampled “1”s and “0”s sampled in the clock signal.

13. The receiver circuit of claim 6 , the compensation logic configured to receive a de-serialized clock signal formed from the clock signal.

14. The receiver circuit of claim 6 , further comprising a configurable delay at an input of the clock sampler.

15. The receiver circuit of claim 6 , further comprising a dithering circuit configured to affect a sample interval of the clock sampler.

16. A method in a digital receiver, the method comprising:

receiving a clock signal;

applying the clock signal to a data receiver;

sampling the clock signal with a delayed version of the clock signal centered on one unit interval of the clock signal to generate a count; and

adjusting an offset voltage of the data receiver based on the count.

17. The method of claim 16 , wherein the clock signal is sampled at a half-rate.

18. The method of claim 16 , wherein the count comprises a combination of sampled even bits and sampled odd bits of the clock signal.

19. The method of claim 18 , wherein the combination is a sum.

20. The method of claim 18 , wherein the delay of sampling the clock signal is dithered.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jul 23, 2020
From: CHEN, XI; NEDOVIC, NIKOLA; GRAY, CARL THOMAS; TELL, STEPHEN G
To: NVIDIA CORP.
Reel/Frame 053290/0934 →
Continuity (2)
Provisional Application 62902232 · Sep 18, 2019
Related Publication 20210083836A1 · Mar 18, 2021