公开(公告)号：US10020967B1

公开(公告)日：2018-07-10

申请号：US15443791

申请日：2017-02-27

Applicant: Intel Corporation

Inventor： Mike Peng Li ,  Hsinho Wu ,  Masashi Shimanouchi

IPC: H04L27/04 ,  H04L27/06 ,  H04L7/02

CPC classification number: H04L27/06 ,  H04L1/00 ,  H04L1/205 ,  H04L25/03031 ,  H04L25/061 ,  H04L27/04

Abstract: A method includes receiving an n-level Pulse Amplitude Modulated (PAM-n) signal at a receiver from a transmitter via a channel. The method also includes determining one or more sampling times of the PAM-n signal. The method further includes determining one or more slicing levels of the PAM-n signal. The method also includes extracting and decomposing jitter in the PAM-n signal for each slicing level of the PAM-n signal to determine one or more jitter components. The method further includes extracting and decomposing noise in the PAM-n signal for each data level of the PAM-n signal to determine one or more noise components. The method also includes adjusting the receiver, the transmitter, the channel, or any combination thereof, based on the one or more jitter components, the one or more noise components, or both.

公开(公告)号：US10476716B2

公开(公告)日：2019-11-12

申请号：US16029900

申请日：2018-07-09

Applicant: Intel Corporation

Inventor： Mike Peng Li ,  Hsinho Wu ,  Masashi Shimanouchi

IPC: H04L27/06 ,  H04L27/04 ,  H04L1/00 ,  H04L25/03 ,  H04L25/06 ,  H04L1/20

Abstract: A method includes receiving an n-level Pulse Amplitude Modulated (PAM-n) signal at a receiver from a transmitter via a channel. The method also includes determining one or more sampling times of the PAM-n signal. The method further includes determining one or more slicing levels of the PAM-n signal. The method also includes extracting and decomposing jitter in the PAM-n signal for each slicing level of the PAM-n signal to determine one or more jitter components. The method further includes extracting and decomposing noise in the PAM-n signal for each data level of the PAM-n signal to determine one or more noise components. The method also includes adjusting the receiver, the transmitter, the channel, or any combination thereof, based on the one or more jitter components, the one or more noise components, or both.

公开(公告)号：US20180316527A1

公开(公告)日：2018-11-01

申请号：US16029900

申请日：2018-07-09

Applicant: Intel Corporation

Inventor： Mike Peng Li ,  Hsinho Wu ,  Masashi Shimanouchi

IPC: H04L27/06 ,  H04L27/04 ,  H04L7/02

Abstract: A method includes receiving an n-level Pulse Amplitude Modulated (PAM-n) signal at a receiver from a transmitter via a channel. The method also includes determining one or more sampling times of the PAM-n signal. The method further includes determining one or more slicing levels of the PAM-n signal. The method also includes extracting and decomposing jitter in the PAM-n signal for each slicing level of the PAM-n signal to determine one or more jitter components. The method further includes extracting and decomposing noise in the PAM-n signal for each data level of the PAM-n signal to determine one or more noise components. The method also includes adjusting the receiver, the transmitter, the channel, or any combination thereof, based on the one or more jitter components, the one or more noise components, or both.

公开(公告)号：US20190190653A1

公开(公告)日：2019-06-20

申请号：US15842331

申请日：2017-12-14

Applicant: Intel Corporation

Inventor： Martin Langhammer ,  Mike Peng Li ,  Masashi Shimanouchi

IPC: H04L1/00 ,  H03M13/15

CPC classification number: H04L1/0058 ,  H03M13/1515 ,  H04L1/0061

Abstract: Network communication systems may employ coding schemes to provide error checking and/or error correction. Such schemes may include parity or check symbols in a message that may add redundancy, which may be used to check for errors. For example, Ethernet may employ forward error correction (FEC) schemes using Reed-Solomon codes. An increase in the number of parity symbols may increase the power of the error-correcting scheme, but may lead to an increased in latencies. Encoders and decoders that may be configured in a manner to produce variable-length messages while preserving compatibility with network standards are described. Decoders described herein may be able to verify long codewords by checking short codes and integrating the results. Encoders described herein may be able to generate codewords in multiple formats without replicating large segments of the circuitry.