USB 3.2 LFSR Descrambler Seed Locking for Faster Decoding

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

The USB 3.2 Gen 2 standard's 23-bit scrambler requires a lengthy process to find the right seed, slowing down data decoding, and existing methods for seed extraction involve additional probing or time-consuming search algorithms.

Innovation Solution

A method using a test and measurement instrument to dynamically extract and validate the seed value of a Linear Feedback Shift Register (LFSR) by rotating a known portion of the LFSR and comparing it with incoming data, eliminating the need for side band signal probing and search algorithms.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Object-affected harmful factors

If a 23-bit scrambler is used in USB 3.2 Gen 2, then EMI reduction is improved, but the time to find the right seed increases significantly

Engineering Contradiction:
ImproveEMI reductionVSAvoidtime to find seed
Core Design Contradiction:
Object-affected harmful factorsVSLoss of time

Solution Approach 1:

The patent applies preliminary action by extracting and validating the seed value during the link training phase before actual data transmission begins. The test and measurement instrument captures the LFSR state during training sequences (such as ULPI training or USB 3.2 Gen 2 training) and computes the seed value in advance, so that when data decoding starts, the seed is already determined and can be used immediately without time-consuming search algorithms.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent uses feedback by continuously monitoring the LFSR state during link training and using this feedback to dynamically extract and validate the seed value. The instrument compares the extracted seed with expected values or validates it through subsequent training sequences to ensure correctness, creating a closed-loop feedback mechanism that ensures accurate seed determination without brute-force search.

Inventive Principle:
Principle #23Feedback

2Productivity

If traditional seed extraction methods are used, then decoding can proceed, but additional probing channels and search algorithms are required

Engineering Contradiction:
Improvedecoding capabilityVSAvoidprobing channels and search algorithms
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent applies the extraction principle by isolating and extracting the seed value from the LFSR state during link training. The test and measurement instrument captures the LFSR register state at specific points during training sequences and computes the seed value directly from this extracted state, eliminating the need for complex search algorithms or additional probing channels that would otherwise be required to determine the seed.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent implements self-service by enabling the link training process itself to provide the necessary information for seed extraction. The training sequences and LFSR state transitions during normal USB initialization serve the dual purpose of both link training and seed determination, allowing the system to self-produce the seed value without requiring external search algorithms or additional measurement channels.

Inventive Principle:
Principle #25Self-service

Data Source

PatentUS20250265161A1Dynamical locking of seed value for LFSR descrambler for USB
Publication Date: 2025.08.21 TEKTRONIX INC
  • US20250265161A1 patent drawing
  • US20250265161A1 patent drawing
  • US20250265161A1 patent drawing

AI summary

A test and measurement instrument has one or more channels to receive data from a device under test through a test fixture, the data being transmitted by the DUT in accordance with the Universal Serial Bus protocol 3.x, and one or more processors to: acquire a state of a linear feedback shift register (LFSR), extract a known portion of the LFSR as an LFSR value; shift the LFSR value a predetermined number of times to produce a new value of the LFSR, compare the new value of the LFSR to a value of incoming data to determine if the incoming data comprises a potential control code, repeat to acquire a predetermined number of potential control codes, compare them to known control code values to determine if they are valid, and if so, lock a seed used to descramble the remaining incoming data.