Fractional Delay Filter Decouples Equalizer Complexity

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current fractionally spaced equalizers (FSEs) in optical communication systems are complex, power-hungry, and costly, limiting the increase in bandwidth and throughput due to their design constraints, particularly in meeting the demands of next-generation optical communication systems requiring multi-terabit per second capacities.

Innovation Solution

The implementation of a non-integer, sub-symbol sampling FSE system that decouples channel equalization functions from fractional delay functions, using a fractional delay filter (FDF) and a single adaptive fractionally spaced equalizer (AFSE) to align received samples with transmitted symbols, reducing power consumption and design complexity by operating at a lower sampling rate and employing a single AFSE per output signal from an analog-to-digital converter (ADC).

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If current fractionally spaced equalizers (FSEs) are used to correct sampling phase errors and minimize noise enhancement, then robustness against sampling phase errors is improved, but device complexity and power consumption increase

Engineering Contradiction:
Improverobustness against sampling phase errorsVSAvoidequalizer complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent segments the equalization process into two independent parts: a fractional delay filter that handles sampling phase alignment, and a simplified integer-spaced equalizer that handles noise enhancement. This segmentation allows each component to be optimized independently, reducing overall complexity while maintaining robustness.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces a fractional delay filter as an intermediary component between the ADC and the equalizer. This intermediary handles the sampling phase correction function, allowing the main equalizer to focus solely on noise enhancement mitigation, thereby reducing its complexity.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Reliability

If oversampling is used to enable fractionally spaced equalization, then tolerance against sampling phase errors is improved, but sampling rate requirements and power consumption increase

Engineering Contradiction:
Improvetolerance against sampling phase errorsVSAvoidpower consumption
Core Design Contradiction:
ReliabilityVSUse of energy by moving object

Solution Approach 1:

The patent separates the sampling phase correction function (handled by fractional delay filter) from the equalization function (handled by integer-spaced equalizer). This allows the system to achieve phase error tolerance without requiring high oversampling rates, thereby reducing power consumption.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent changes the sampling rate parameter from high oversampling rates to lower rates that still satisfy the Nyquist criterion. By combining this parameter change with the fractional delay filter, the system maintains phase error tolerance while reducing power consumption.

Inventive Principle:
Principle #35Parameter changes

3Reliability

If fractionally spaced equalizers are used to minimize noise enhancement from spectral nulls, then signal quality is improved, but device complexity and cost increase

Engineering Contradiction:
Improvenoise enhancement minimizationVSAvoidequalizer complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent divides the equalization task into two separate components: fractional delay filtering for phase alignment and integer-spaced equalization for noise enhancement minimization. This segmentation allows the use of simpler integer-spaced equalizer structure while achieving the same noise enhancement minimization效果.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The fractional delay filter acts as an intermediary that pre-aligns the signal phases before the integer-spaced equalizer processes the signal. This intermediary function enables the use of simpler equalizer structure while maintaining the ability to minimize noise enhancement from spectral nulls.

Inventive Principle:
Principle #24Intermediary (Mediator)

4Productivity

If higher sampling rates are used to increase bandwidth and throughput capacity, then capacity is improved, but device complexity and power consumption increase

Engineering Contradiction:
Improvebandwidth and throughput capacityVSAvoidsignal processing circuitry complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent changes the sampling rate parameter from high oversampling rates to lower rates that still satisfy the Nyquist criterion. This parameter change reduces device complexity and power consumption while maintaining the ability to achieve high bandwidth and throughput capacity through efficient signal processing.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP2992654B1Low complexity, adaptive, fractionally spaced equalizer with non-integer sampling
Publication Date: 2017.12.27 HUAWEI TECH CO LTD
  • EP2992654B1 patent drawingFigure 1~3
  • EP2992654B1 patent drawingFigure 4
  • EP2992654B1 patent drawingFigure 5~6

AI summary

An apparatus comprising a memory and a processor coupled to the memory, wherein the memory includes instructions that when executed by the processor cause the apparatus to perform the following: receive an incoming signal at a sampling rate that is greater than a symbol rate associated with the incoming signal, replicate a plurality of data streams from the incoming signal, apply a plurality of fractional delays to the plurality of data streams correspondingly, generate data blocks from each of the fractionally delayed data streams, and perform an adaptive equalization on each of the data blocks sequentially, wherein the fractional delay is applied to the data streams independently of the adaptive equalization, and wherein the adaptive equalization implements taps spaced at a fraction of a symbol interval associated with the incoming signal.