GFDM Transmultiplexer Complexity Reduction via Time Domain Over-Sampling

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

GFDM transmission techniques face high complexity and non-orthogonality issues due to over-sampling in the frequency domain, leading to inefficiencies and increased operational complexity, especially when implementing transmultiplexers for high values of M or K.

Innovation Solution

A method that transforms data symbols from the frequency domain to the time domain, followed by cyclic repetition and filtering with a shaping filter, reducing complexity by eliminating the need for large inverse transforms and achieving quasi-orthogonality through a modified square-root raised cosine filter with optimized span parameters.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Manufacturing precision

If over-sampling is implemented in the frequency domain sub-carrier by sub-carrier, then spectrum improvement is achieved, but device complexity increases significantly

Engineering Contradiction:

Improvespectrum qualityVSAvoidtransmultiplexer complexity

Core Design Contradiction:

Manufacturing precisionVSDevice complexity

Solution Approach 1:

The patent inverts the conventional GFDM processing order by performing over-sampling in the time domain first, then applying filtering in the frequency domain, rather than the traditional approach of frequency domain over-sampling followed by time domain filtering. This inversion reduces computational complexity while maintaining spectral properties

Inventive Principle:

Principle #13The other way round (Inversion)

Solution Approach 2:

The patent segments the block of M×K data symbols into M columns, where each column corresponds to a sub-carrier. By processing each column independently through time domain over-sampling and then applying frequency domain filtering, the system achieves spectral improvement with reduced overall complexity compared to processing the entire block simultaneously in the traditional manner

Inventive Principle:

Principle #1Segmentation

2Device complexity

If GFDM is implemented with frequency domain over-sampling, then orthogonality is lost, but if direct implementation is used, then operational complexity becomes too high

Engineering Contradiction:

Improveoperational complexityVSAvoidorthogonality

Core Design Contradiction:

Device complexityVSReliability

Solution Approach 1:

The patent introduces a shaping filter as an intermediary element that operates in the frequency domain after time domain over-sampling. This filter serves as a mediator that restores orthogonality properties by shaping the spectral content appropriately, while the time domain over-sampling handles the expansion without directly causing orthogonality loss

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

By inverting the processing sequence and performing time domain over-sampling before frequency domain filtering, the patent avoids the orthogonality degradation that occurs with conventional frequency domain over-sampling, while still achieving the necessary signal expansion for GFDM operation

Inventive Principle:

Principle #13The other way round (Inversion)

3Productivity

If large blocks of data symbols are processed, then transmission capacity increases, but complexity of implementation becomes prohibitive

Engineering Contradiction:

Improvetransmission capacityVSAvoidimplementation complexity

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The patent segments the M×K block of data symbols into M independent columns, each representing a sub-carrier with K time slots. This column-wise segmentation allows independent processing of each sub-carrier through time domain over-sampling, followed by efficient frequency domain filtering, thereby managing complexity while maintaining high transmission capacity through large block processing

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent employs dynamic processing where the over-sampling factor N can be adjusted independently for each sub-carrier column, and the shaping filter parameters can be optimized based on channel conditions. This dynamic approach allows flexible adaptation to different transmission requirements while managing computational complexity through selective processing

Inventive Principle:

Principle #15Dynamics

Data Source

PatentUS10103922B2Methods, devices pertaining to transmission of multi-carrier signal

Publication Date: 2018.10.16 ORANGE SA

AI summary

A method is provided for transmitting complex data symbols, supplying a multiple carrier signal. The method includes shaping at least one block of M×K complex data symbols, where M>1 and K>1, implementing the following acts: for at least one column of the block, conversion of the M complex data symbols of the column from the frequency domain to the time domain, supplying N converted symbols; cyclic repetition of the N converted symbols, supplying NK repeated symbols; cyclic repetition of the transformed N symbols, supplying NK repeated symbols; filtering the NK repeated symbols via a shaping filter, supplying NK filtered symbols; and summation of the obtained filtered symbols for the various columns of the block, supplying NK time samples.