Composite Synchronization Sequences for 5G PRACH

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

In 5G radio communication systems, the need for special large discrete Fourier transforms in ASICs for each antenna branch is undesirable, and existing methods for generating synchronization sequences face challenges with coexistence of different signal numerologies and limited sequence lengths with good correlation properties.

Innovation Solution

A method for generating a composite synchronization sequence by combining Zadoff-Chu and m-sequences of unequal lengths, extending them to a common length in the frequency domain, and then multiplying them element-wise in the time domain to achieve sequences with good auto- and cross-correlation properties, allowing for a larger number of sequences with desired properties.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If a special long OFDM symbol with large DFT is used for PRACH to handle large round-trip delay, then detection accuracy is improved, but device complexity and implementation cost increase

Engineering Contradiction:
Improvedetection accuracyVSAvoidcomplexity of large DFT in ASIC
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent segments the long synchronization sequence into multiple shorter sequences (first sequence and second sequence of different lengths). Instead of using a single long sequence requiring large DFT, the system transmits multiple shorter sequences that can be processed with smaller, more manageable DFT sizes in ASIC, thereby reducing implementation complexity while maintaining detection capability for large round-trip delays

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent transitions from a single-dimensional long sequence approach to a multi-dimensional approach by combining sequences of different lengths (e.g., different Zadoff-Chu root indices and different m-sequence lengths). This dimensional diversity in sequence length and type allows the system to achieve long effective sequence duration without requiring a single large DFT operation

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

2Reliability

If sequences of different lengths are combined, then the number of sequences with good correlation properties increases, but the complexity of sequence generation and processing increases

Engineering Contradiction:
Improvenumber of sequences with good correlation propertiesVSAvoidcomplexity of sequence generation
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent changes key parameters of the synchronization sequences, specifically using sequences of different lengths (e.g., first sequence length of 839 or 139, second sequence length of 127 or 63). By varying these length parameters and combining them through element-wise multiplication after appropriate zero-padding, the system generates a larger pool of sequences with good auto- and cross-correlation properties without excessively increasing generation complexity

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent creates composite synchronization sequences by combining two different sequence types (e.g., Zadoff-Chu sequence and m-sequence) of different lengths. This composite approach leverages the complementary properties of different sequence families, achieving robust correlation properties that neither sequence type alone could provide, while the combination method keeps processing complexity manageable

Inventive Principle:
Principle #40Composite materials

Data Source

PatentEP3577804B1Combining synchronization sequences of different lengths
Publication Date: 2021.07.28 TELEFONAKTIEBOLAGET LM ERICSSON (PUBL)
  • EP3577804B1 patent drawingFigure 1
  • EP3577804B1 patent drawingFigure 2~3
  • EP3577804B1 patent drawingFigure 4

AI summary

A system and method for generating a composite synchronization sequence in a communication system. In one embodiment, the apparatus (110, 200, 120, 300) is configured to provide a first synchronization sequence and a second synchronization sequence in a first domain, transform the first synchronization sequence and the second synchronization sequence into a second domain, and extend the first synchronization sequence and the second synchronization sequence in the second domain to a common length to produce an extended first synchronization sequence and an extended second synchronization sequence. The apparatus (110, 200, 120, 300) is also configured to transform the extended first synchronization sequence and the extended second synchronization sequence into the first domain, and multiply elementwise the extended first synchronization sequence by the extended second synchronization sequence in the first domain to obtain a composite synchronization sequence. The extension proposed is an interpolation in time domain through zero padding in frequency domain. Typically a Zadoff-Chu sequence with different cyclic shifts is combined with a m-sequence. Such a combination increases the number of available sequences for PRACH with good auto- and cross-correlation properties.