NR Wideband Sync Detection via Sub-band Extraction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

In New Radio (NR) wireless communication networks, the detection of synchronization signals with unknown frequency locations poses challenges due to the absence of always-on Cell-specific Reference Signals, making traditional spectrum estimation inefficient, and requiring radio network devices to perform complex grid searches for synchronization.

Innovation Solution

A method involving sampling the received wideband signal over a band of interest in both frequency and time, converting it to the frequency domain, extracting sub-bands corresponding to possible sync locations and frequency offsets, and applying time-domain matched filtering to detect synchronization signals without prior knowledge of their exact positions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If traditional spectrum estimation methods are used for sync signal detection in NR, then the detection process becomes inefficient, but performing a full grid search improves detection accuracy

Engineering Contradiction:
Improvesync signal detection accuracyVSAvoiddetection efficiency
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The patent segments the wideband signal into multiple sub-bands corresponding to different possible sync signal locations and frequency offsets. Instead of performing a full grid search across the entire bandwidth, the method divides the search space into manageable sub-bands, processes them separately, and combines results. This segmentation approach maintains detection accuracy while significantly improving efficiency by avoiding redundant computations across the full bandwidth.

Inventive Principle:
Principle #1Segmentation

2Measurement precision

If radio network devices record and process large bandwidth signals, then detection accuracy improves, but energy consumption increases

Engineering Contradiction:
Improvesync signal detection accuracyVSAvoiddevice energy consumption
Core Design Contradiction:
Measurement precisionVSUse of energy by moving object

Solution Approach 1:

The patent extracts only the necessary sub-bands from the wideband signal that are likely to contain sync signals, based on expected frequency locations and offsets. Rather than processing the entire wideband signal, the method identifies and extracts relevant frequency regions, processes them for sync detection, and discards the rest. This extraction approach maintains detection accuracy for the target signals while dramatically reducing the computational load and energy consumption associated with processing the full bandwidth.

Inventive Principle:
Principle #2Taking out (Extraction)

3Measurement precision

If the SSB covers larger bandwidth in NR compared to LTE, then frequency resolution improves, but the complexity of grid search for sync detection increases

Engineering Contradiction:
Improvefrequency resolutionVSAvoidgrid search complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent performs preliminary actions by first transforming the wideband signal to the frequency domain and identifying potential sync signal locations based on known SSB structures and frequency grids. Before conducting the full search, the method pre-identifies candidate sub-bands where sync signals are likely to be present, based on the larger bandwidth characteristics of NR SSB. This preliminary identification step reduces the subsequent search complexity while maintaining the frequency resolution benefits of the wider bandwidth.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP3602851B1NR wideband sync detection
Publication Date: 2022.09.28 TELEFONAKTIEBOLAGET LM ERICSSON (PUBL)
  • EP3602851B1 patent drawingFigure 1
  • EP3602851B1 patent drawingFigure 2
  • EP3602851B1 patent drawingFigure 3a~3b

AI summary

A network sync signal with unknown frequency location is detected by sampling the received signal over a band of interest in frequency, and over the repetition period of the sync signal in time. The signal is converted to the frequency domain. Sub-bands of the frequency-domain signal, corresponding to different possible sync locations and frequency offsets, are extracted and converted to the time domain, where the sync signal is searched over the reception window length using time-domain matched filtering.