Channel Estimation Using 1D and 2D Filters

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current radio receiver systems face significant delays in channel estimation, particularly due to the complexity of two-dimensional Wiener filters, which contribute substantially to the Time to Audio (TTA) metric, exceeding the minimum acceptable standards.

Innovation Solution

Implementing a parallel processing method using one-dimensional (1D) channel estimation for the Fast Access Channel (FAC) followed by two-dimensional (2D) estimation, where the 1D filter completes initial decoding before the 2D filter, reducing overall processing time and TTA.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If two-dimensional Wiener filter is used for channel estimation, then channel estimation accuracy is improved, but Time to Audio (TTA) delay increases

Engineering Contradiction:
Improvechannel estimation accuracyVSAvoidTime to Audio (TTA) delay
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent divides the channel estimation process into two separate stages: first a 1D filter is applied in one dimension (e.g., time or frequency), followed by a 2D filter in the subsequent stage. This segmentation allows the system to achieve accurate 2D channel estimation while reducing the immediate processing delay, as the 1D filter provides initial estimation results faster than a complete 2D filter would require.

Inventive Principle:
Principle #1Segmentation

2Measurement precision

If two-dimensional Wiener filter is used for channel estimation, then channel estimation accuracy is improved, but device complexity increases

Engineering Contradiction:
Improvechannel estimation accuracyVSAvoidfilter processing complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent segments the complex 2D Wiener filter operation into two simpler sequential operations: a 1D filter operation followed by a 2D filter operation. This segmentation reduces the computational complexity at any given moment while still achieving the benefit of 2D channel estimation, as each stage processes the data with reduced computational burden compared to applying a full 2D filter directly.

Inventive Principle:
Principle #1Segmentation

3Loss of time

If one-dimensional filter is used for channel estimation, then processing time is reduced, but channel estimation accuracy deteriorates

Engineering Contradiction:
Improveprocessing timeVSAvoidchannel estimation accuracy
Core Design Contradiction:
Loss of timeVSMeasurement precision

Solution Approach 1:

The patent applies a 1D filter as a preliminary action before applying the 2D filter. This preliminary 1D filtering provides initial channel estimation results that are then refined by the subsequent 2D filtering stage, achieving both reduced processing time (from the faster 1D operation) and maintained accuracy (through the refinements of the 2D operation).

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP3534580B1Method and apparatus to reduce delays in channel estimation
Publication Date: 2021.04.28 NXP BV
  • EP3534580B1 patent drawingFigure 1
  • EP3534580B1 patent drawingFigure 2
  • EP3534580B1 patent drawingFigure 3

AI summary

A signal processor and method of signal processing to reduce delays in channel estimation is described. In the method, the following operations are performed: receiving a signal at a signal processor (200); estimating a first channel from the received signal with a one-dimensional, 1D, frequency filter (304) in a frequency domain to generate an interim output; decoding said interim output (406) to generate decoded interim output; estimating (414,418) at least a second channel from the decoded interim output with a two-dimensional, 2D, filter (100) to produce a final output (420).