Minimum Phase FIR Filter Coefficient Generation for Audio Systems

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing filter generation apparatuses are insufficient for multi-speaker systems and crossover processing, requiring increased memory capacity for storing parameters related to frequency band division and amplitude correction.

Innovation Solution

A filter generation apparatus that acquires and processes filter coefficients to generate approximation filters by converting linear phase FIR filters to minimum phase FIR filters, using cut-out and rounding processing with a window function, and storing coefficients only when the error is within a predetermined threshold, thereby reducing memory usage.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If a filter generation apparatus is used for multi-speaker systems and crossover processing, then audio quality and frequency band division are improved, but memory capacity requirements increase

Engineering Contradiction:
Improveaudio qualityVSAvoidmemory capacity
Core Design Contradiction:
ReliabilityVSQuantity of substance

Solution Approach 1:

The patent transforms the filter coefficient representation from linear phase to minimum phase domain, changing the mathematical parameters to achieve a more compact form that requires less memory storage while maintaining audio processing functionality

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent extracts only the essential filter coefficients needed for approximation, storing them in a coefficient storage unit, and generates approximation filters on-demand through cut-out processing and window function application, rather than storing all possible filter parameters

Inventive Principle:
Principle #2Taking out (Extraction)

2Adaptability or versatility

If filter coefficients are stored for each processing parameter in multi-speaker systems, then frequency band division and amplitude correction functions are improved, but device complexity increases

Engineering Contradiction:
Improvefrequency band division functionVSAvoidsystem complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The minimum phase filter coefficient storage unit serves multiple purposes: it stores coefficients for both crossover processing and amplitude correction, and enables generation of approximation filters for different frequency bands through a unified process, reducing the need for separate storage for each function

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

The patent segments the filter coefficient data into minimum phase components that can be independently processed and recombined, allowing flexible generation of different frequency band filters through cut-out processing and window function application

Inventive Principle:
Principle #1Segmentation

Data Source

PatentUS11792571B2Filter generation apparatus and recording medium
Publication Date: 2023.10.17 FAURECIA CLARION ELECTRONICS CO LTD
  • US11792571B2 patent drawing
  • US11792571B2 patent drawing
  • US11792571B2 patent drawing

AI summary

A filter generation apparatus includes: a coefficient calculation unit that performs IFFT on a frequency characteristic of a frequency domain filter and calculates a filter coefficient of a linear phase FIR filter; a conversion unit that converts the filter coefficient to a filter coefficient of a minimum phase FIR filter; a coefficient generation unit that generates a filter coefficient of an approximation filter by executing cut-out processing to a filter length set by a filter length setting unit and rounding processing performed by a window function on the filter coefficient; an error calculation unit that calculates an error of a frequency characteristic corresponding to the approximation filter with respect to the frequency characteristic of the frequency domain filter; and a filter storage unit that stores the filter coefficient therein as a filter coefficient that specifies a new filter when the error is equal to or less than a predetermined threshold value.