Digital Audio Signal Processing Using FFT Vector Grouping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current digital signal processors (DSPs) lack libraries for fast calculations of real Fourier Transforms, relying on complex FFTs, which increases computational load for filtering operations in digital audio signal processing, especially for spatialization of sound.
Innovation Solution
A method that groups real and imaginary parts of input signals to form complex vectors, applies filtering in Fourier space, and performs inverse FFT to generate output signals, significantly reducing calculation time by optimizing the use of existing FFT libraries.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If complex FFT libraries are used for real signal processing, then the DSP can perform Fourier Transforms, but the computational load increases and processing efficiency decreases
Solution Approach 1:
The patent combines two real input signals into a single complex signal by treating one as the real part and the other as the imaginary part. This merging allows both signals to be processed simultaneously through a single complex FFT, reducing the number of required calculations while maintaining full processing capability for both original signals.
2Measurement precision
If separate FFT calculations are performed for each input signal, then accurate filtering is achieved, but the number of FFT calculations increases significantly
Solution Approach 1:
The patent merges multiple real signals into complex signals to reduce the number of FFT operations. By processing multiple signals through shared FFT calculations and then separating the results through appropriate mathematical operations, the patent maintains filtering accuracy while significantly reducing computation time.
Solution Approach 2:
The patent segments the processing into distinct phases: combining real signals into complex signals for efficient FFT processing, performing the FFT in the frequency domain, and then separating the results back into real output signals. This segmentation allows optimization at each stage while preserving overall accuracy.
3Productivity
If real Fourier Transform libraries are implemented, then calculation optimization is achieved, but device complexity increases
Solution Approach 1:
The patent makes the existing complex FFT library multi-functional by applying it to real signal processing through complex number construction. Instead of requiring a separate real FFT library, the complex FFT library is used in a versatile manner that handles real signals by embedding them in complex form, thus achieving optimization without increasing device complexity.
Data Source
Figure 1~2
Figure 3
Figure 4
AI summary
The invention relates to a method for processing a signal, in particular a digital audio signal, suitable for being implemented by a digital signal processor (DSP) having libraries for calculating Fourier transforms from the complex number space to the complex number space, for digitally processing P input signals, P being an integer at least equal to 2, more particularly for filtering said P input signals by the convolution of sampled fast Fourier transforms (FFT), thus obtaining Q output signals, Q being an integer at least equal to 2. According to the invention, the method includes at least the following steps: - grouping said P input signals by twos, one representing the real portion and the other the imaginary portion of a complex number, thus defining one or more input vectors, - filtering said input vector or vectors, passing through the Fourier space, thus generating one or more output vectors, which are complex numbers, the real portion and the imaginary portion of said vector or each one of said output vectors respectively representing one of said Q output signals.