Adaptive MIMO Filtering in Dynamically Adapted Transform Domains

Simple SummaryContent extracted from patent full text and abstract with AI.

This invention describes a method and device for adaptive multiple-input and multiple-output (MIMO) filtering, especially in systems with dynamically changing spatial and temporal transformations. The core idea is to adapt both the temporal and spatial transformations of the input and output signals such that the relevant correlation matrices become block-diagonal. This provides efficient decoupling of the MIMO system and allows for improved filter adaptation. The method iteratively optimizes the transformation matrices based on the actual signal data, enabling the filtering process to adapt flexibly to different environments and operating conditions, such as changing room acoustics or sensor configurations.

Use CasesContent extracted from patent full text and abstract with AI.

• Acoustic echo cancellation in teleconference and hands-free systems
• Noise suppression and source separation in multichannel audio environments
• Multichannel active noise control for automotive or industrial applications
• Adaptive sound reproduction in smart speaker arrays or immersive audio systems
• Signal separation and processing in sensor arrays for communication devices (e.g., beamforming microphones)
• Wireless MIMO communication systems where channel conditions change dynamically

BenefitsContent extracted from patent full text and abstract with AI.

• Improved adaptive filtering performance in complex, dynamically changing environments
• Reduced computational complexity thanks to block-diagonalization, enabling efficient implementation
• Better numerical stability and faster convergence in adaptive algorithms
• Enhanced separation and decoupling of multichannel signals, improving noise or echo suppression
• Automatic adaptation to changes in system properties or environments (e.g., movement of microphones or speakers, or changes in room acoustics)
• Flexible adjustment of computational load according to available processing power

Technical Classifications (CPCs)

Inventors & Applicants

Patent Abstract

The method involves performing temporal and spatial transformation of input signals and desired output signals. A coefficient matrix is formed according to a relation of correlation matrices of the input signals or the input and output signals in a spatiotemporally transformed area. The correlation matrices are described using spatial transformation matrices according to a relation of corresponding correlation matrices with analog block structures in a temporally transformed area. The transformation matrices are optimized such that the correlation matrices are respectively block-diagonal. An independent claim is also included for a device for adaptive multiple-input and multiple-output (MIMO) filtering in dynamically adapted transformation areas, comprising a block for temporal transformation of input signals.