Steering Vector Calculation Using Power Spectral Density
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In beamforming technologies, updating the steering vector (SV) for adaptive beamforming is challenging due to the difficulty in measuring impulse responses over time, which increases the load on the measurer and affects the signal-noise ratio.
Innovation Solution
An information processing device that calculates a filter for forming a beam in a different direction using frequencies and steering vector information, allowing for the calculation of a steering vector without relying on measurement values of impulse responses, thereby reducing the load on the measurer.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If impulse response measurement is performed to update the steering vector, then the steering vector accuracy is improved, but the measurer's workload increases
Solution Approach 1:
The system uses the sound signals already being collected by the microphone array for self-diagnosis and steering vector estimation. The sound signal acquisition unit acquires sounds including speech and noise, and the analysis unit processes these signals to estimate the steering vector without requiring external measurement equipment or additional measurer intervention.
Solution Approach 2:
The sound signal acquisition unit serves multiple functions: it collects sounds for beamforming processing and simultaneously provides data for steering vector estimation. The analysis unit performs both frequency analysis for beamforming and impulse response estimation for steering vector calculation, eliminating the need for separate measurement processes.
2Adaptability or versatility
If impulse response measurement is performed over time to update the steering vector, then the adaptability to changing sound directions is improved, but the measurement difficulty increases
Solution Approach 1:
The system continuously estimates the steering vector by processing the ongoing sound signals acquired from the environment. The analysis unit continuously analyzes the frequency characteristics of acquired sounds and updates the impulse response estimate, enabling real-time adaptation to changing sound directions without interrupting the normal operation.
Solution Approach 2:
The patent introduces an intermediate estimation process that uses the power spectral density of acquired sounds as a mediator to derive the impulse response. Instead of directly measuring impulse response which is difficult, the system measures power spectral density from available sounds and uses this as an intermediary to estimate the steering vector through mathematical relationships.
Data Source
AI summary
An information processing device includes a sound signal acquisition unit that acquires sound signals outputted from a mic array, an analysis unit that analyzes frequencies of the sound signals, an information acquisition unit that acquires predetermined information indicating a steering vector in a first direction as a direction from the mic array to a target sound source, and a calculation unit that calculates a filter for formation in a second direction as a direction different from the first direction based on the frequencies and the information indicating the steering vector in the first direction and calculates a steering vector in the second direction by using an expression indicating a relationship between the calculated filter and the steering vector in the second direction.