HOA Signal Compression via Frequency Sub-band Decomposition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Higher Order Ambisonics (HOA) representations require high bit rates for transmission, making them unsuitable for applications like audio streaming to mobile devices, where lower data rates such as 128 kbit/s are necessary, due to the large number of expansion coefficients and high bit rate requirements for practical applications.
Innovation Solution
A low-bit rate compression method for HOA representations is introduced, which decomposes the HOA signal into frequency sub-bands, using a truncated HOA representation and predicted directional subband signals. This involves selecting a small number of coefficient sequences, de-correlating them, and representing directional subband signals with parametric predictions, eliminating the need for extra side information during decompression.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If conventional HOA compression methods are used, then audio quality is maintained, but data rate remains too high (above 256 kbit/s) for mobile streaming applications
Solution Approach 1:
The HOA signal is divided into multiple frequency sub-bands, allowing different processing strategies for different frequency ranges. This segmentation enables more efficient compression by applying appropriate coding methods to each sub-band, achieving lower overall bit rates while preserving audio quality.
Solution Approach 2:
The patent transforms the HOA representation by changing parameters such as converting to a different basis (e.g., from spherical harmonics to an alternative representation), applying de-correlation transforms, and using parametric predictions. These parameter changes reduce redundancy and enable more efficient encoding at lower bit rates.
2Measurement precision
If the number of expansion coefficients is increased to improve spatial resolution, then HOA representation accuracy improves, but transmission bit rate increases quadratically
Solution Approach 1:
The patent extracts and separately encodes only the most significant HOA coefficients that contribute to spatial resolution, while representing less significant coefficients through prediction or approximation. This selective extraction maintains spatial resolution where needed while reducing the total number of coefficients that require explicit transmission.
Solution Approach 2:
Instead of encoding all coefficients at full precision, the patent applies partial encoding strategies where only essential coefficients are transmitted with high precision, while others are reconstructed with lower precision through prediction methods, achieving a balance between spatial resolution and bit rate.
3Productivity
If directional subband signals are represented with parametric predictions, then coding efficiency increases, but complexity of the encoding process increases
Solution Approach 1:
The patent performs preliminary processing steps such as de-correlation transforms and sub-band decomposition before the main encoding stage. These preliminary actions prepare the signal in a form that is more amenable to efficient parametric prediction, reducing the overall encoding complexity despite the additional initial processing steps.
Data Source
AI summary
Encoding of Higher Order Ambisonics (HOA) signals commonly results in high data rates. A method for low bit-rate encoding frames of an input HOA signal having coefficient sequences comprises computing (s110) a truncated HOA representation (CT(k)), determining (s111) active coefficient sequences (IC,ACTT(k)), estimating (s16) candidate directions (MDIR(k)), dividing (s15) the input HOA signal into a plurality of frequency subbands (f1, . . . , fF), estimating (s161) for each of the frequency subbands a subset of candidate directions (MDIR(k)) as active directions (MDIR(k,f1), . . . , MDIR(k,fF)) and for each active direction a trajectory, computing (s17) for each frequency subband directional subband signals from the coefficient sequences of the frequency subband according to the active directions, calculating (s18) for each frequency subband a prediction matrix (A(k,f1), . . . , A(k,fF)) that can be used for predicting the directional subband signals from the coefficient sequences of the frequency subband using the respective active coefficient sequences (K)), and encoding (s19) the candidate directions, active directions, prediction matrices and truncated HOA representation.