USAC Window Sequence Mode Switching for TDA Cancellation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional Unified Speech and Audio Codec (USAC) methods for window sequence processing during mode switching between Linear Prediction Domain (LPD) and Frequency Domain (FD) modes do not effectively improve coding efficiency and are prone to block artifacts and Time Domain Aliasing (TDA).
Innovation Solution
The USAC employs a mode switching unit to switch between LPD and FD modes, with an encoding/decoding unit that applies a window sequence based on the switched mode, performing overlap-add operations using a folding point to cancel TDA and optimize window sequences for improved coding efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional window sequence processing is used during mode switching between LPD and FD modes, then the codec structure remains simple, but block artifacts and Time Domain Aliasing occur and coding efficiency is not improved
Solution Approach 1:
The patent applies preliminary action by pre-defining multiple window sequence types (first type, second type, third type) corresponding to different mode switching scenarios. The mode switching unit selects the appropriate window sequence in advance based on the switching direction (LPD to FD or FD to LPD) and the LPD mode of the current frame, preventing block artifacts and TDA before they occur during decoding.
Solution Approach 2:
The patent changes parameters by varying the window sequence type based on mode switching conditions. When switching from LPD to FD mode, a first window sequence is used; when switching from FD to LPD mode, a second or third window sequence is used depending on the LPD mode. This parameter change optimizes the transition characteristics and eliminates harmful artifacts.
2Reliability
If mode switching between LPD and FD modes is implemented, then coding performance is improved, but complex window sequence processing is required to avoid artifacts
Solution Approach 1:
The patent segments the window sequence processing into distinct types based on mode switching scenarios. By dividing the processing into first type, second type, and third type window sequences, each optimized for specific conditions, the complexity is managed through structured classification rather than monolithic complex processing.
Solution Approach 2:
The patent introduces a mode switching unit as an intermediary between the coding/decoding units. This intermediary component handles the complex window sequence selection logic, managing the transitions between LPD and FD modes and selecting appropriate window sequences, thereby simplifying the overall system architecture while maintaining high coding performance.
3Reliability
If overlap-add operation is performed based on folding point, then Time Domain Aliasing is canceled and coding gain is improved, but the processing complexity increases
Solution Approach 1:
The patent applies preliminary action by pre-defining the folding point positions and overlap-add parameters for each window sequence type. The mode switching unit determines the appropriate overlap-add configuration in advance based on the mode switching direction, allowing the decoding unit to perform the operation efficiently without real-time complex calculations.
Data Source
AI summary
A Unified Speech and Audio Codec (USAC) that may process a window sequence based on mode switching is provided. The USAC may perform encoding or decoding by overlapping between frames based on a folding point when mode switching occurs. The USAC may process different window sequences for each situation to perform encoding or decoding, and thereby may improve a coding efficiency.