Binary Arithmetic Decoder Using Parallel Renormalization in CABAC

Solution Overview

Problem

Conventional CABAC decoders require long processing times due to repeated renormalization loops in decoding encoded data, which hinders the speed and efficiency of the decoding process.

Innovation Solution

A binary arithmetic decoder is designed with parallel renormalization processing units that update range lengths and offsets simultaneously with LPS/MPS determination, allowing for immediate selection of renormalized values once a predetermined section length is reached, thereby reducing processing time.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Speed

If sequential renormalization processing is performed after LPS/MPS determination, then processing accuracy is maintained, but decoding speed decreases due to repeated loops

Engineering Contradiction:

Improvedecoding speedVSAvoidprocessing time

Core Design Contradiction:

SpeedVSLoss of time

Solution Approach 1:

The patent performs renormalization processing in parallel with LPS/MPS determination by pre-calculating renormalized range length and offset values before the determination is complete. This preliminary action eliminates the need for sequential waiting, allowing the decoder to immediately use pre-computed values once determination finishes, thereby reducing total processing time while maintaining accuracy

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent implements continuous parallel execution of renormalization processing alongside LPS/MPS determination. Instead of interrupting renormalization to wait for determination results, the system maintains continuous useful action by computing both processes simultaneously, ensuring no idle time is lost and decoding operations proceed without interruption

Inventive Principle:

Principle #20Continuity of useful action

2Measurement precision

If renormalization loops are repeated multiple times to ensure accuracy, then decoding precision is improved, but processing efficiency deteriorates

Engineering Contradiction:

Improvedecoding precisionVSAvoidprocessing efficiency

Core Design Contradiction:

Measurement precisionVSProductivity

Solution Approach 1:

The patent performs renormalization processing in parallel with LPS/MPS determination by pre-calculating renormalized range length and offset values before the determination is complete. This preliminary action eliminates the need for sequential waiting, allowing the decoder to immediately use pre-computed values once determination finishes, thereby reducing total processing time while maintaining accuracy

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent implements continuous parallel execution of renormalization processing alongside LPS/MPS determination. Instead of interrupting renormalization to wait for determination results, the system maintains continuous useful action by computing both processes simultaneously, ensuring no idle time is lost and decoding operations proceed without interruption

Inventive Principle:

Principle #20Continuity of useful action

Data Source

PatentUS11134249B2Binary arithmetic decoder and binary arithmetic decoding device

Publication Date: 2021.09.28 NTT ELECTORNICS CORP

AI summary

A binary arithmetic device includes an LPS/MPS determining unit that determines, using a context variable, a range length, and an offset, whether a code is an inferior probability code or a superior probability code, a renormalization processing unit that performs renormalization processing on the range length and the offset, and a context-variable calculating unit that derives the binary data of the code using a determination result and updates the context variable according to the determination result. The renormalization processing unit 15 includes a first renormalizing unit and a second renormalizing unit and a selecting unit that selects, according to the determination result, an output of the first renormalizing unit or an output of the second renormalizing unit.