Audio Jitter Buffering for Multiple Description Packets
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Audio jitter buffers in real-time communication systems can only process single description coding bitstreams, leading to loss of multiple description coding bitstream data and deteriorating audio quality.
Innovation Solution
A method and apparatus for processing audio coding data packets that parse the packets to obtain timestamp and description index values, determining identical packets in the buffer, and writing non-identical packets into the buffer, allowing for multiple description coding bitstreams to be processed without loss.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the audio jitter buffer buffers only one audio coding data packet for one timestamp, then the buffer management is simple, but multiple description coding bitstream data is lost
Solution Approach 1:
The patent segments the buffer management by introducing description index values as an additional dimension for identifying data packets. Instead of treating all packets with the same timestamp as identical, the system now segments them by their description index values, allowing multiple packets with the same timestamp but different description indices to be buffered separately. This segmentation resolves the contradiction by maintaining data completeness while managing complexity through structured organization.
Solution Approach 2:
The patent adds another dimension (description index value) to the buffer identification criteria beyond just timestamp. By transitioning from a single-dimension (timestamp only) to a two-dimension (timestamp + description index value) identification system, the patent enables the buffer to distinguish between multiple packets with the same timestamp, thereby preventing data loss while maintaining manageable complexity through the additional discriminatory dimension.
2Reliability
If the audio jitter buffer buffers only one audio coding data packet for one timestamp, then the processing logic is simple, but audio data is lost
Solution Approach 1:
The processing logic is segmented by introducing description index value comparison as an additional step. When a new packet arrives, the system segments the checking process into: (1) check timestamp, (2) check description index value. This segmentation allows the system to process multiple packets with the same timestamp without creating complex conditional logic, as each packet is evaluated against both criteria independently.
Solution Approach 2:
By adding description index value as another dimension to the processing logic, the patent transforms the decision-making process from a single-criterion check to a two-criterion check. This dimensional expansion simplifies the overall logic by providing clear, rule-based differentiation between packets, avoiding the need for complex algorithms to determine which packets to retain or discard.
3Reliability
If the buffer stores multiple packets with the same timestamp, then all MDC bitstream data is preserved, but the buffer space requirement increases
Solution Approach 1:
The patent implements a discarding and recovering mechanism where packets are selectively retained or discarded based on their description index values and timestamp combinations. The buffer recovers optimal space utilization by identifying and removing duplicate packets (same timestamp and description index value) while preserving unique packets. This principle resolves the contradiction by ensuring data preservation for unique packets while discarding redundant ones to maintain efficient buffer space usage.
Data Source
AI summary
A method and an apparatus for processing an audio coding data packet are disclosed. The method includes: parsing an audio coding data packet to obtain data packet information of the audio coding data packet, where the data packet information includes a timestamp and a description index value of the audio coding data packet, audio data corresponding to the audio coding data packet includes at least one description, and the description index value is an index of the description; determining whether a data packet whose data packet information is identical to data packet information of the audio coding data packet is buffered in a data packet buffer; and if not, writing the audio coding data packet into the data packet buffer based on the data packet information of the audio coding data packet.