Selective Speech Frame Discarding in Mobile Transport Networks

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

In IP-based mobile communication networks, high packet loss rates and jitter lead to robotic voice due to the discarding of speech frames during overload situations, as existing technologies discard frames without considering their content, resulting in the loss of valuable speech information.

Innovation Solution

A device and method for selectively discarding speech frames based on their content, specifically erroneous or silent frames, using a discarding unit that evaluates Frame Quality Indicators (FQI) and Voice Activity Detection (VAD) flags to prioritize the transmission of more valuable speech frames, thereby mitigating overload situations and improving speech quality.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If speech frames are discarded during overload situations without content evaluation, then network overload is mitigated, but speech quality deteriorates due to loss of valuable speech information

Engineering Contradiction:
Improvenetwork capacity utilizationVSAvoidspeech quality
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The patent applies local quality by evaluating and differentiating the quality of individual speech frames based on their content characteristics. The discarding unit analyzes specific properties of each frame (such as voice activity, frame type, importance) and makes selective discarding decisions, rather than treating all frames uniformly. This allows the system to preserve high-quality important frames while discarding less important ones during overload, thus maintaining speech quality while managing network capacity.

Inventive Principle:
Principle #3Local quality

2Reliability

If all speech frames are buffered during overload situations, then speech quality is maintained, but network resources are wasted and overload conditions worsen

Engineering Contradiction:
Improvespeech qualityVSAvoidnetwork efficiency
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The patent implements selective discarding of speech frames based on their content evaluation. The discarding unit identifies and discards less important frames (such as silent periods, duplicate frames, or low-priority traffic) while preserving important speech frames. This selective approach allows the network to free up resources efficiently without compromising speech quality, as the discarded frames are those least likely to affect perceived quality.

Inventive Principle:
Principle #34Discarding and recovering

3Reliability

If content-based evaluation of speech frames is implemented, then selective discarding improves speech quality, but device complexity increases

Engineering Contradiction:
Improvespeech qualityVSAvoidrouting device complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent applies preliminary action by pre-evaluating speech frames for their content characteristics and importance before making discarding decisions. The discarding unit analyzes frame properties (voice activity detection, frame type identification, importance assessment) in advance, so that when overload occurs, the system can quickly make informed discarding decisions without complex real-time analysis. This preprocessing approach reduces the computational burden during critical overload situations.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP3364615B1Device and method for forwarding or routing speech frames in a transport network of a mobile communications system
Publication Date: 2022.07.27 TELEFONICA GERMANY GMBH & CO
  • EP3364615B1 patent drawingFigure 1
  • EP3364615B1 patent drawingFigure 2
  • EP3364615B1 patent drawingFigure 3

AI summary

The present invention relates to a device for forwarding or routing speech frames in a transport network of a mobile communications system, the device comprising: a buffer for storing incoming speech frames of a plurality of users of the mobile communications system, and a discarding unit for discarding speech frames, characterised in that the discarding unit is adapted to discard a speech frame stored in the buffer on basis of the speech frame's content.