De-jitter Buffer for 5G TSN Deterministic Latency

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Wireless communication networks in industrial settings face significant latency variations due to radio channel conditions and network routing uncertainties, leading to jitter issues that affect the deterministic delivery of Time-Sensitive Networking (TSN) frames, which existing solutions address inadequately by applying a single worst-case maximum packet-hold time across all user equipment (UEs), resulting in unnecessary latency and energy consumption.

Innovation Solution

Customizing the maximum packet-hold time for each UE or group of UEs based on environment data using machine learning models, allowing for optimized packet buffering that minimizes latency and energy consumption by tailoring the buffering time to the specific radio channel quality of each UE.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If a single worst-case maximum packet-hold time is applied across all user equipment, then deterministic latency is ensured for all UEs, but unnecessary latency and energy consumption occur for UEs with better radio channel conditions

Engineering Contradiction:
Improvedeterministic latencyVSAvoidunnecessary latency
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent applies local quality by customizing the maximum packet-hold time for each UE or group of UEs based on their specific environment data (radio channel quality, location, device characteristics). Instead of using a uniform worst-case value for all UEs, the system tailors the packet-hold time to match the actual conditions of each UE, thereby eliminating unnecessary latency for UEs with better channel conditions while maintaining deterministic latency guarantees where needed.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The patent implements dynamics by making the maximum packet-hold time adaptive rather than static. The system uses machine learning models to dynamically adjust the packet-hold time based on real-time environment data, allowing the de-jitter buffer to respond to changing radio channel conditions, UE locations, and network circumstances, thus optimizing latency performance while maintaining reliability.

Inventive Principle:
Principle #15Dynamics

2Reliability

If a single worst-case maximum packet-hold time is applied across all user equipment, then deterministic latency is ensured, but energy consumption increases due to unnecessary buffering

Engineering Contradiction:
Improvedeterministic latencyVSAvoidenergy consumption
Core Design Contradiction:
ReliabilityVSUse of energy by moving object

Solution Approach 1:

The patent reduces energy consumption by applying local quality - each UE receives a customized maximum packet-hold time matched to its specific environment. UEs with excellent radio channel conditions and stable locations receive shorter packet-hold times, allowing them to drain buffers more quickly and consume less energy, while UEs in challenging environments receive longer packet-hold times to ensure deterministic latency delivery.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The system dynamically adjusts packet-hold times based on real-time environment monitoring, allowing UEs to transition between different buffering states. When channel conditions improve, the system reduces packet-hold time, enabling faster buffer drainage and lower energy consumption. This dynamic adaptation ensures energy efficiency is optimized alongside reliability guarantees.

Inventive Principle:
Principle #15Dynamics

3Reliability

If maximum packet-hold time is increased to handle worst-case latency, then all packets can be delivered within the buffer, but latency increases for packets that don't require such long buffering

Engineering Contradiction:
Improvepacket delivery guaranteeVSAvoidpacket latency
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent resolves this contradiction by applying local quality - different maximum packet-hold times are assigned to different UEs based on their specific characteristics. UEs with stable radio conditions, favorable locations, and reliable device performance receive shorter packet-hold times, enabling faster packet delivery. UEs experiencing challenging conditions receive longer packet-hold times to ensure all packets are delivered within the buffer, thus matching the buffer duration to actual needs rather than applying a universal worst-case value.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The system changes the packet-hold time parameter dynamically based on environment data and machine learning model predictions. By adjusting this critical parameter according to actual UE conditions rather than using a fixed worst-case value, the system optimizes the balance between packet delivery guarantee and latency performance for each individual UE.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP3895398B1Communication system with de-jitter buffer for reducing jitter
Publication Date: 2024.06.26 TELEFONAKTIEBOLAGET LM ERICSSON (PUBL)
  • EP3895398B1 patent drawingFigure 1
  • EP3895398B1 patent drawingFigure 2
  • EP3895398B1 patent drawingFigure 3

AI summary

A de-jitter function for holding-and-forwarding packets such that the packets are delivered with an agreed fixed latency. The de-jitter function can be placed at the edge of a virtual 5G TSN switch (e.g. the de-jitter function can be deployed as part of a UPF for uplink (UL) packets and/or it can be deployed as part of a user equipment (UE) for downlink (DL) packets). By using the de-jitter function, the TSN can consider the wireless network as having a consistent, deterministic latency with no jitter. (figure 2)