Carrier Load Balancing via Predicted Data Flow Volume

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing load balancing procedures in communications networks fail to achieve well-balanced total loads between carriers, as they primarily rely on channel quality and number of wireless devices, leading to imbalances and subsequent repeated load balancing procedures.

Innovation Solution

A system that predicts future data flow volumes by classifying data flows into intervals using machine learning on traffic properties, and determines offloading based on current channel quality and predicted volumes to evenly distribute wireless devices between carriers.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of manufacture

If wireless devices are handed over between carriers based only on channel quality and number of devices, then the procedure is simple to implement, but the resulting total load is not well balanced between carriers

Engineering Contradiction:

Improveease of implementationVSAvoidload balancing precision

Core Design Contradiction:

Ease of manufactureVSManufacturing precision

Solution Approach 1:

The system performs preliminary classification of data flows into volume intervals and predicts future data volumes before making handover decisions. This allows the load balancing to anticipate future load conditions rather than reacting to current state only, thereby achieving better load balance while maintaining a systematic decision process

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent segments data flows into different volume intervals and applies different handover strategies to different segments. By classifying data flows based on volume characteristics and applying targeted offloading rules to specific segments, the system achieves more precise load balancing than uniform approaches

Inventive Principle:

Principle #1Segmentation

2Ease of operation

If small data flow devices are handed over to balance device counts, then the number of devices per carrier is equalized, but the total load remains imbalanced and triggers repeated load balancing procedures

Engineering Contradiction:

Improvedevice distribution uniformityVSAvoidnetwork efficiency

Core Design Contradiction:

Ease of operationVSProductivity

Solution Approach 1:

The system changes the decision parameters from solely device count and channel quality to include predicted future data volume. By incorporating volume predictions as a new parameter, the handover decisions account for the actual load contribution of each device, preventing futile handovers of small data flow devices and reducing repeated load balancing operations

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The system uses feedback from predicted future data volumes to adjust handover decisions. By continuously monitoring and predicting data flow volumes, the system can identify which devices will significantly contribute to load on the target carrier, thereby making more informed decisions that prevent load imbalance and reduce the need for repeated procedures

Inventive Principle:

Principle #23Feedback

Data Source

PatentUS11197195B2Load balancing of data traffic in a communications network

Publication Date: 2021.12.07 TELEFONAKTIEBOLAGET LM ERICSSON (PUBL)

AI summary

There is provided mechanisms for load balancing data traffic between at least two carriers in a communications network supporting the at least two carriers. A method is performed by a system. The method comprises obtaining current channel quality information for a wireless device of the at least two carriers. The method comprises predicting a future volume of a data flow of the wireless device. The method comprises determining whether to offload the wireless device to a second carrier of the at least two carriers or not according to the current channel quality information and the predicted future volume.