Transceiver Deployment Configuration Using Staged ML Training

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing methods for configuring transceiver deployment in challenging environments, such as buildings with obstructing structures, are labor-intensive and time-consuming, and existing machine learning training processes are inefficient.

Innovation Solution

Utilize a curriculum learning approach with multiple stages of training for machine learning models, starting with simplified environments and progressively increasing complexity, combined with reinforcement learning to optimize transceiver deployment configurations.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If physical testing is used to determine transceiver deployment configuration, then coverage effectiveness can be verified, but the process becomes cost and labour intensive and disruptive to users

Engineering Contradiction:
Improvecoverage effectivenessVSAvoiddeployment efficiency
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The patent creates a digital twin (virtual copy) of the physical environment including obstacles and transceivers. This digital model allows simulation and verification of deployment configurations without physical testing, eliminating the need to disrupt actual users while maintaining coverage verification capability.

Inventive Principle:
Principle #26Copying

Solution Approach 2:

The patent introduces a ray-tracing simulation as an intermediary between deployment configuration and coverage verification. This simulation acts as a mediator that predicts coverage outcomes without requiring physical testing, thus avoiding disruption to users while still verifying effectiveness.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Ease of operation

If regular spatial configuration is used for transceiver deployment, then deployment is simple and systematic, but coverage effectiveness in environments with obstructing structures is reduced

Engineering Contradiction:
Improvedeployment simplicityVSAvoidcoverage effectiveness
Core Design Contradiction:
Ease of operationVSReliability

Solution Approach 1:

The patent optimizes transceiver positions locally based on environmental characteristics. Instead of uniform spacing, each transceiver position is individually adjusted according to local obstacle distributions and coverage requirements, achieving both effectiveness and reasonable deployment simplicity.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The patent performs preliminary simulation and optimization of transceiver positions before actual deployment. By pre-calculating optimal positions using ray-tracing simulations, the system avoids trial-and-error physical testing while ensuring coverage effectiveness in complex environments.

Inventive Principle:
Principle #10Preliminary action

3Reliability

If ML model training uses complex operating environments from the start, then the model learns realistic scenarios, but training time and computational resources increase significantly

Engineering Contradiction:
Improvemodel learning accuracyVSAvoidtraining time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent segments the training process into multiple stages with increasing complexity. Training begins with simplified environments and progressively introduces more complex scenarios, allowing the model to build foundational knowledge before learning intricate patterns, thereby reducing overall training time.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent performs preliminary training on simplified environments before transitioning to complex realistic scenarios. This preliminary action allows the model to converge faster on basic patterns, reducing the total training time required when eventually training on complex environments.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP4584983B1Method and apparatus for network coverage configuration
Publication Date: 2025.12.10 TELEFONAKTIEBOLAGET LM ERICSSON (PUBL)
  • EP4584983B1 patent drawingFigure 1
  • EP4584983B1 patent drawingFigure 2
  • EP4584983B1 patent drawingFigure 3A

AI summary

Methods and apparatus for network coverage configuration are provided. A computer-implemented method for configuring network coverage using a ML agent hosting a ML model comprises obtaining first training data and second training data relating to an operating environment. The method further comprises training the ML model in a first training stage using the first training data. When the ML model satisfies a first performance criterion, the first training stage ends, and the ML model is trained in a second training stage using the second training data. When the ML model satisfies a second performance criterion, the second training stage ends. The ML model is then used to generate a deployment configuration for configuring the spatial deployment of a plurality of transceivers to provide network connection capability to the operating environment