5G Path Loss Propagation Models for RSRP Using 4G User Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing path loss prediction models for 5G networks are inefficient and costly due to the need for extensive drive tests and lack of computational efficiency in deterministic methods, making accurate signal strength prediction difficult in complex propagation environments.
Innovation Solution
A system and method using machine learning, specifically Artificial Neural Networks (ANN) and Random Forest techniques, leverage actual 4G user data to predict Reference Signal Received Power (RSRP) with error correction, optimizing the prediction process and improving accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If drive test data is used to extract parameters for empirical path loss models, then measurement precision is improved, but loss of time and productivity deteriorate due to multiple time-consuming iterations
Solution Approach 1:
The system performs preliminary path loss model generation using machine learning algorithms before actual path loss prediction is needed. By pre-training the ML model with available data and pre-processing propagation environment parameters, the system eliminates the need for multiple iterative drive tests, achieving both high accuracy and time efficiency
Solution Approach 2:
The patent replaces the mechanical drive test process with a computational machine learning approach. Instead of physically conducting repeated drive tests to extract parameters, the system uses ML algorithms to learn path loss patterns from data, substituting physical measurement iterations with computational processing that is significantly faster
2Measurement precision
If deterministic models like ray tracing are used for path loss prediction, then measurement precision is improved, but productivity deteriorates due to prohibitive computation time
Solution Approach 1:
The system uses a machine learning model that is computationally lighter and faster than deterministic ray tracing models. The ML model provides sufficiently accurate predictions without the prohibitive computation time of deterministic methods, effectively replacing expensive computational objects with cheaper, faster alternatives
Solution Approach 2:
The patent changes the computational parameters and approach from deterministic ray tracing to probabilistic machine learning. This parameter change transforms the prediction process from computationally intensive deterministic calculations to efficient probabilistic inference, achieving both accuracy and speed
3Measurement precision
If complex propagation environment mechanisms are modeled using deterministic methods, then measurement precision is improved, but device complexity increases making implementation difficult
Solution Approach 1:
The system extracts the essential path loss prediction function from complex deterministic models and implements it through a simplified machine learning framework. By taking out only the necessary prediction capability and removing unnecessary computational complexity, the system achieves accurate predictions with simpler implementation
Solution Approach 2:
The patent creates a simplified copy of the path loss prediction functionality using machine learning models. Instead of implementing full deterministic ray tracing, the system creates an ML-based copy that captures the essential prediction behavior with much lower complexity, making it easier to implement and deploy
Data Source
AI summary
The present disclosure provides a system and a method for generating a path loss propagation model through machine learning. The system generates a path loss propagation model for fifth generation (5G) networks for network planning. The path loss model predicts a reference signal received power/signal to noise interference ratio (RSRP/SINR) by leveraging a fourth generation (4G) user data.