Machine Learning Frequency Selection for HF Radio Propagation

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Communication systems employing HF radios face challenges such as a lack of skilled users, stale data in Automatic Link Establishment (ALE), and increased latency and overhead as data rates increase, necessitating near-real-time frequency management with minimal impact on throughput and Size, Weight, Power, and Cost (SWAP-C) considerations.

Innovation Solution

The implementation of a machine learning model that infers the best frequency for communication devices based on weather and space weather information, user inputs, and propagation parameters, using a weighted combination of inputs to select and tune the frequency, with the ability to update the model based on feedback.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If expert users with in-depth knowledge are used to select carrier frequencies, then frequency selection accuracy is improved, but system complexity and operational difficulty increase

Engineering Contradiction:

Improvefrequency selection accuracyVSAvoidoperational difficulty

Core Design Contradiction:

Measurement precisionVSEase of operation

Solution Approach 1:

The system performs self-service by automatically selecting optimal carrier frequencies using a machine learning model that processes propagation data, eliminating the need for expert human operators to manually analyze multiple factors and make frequency selection decisions

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The patent replaces the mechanical system of human expert analysis with an automated machine learning-based propagation prediction system that processes weather report information, predicted weather information, and other parameters to determine optimal frequencies

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Measurement precision

If pre-mission planning using propagation prediction programs is used, then frequency selection is improved, but time consumption and latency increase

Engineering Contradiction:

Improvefrequency selection accuracyVSAvoidtime consumption

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The system performs preliminary action by pre-processing and storing propagation data, weather information, and model training offline before mission execution, enabling rapid real-time frequency selection during actual operations without repeated heavy computations

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent replaces traditional computation-intensive propagation prediction programs with a machine learning model that has been pre-trained to provide faster predictions, reducing computational overhead and latency

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

3Productivity

If Automatic Link Establishment with recent link information is used, then communication link establishment is improved, but data staleness and reliability issues occur

Engineering Contradiction:

Improvelink establishment speedVSAvoiddata freshness

Core Design Contradiction:

ProductivityVSReliability

Solution Approach 1:

The system merges multiple data sources including weather report information, predicted weather information, and measured propagation data to create a more reliable and comprehensive basis for frequency selection, reducing reliance on any single potentially stale data source

Inventive Principle:

Principle #5Merging (Combining)

Solution Approach 2:

The system implements feedback mechanisms by continuously monitoring actual link performance and using this information to update the machine learning model, ensuring that the system adapts to changing propagation conditions and maintains reliability

Inventive Principle:

Principle #23Feedback

4Measurement precision

If traditional propagation prediction methods are used, then accuracy is maintained, but device complexity and SWAP-C constraints are violated

Engineering Contradiction:

Improveprediction accuracyVSAvoidsystem complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent replaces complex traditional propagation prediction computational systems with a machine learning model that, once trained, provides accurate predictions with lower computational overhead, reducing device complexity and SWAP-C requirements

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system performs preliminary action by completing the computationally intensive model training phase beforehand, allowing the deployed system to use the trained model for efficient real-time predictions without requiring heavy computational resources during operation

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS12041010B2Systems and methods for supporting near-real time propagation prediction and frequency management

Publication Date: 2024.07.16 L3HARRIS GLOBAL COMMUNICATIONS INC

AI summary

Systems and methods for operating a system. The methods may comprise: using a machine learning model to identify which frequency of a plurality of possible frequencies that can be used by a communication device for wireless communications is an inferred best frequency based on weather report information and/or predicted weather information; performing operations to select a best frequency from the plurality of possible frequencies using the inferred best frequency and a frequency value selected by a link establishment process of the communication device; and causing the communication device to communicate signals with the best frequency which was selected.