Radio Wave UAV Detection with Spectrograms and Receive Beamforming
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing drone detection technologies face challenges in efficiently detecting and identifying unmanned aerial vehicles (UAVs) due to their operation in unlicensed bands with mixed signals and varying protocols, making it difficult to accurately determine their presence, direction, and type, especially in conditions of limited visibility.
Innovation Solution
A method and apparatus utilizing radio wave measurement and artificial intelligence (AI) to generate a spectrogram, identify signal regions, and perform receive beamforming to detect and classify UAVs based on signal characteristics, including frequency, bandwidth, and signal quality, enhancing detection speed and accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If radio wave measurement and AI are used to detect UAVs in unlicensed bands with mixed signals, then detection accuracy and speed are improved, but device complexity increases
Solution Approach 1:
The detection process is divided into distinct stages: signal acquisition, spectrogram generation, signal region identification, receive beamforming, and UAV classification. This segmentation allows complex detection tasks to be broken down into manageable modules, improving accuracy while controlling complexity through structured processing
Solution Approach 2:
A spectrogram is introduced as an intermediary representation between raw radio wave signals and UAV detection results. The spectrogram transforms time-frequency signal characteristics into a visual format that facilitates automated analysis and AI processing, bridging the gap between raw data and detection outcomes
2Measurement precision
If receive beamforming is performed to identify UAV type from signals, then classification accuracy is improved, but processing time increases
Solution Approach 1:
Receive beamforming weights are pre-calculated based on signal direction information obtained from the spectrogram. By preparing the beamforming configuration in advance based on directional data, the actual classification process can proceed more quickly without sacrificing accuracy
Solution Approach 2:
The system dynamically adjusts the application of receive beamforming based on signal quality assessment. When signal quality is sufficient, beamforming is applied to maximize classification accuracy; when signal quality is poor or processing time is critical, the system can bypass beamforming to reduce processing time
3Productivity
If signal regions are determined based on spectrogram patterns to detect UAV direction, then detection speed is improved, but measurement precision may deteriorate
Solution Approach 1:
The system identifies and processes only the most relevant signal regions in the spectrogram that contain UAV characteristics, rather than analyzing the entire frequency-time spectrum. This partial action approach maintains detection speed while focusing computational resources on critical areas that contribute most to direction detection accuracy
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The proposed solution effectively detects and identifies UAVs by improving search speed and accuracy, enabling real-time detection and classification of UAVs even in conditions of limited visibility, such as dark nights or adverse weather.
Implementation Method 1
generating a spectrogram
Implementation Method 2
performing receive beamforming on the signals from the first UAV based on the receive beamforming weight
Data Source
AI summary
The detection and identification of unmanned aerial vehicles (UAVs) from a radio wave measurement result based on artificial intelligence (AI) are provided. A method of operating an apparatus to detect unmanned aerial vehicles (UAVs) includes generating a spectrogram, determining a first region to find a direction of the UAVs in the spectrogram, determining a direction of a first UAV of the UAVs based on signal values in the first region, determining a second region to identify a type of the first UAV in the spectrogram, and identifying the type of the first UAV based on signal values in the second region.