UAV Network Parameter Optimization via Neural Signal Quality Maps
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current network optimization methods for UAV-based communication networks are inadequate as they fail to effectively consider the unique characteristics of UAVs, such as altitude and mobility, leading to suboptimal communication performance due to limited parameter optimization and environmental reflection.
Innovation Solution
A method utilizing a neural network to determine optimal network parameters by collecting training data that includes UAV characteristic parameters and environmental factors, allowing for the generation of signal quality maps to enhance communication performance in UAV-based networks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If network optimization is performed using only specific parameters, then the optimization process is simple, but optimization performance deteriorates because characteristics in various communication environments cannot be reflected
Solution Approach 1:
The patent applies parameter changes by expanding the optimization parameters from specific parameters to multiple parameters including network parameters, UAV characteristic parameters, and environmental characteristic parameters. This allows the neural network to capture diverse communication environment characteristics while maintaining a systematic optimization framework.
Solution Approach 2:
The patent introduces another dimension by categorizing parameters into three distinct dimensions: network parameters (e.g., distance, location, antenna angle), UAV characteristic parameters (e.g., altitude, speed, mobility), and environmental characteristic parameters. This dimensional classification enables comprehensive optimization without overwhelming complexity.
2Reliability
If multiple parameters including UAV characteristics and environmental factors are considered, then optimization performance improves, but the complexity of data collection and processing increases
Solution Approach 1:
The patent uses a neural network as an intermediary that processes the complex relationships between multiple parameters. The neural network receives network parameters, UAV characteristic parameters, and environmental characteristic parameters as inputs and outputs optimized network parameters, thereby managing the complexity of data processing automatically.
Solution Approach 2:
The patent implements feedback mechanisms where the neural network continuously learns from communication data and adjusts its predictions. The system transmits initiation signals to collect training data, processes this data through the neural network, and uses the results to optimize network parameters dynamically, creating a closed-loop optimization system.
3Measurement precision
If training data collection is performed continuously, then prediction accuracy improves, but communication overhead and time consumption increase
Solution Approach 1:
The patent applies periodic action by transmitting initiation signals at determined intervals to trigger data collection. The transmission periodicity can be adjusted based on the prediction accuracy of the neural network, allowing the system to balance between maintaining high accuracy and reducing communication overhead.
Solution Approach 2:
The system performs self-service optimization where the neural network automatically processes collected data and generates optimized network parameters without requiring manual intervention. The communication nodes autonomously collect data, train the neural network, and apply the optimized parameters, reducing the time and resources needed for manual optimization.
Data Source
AI summary
A method of a first communication node may comprise: transmitting, to a second communication node, an initiation signal indicating to initiate a collection procedure of training data for a neural network; transmitting, to the second communication node, an information signal including network parameters of the first communication node; receiving, from the second communication node, the training data in response to the information signal; training the neural network using the training data; determining optimal network parameters using the trained neural network; and performing communication with the second communication node using the optimal network parameters.