Neural Network Channel Estimation Using Pre-Trained OSCC Parameters
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional channel estimation methods for antenna array systems face high computational complexity, especially with massive MIMO systems, making real-time processing challenging, and require costly user-specific training and parameter sets for optimal performance.
Innovation Solution
A neural network-based channel estimation approach that uses a central network entity to classify On-Site Channel Characteristics (OSCC) and store class-specific neural network parameters, allowing for efficient channel estimation without the need for user-specific training, by leveraging pre-calculated weights and parameters from a central pool.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional channel estimation methods are used for massive MIMO systems, then channel estimation accuracy can be maintained, but computational complexity increases significantly making real-time processing unaffordable
Solution Approach 1:
The patent applies preliminary action by pre-training neural network parameters offline for different channel scenarios. The trained parameters are stored and later retrieved during real-time operation, avoiding the need for complex real-time computations while maintaining accurate channel estimation performance.
2Measurement precision
If user-specific training is performed for optimal channel estimation performance, then estimation accuracy improves, but training effort and system complexity increase
Solution Approach 1:
The patent implements universality by creating a centralized parameter repository that stores pre-trained neural network parameters for multiple channel scenarios. Any user can retrieve the appropriate pre-trained parameters based on their channel conditions, eliminating the need for individual user-specific training while maintaining optimal performance across different scenarios.
3Productivity
If the number of MIMO receive antennas and modulation order are increased, then system capacity improves, but the complexity of joint data detection and channel estimation increases exponentially
Solution Approach 1:
The patent extracts the complex joint detection and estimation function into a separate neural network module with pre-trained parameters. This allows the main system to handle increased antennas and modulation orders while the neural network handles the computationally intensive processing using efficiently stored parameters, preventing exponential complexity growth.
Data Source
AI summary
It is provided a method, comprising identifying a value of an onsite channel characteristic of a receive channel; requesting a neural network parameter, wherein the request comprises an indication of the onsite channel characteristic; monitoring if the neural network parameter is received in response to the request; estimating the receive channel by a neural network using the neural network parameter if the neural network parameter is received.