MIMO Channel Prediction Model Using Reduced Training Data Dimensions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional AI-based channel prediction technologies in MIMO communication systems face excessive overhead in training, leading to inaccurate channel state information due to feedback delays and user equipment mobility, which reduces transmission rates.
Innovation Solution
An electronic device with processing circuitry that acquires training data from multiple antennas, trains a channel prediction model using reduced data dimensions, and estimates channel values for resource blocks based on pilot signals, employing a Multi-Layer Perceptron to minimize prediction errors, thereby reducing training overhead and improving accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional AI-based channel prediction technology is used, then channel prediction accuracy is improved, but training overhead (time and resources) increases excessively
Solution Approach 1:
The patent extracts only the essential features needed for channel prediction from the full channel state information, focusing on key temporal and spatial characteristics. This selective extraction reduces the dimensionality of training data while preserving the most informative aspects for prediction accuracy.
Solution Approach 2:
The patent transforms the training data by changing parameters such as reducing temporal resolution (using fewer time slots) and spatial resolution (selective antenna sampling). This parameter transformation maintains prediction accuracy while significantly reducing the volume of training data required.
2Productivity
If channel state information is updated frequently to maintain accuracy, then transmission rate is improved, but feedback delay and mobility cause the information to become outdated
Solution Approach 1:
The patent performs preliminary channel prediction using historical data and learned patterns to estimate future channel states before actual measurements are available. This allows the system to prepare transmission parameters in advance based on predicted channel conditions, maintaining high transmission rates without waiting for real-time feedback.
Solution Approach 2:
The patent implements a feedback mechanism where prediction errors are continuously monitored and used to retrain and refine the prediction model. This closed-loop feedback ensures that the model adapts to changing channel conditions caused by mobility while maintaining accurate predictions for optimal transmission rates.
Data Source
AI summary
An electronic device which includes processing circuitry configured to acquire training data corresponding to multiple antennas, the training data having a label and features, the label including a first channel value for a first time slot, and the features including a plurality of second channel values for a plurality of time slots before the first time slot, train a channel prediction model based on the training data to obtain a trained channel prediction model, and obtain a channel prediction value for a prediction time based on the trained channel prediction model.