Time-Domain CSI Compression for Low-Overhead AI Positioning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional AI models require large amounts of air-interface resources to transmit time-domain channel state information due to its high data size, which is typically 256, leading to inefficient resource utilization.
Innovation Solution
Perform processing on time-domain channel state information to reduce its data size before transmission, using methods such as truncation, filtering, normalization, or quantization, and then send the processed information to a target model for positioning or feature extraction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If time-domain channel state information is transmitted directly for AI model processing, then positioning precision is maintained, but air-interface resource consumption increases significantly
Solution Approach 1:
The patent applies preliminary action by performing processing on the time-domain channel state information at the terminal before transmission to the network side. The terminal performs truncation, filtering, or other processing to obtain processed channel state information with reduced data size, thereby reducing air-interface resource consumption while maintaining positioning precision through the AI model's ability to extract essential features from the processed data.
Solution Approach 2:
The patent applies the extraction principle by extracting only the essential features from the time-domain channel state information. Through truncation of less significant parts or extraction of key characteristics, the patent reduces the data volume transmitted over the air interface while preserving the critical information needed for accurate positioning by the AI model.
2Quantity of substance
If the data size of time-domain channel state information is reduced through processing, then air-interface resource consumption decreases, but information completeness may be compromised
Solution Approach 1:
The patent applies parameter changes by modifying the characteristics of the time-domain channel state information through processing operations. The terminal changes parameters such as data length, filtering thresholds, or compression ratios to achieve an optimal balance between reducing data size for efficient transmission and maintaining sufficient information completeness for accurate positioning.
Solution Approach 2:
The patent applies feedback by implementing a mechanism where the network side receives the processed channel state information and provides feedback to the terminal. This feedback loop allows the terminal to adjust its processing parameters dynamically, ensuring that information completeness is maintained while optimizing data size reduction based on actual transmission needs and positioning requirements.
Data Source
AI summary
This application discloses an information transmission method, an information processing method, an apparatus, a terminal, and a network-side device, and belongs to the field of communication technologies. The information transmission method according to embodiments of this application includes: a first device performs first processing on first time-domain channel state information, to obtain second time-domain channel state information, where a data size of the first time-domain channel state information is greater than a data size of the second time-domain channel state information; and the first device sends first information to a first node, where the first information is related information of the second time-domain channel state information, where the first time-domain channel state information is time-domain channel state information obtained by the first device through measurement based on a reference signal.