ACQI Decoding Confidence Detection in LTE Cat-M
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In LTE CAT-M systems, Aperiodic Channel Quality Indication (ACQI) is transmitted unreliably without CRC or acknowledgement, leading to potential incorrect MCS selection for downlink transmissions, which can result in decoding errors due to the lack of confidence in ACQI reporting, especially when the number of ACQI bits is less than 12.
Innovation Solution
A confidence metric is designed to measure the reliability of ACQI decoding, and a machine learning-based approach is used to generate a confidence metric threshold model, allowing only reports with a confidence metric above a certain threshold to be considered for MCS selection, thereby reducing the probability of incorrect reporting.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If ACQI is transmitted without CRC or acknowledgement to reduce overhead, then transmission efficiency is improved, but reliability of ACQI reporting deteriorates
Solution Approach 1:
The patent replaces the traditional mechanical CRC check mechanism with a machine learning-based confidence metric evaluation system. The base station uses trained ML models to assess the reliability of received ACQI bits without requiring additional CRC bits, thus maintaining transmission efficiency while improving reliability assessment capability.
Solution Approach 2:
The patent changes the parameter used for reliability assessment from binary CRC pass/fail to a continuous confidence metric value. This allows for more nuanced reliability evaluation and enables threshold-based filtering of low-confidence ACQI reports, resolving the contradiction between efficiency and reliability.
2Quantity of substance
If the number of ACQI bits is reduced to save resources, then resource consumption is improved, but decoding accuracy deteriorates
Solution Approach 1:
The patent introduces confidence metric values as an intermediary between the received ACQI bits and the MCS selection process. This intermediary allows the system to evaluate the quality of limited ACQI bits and make informed decisions about their reliability, compensating for the reduced bit quantity.
Solution Approach 2:
The patent performs preliminary confidence metric calculation and threshold comparison before using ACQI bits for MCS selection. This preliminary action filters out low-confidence reports, ensuring that only reliable ACQI information is used for critical scheduling decisions, thereby maintaining decoding accuracy despite reduced bit quantity.
3Ease of operation
If traditional threshold modeling methods are used for ACQI decoding, then implementation simplicity is maintained, but accuracy of confidence measurement deteriorates
Solution Approach 1:
The patent replaces traditional analytical threshold modeling with machine learning-based threshold modeling. The ML models are trained offline using simulation or measurement data, and the trained models are then deployed at the base station for real-time confidence metric evaluation, achieving high accuracy while maintaining operational simplicity.
Solution Approach 2:
The patent performs extensive threshold modeling and model training in advance (offline phase), so that during actual operation (online phase), the base station only needs to evaluate pre-computed confidence metrics against stored thresholds. This preliminary action separates the complex modeling work from real-time operation, maintaining simplicity while improving accuracy.
Data Source
AI summary
According to an example embodiment, a method is provided including: decoding, at a base station of a wireless system, one or more bits of an uplink transmission from a user equipment, wherein the one or more bits are indicative of a channel quality; calculating a confidence metric corresponding to a reliability of the decoding; and causing a channel quality report to be generated based at least on the calculated confidence metric and a value for each system configuration parameter in a limited set of one or more system configuration parameters associated with the base station.