RL Neural Network MIMO Detector Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In wireless communication systems, particularly in MIMO systems like LTE and 5G NR, there is a trade-off between detector complexity and error rate, with high complexity detectors offering low error rates but high power consumption, and low complexity detectors resulting in higher error rates, necessitating a dynamic approach to select the appropriate detector based on channel conditions.
Innovation Solution
A reinforcement learning (RL) neural network is used to extract features from channel and LLR data across multiple resource elements, enabling the selection of the most suitable symbol detector for each element, thereby optimizing complexity and error rate, and reducing power consumption by employing the least complex detector necessary for successful decoding.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a high complexity detector (e.g., ML) is used, then error rate is reduced, but power consumption and complexity increase
Solution Approach 1:
The patent implements dynamic detector selection where the system switches between different detector types (ML, MMSE, etc.) based on real-time channel conditions. The detector choice is not fixed but adapts dynamically to current transmission conditions, allowing the system to use high-complexity ML detectors only when necessary for maintaining low error rates, while using lower-complexity detectors in favorable conditions.
Solution Approach 2:
The system changes the operational parameters by selecting different detector algorithms based on channel quality metrics. When channel conditions deteriorate below certain thresholds, the system transitions from low-complexity to high-complexity detectors, effectively changing the detection parameter to maintain reliability while optimizing overall system performance.
2Device complexity
If a low complexity detector (e.g., MMSE) is used, then power consumption and complexity are reduced, but error rate increases
Solution Approach 1:
The system dynamically adjusts detector complexity based on channel conditions rather than using a fixed low-complexity detector. This allows the system to maintain low complexity in good channel conditions while automatically increasing complexity when reliability requirements demand it.
Solution Approach 2:
The detector selection mechanism changes the operational parameter (detector type) based on channel quality, allowing the system to use simple MMSE detectors in favorable conditions and switch to more complex ML detectors when error rates would otherwise become unacceptable.
3Device complexity
If a static detector is used for all resource elements, then device complexity is simplified, but adaptability to varying channel conditions deteriorates
Solution Approach 1:
The patent implements a dynamic detector selection mechanism that evaluates channel conditions for each resource element and selects the appropriate detector type accordingly. This dynamic approach replaces static detector selection, allowing the system to adapt to varying channel conditions while maintaining manageable complexity through automated decision-making.
Solution Approach 2:
The system performs self-service by automatically selecting appropriate detectors based on its own channel condition assessments. The detector selection mechanism uses embedded channel quality metrics to make autonomous decisions about which detector to employ, eliminating the need for external control while achieving adaptability.
Data Source
AI summary
A method and system for selecting a symbol detector are herein provided. A method includes extracting a first set of features for a k-th resource element (RE), where k is an integer greater than one, extracting a second set of features from a first RE to a (k−1)th RE, and selecting a symbol detector for the k-th RE using a reinforcement learning (RL) neural network based on the extracted first set of features and the extracted second set of features.