Machine-Learned Reference Signal Timing to Reduce 5G Feedback Overhead
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional 5G systems with large antenna arrays face significant challenges in managing overhead of channel estimation and feedback, which increases bandwidth, time, and compute resources, while maintaining system performance.
Innovation Solution
Implementing a trained machine learning model to determine reference symbol transmission times based on signal feedback, using contextual bandit learning to optimize downlink throughput and uplink overhead, thereby reducing computational resources and bandwidth overhead.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If large antenna arrays are used for beamforming and MIMO techniques, then downlink throughput and signal strength are improved, but overhead of channel estimation and feedback increases linearly
Solution Approach 1:
The patent extracts only the essential channel state information needed for beamforming adjustments rather than transmitting complete channel estimates. By selecting and transmitting only critical reference symbols that capture the most important channel variations, the system reduces feedback overhead while maintaining the ability to adjust beamforming weights effectively.
Solution Approach 2:
The patent segments the channel estimation process into periodic reference symbol transmissions interspersed with data transmissions. Instead of continuous channel estimation feedback, the system divides communication into frames with periodic reference symbols, reducing the frequency and volume of feedback while still capturing channel variations over time.
2Measurement precision
If periodic reference symbol exchange is used to adjust beamforming, then beamforming accuracy is improved, but bandwidth and time resources are consumed
Solution Approach 1:
The patent implements periodic reference symbol transmission within communication frames, where reference symbols are sent at regular intervals rather than continuously. This periodic approach maintains beamforming accuracy by capturing channel variations at sufficient intervals while freeing up time and bandwidth resources for data transmission during non-reference symbol periods.
Solution Approach 2:
The system dynamically adjusts beamforming weights based on periodically received reference symbols rather than using static weights. The base station updates precoding matrices at each reference symbol occasion, allowing beamforming to adapt to changing channel conditions while maintaining a balance between accuracy and resource consumption.
Data Source
AI summary
Aspects of the present disclosure relate to determining reference symbol transmission times. In some examples, a method for determining reference symbol transmission times for cellular communications includes receiving signal feedback based on a wireless communication channel between a wireless communication device and a base station, identifying a periodic exchange of reference symbols that are used to adjust beamforming between the wireless communication device and the base station, generating a vector based on the signal feedback, and providing the vector as an input to a trained machine learning model. A training of the trained machine learning model includes calculating a plurality of rewards for a respective plurality of transmission time delays. The plurality of rewards are each calculated based on a function of downlink throughput and uplink overhead. The function of downlink throughput and uplink overhead are based upon a priority level of the wireless communication device.