Adaptive DMRS Pattern Design for Uplink Channel Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current deterministic de-modulation reference signal (DMRS) patterns in 3GPP Release 15 are inflexible and unable to adapt to changing radio conditions, leading to either unnecessary overhead or reduced channel estimation performance in uplink transmission.
Innovation Solution
Implementing an adaptive DMRS pattern design that uses a predictor function, such as a recursive neural network, to optimize DMRS patterns in both time and frequency domains, reducing the number of DMRS resource elements while maintaining channel estimation performance, and dynamically signaling the optimal pattern to user equipment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If more DMRS pilots are used, then channel estimation performance is improved, but uplink overhead increases
Solution Approach 1:
The patent implements dynamic DMRS pattern selection where the system adapts the DMRS configuration based on current radio conditions. The network entity determines appropriate DMRS patterns from multiple candidate patterns and signals them to the UE, allowing the system to transition from static deterministic patterns to dynamic adaptive patterns that optimize the balance between channel estimation performance and uplink overhead.
Solution Approach 2:
The patent changes multiple DMRS parameters simultaneously including time-domain positions, frequency-domain positions, and density configurations. By defining multiple candidate patterns with different parameter combinations and selecting among them based on radio conditions, the system can adjust the number and distribution of DMRS pilots to achieve optimal channel estimation while minimizing overhead.
2Adaptability or versatility
If deterministic DMRS patterns are used, then implementation is simple, but adaptability to channel variations is poor
Solution Approach 1:
The patent pre-defines multiple candidate DMRS patterns with different time-frequency configurations before actual transmission. These candidate patterns are prepared in advance and stored, allowing the system to quickly select from pre-prepared options based on current radio conditions without complex real-time calculations, thus maintaining relatively simple implementation while achieving good adaptability.
Solution Approach 2:
The system uses feedback mechanisms where the network entity monitors radio conditions and determines appropriate DMRS patterns from candidate patterns, then signals the selection to the UE. This closed-loop feedback approach enables adaptive pattern selection that responds to changing channel conditions while keeping the complexity manageable through standardized signaling procedures.
3Speed
If RRC reconfiguration is used to adapt DMRS parameters, then configuration can be updated, but response time is slow
Solution Approach 1:
The patent segments the DMRS configuration into two parts: semi-static parameters configured via RRC and dynamic parameters indicated via faster signaling. The candidate patterns are pre-configured through RRC, but the specific pattern selection is dynamically indicated through faster signaling mechanisms, allowing quick adaptation to channel changes while maintaining configuration reliability through the layered approach.
Solution Approach 2:
The patent introduces an intermediary layer of pre-configured candidate patterns that act as a bridge between RRC configuration and dynamic adaptation. Instead of directly reconfiguring all parameters through slow RRC signaling, the system uses pre-prepared pattern candidates as intermediaries, allowing faster selection and switching based on current radio conditions while maintaining the reliability of RRC-configured parameters.
Data Source
AI summary
There are provided measures for adaptive de-modulation reference signal pattern design. Such measures exemplarily comprise receiving, from a communication entity, in a physical channel, at least one reference signal representative of radio conditions in at least one portion of said physical channel, and predicting, based on said at least one reference signal and a model of said physical channel, a first channel estimation representative of radio conditions in a total of said physical channel.