CSI Feedback Processing Units for Wireless Communication
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In wireless communication systems, particularly in multi-antenna technologies like 3GPP LTE and NR, the increasing complexity of CSI feedback leads to high computing power requirements, causing inefficiencies and waste, especially when different CSI reportings have varying computing power needs.
Innovation Solution
The method involves using two types of processing units to handle Type I and Type II CSI reportings separately, allowing the UE to determine which CSI reportings to update based on priorities and thresholds, optimizing the use of computing resources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If ML/AI-based CSI feedback techniques are introduced to compress CSI and address feedback accuracy and overhead, then CSI feedback accuracy and efficiency are improved, but computing power requirements and UE complexity increase significantly
Solution Approach 1:
The patent segments CSI reportings into different types (Type I and Type II) with different computing power requirements. Type I CSI reportings use first-type processing units while Type II CSI reportings use second-type processing units. This segmentation allows the UE to manage computing resources more efficiently by not allocating high computing power to all CSI reportings uniformly, thus reducing overall UE complexity while maintaining feedback accuracy where needed.
Solution Approach 2:
The patent applies local quality by assigning different processing qualities to different CSI reportings based on their specific requirements. Not all CSI reportings require the same computing power, so the system provides differentiated processing levels locally at the UE, optimizing the balance between feedback accuracy and computing resource consumption.
2Reliability
If the UE updates all N CSI reportings simultaneously to ensure complete feedback, then feedback completeness is improved, but computing power waste increases due to uniform processing of all reportings
Solution Approach 1:
The patent segments CSI reportings into different categories based on their computing power requirements. By dividing the N CSI reportings into Type I and Type II groups, the UE can selectively update only the necessary reportings in each group based on available computing resources, avoiding the waste of processing all reportings uniformly while maintaining feedback completeness for critical reportings.
Solution Approach 2:
The patent applies partial action by updating only a subset of CSI reportings when computing resources are limited, rather than attempting to update all N reportings simultaneously. The UE can prioritize updating Type I CSI reportings or a subset of Type II reportings based on availability, achieving sufficient feedback without excessive computing power consumption.
3Device complexity
If the UE uses a single processing unit for all CSI reportings to simplify architecture, then device complexity is reduced, but computing power allocation efficiency deteriorates
Solution Approach 1:
The patent segments processing units into first-type processing units for Type I CSI reportings and second-type processing units for Type II CSI reportings. This segmentation, while increasing structural complexity, enables efficient computing power allocation by matching the right processing capabilities to the right CSI reporting types, avoiding the inefficiency of uniform processing.
Solution Approach 2:
The patent implements local quality in the processing architecture by providing different processing capabilities at different locations (for different CSI reporting types). This allows optimized computing power allocation where Type I reportings receive appropriate processing from first-type units and Type II reportings receive processing from second-type units, maximizing overall productivity.
Data Source
AI summary
A first node receives N CSI configuration information; updates a CSI reporting in a first CSI reporting subset; transmits a first information block; the N CSI configuration information is respectively used to determine N CSI reportings, and the first CSI reporting subset is a subset of the N CSI reportings; the first information block comprises at least one CSI reporting of the N CSI reportings; any Type I CSI reporting in the N CSI reportings corresponds to a first-type integer, and any Type II CSI reporting in the N CSI reportings corresponds to a second-type integer, the N CSI reportings respectively correspond to N priorities, and the N priorities are used to determine a number of CSI reporting(s) comprised in the first CSI reporting subset. The above method configures different processing units for CSI reportings with different CSI computing power requirements, thus avoiding the waste of computing power.