Massive MIMO Beam Domain Channel Model Using Refined Sampling Steering Vector Matrix
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current massive MIMO systems face challenges in accurately acquiring statistical channel information due to limited antenna sizes and increased user density, leading to errors in instantaneous channel estimation and limited time-frequency resources, which affects downlink multi-user precoding transmission.
Innovation Solution
A method for acquiring massive MIMO beam domain a priori and a posteriori statistical channel information using a refined sampling steering vector matrix with more steering vectors than antennas, transforming pilot signals and channel information into a refined beam domain, and solving for channel energy matrices using sampling statistics.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If a conventional DFT matrix-based beam domain channel model is used, then the model is simple and easy to implement, but it deviates from the actual physical channel model to a considerable extent due to limited antenna size
Solution Approach 1:
The patent changes the fundamental parameter of the beam domain transformation by replacing the DFT matrix with a refined sampling steering vector matrix that has more steering vectors than antennas. This parameter change allows the model to better represent the actual physical channel while maintaining computational feasibility through the specific structure of the refined sampling approach.
Solution Approach 2:
The patent introduces an additional dimension by using more steering vectors than the number of antennas. The refined sampling steering vector matrix has dimensions that exceed the antenna count, adding a new dimensional aspect to the beam domain representation that captures more physical channel characteristics without requiring more physical antennas.
2Quantity of substance
If the number of user antennas occupying the same time-frequency resources is increased, then user density is improved, but time-frequency resources for pilots are limited leading to errors in instantaneous channel estimation
Solution Approach 1:
The patent performs preliminary action by acquiring a priori statistical channel information before actual data transmission. The refined beam domain a priori statistical model is established in advance, allowing the system to prepare robust precoding strategies before channel conditions deteriorate or change, thereby maintaining accuracy even with limited pilot resources.
Solution Approach 2:
The patent implements feedback by using both a priori statistical information and a posteriori instantaneous channel information. The system continuously updates and refines channel estimates by combining prior statistical knowledge with current channel measurements, creating a feedback loop that improves estimation accuracy despite resource constraints.
3Productivity
If downlink multi-user precoding transmission is implemented, then spectral efficiency is improved, but mobility of users poses significant challenges requiring robust transmission methods
Solution Approach 1:
The patent performs preliminary action by establishing the refined beam domain a priori statistical channel model in advance of actual transmission. This pre-established model provides a foundation for robust precoding that can adapt to user mobility, allowing the system to maintain reliability while achieving high spectral efficiency through multi-user precoding.
Data Source
AI summary
Disclosed are a method and system for acquiring massive MIMO beam domain statistical channel information. A refined beam domain channel model involved in the disclosed method is based on a refined sampling steering vector matrix. Compared with a traditional DFT matrix-based beam domain channel model, when antenna size is limited, said model is closer to a physical channel model, and provides a model basis for solving the problem of the universality of massive MIMO for various typical mobile scenarios under a constraint on antenna size. The present invention provides a method for acquiring massive MIMO refined beam domain a priori statistical channel information and a posteriori statistical channel information, the a posteriori statistical channel information comprising mean and variance information of the a posteriori channel. The method of the present invention has low complexity, can be applied to an actual massive MIMO system, provides support for a robust precoding transmission method, and has large application value.