Partitioned Beamforming Weights for Distributed MIMO Complexity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current beamforming schemes in distributed MIMO systems face challenges in achieving high performance without excessive computational complexity, particularly in scenarios with non-static and non-calibrated antenna arrays, and are sensitive to estimation errors and interference.
Innovation Solution
The proposed solution involves a partitioned beamforming scheme that performs eigenbeamforming for each remote radio unit based on second-order statistics and single-user MIMO optimization between remote radio units, optionally using an interference-aware eigenbeamforming scheme to minimize total transmit power and ensure a pre-defined SINR constraint.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Power
If conventional beamforming schemes are applied in distributed MIMO systems, then antenna gain can be maximized, but computational complexity becomes excessive
Solution Approach 1:
The patent segments the beamforming problem into two independent parts: (1) eigenbeamforming at each remote radio unit based on second-order statistics, and (2) single-user MIMO optimization between remote radio units. This segmentation allows each part to be solved separately with reduced computational burden compared to conventional centralized beamforming schemes.
Solution Approach 2:
The patent changes the statistical parameter from first-order (channel matrix) to second-order statistics (covariance matrix) for eigenbeamforming. This parameter change enables the system to achieve robust beamforming with reduced computational complexity by working with correlation information rather than full channel state information.
2Reliability
If beamforming is applied to non-calibrated distributed antenna arrays, then system performance can be improved, but sensitivity to estimation errors increases
Solution Approach 1:
The patent transitions from using first-order channel statistics to second-order statistics (covariance matrices) for beamforming. This parameter change makes the system more robust to estimation errors because second-order statistics capture correlation information that is less sensitive to instantaneous channel estimation inaccuracies in non-calibrated distributed arrays.
Solution Approach 2:
The patent performs eigenbeamforming based on second-order statistics as a preliminary step before single-user MIMO optimization. This preliminary action pre-processes the channel information in a way that reduces sensitivity to subsequent estimation errors in the optimization phase.
3Productivity
If distributed MIMO antenna arrays are used, then spectral efficiency can be enhanced, but interference management becomes more difficult
Solution Approach 1:
The patent segments interference management into two stages: (1) eigenbeamforming at each remote radio unit that inherently provides spatial separation, and (2) single-user MIMO optimization that coordinates between units. This segmentation simplifies interference management compared to conventional centralized approaches by localizing the interference mitigation at each unit while maintaining coordination benefits.
Data Source
AI summary
According to an aspect, there is provided an apparatus for performing beamforming optimization for a MIMO architecture. The apparatus maintains, in a database, channel state information of a plurality of radio channels comprising first and second order statistics. The apparatus calculates, separately for each of two or more remote radio units using an eigenbeamforming scheme, first sets of beamforming weights based on the second order statistics. Then, the apparatus performs, for each terminal device, single-user MIMO optimization between the two or more remote radio units to maximize a pre-defined metric, being a mutual information metric or signal-to-interference-plus-noise, based on first sets of beamforming weights and the first order statistics. The result, for each terminal device, is a second set of beamforming weights. The apparatus causes transmitting data using beams formed by applying both first and second sets of optimized beamforming weights at the two or more remote radio units.