MIMO Codebook Sidelobe Leakage Reduction via Reshaping Matrix
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current DFT-based codebooks for DL MU-MIMO suffer from strong sidelobe interference, which limits the Signal to Interference plus Noise Ratio (SINR) and degrades the MU-MIMO gain due to co-channel interference, especially in Non-Line-of-Sight (NLOS) scenarios.
Innovation Solution
A new codebook is generated by applying a reshaping matrix to the existing DFT codebook, specifically using nulling precoding weights or tapered windows to reduce sidelobe leakage, resulting in a codebook with near-to-zero sidelobe leakage and wider angle coverage, suitable for NLOS scenarios.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If a DFT-based codebook is used for DL MU-MIMO, then the precoding structure is simple and easy to implement, but strong sidelobe interference occurs causing degraded MU-MIMO gain
Solution Approach 1:
The patent extracts and eliminates the harmful sidelobe components from the DFT codebook by applying a reshaping matrix that nulls out sidelobe leakage. This separates the main lobe (useful signal) from the sidelobe (interference) and removes the harmful part while preserving the simple codebook structure.
Solution Approach 2:
The patent converts the harmful sidelobe leakage into a benefit by using the reshaping matrix to create nulls in sidelobe directions. The same mathematical structure that originally created interference is now used to eliminate it, turning the harmful characteristic into a useful interference-rejection mechanism.
2Length of moving object
If a DFT-based codebook is used for DL MU-MIMO, then the main-lobe beamwidth is narrow providing good directional accuracy, but sidelobe leakage causes strong co-channel interference in NLOS scenarios
Solution Approach 1:
The patent applies different characteristics to different parts of the beam pattern. The main lobe maintains narrow beamwidth for good directional accuracy, while the sidelobes are modified to have near-zero leakage through the reshaping matrix. This creates local quality differences within the overall beam pattern.
Solution Approach 2:
The patent adds a new dimension to the codebook structure by introducing the reshaping matrix that operates in the spectral domain. This transforms the codebook from a simple DFT-based structure to a enhanced structure with additional degrees of freedom for controlling sidelobe leakage while maintaining main lobe characteristics.
3Object-generated harmful factors
If a codebook with reduced sidelobe leakage is generated, then co-channel interference is minimized, but the codebook structure becomes more complex
Solution Approach 1:
The patent performs preliminary action by pre-calculating and storing the reshaping matrix that nulls sidelobe leakage. Instead of computing complex precoding weights in real-time, the system uses a pre-generated codebook with built-in interference rejection capabilities, reducing computational complexity at the transmitter.
Solution Approach 2:
The patent creates a copy of the DFT codebook structure but modifies it through the reshaping matrix. The new codebook maintains the same size and indexing as the original DFT codebook, making it compatible with existing UE implementations, while adding sidelobe suppression capabilities through the modified structure.
4Adaptability or versatility
If a codebook with wider angle coverage is used, then NLOS scenario performance is improved, but the main-lobe beamwidth increases reducing directional accuracy
Solution Approach 1:
The patent segments the beam pattern into distinct functional regions: the main lobe for directional signal transmission and the sidelobe regions for interference suppression. By controlling these segments independently through the reshaping matrix, the system achieves both narrow main-lobe beamwidth and wide-angle coverage capability.
Solution Approach 2:
The patent creates a composite codebook structure that combines the characteristics of the original DFT codebook with the reshaping matrix modifications. This composite structure integrates the narrow beamwidth advantage of DFT with the interference rejection capability of the reshaped version, achieving both directional accuracy and NLOS adaptability.
Data Source
AI summary
A method in a network node for Multiple Input Multiple Output (MIMO) is provided. The method comprises: obtaining a precoding matrix indicator (PMI) for a first codebook for use in a Single User Multiple Input Multiple Output (SU-MIMO) transmission; determining a precoding matrix for a second codebook, based on the obtained precoding matrix indicator; and selecting the determined precoding matrix for the second codebook in response to determining that an User Equipment (UE) is scheduled for a Multi-User (MU)-MIMO transmission. A network node for performing this method is also provided.