Digital Self-Interference Estimation for Full-Duplex Radio
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional linear digital self-interference estimation methods for Full-Duplex Radio (FDR) networks are complex and unable to accurately estimate self-interference signals, particularly due to signal memory effects caused by RF amplifiers, leading to suboptimal interference cancellation.
Innovation Solution
A digital self-interference estimation method that involves receiving multiple self-interference signals and ideal transmitting signals at different timings, creating a signal adjusting vector to estimate subsequent self-interference signals using a linear dynamic model, allowing for interference cancellation without complex operations and accounting for signal memory effects.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If Maximum Likelihood estimation algorithm is used for self-interference signal estimation, then estimation accuracy is improved, but operational complexity increases significantly
Solution Approach 1:
The patent changes the estimation approach from Maximum Likelihood (iterative optimization) to Linear Minimum Mean Square Error (LMMSE) estimation. This parameter change in the estimation methodology reduces computational complexity while maintaining acceptable accuracy by using linear operations instead of iterative optimization, directly addressing the contradiction between estimation accuracy and operational complexity
Solution Approach 2:
The patent employs a simplified estimation model that uses readily available pilot signals and channel state information to compute self-interference estimates through simple linear operations. This approach uses 'cheap' computational resources (basic matrix operations) instead of expensive iterative algorithms, reducing operational complexity while achieving sufficient estimation accuracy for interference cancellation
2Device complexity
If overall network environment is used for estimation, then estimation process is simplified, but ability to account for device-specific signal memory effects is lost
Solution Approach 1:
The patent segments the self-interference estimation process into two parts: (1) using overall network environment information for the general estimation framework, and (2) incorporating device-specific signal memory effects through separate channel impulse response characterization. This segmentation allows both global and local factors to be considered without significantly increasing complexity
Solution Approach 2:
The patent performs preliminary characterization of the device-specific channel impulse response and signal memory effects during calibration phases using pilot signals. This preliminary action captures device-specific characteristics in advance, allowing the main estimation process to use pre-computed parameters rather than calculating everything in real-time, thus maintaining simplicity while improving accuracy
Data Source
AI summary
A wireless communication device and a digital self-interference estimation method thereof are provided. The wireless communication device, at respective timings, receives a plurality of self-interference signals and generates a plurality of ideal transmitting signals. The wireless communication device calculates a signal adjusting vector based on the self-interference signals and the ideal transmitting signals at different timings. The wireless communication device generates a main ideal transmitting signal at a main timing, and calculates, based on the signal adjusting vector, a main self-interference signal corresponding to the main timing according to the received self-interference and the main ideal transmitting signal.