Distributed Antenna Combining Reduces Computational Complexity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current antenna combining methods for wireless communication receivers, such as Maximum Ratio Combining (MRC) and Interference Rejection Combining (IRC), face challenges in achieving optimal performance due to computational complexity and cost considerations, especially with large antenna arrays, where IRC requires extensive parallel processing and MRC is inferior in multi-user scenarios.
Innovation Solution
A distributed antenna combining system with two stages: a first antenna combining circuit applying a low-complexity algorithm like MRC near the radio receivers, and a second circuit applying a high-complexity algorithm like IRC in the baseband section, optimizing signal processing and reducing computational load by transforming signals into virtual streams.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If IRC combining is used to optimize SINR, then signal quality is improved, but computational complexity increases due to matrix inversion requirements
Solution Approach 1:
The patent segments the antenna combining process into two distinct stages: a first combining stage that performs initial signal combination with lower computational complexity, and a second combining stage that refines the combination to achieve optimal SINR. This segmentation allows the system to distribute computational load across different processing units and time periods, making the overall IRC implementation feasible for large antenna arrays.
2Reliability
If fully parallel implementation of IRC is used, then optimal layer signal extraction is achieved, but cost and power consumption increase
Solution Approach 1:
The patent divides the parallel processing requirements into two stages, where the first combining circuit performs initial processing on all antenna signals, and the second combining circuit performs refined processing on the combined outputs. This reduces the peak parallel processing requirements compared to fully parallel IRC implementation while maintaining optimal signal extraction performance.
Solution Approach 2:
The patent transforms the problem from a single-stage high-dimensional parallel processing challenge into a two-stage process where the first stage reduces dimensionality by combining N antenna signals into intermediate results, and the second stage operates on this reduced dimensionality space. This dimensional transformation reduces hardware complexity while preserving performance.
3Device complexity
If MRC combining is implemented in radio receivers, then computational load is reduced, but receiver performance deteriorates in multi-user scenarios
Solution Approach 1:
The patent implements a two-stage combining approach where the first stage can use simpler combining methods to reduce computational load at the radio receiver level, while the second stage applies more sophisticated combining techniques to restore and optimize performance in multi-user scenarios. This segmented approach allows performance optimization without overwhelming the radio receivers.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A receiver (10) featuring a distributed antenna combining system performs two stages of antenna combining. An antenna array (30) receives (102) L streams, or information flows, from UEs (20), and outputs antenna signals on N antenna ports. N parallel radio receivers (40) frequency convert (104) the N received signals to baseband. A first antenna combining circuit (50) applies (106) a relatively low-complexity form of antenna combining (e.g., MRC) to the N radio signals to generate K streams of virtual information. A second antenna combining circuit (60) applies (108) a relative high-complexity form of antenna combining (e.g., IRC) to the K virtual streams to generate L information streams. In one embodiment, antenna elements may have different polarization directions, and the first antenna combining circuit (50) generates two, differently-polarized virtual information streams for each UE (20) stream.