RF Auto-Correlation Estimation Using Power Sensors and Phase Shifts
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional systems for estimating the auto-correlation matrix in transceivers require high-speed analog-to-digital converters, which are expensive in terms of processing hardware, power, and cost, making them inefficient.
Innovation Solution
The transceiver utilizes power sensors and phase shifters to measure power in the RF domain, eliminating the need for high-speed digitization and baseband processing, by shifting signals at different phase offsets and combining them to estimate the auto-correlation matrix.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If high-speed analog-to-digital converters are used to digitize incoming analog signals in the voltage domain, then the auto-correlation matrix can be calculated using statistical methods, but the processing hardware cost, power consumption, and system cost increase significantly
Solution Approach 1:
The patent extracts only the essential information needed for auto-correlation matrix estimation by using power sensors to measure signal power directly in the RF domain, rather than converting the entire signal waveform to digital form. This extraction approach obtains the necessary statistical information (power values) without the overhead of high-speed ADCs and complex baseband processing hardware.
Solution Approach 2:
The patent replaces the electronic/mechanical system of high-speed ADCs and digital signal processors with a simpler measurement system using power sensors that operate directly in the RF domain. This substitution eliminates the need for voltage-domain digitization while still providing the necessary information for auto-correlation estimation through power measurements.
2Productivity
If high-speed analog-to-digital converters are used to digitize incoming analog signals, then real-time calculation of the auto-correlation matrix is possible, but power consumption increases significantly
Solution Approach 1:
The patent extracts only the power information from the RF signals using power sensors, rather than converting the complete signal waveform to digital form. This extraction of essential information maintains real-time processing capability while dramatically reducing power consumption by avoiding the energy-intensive operations of high-speed ADCs and digital signal processing.
Solution Approach 2:
The power sensors directly measure the power of RF signals in their native domain without requiring conversion to the voltage domain. This self-service approach allows the measurement system to operate in the domain where the signals naturally exist, eliminating the need for energy-consuming domain conversion and baseband processing.
3Measurement precision
If conventional voltage-domain processing is used to estimate the auto-correlation matrix, then accurate statistical methods can be applied, but the system cost increases due to expensive processing hardware
Solution Approach 1:
The patent uses inexpensive power sensors instead of expensive high-speed ADCs and digital processing hardware. The power sensors provide the necessary measurement capability at a fraction of the cost of conventional voltage-domain processing equipment, making the system more economically viable while still achieving accurate auto-correlation matrix estimation.
Solution Approach 2:
The patent substitutes expensive electronic processing hardware (ADCs, FPGAs, DSPs) with simpler, less costly power measurement systems that operate directly in the RF domain. This substitution maintains estimation accuracy while dramatically reducing system cost and complexity.
Data Source
Figure 1
Figure 2A
Figure 2B
AI summary
Techniques are provided for estimating an auto-correlation matrix for a transceiver that is coupled to a multi-antenna array. A different power sensor measures the power of the signals received at a corresponding antenna of the multi-antenna array. The signals received at one antenna of each unique pair of antennas are shifted in phase using a plurality (e.g., three) of different phase settings to generate a plurality of different shifted signals for each unique pair of antennas. Each of one or more power sensors, for each unique pair of antennas, measures different combined powers based on the shifted signals and the RF signals received at the other antenna of the unique pair. A module uses the different combined powers with the power measured for each antenna to estimate the auto-correlation matrix.