Space-Time Interleaved Matched-Filter Direction Finding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems for processing signals from array antennas face challenges in efficiently determining signal direction at sampling frequencies lower than the Nyquist frequency, leading to computational costs and numerical errors due to interpolation and potential aliasing issues.
Innovation Solution
A system that includes a receiving array antenna with a digitizer and processing unit, which samples and digitizes analog signals, calculates adjusted sampling times based on antenna element positions and hypothesized signal directions, and fits these samples with sine and cosine functions to determine signal direction without interpolation, allowing for undersampling below the Nyquist frequency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If interpolation is used to estimate signal amplitude at non-sampling times, then time delay can be achieved, but computational cost increases and numerical errors are introduced
Solution Approach 1:
The patent extracts the interpolation operation from the signal processing chain and replaces it with direct use of sampled values. By formulating the direction finding as a fit between sampled data and sinusoidal models, the system eliminates the need for amplitude estimation at non-sampling instants, thereby removing the source of numerical errors while maintaining time delay capability through phase information.
Solution Approach 2:
The patent substitutes the mechanical interpolation process with a mathematical model-fitting approach. Instead of mechanically estimating amplitudes at intermediate times through interpolation algorithms, the system uses direct fitting of sinusoidal functions to the sampled data, replacing a computationally intensive mechanical process with an efficient mathematical operation.
2Reliability
If sampling frequency is increased to meet Nyquist frequency, then aliasing is avoided, but sampling cost and complexity increase
Solution Approach 1:
The patent changes the fundamental parameter of sampling frequency from Nyquist rate to a lower sub-Nyquist rate. By combining this parameter change with a model-fitting approach that can resolve directions even with undersampled data, the system achieves reliable direction finding without requiring high sampling frequencies, thereby reducing converter complexity and power consumption.
3Ease of operation
If interpolation is used for time delay, then directional signal processing is enabled, but computational cost becomes prohibitive
Solution Approach 1:
The patent removes the interpolation step from the directional signal processing pipeline. By directly fitting sinusoidal models to the sampled values from multiple antenna elements, the system achieves time delay and direction finding without the computationally expensive interpolation operation, significantly improving processing efficiency.
Solution Approach 2:
The patent performs preliminary modeling of the signal as sinusoidal functions before actual direction finding. By pre-establishing the sinusoidal model form and fitting it directly to sampled data, the system prepares the processing in advance to avoid costly post-sampling interpolation operations.
Data Source
AI summary
A system for determining a direction for an electromagnetic signal. The system includes a receiving array antenna, including a plurality of antenna elements, each antenna element having a position in the receiving array antenna; a digitizer configured to receive an analog signal from each of the antenna elements, and to sample and digitize each of the analog signals to form a sequence of digitized samples from each of the analog signals; and a processing unit. The processing unit is configured to receive the sequences of sample values from the digitizer; and, for each direction of a plurality of hypothesized directions: combine the sample values from the plurality of antenna elements to form a single time record; fit the single time record with a combination of one or more functions of time; and identify a direction at which a measure of the magnitude of the linear combination is greatest.