Colocated MIMO Radar Target Detection via Serial Cancellation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing colocated MIMO radar systems face challenges in accurately detecting multiple targets and estimating their spatial coordinates due to the computational complexity of multidimensional optimization problems and the potential for missing weaker target echoes.
Innovation Solution
The proposed method employs novel algorithms for estimating the parameters of multiple overlapped sinusoids or complex exponentials affected by additive noise, using one-dimensional fast Fourier transforms and cost function maxima searches in frequency, azimuth, or elevation domains, and incorporates serial cancellation and spatial folding techniques to enhance detection accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If optimal maximum likelihood techniques are employed for target detection and spatial coordinate estimation, then measurement precision is improved, but device complexity increases due to complicated multidimensional optimization problems and huge computational effort
Solution Approach 1:
The patent segments the complex multidimensional optimization problem into simpler one-dimensional search problems. Instead of performing joint optimization over range, azimuth, and elevation simultaneously, the method first performs 1D FFT in the range domain to obtain range estimates, then uses these estimates to simplify the angular domain optimization. This segmentation reduces the computational dimensionality while maintaining estimation accuracy.
Solution Approach 2:
The patent extracts the range estimation step as a separate preliminary operation using 1D FFT before performing angular estimation. By extracting and solving the range dimension first, the remaining angular estimation problem becomes simpler and requires less computational effort, effectively separating the difficult multidimensional optimization into manageable sequential steps.
2Measurement precision
If multidimensional Fast Fourier Transform is computed for signal spectral analysis, then measurement precision is improved, but productivity decreases due to significant computational effort required
Solution Approach 1:
The patent segments the multidimensional FFT operation into sequential 1D FFT operations. First, a 1D FFT is performed in the range domain to obtain range profiles. Then, for each detected target, separate 1D FFTs are performed in the azimuth and elevation domains. This segmentation transforms a computationally intensive ND-FFT into multiple efficient 1D-FFT operations, significantly improving processing speed while maintaining spectral analysis accuracy.
3Measurement precision
If spectral analysis is performed to detect targets, then measurement precision is improved, but loss of information occurs when weaker target echoes are hidden by stronger echoes
Solution Approach 1:
The patent extracts and removes the contribution of strong targets from the received signal before detecting weaker targets. After detecting a strong target and estimating its parameters (range, azimuth, elevation), the method subtracts this target's echo from the original signal. This extraction process eliminates the masking effect, allowing weaker targets that were previously hidden to be detected in the residual signal.
Solution Approach 2:
Instead of directly detecting all targets simultaneously from the original signal where strong targets mask weak ones, the patent inverts the approach by iteratively detecting strong targets first, removing their contributions, and then detecting remaining targets in the residual signal. This inversion of the detection sequence allows weak targets to be revealed by eliminating the dominant strong target interference.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Method for detection of targets through a MIMO FMCW radar equipped with a plurality of transmitting (TX) and receiving (RX) antennas, wherein each couple of said TX and RX Method for detection of targets through a MIMO FMCW radar equipped with a plurality of transmitting (TX) and receiving (RX) antennas, wherein each couple of said TX and RX antennas is replaced with the equivalent virtual antenna; said MIMO FMCW radar is arranged to generate real or complex signals in response to a propagation scenario including a plurality of point targets; said method includes: estimation of the spectrum of said signal and its first derivatives, for each of said virtual antennas through FFT calculation with an oversampling factor, acquisition of a sub-set of a number of virtual antennas spectra and of the corresponding derivatives and recursive execution of calculation of an estimate of the parameters of the most dominant point target, including phase, amplitude and frequency for each of said virtual antennas of said sub-set on the basis of said spectrum, and its derivatives, for each of said virtual antennas of said sub-set, cancellation of said most dominant point target and computation of a residual spectrum, computation of an energy of said residual spectrum and return to (Step 3) until said energy is over a predetermined threshold ( T STDREC ), otherwise computation of an average energy and fusion of ranges information of said point targets detected by said number of virtual antennas to list said point targets according to said average energy in a decreasing order.