.png){kind=icon}
.png){kind=icon}
Return
High-dimensional robust nested parallel streaming adaptive processor
.png){kind=icon}
PRE
AI
DOI：10.23919/JSEE.2025.000042.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
The paper presents a full-exchange streaming adaptive processor architecture with nested parallel sampling covariance matrix estimation and adaptive weight computation, to achieve polarization-space-time adaptive processing (P-STAP), adaptive digital beamforming (ADBF) and multiple side-lobe canceller (MSLC) within a configurable computational framework. An effective method for real-time numerical calculation is proposed by reducing truncation errors, ensuring robust convergence, and low time complexity through the lower-diagnoal-lower transpose (LDLT) right multiplication iterative (LDLT-RMI) method. To fully implement P-STAP for high-intensity interference clutter adaptive robust suppression, a series of intellectual property (IP) cores are designed for high-order, high-condition symmetric positive definite matrix inversion. The adaptive processor can operate on a frame-structured data flow, accommodating various space-time-frequency-polarization multi-domain combinations, which supports 8 to 100 channels with a dynamic range of up to 80 dB. Even the interferences reach a high intensity of 65 dB, the adaptive processor can still work stably.
Keywords:
nested parallel streaming adaptive processing (P-STAP)
configurable computational framework
polarization-space-time adaptive processing
high-order high-condition matrix inversion
Journal
J
IF:
2.1
Papers:
84
Citations:
2.6K