.png){kind=icon}
.png){kind=icon}
Return
Multi-Frame Track-Before-Detect Algorithm for Radar Range-Spread Targets
.png){kind=icon}
PRE
AI
DOI：10.1109/tsp.2026.3717767.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Traditional multi-frame track-before-detect (MF-TBD) algorithms, originally designed for point-like targets, can enhance the detection and tracking of weak targets by integrating radar measurements over consecutive frames. However, with advances in radar resolution, the detection of range-spread targets (RSTs), whose energy is dispersed across unknown scattering centers (SCs), has become increasingly important. The MF-TBD algorithm applied to RSTs may suffer performance degradation due to the insufficient integration of target energy. We make two contributions to address this problem. First, we propose an MF-TBD algorithm tailored for RSTs (RST-MF-TBD) based on the multi-frame joint estimation of the target state sequence and the spatial distribution of SCs using the maximum likelihood (ML) criterion. Second, since the proposed RST-MF-TBD is a general procedure without specific model assumptions, we derive its explicit equations for the frequently encountered case of zero-mean Gaussian noise. Furthermore, we develop an efficient implementation of RST-MF-TBD by incorporating a sparse representation (SR)-based estimation strategy for the prior knowledge of target SCs, achieving both tractable computational complexity and superior detection performance. Finally, numerical simulations and real-measured data are used to validate the effectiveness of the proposed RST-MF-TBD algorithm across different target scattering models and in comparison with competitive methods.
Keywords:
Multi-frame track-before-detect
range spread targets
sparse representation
Journal
I
IF:
5.8
Papers:
276
Citations:
0