A Dual-Pointer Flattening Architecture for Asymmetric Resampling in Bernoulli Particle Filters
Year: 2026
Type: conference-paper
Abstract: Systematic resampling is a critical computational bottleneck for Bernoulli particle filters in monopulse radar target tracking. Conventional hardware implementations often encounter high execution latency due to nested loop structures and data-dependent branch jumps. To address this issue, this paper proposes a parallel hardware architecture based on dual-pointer loop flattening logic. By exploiting the inherent monotonicity of sampling sequences, the design flattens the nested search into a single-layer structure with an initiation interval of 1. Implementation results show that the architecture completes the resampling task for a scale of 1500 to 1000 particles within 0 120. ms , achieving 99 9. % functional consistency with the software reference model. Furthermore, the O(Nin+ Nout) linear complexity achieved via loop flattening contributes to the architecture's robust scalability for diverse particle scales. This design enhances processing efficiency while maintaining low resource overhead, providing a practical reference for the hardware implementation of real-time radar tracking systems.
Authors: Jianying Tian, Ziqiao Yuan
Language: English
Subfield: Hardware and Architecture
Field: Computer Science
Domain: Physical Sciences
Open Access status: hybrid
OpenAlex ID: W7165552477
