Thiran Filters for Wideband DSP-Based Multi-Beam True Time Delay RF Sensing Applications

Area of Science:

• Electrical Engineering
• Signal Processing
• Electromagnetics

Background:

• Direction of Arrival (DOA) estimation is crucial for radio frequency (RF) sensing, communications, and imaging.
• Existing wideband beamforming systems often require high filter orders, increasing circuit complexity.
• Thiran fractional delay filters offer a trade-off between usable bandwidth and circuit complexity.

Purpose of the Study:

• To introduce a novel algorithm for wideband true time delay digital delay Vandermonde matrix (DVM) factorization.
• To leverage Thiran fractional delays for realizing RF sensors with multiple DOA support.
• To reduce circuit complexity in multi-beam digital wideband beamforming systems.

Main Methods:

• Development of a digital DVM algorithm utilizing sparse matrix factorization.
• Implementation of Thiran fractional delays for approximating true time delays.
• Evaluation of Thiran filters with fractional sampling periods and orders 3 and 4 on Xilinx FPGA RF-SoC.

Main Results:

• The proposed algorithm enables multiple simultaneous RF beams, crucial for low-complexity sensing.
• Thiran filters demonstrate favorable phase and group delay responses, especially with a temporal oversampling factor of three.
• Preliminary FPGA results show potential for DVM factorization realization using Application-Specific Integrated Circuit (ASIC) technology.

Conclusions:

• The developed digital DVM algorithm with Thiran fractional delays offers a low-complexity solution for multi-beam RF sensing.
• Thiran filters provide an effective method for achieving wideband beamforming with reduced hardware requirements.
• The findings pave the way for efficient implementation of advanced RF sensing and DOA estimation systems.