Parameter Estimation of Multiple Frequency-Hopping Signals with Two Sensors

Area of Science:

• Signal Processing
• Array Signal Processing
• Electromagnetics

Background:

• Estimating direction of arrival (DOA) and sorting signals from multiple sources is crucial in various applications.
• Existing methods often require complex or expensive sensor arrays.
• Wideband sources with time-varying frequencies present unique challenges.

Purpose of the Study:

• To develop a parameter estimation method for multiple wideband sources with time-varying frequencies.
• To achieve accurate two-dimensional (2-D) DOA estimation and signal sorting.
• To utilize a low-cost circular synthetic array (CSA) with only two rotating sensors.

Main Methods:

• Decomposition of received data into separated groups for individual source DOA estimation.
• Application of the expectation maximization (EM) algorithm, comprising expectation-step (E-step) and maximization (M-step).
• Development of closed-form maximum-likelihood (ML) estimation formulae for DOA based on phase data and complex responses.

Main Results:

• Joint determination of DOAs and sorting of multiple signals through iterative E-step and M-step execution.
• Optimal DOA estimation achieved using ML estimation based on phase data.
• Directional ambiguity successfully addressed using ML estimation on received complex responses.

Conclusions:

• The proposed EM-based method effectively estimates parameters for multiple wideband sources using a cost-effective CSA.
• The derived closed-form ML formulae provide optimal and accurate DOA estimation.
• Simulations confirm the method's performance and its ability to handle directional ambiguity.