Optimization of Sparse Cross Array Synthesis via Perturbed Convex Optimization.

Area of Science:

• Acoustics
• Signal Processing
• Array Signal Processing

Background:

• Three-dimensional (3-D) imaging sonar systems typically require large planar arrays, leading to high hardware costs.
• Cross arrays offer a sensor-efficient alternative for 3-D imaging by using two perpendicular linear arrays.
• Aperiodic sparse arrays can further reduce sensor requirements.

Purpose of the Study:

• To propose an optimized method for sparse cross array synthesis.
• To simplify beamforming for cross arrays in both near-field and far-field conditions.
• To minimize the number of active sensors while optimizing the beam pattern.

Main Methods:

• Simplified multi-frequency beamforming for cross arrays.
• A perturbed convex optimization algorithm for sparse cross array synthesis.
• Utilizing first-order Taylor expansion for position perturbations to optimize beam patterns and sensor count.

Main Results:

• Successfully synthesized a sparse cross array with 45 + 45 sensors from an initial 100 + 100 sensor array.
• Demonstrated superior effectiveness compared to existing methods for sparse cross array synthesis.
• Achieved optimal results in both near-field and far-field scenarios.

Conclusions:

• The proposed method provides an effective approach for sparse cross array synthesis.
• This technique significantly reduces sensor count for 3-D imaging sonar systems.
• The method offers improved performance and efficiency for sonar applications.