Sparse nonlinear optimization for signal processing and communications

<p>"This dissertation proposes three classes of new sparse nonlinear optimization algorithms for network echo cancellation (NEC), 3-D synthetic aperture radar (SAR) image reconstruction, and adaptive turbo equalization in multiple-input multiple-output (MIMO) underwater acoustic (UWA) communications, respectively.</p> <p>For NEC, the proposed two proportionate affine projection sign algorithms (APSAs) utilize the sparse nature of the network impulse response (NIR). Benefiting from the characteristics of l₁-norm optimization, affine projection, and proportionate matrix, the new algorithms are more robust to impulsive interferences and colored input than the conventional adaptive algorithms.</p> <p>For 3-D SAR image reconstruction, the proposed two compressed sensing (CS) approaches exploit the sparse nature of the SAR holographic image. Combining CS with the range migration algorithms (RMAs), these approaches can decrease the load of data acquisition while recovering satisfactory 3-D SAR image through l₁-norm optimization.</p> <p>For MIMO UWA communications, a robust iterative channel estimation based minimum mean-square-error (MMSE) turbo equalizer is proposed for large MIMO detection. The MIMO channel estimation is performed jointly with the MMSE equalizer and the maximum a posteriori probability (MAP) decoder. The proposed MIMO detection scheme has been tested by experimental data and proved to be robust against tough MIMO channels."--Abstract, page iv.</p>

Name thesis:degree_name

Ph. D. in Electrical Engineering

Grantor

Missouri University of Science and Technology

Year dc:date.available

2016

Author dc:creator

• Yang, Zengli

• 3-D radar imaging
• Adaptive turbo equalization
• Compressed sensing (CS)
• Sparse nonlinear optimization
• Underwater acoustic (UWA) communications
• Electrical and Computer Engineering

Repository record dc:identifier

https://scholarsmine.mst.edu/doctoral_dissertations/2267

OAI identifier oai:identifier

oai:scholarsmine.mst.edu:doctoral_dissertations-3269