Dimensional Signal

Dimensional signal processing focuses on efficiently handling high-dimensional data by leveraging its underlying low-dimensional structure. Current research emphasizes developing methods to acquire and reconstruct these signals from limited, often noisy or quantized, measurements, employing techniques like compressive sensing, generative models (including diffusion and score-based models), and reinforcement learning to optimize measurement strategies and reconstruction algorithms. This work is crucial for advancing various fields, including medical imaging, robotic manipulation, and signal processing in resource-constrained environments, by enabling the acquisition and analysis of complex data with reduced computational cost and improved accuracy.