Compressed Sensing and Parallel Acquisition

Due to several practical limitations, a single sensor often does not provide enough measurements for a source signal to be recovered. So-called parallel acquisition systems can resolve this problem by allowing 1) simultaneous data acquisition in multiple sensors and 2) provision of more measurements. Parallel acquisition techniques have been applied to solve various practical issues including
measurement acquisition time reduction (e.g., parallel magnetic resonance imaging, pMRI), power consumption reduction in sensors (e.g., wireless sensor network), a recovery of higher-resolution or higher-dimensional signal (e.g., multi-view imaging or light eld imaging), etc. At the same time, an emerging random sub-sampling theory, compressed sensing (CS), has a great potential to maximize the sampling performance in parallel acquisition systems. This talk introduces our recent theoretical findings regarding the gain (in terms of the number of measurements required) powered by increasing the number of sensors for CS in parallel acquisition systems, which depends on sensor pro ⋀ le (i.e., environmental/geometric condition from a single source to each sensor) and
sampling strategies. Furthermore, the specific application of our results in pMRI motivates a new signal recovery framework (CS SENSitivity Encoding (SENSE) pMRI reconstruction promoting joint sparsity (JS), JS CS SENSE) and provides a number of suggestions to coil sensitivity pro ⋀ le design.
Il Yong (Bobby) Chun is currently a post-doc at Purdue University Department of Mathematics under the supervision of Professor Ben Adcock. He received his PhD at Purdue University in West Lafayette, IN in August 2015, and his Bachelor of Engineering at Korea University in Seoul, South Korea in February 2009. His research is in compressed sensing in medical imaging, model-based computational medical imaging, algorithm acceleration in image processing, and image analysis for medical imaging and has been published in IEEE Transactions on Computational Imaging and the Journal of Developmental Neuropsychology in addition to the numerous papers he has presented.