Research Fellow  |  Lu, Chun-Shien  
Research Descriptions

My current research interests mainly focus on (1) Security and Privacy in Multimedia and Sensor Network and (2) Compressed Sensing (algorithm design and applications). Security and Privacy are long-studied problems but we focus on relevant and common issues, including content authentication, integrity detection, and privacy-preserving, in Multimedia and Sensor Network, in addition to our previous researches on multimedia data hiding/watermarking and hashing. As for sensor network security, we have developed a non-interactive key pre-distribution mechanism, a message authentication method, and a privacy-preserving query scheme. We are currently trying to build a secure sensor network system composed of several security components. Recently, with an eye on the fact that compressed sensing (CS) is a revolutionary technology of simultaneously sensing and compressing signals, and builds a new sampling theorem beyond the Nyquist rate, we study the fundamental issues, including dictionary learning for sparsifying signals, and more accurate and fast CS recovery, in compressed sensing of signals and images. We have developed a distributed compressive video sensing (DCVS) method to simultaneously sensing and compressing videos. We also have presented a compressed image sensing (CIS) method for turbo fast recovery of images from (far) fewer measurements. We are currently studying a fast Orthogonal Matching Pursuit (FOMP) algorithm by reformulating OMP in terms of refining L2-norm solutions in a greedy manner. Our FOMP method not only provides theoretic guarantee of recovery based on Mutual Incoherence Property (MIP) but also provides a more practical exact recovery analysis via order statistics. We plan to investigate CS-based applications, including sparse representation and compressive sensing problems related to multimedia, node replica detection in sensor network, space shift keying in MIMO, etc., based on exploiting the sparsity of signals we would like to explore.

*For those that can read Chinese, please also see the Chinese version.