Welcome Prof. Yunghsiang S. Han, ​from Dongguan University of Technology, Taiwan, China to be Keynote Speaker!

1573546952793237.jpg

Prof. Yunghsiang S. Han, IEEE Fellow, School of Electrical Engineering & Intelligentization, 

Dongguan University of Technology, Taiwan, China (click)


Research Area: error-control coding, wireless networks, and security


Title: Codes over Rings for Distributed Storage Systems

Abstract: Array codes have been widely used in distributed storage systems such as RAID (Redundant Array of Independent Disks). Binary maximum distance separable (MDS) array codes are constructed by encoding k information columns into r parity columns, in which each element in a column is a bit, such that any k out of the k+r columns suffice to recover all information bits. Binary MDS array codes have low computational complexity since the encoding and decoding procedures only involve XOR operations. In addition to the computational complexity, it is critical to improve repair performance in big data storage applications. Specifically, if a single column fails, the goal is to minimize the repair bandwidth by downloading the least amount of bits from d healthy columns, where k<= d<= k+r-1. If one column of an MDS code is failed, it is known that we need to download at least 1/(d-k+1) fraction of the data stored in each of d healthy columns. In this talk, we present a generic binary cyclic ring for designing binary MDS array codes.  We show that EVENODD codes and RDP codes can be viewed as array codes over a special binary cyclic ring with encoding matrix being a Vandermonde matrix, and propose a fast decoding method for EVENODD codes and RDP codes based on an LU factorization of Vandermonde matrix. Moreover, we also present a new construction of Rabin-like codes over the special binary cyclic ring with encoding matrix being a Cauchy matrix and give a fast decoding method for the proposed codes. More importantly, we propose two explicit constructions of new array codes over the generic binary cyclic ring that have optimal repair bandwidth.