Yi Jiang, Professor at Fudan University, Shanghai, China, passed away on October 2024, at the age of 45.
Yi was born in Yixing, Jiangsu, China, in 1978. He attended the University of Science and Technology of China in 1996 to study Electronic Engineering and Information Science. After obtaining his Bachelor degree, he continued his study at the University of Florida, and obtained his M.S. and Ph.D. degrees under the supervision of Professor Jian Li. He then received his postdoc training at the University of Colorado Boulder, under the mentorship of Professor Mahesh Varanasi. After working in industry at NextWave, Qualcomm, IAA and Silvus for a few years in California, he returned to his home country and joined the faculty of Fudan University.
Yi has made a number of impactful contributions in the areas of signal processing, wireless communications, and information theory. He invented the geometric mean decomposition (GMD) and the generalized triangular decomposition (GTD) of complex matrices. For matrix computation, GTD can be applied to solve inverse eigenvalue problems efficiently, and hence has many applications not only in wireless communications, but also in control theory, principal component analysis, particle physics, circuit theory, and other areas. In multiple-input-multiple-output (MIMO) wireless communications, GMD-based methods have led to influential MIMO transceiver design that provides an alternative approach besides the standard singular value decomposition (SVD), to decompose, in a strictly capacity lossless sense, a MIMO channel into multiple subchannels with identical capacities. In another influential work, contrary to a common misperception that ZF and MMSE equalizers for MIMO systems are asymptotically equivalent at high SNR, Yi conducted an in-depth analysis to reveal that the output SNRs of these two equalizers exhibit a gap, which converges with probability one to an F-distributed random variable as input SNR goes to infinity. Based on this result, he further provided the solution to a long-standing open problem of the diversity-multiplexing tradeoff of V-BLAST with ordered decoding.
Professor Yi Jiang was greatly admired and loved, and will be sorely missed by his family, friends, colleagues, and students.
