THE NORTH JERSEY CHAPTER OF THE IEEE COMMUNICATIONS SOCIETY
and
THE CENTER FOR COMMUNICATIONS AND SIGNAL PROCESSING RESEARCH
of
NEW JERSEY INSTITUTE OF TECHNOLOGY
Present
Multi-Shot Approaches to
Multiuser Separation and Detection in Asynchronous CDMA Systems
by
HONGYA GE
NEW JERSEY INSTITUTE OF TECHNOLOGY
WHEN: Tuesday, October 10, 1995, 5:45 p.m. (refreshments at 5:30)
WHERE: Room 202 of the Electrical and Computer Engineering Center, New Jersey
Institute of Tech.
ABSTRACT
Code Division Multiple Access (CDMA) is an accepted scheme for future high-
capacity digital wireless/cellular communications. The need for a large
number of users dictates the implementation of a semi-orthogonal set of
signature waveforms, such as Gold or Kasami codes. The cross-correlation
between the users' codes will cause multiuser interference (MI). The detection
performance of a conventional digital receiver that ignores the existence of
other users degrades quickly with increasing interference. Further, the
existence of many strong users will heavily interfere with the desired weak
user. This is referred to as the near-far problem in CDMA systems. Hence a
multiuser detector is necessary for multiple-access communication systems. In
this work, we propose a new computationally simple scheme (termed the multi-
shot approach) to separate and detect multiuser signals in asynchronous CDMA
communication systems. First we explore the structure inherent in the matrix
decomposition of the properly arranged data, obtained from multi-shot matched
filtering. Then we make a matrix approximation to make the tentative multiuser
information bit estimation trackable and simple. Further matrix filtering and
de-biasing, based on the inherent structure in the data model, are suggested
to improve the overall detection performance. Simulation results support the
newly proposed detection scheme.
BIOGRAPHY
Hongya Ge received the B.S. degree from the University of Electronic Science
and Technology of China (UESTC) in 1982, the M.S. degree from the Nanjing
University of Aeronautics and Astronautics in 1985, and the Ph.D. degree from
the University of Rhode Island in 1994--all in electrical engineering. From
1986 to 1990 she was with the Dept. of Information and Electronics, Zhejiang
University, as a lecturer and researcher in numerical analysis, communication
theory, and radar signal processing. She has been with the Department of
Electrical and Computer Engineering, NJIT, since February 1995. Her research
interests are in the areas of statistical/digital signal processing (detection
and estimation), communication theory, numerical analysis, and approximation
theory.
For more information, contact Professor N. Ansari, (201) 596-3670,
e-mail: ang@faraday.njit.edu