Multiuser Joint Detection

Multiuser detection could achieve a significant capacity improvement over conventional detection. In our previous work, we have showed that the breadth first reduced complexity multiuser detector can achieve near optimum performance at a very low complexity when the receiver knows all the channel parameters and correlation matrix.

Selected Papers:

• L. Wei, P. Alexander, C. B. Schlegel and L. K. Rasmussen, " A Detector System for a Telecommunications System", Australian Provisional Patent, Nov. 1994.
• L. Wei, L. K. Rasmussen and R. Wyrwas, "Near Optimum Tree-Search Detection Schemes for Bit-Synchronous Multiuser CDMA Systems Over Gaussian and Two-Path Rayleigh Fading Channels", IEEE Trans. on Communications, Vol.COM-45, June 1997, pp.691-701.
• L. Wei and L. K. Rasmussen "A Near Ideal Whitening Filter for an Asynchronous time-variant CDMA System", IEEE Trans. on Communications, Vol.COM-44, Oct. 1996, pp.1355-1362.
• L.Wei and C.Schlegel, "Synchronous DS-SSMA System with Improved Decorrelating Decision-Feedback Multiuser Detection"' IEEE Trans. on Vehicular Technology, (special issue on PCN), Aug., 1994, pp.767-772.

Here is a glimpse of the bit error performance of near optimal detectors as a function of the number of users (K) for bit energy to noise power ratio (Eb/No) of 7 dB, where binary random spreading codes are used

The above curve is from the paper (L. Wei, L. K. Rasmussen and R. Wyrwas, "Near Optimum Tree-Search Detection Schemes for Bit-Synchronous Multiuser CDMA Systems Over Gaussian and Two-Path Rayleigh Fading Channels", IEEE Trans. on Communications, Vol.COM-45, June 1997, pp.691-701.)

Near Optimal Low Complexity Multiuser Detection

This research concerns the understanding of near optimal low complexity multiuser detection. Current research has been focused on how to prove theoretically that the breadth first detector is asymptotically optimal (i.e., P(e) for our detector is equal to P(e) of the optimal detector when the bit energy to noise power ratio tends to infinity). Work done by Forney, Anderson and etc, provides an essential foundation for this research.

Selected Papers:

• L. Wei and H.QI, "Near Optimal Limited Search Decoding: part 1. Finite Interference Channels", (submitted).
• H. Jones and L.Wei, "Computation of Minimum Complexity for Near Optimal Synchronous CDMA Multiuser Detection", Proceeding of IEEE Globecom, 1998, Sydney, accepted.
• L. Wei,H.Qi, "Performance Upper Bounds for Breadth First Search Detection over Finite Interference Channels", Proceeding of IEEE International Symposium on Information Theory (ISIT'97), Ulm, Germany, June 29-July 4, pp.495.
• L. Wei, "An upper bound for correct path loss probability in the M-algorithm over ISI channels", Proceeding of IEEE Information Theory Workshop (ISW'97), Longyearbyen, Norway, July 6-12, pp.31-32.

Performance Analysis Techniques for Multiuser Systems

This research concerns how to determine the performance of optimal multiuser detection for coded and uncoded CDMA systems. For uncoded CDMA systems we have a simple way to compute the bounds on the BER of the optimal decoder. For coded CDMA systems, we know much more about the key performance indicator-- the minimum distance of coded systems.

Selected Papers:

• L. Wei, "A Simple Way to Compute Minimum Euclidean Distance for Synchronous Coded Multiuser Systems",IEEE Communications Letters, (accepted).
• L. Wei and R. Jana, "Properties of Minimum Euclidean Distance for Trellis Coded Multiuser systems", (submitted)
• L. Wei and R. Jana, "Performance Bounds for Optimum Multiuser DS-CDMA Systems", IEEE Trans. on Communications, accepted.
• C. Schlegel and L. Wei, "A Simple Way to Compute the Minimum Distance in Multiuser CDMA Systems", IEEE Trans. on Communications, Vol.COM-45, May 1997, pp.532-536.

Adaptive & Robust Near Optimal Low Complexity Multiuser Detection.

How to implement the near optimal multiuser detector is the key focusing point of this project. We are focusing on using adaptive algorithms to directly estimate the parameters of the whitening matched filter. The whitening matched filter was proposed by Forney in 1974. How to directly estimate its parameters is still an open problem. If it can be done, then the receiver complexity will be dramatically reduced. Also we are applying the concept of "robust" (or min-max) to reduce the effect of estimation errors on the system performance.

Selected Papers:

• Z. Li and L.Wei, "Breadth First Algorithm with Adaptive Forney Structure for ISI Channels", Proceeding of IEEE Globecom, 1998, Sydney.
• R. Jana and L. Wei, "Robust Multiuser CDMA Receivers Using Sequential Decoding", proceedings of IEEE International Symposium on Signal Processing and its application (ISSPA), Gold Coast, Australia, Aug. 1996, pp.176-179.

Much of research attention has also been focused on adaptive and robust iterative multiuser detection.

Top