Digital Signal Processing Reference
In-Depth Information
Table 2.
Number of Iterations needed by NNTCTG-MUD and WCSANN-MUD
Average Iterations
10
−
)
3
BER(
EN
/
NNTCTG
WCSANN
b
o
6 dB
14.997
21.663
5.9
7 dB
11.5
19.6
2.1
8 dB
10.1
18.8
0.71
9 dB
9.9
16.9
0.25
10 dB
9.7
16.0
0.17
11 dB
9.2
15.5
0.06
Although we have seen that WCSANN-based MUD detector outperforms the
NNTCTG-based MUD detector, but it is still a question how much additional iteration
is needed in exchange for better performance. It is worth comparing this in
Table 2
,
where the number of iterations need to reach the steady state by NNTCTG and
WCSANN are shown from
EN
/
=
6
to 11dB. It is very clear that WCSANN
needs 7~8 additional iterations to exchange for her better performance.
b
o
5
Conclusions
In this paper we have proposed a novel multi-user detection scheme, which makes use
of WCSANN optimization algorithm in MUD. The new method resulted in better
performance than NNTCTG detector algorithms, namely 1..2dB gain in performance
can be achieved, only 7..8 additional iteration is needed. In the future, we would like
to develop a more sophisticated, adaptive version of the algorithm.
Acknowledgments.
This paper is supported by Natural Science Fund of Anhui Prov-
ince of China (050420101).
References
1.
Verdu, S.: Minimum probability of error for asynchronous Gaussian multiple-access chan-
nels. IEEE Trans. Inform.Theory IT-32, 85-96 (1986)
2.
Varanasi, M.K., Azhang, B.: Multistage Detection in Asynchronous Code-Division Multiple
Access Communications. IEEE Trans. on Comm. 38, 509-519 (1990)
3.
Yoon, S.H., Rao, S.: Multiuser detection in CDMA based on the annealed neural network.
In: IEEE Int. Conf. Neural Networks, vol. 4, pp. 2124-2129 (1996)
4.
Kechriotis, G., Manolakos, E.S.: A Hybrid Digital Signal Processing-Neural Network
CDMA Mulituser Detector Scheme. IEEE Trans. Cicuits and Systems 43(2), 96-104 (1996)
5.
Wang, B., He, Z., Nie, J.: To Implement the CDMA Multiuser Detector by Using Transient-
ly Chaotic Neural Network. IEEE Trans. Aerospace and Elec. Sys. 33, 1068-1071 (1997)
6.
Tan, Y., Deng, C., Wang, B., He, Z.: A Neural Network with Transient Chaos and Time-
Varying Gain and its Application to Optimization Calculation. Acta Electronica Sini-
ca 26(7), 123-127 (2005)
