Modified Hopfield Neural Network for CDMA Multiuser Detection
We investigate the application of modified Hopfield neural networks (HNNs) based on Annealing Techniques to the problem of multiuser detection in spread spectrum CDMA communication systems. It is shown that the NP-complete problem of minimizing the objective function of the optimal multiuser detector (OMD) can be translated into minimizing an HNN “energy” function, thus allowed to take advantage of the ability of HNNs to perform very fast gradient descent algorithms in analog hardware. The performance of the proposed HNN receiver is evaluated via computer simulations and compared to that of the general HNN as well as to that of the OMD for CDMA transmission cases. It is shown that the modified HNN detection scheme exhibits a number of attractive properties and that it provides in fact more powerful performance than the general HNN scheme or the OMD scheme.
KeywordsCode Division Multiple Access Successive Interference Cancellation Hopfield Neural Network Code Division Multiple Access System Hopfield Network
Unable to display preview. Download preview PDF.
- 1.Verdu, S.: Multiuser Detection. Cambridge University Press, Cambridge (1984)Google Scholar
- 2.Yoon, S.H., Rao, S.: Multiuser detection in CDMA Based on the Annealed Neural Network. IEEE Int. Conf. Neural Networks 4, 2124–2129 (1996)Google Scholar
- 3.Chen, D.C., Sheu, B.J.: A Compact Neural network Based CDMA Receiver for Multimedia Wireless Communication. IEEE Conf. Comp. Design, 99–103 (1996)Google Scholar
- 4.Kechriotis, G.I., Manolakos, E.S.: Implementing the Optimal CDMA Multiuser Detector with Hopfield Neural network. IEEE Workshop on Appl. Neural Networks and telecomm., 60–66 (1993)Google Scholar
- 7.Jeney, G., Levendovezky, J., Imre, S., Pap, L.: Comparison of Different Neural Network Based Multiuser Detector. In: EUNICE 2000, Enschede, pp. 117–123 (2000)Google Scholar