Please Note:

The Open Access Repository has moved to a new authentication system as of the 1st of November.

Account holders will now be able to login using their University of Tasmania credentials.
If you have trouble logging in please email us on so we can assist you.

Public users can still access the records in this repository as normal

Open Access Repository

Complex-value recurrent neural networks for global optimization of beamforming in multi-symbol MIMO communication systems


Downloads per month over past year

Jiang, D (2007) Complex-value recurrent neural networks for global optimization of beamforming in multi-symbol MIMO communication systems. In: International Conference on Signal processing and Communication Systems, 17-19 Dec. 2007, Gold Coast, Australia.

4803.pdf | Download (171kB)
Available under University of Tasmania Standard License.


Multiple antennas at transmitter and receiver
can be used to improve communication efficiency by
canceling channel noises using the correlated information
among the signals transmitted from different antennas. In
this paper, a novel approach is proposed for this problem
for another interesting case where multiple symbols are
used to make the best use of the multiple antenna channel.
Such an issue cannot be converted into a convex optimization
problem. It can be considered as a generalization
of the vector optimization problem on Grassmannian
manifold to that one a complex Stiefel manifold, which has
not been well considered yet. The proposed algorithm is
based on the gradient search on a complex Stiefel manifold
of a non-convex problem to maximize the system signal
to noise ratio. With appropriately defined Riemannian
metric on this manifold, a neat formula has been developed
for the gradient function. It is proved that the proposed
algorithm converges to the global optimum. This algorithm
can also be implemented into recurrent neural network
to facilitate real-time computation. Its parallel structure
can be realized using analog circuits. Furthermore, a
modified gradient flow defined on the non-compact Stiefel
manifold is also developed, which is robust against any
initial condition error. The corresponding recurrent neural
network is also discussed. Simulation experiments are
included to demonstrate the advantages of the proposed

Item Type: Conference or Workshop Item (Paper)
Page Range: 1-Aug
Additional Information:

© 2008 DSP for Communication Systems,

Date Deposited: 07 Apr 2008 14:58
Last Modified: 18 Nov 2014 03:36
Item Statistics: View statistics for this item

Actions (login required)

Item Control Page Item Control Page