LMI approach to the global robust stability of a larger class of neural networks with delay

Sufficient conditions in the form of linear matrix inequality for the uniqueness and global asymptotic stability of the equilibrium point of a large class of uncertain neural networks with delay are presented. The conditions are based on norm-bounded uncertainties. An example is given to show the effectiveness of the obtained results. A comparison is made between the present approach and an earlier approach due to Lu, Rong and Chen. An error is corrected in an earlier publication.