Chaotic analog associative memory

摘要

In this paper, we propose a chaotic analog associative memory (CAAM). It can deal with associations of analog patterns including common patterns. Although a lot of associative memories have been proposed, most of them can not deal with analog patterns. Moreover, they can not deal with one-to-many associations because the stored common patterns cause superimposed patterns. In contrast, the proposed CAAM can realize associations of analog patterns because it is based on a multi-winners self-organizing neural network (MWSONN) which can deal with analog patterns. The proposed model can also realize one-to-many associations by using chaotic neurons. We carried out a series of computer simulations and confirmed the effectiveness of the proposed model. The proposed model has the following features: (1) it can deal with associations of analog patterns; (2) it can deal with one-to-many associations; (8) it has robustness for noisy input and neuron damage.