细胞神经网络学习00

Cellular Neural Networks: Theory

LEON 0. CHUA, FELLOW, IEEE, AND LIN YANG, STUDENT MEMBER, IEEE

神经网络结构

 fig1：一个二维的神经细胞网络，电路尺寸4x4，这个正方形是被称为 细胞 的电路单元，细胞间的链接表明在相连细胞间有关联。

单个细胞定义“邻居”

 fig2：shows 3 neighborhoods of the same cell (located at the center and shown shaded) with r =1, 2 and 3, respectively. Usually, we call the r =1 neighborhood a “3 x 3 neighborhood,” the r = 2 neighborhood a “5 X 5 neighborhood,” and the r = 3 neighborhood a “7 X 7 neighborhood.”

        由定位的C(i,j)以及其半径r 定义细胞周围邻居细胞数量。如图r=1时，邻居数为8；r=2时，邻居数为24。

         the neighborhood system defined above exhibits a symmetry property in the sense that if C( i, j) N,( k, I), then C( k, I) Nr( i, j), for all C(i, j) and C(k, I) in a cellular neural network.      

单个细胞结构

 fig3：An example of a cell circuit. Vu = input, Vx = state, Vy = output.

 

        Observe from Fig. 3 that each cell C(i, j) contains one independent voltage source Eij, one independent current source1,  one linear capacitor C, two linear resistors Rx, and Ry, and at most 2m linear voltage-controlled current sources whch are coupled to its neighbor cells via the controlling input voltage Vukl, and the feedback from the output voltage Uykl, of each neighbor cell C(k, I), where m is equal to the number of neighbor cells.

 

.Without loss of generality,the cell circuit architecture in Fig3 will be used throughout this paper