Multidimensional system

Multidimensional systems or m-D systems are the necessary mathematical background for modern digital image processing with many applications in biomedicine, X-ray technology and satellite communications. There are also some studies combining m-D systems with partial differential equations (PDEs). == Linear multidimensional state-space model ==

A state-space model is a representation of a system in which the effect of all "prior" input values is contained by a state vector. In the case of an m-d system, each dimension has a state vector that contains the effect of prior inputs relative to that dimension. The collection of all such dimensional state vectors at a point constitutes the total state vector at the point. Consider a uniform discrete space linear two-dimensional (2d) system that is space invariant and causal. It can be represented in matrix-vector form as follows: Represent the input vector at each point (i,j) by u(i,j), the output vector by y(i,j) the horizontal state vector by R(i,j) and the vertical state vector by S(i,j). Then the operation at each point is defined by: : \begin{align} R(i+1,j) & = A_1R(i,j) + A_2S(i,j) + B_1u(i,j) \\ S(i,j+1) & = A_3R(i,j) + A_4S(i,j) + B_2u(i,j) \\ y(i,j) & = C_1R(i,j) +C_2S(i,j) + Du(i,j) \end{align} where A_1, A_2, A_3, A_4, B_1, B_2, C_1, C_2 and D are matrices of appropriate dimensions. These equations can be written more compactly by combining the matrices: : \begin{bmatrix} R(i+1,j) \\ S(i,j+1) \\ y(i,j) \end{bmatrix} = \begin{bmatrix} A_1 & A_2 & B_1 \\ A_3 & A_4 & B_2 \\ C_1 & C_2 & D \end{bmatrix} \begin{bmatrix} R(i,j) \\ S(i,j) \\ u(i,j) \end{bmatrix} Given input vectors u(i,j) at each point and initial state values, the value of each output vector can be computed by recursively performing the operation above. == Multidimensional transfer function ==

A discrete linear two-dimensional system is often described by a partial difference equation in the form: \sum_{p,q=0,0}^{m,n}a_{p,q}y(i-p,j-q) = \sum_{p,q=0,0}^{m,n}b_{p,q}x(i-p,j-q) where x(i,j) is the input and y(i,j) is the output at point (i,j) and a_{p,q} and b_{p,q} are constant coefficients. To derive a transfer function for the system the 2d Z-transform is applied to both sides of the equation above. : \sum_{p,q=0,0}^{m,n} a_{p,q}z_1^{-p}z_2^{-q}Y(z_1,z_2) = \sum_{p,q=0,0}^{m,n}b_{p,q}z_1^{-p}z_2^{-q}X(z_1,z_2) Transposing yields the transfer function T(z_1,z_2): : T(z_1,z_2) = {Y(z_1,z_2) \over X(z_1,z_2)} = {\sum_{p,q=0,0}^{m,n}b_{p,q}z_1^{-p}z_2^{-q} \over \sum_{p,q=0,0}^{m,n}a_{p,q}z_1^{-p}z_2^{-q}} So given any pattern of input values, the 2d Z-transform of the pattern is computed and then multiplied by the transfer function T(z_1,z_2) to produce the Z-transform of the system output. == Realization of a 2d transfer function ==

Often an image processing or other md computational task is described by a transfer function that has certain filtering properties, but it is desired to convert it to state-space form for more direct computation. Such conversion is referred to as realization of the transfer function. Consider a 2d linear spatially invariant causal system having an input-output relationship described by: : Y(z_1,z_2) = {\sum_{p,q=0,0}^{m,n}b_{p,q}z_1^{-p}z_2^{-q} \over \sum_{p,q=0,0}^{m,n}a_{p,q}z_1^{-p}z_2^{-q}}X(z_1,z_2) Two cases are individually considered 1) the bottom summation is simply the constant 1 2) the top summation is simply a constant k. Case 1 is often called the "all-zero" or "finite impulse response" case, whereas case 2 is called the "all-pole" or "infinite impulse response" case. The general situation can be implemented as a cascade of the two individual cases. The solution for case 1 is considerably simpler than case 2 and is shown below. Example: all zero or finite impulse response : Y(z_1,z_2) = \sum_{p,q=0,0}^{m,n}b_{p,q}z_1^{-p}z_2^{-q}X(z_1,z_2) The state-space vectors will have the following dimensions: : R (1 \times m),\quad S (1 \times n),\quad x (1 \times 1) and y (1 \times 1) Each term in the summation involves a negative (or zero) power of z_1 and of z_2 which correspond to a delay (or shift) along the respective dimension of the input x(i,j). This delay can be effected by placing 1’s along the super diagonal in the A_1. and A_4 matrices and the multiplying coefficients b_{i,j} in the proper positions in the A_2. The value b_{0,0} is placed in the upper position of the B_1 matrix, which will multiply the input x(i,j) and add it to the first component of the R_{i,j} vector. Also, a value of b_{0,0} is placed in the D matrix which will multiply the input x(i,j) and add it to the output y. The matrices then appear as follows: : A_1 = \begin{bmatrix}0 & 0 & 0 & \cdots & 0 & 0 \\ 1 & 0 & 0 & \cdots & 0 & 0 \\ 0 & 1 & 0 & \cdots & 0 & 0 \\ \vdots & \vdots & \vdots & \ddots & \vdots & \vdots \\ 0 & 0 & 0 & \cdots & 0 & 0 \\ 0 & 0 & 0 & \cdots & 1 & 0 \end{bmatrix} : A_2 = \begin{bmatrix}0 & 0 & 0 & \cdots & 0 & 0 \\ 0 & 0 & 0 & \cdots & 0 & 0 \\ 0 & 0 & 0 & \cdots & 0 & 0 \\ \vdots & \vdots & \vdots & \ddots & \vdots & \vdots \\ 0 & 0 & 0 & \cdots & 0 & 0 \\ 0 & 0 & 0 & \cdots & 0 & 0 \end{bmatrix} : A_3 = \begin{bmatrix} b_{1,n} & b_{2,n} & b_{3,n} & \cdots & b_{m-1,n} & b_{m,n} \\ b_{1,n-1} & b_{2,n-1} & b_{3,n-1} & \cdots & b_{m-1, n-1} & b_{m,n-1} \\ b_{1,n-2} & b_{2,n-2} & b_{3,n-2} & \cdots & b_{m-1, n-2} & b_{m,n-2} \\ \vdots & \vdots & \vdots & \ddots & \vdots & \vdots \\ b_{1,2} & b_{2,2} & b_{3,2} & \cdots & b_{m-1,2} & b_{m,2} \\ b_{1,1} & b_{2,1} & b_{3,1} & \cdots & b_{m-1,1} & b_{m,1} \end{bmatrix} A_4 = \begin{bmatrix}0 & 0 & 0 & \cdots & 0 & 0 \\ 1 & 0 & 0 & \cdots & 0 & 0 \\ 0 & 1 & 0 & \cdots & 0 & 0 \\ \vdots & \vdots & \vdots & \ddots & \vdots & \vdots \\ 0 & 0 & 0 & \cdots & 0 & 0 \\ 0 & 0 & 0 & \cdots & 1 & 0 \end{bmatrix} : B_1 = \begin{bmatrix}1 \\ 0 \\ 0\\ 0\\ \vdots \\ 0 \\ 0 \end{bmatrix} : B_2 = \begin{bmatrix} b_{0,n} \\ b_{0,n-1} \\ b_{0,n-2} \\ \vdots \\ b_{0,2} \\ b_{0,1} \end{bmatrix} : C_1 = \begin{bmatrix} b_{1,0} & b_{2,0} & b_{3,0} & \cdots & b_{m-1,0} & b_{m,0} \\ \end{bmatrix} : C_2 = \begin{bmatrix}0 & 0 & 0 & \cdots & 0 & 1 \\ \end{bmatrix} : D = \begin{bmatrix}b_{0,0} \end{bmatrix} == References ==