F.N. Valvi,V.S. Geroyannis
Abstract:We present explicit inverses of two Brownian--type matrices, which are defined as Hadamard products of certain already known matrices. The matrices under consideration are defined by $3n-1$ parameters and their lower Hessenberg form inverses are expressed analytically in terms of these parameters. Such matrices are useful in the theory of digital signal processing and in testing matrix inversion algorithms.
What problem does this paper attempt to address?
The problem that this paper attempts to solve is to provide explicit inverse matrix expressions for two types of Brownian - type matrices. Specifically, the author studied two matrices \(A_1\) and \(A_2\) with specific structures, which are defined by the Hadamard product of known matrices and have \(3n - 1\) parameters. These matrices are of great significance in digital signal processing and testing matrix inversion algorithms.
### Specific Problem Description
1. **Matrix Definition**:
- Matrices \(A_1\) and \(A_2\) are defined as follows respectively:
\[
A_1=K\circ G_n
\]
\[
A_2 = N\circ G_n
\]
Here, the symbol \(\circ\) represents the Hadamard product (element - wise multiplication), and \(K\), \(N\) and \(G_n\) are known special matrices.
2. **Matrix Structure**:
- \(K\) and \(N\) are symmetric matrices, and their elements are:
\[
\kappa_{ij}=k_i,\quad i\leq j
\]
\[
\nu_{ij}=k_j,\quad i\leq j
\]
- \(G_n\) is a lower triangular matrix, and its elements are:
\[
\beta_{ij}=\begin{cases}b_j, & i\leq j\\a_j, & i > j\end{cases}
\]
3. **Objective**:
- The main objective of the paper is to derive the explicit inverse matrices of \(A_1\) and \(A_2\) and give the specific forms of these inverse matrices.
- At the same time, study the numerical complexity of these inverse matrices to evaluate the computational efficiency.
### Solution
To achieve the above objectives, the author carried out the following work:
1. **Derivation of Inverse Matrix Formulas**:
- Through a series of row transformation operations, \(A_1\) and \(A_2\) are transformed into simpler matrix forms (such as diagonal matrices or identity matrices), thereby deriving the analytical expressions of the inverse matrices.
- For \(A_1\), its inverse matrix \(A_1^{-1}\) is a lower Hessenberg matrix, and its elements are determined by \(3n - 1\) parameters.
- For \(A_2\), its inverse matrix \(A_2^{-1}\) is also a lower Hessenberg matrix, but the specific element expressions are different.
2. **Numerical Complexity Analysis**:
- The computational complexity of solving \(A_1^{-1}\) and \(A_2^{-1}\) is calculated. The results show that the calculation can be completed within \(O(n^2)\) multiplications/divisions and \(O(n)\) additions/subtractions using a recursive algorithm.
- Through numerical experiments, it is verified that the recursive algorithm is much faster than the LU decomposition method, especially when the matrix order is large (for example, \(n = 256\) to \(n=1024\)).
### Application Background
This type of matrix has important applications in the field of digital signal processing, especially for non - stationary signal modeling. In addition, these matrices are also widely used to test the performance and accuracy of matrix inversion algorithms.
In summary, this paper solves the problem of how to efficiently solve the inverse matrices of two types of matrices with special structures, and provides detailed theoretical derivations and numerical experiment results, providing a valuable reference for research and applications in related fields.