Signal Processing Blockset™

Matrix Square - Compute square of input matrix

Library

Math Functions / Matrices and Linear Algebra / Matrix Operations

dspmtrx3

Description

The Matrix Square block computes the square of an M-by-N input matrix, u, by premultiplying with the Hermitian transpose.

y = u' * u					% Equivalent MATLAB code

A length-M 1-D vector input is treated as an M-by-1 matrix. For both sample-based and frame-based inputs, output y is sample based with dimension N-by-N.

Applications

The Matrix Square block is useful in a variety of applications:

General matrix squares — The Matrix Square block computes the output matrix, y, without explicitly forming u'. It is therefore more efficient than other methods for computing the matrix square.
Sum of squares — When the input is a column vector (N=1), the block's operation is equivalent to a multiply-accumulate (MAC) process, or inner product. The output is the sum of the squares of the input, and is always a real scalar.
Correlation matrix — When the input is a row vector (M=1), the output, y, is the symmetric autocorrelation matrix, or outer product.

Dialog Box

Supported Data Types

Double-precision floating point
Single-precision floating point