Method and apparatus for determining indices of orthogonal basis vectors
An orthonormal basis and index technology, applied in the field of determining the index of the orthonormal basis vector, can solve the problem of large feedback overhead of the orthonormal basis vector.
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
example 1
[0226] In Example 1, the following conditions are satisfied: (1) each layer is selected independently; (2) the 2L rows of the transform domain matrix jointly select the index of the orthogonal basis vector; (3) the oversampling factor O=1; (4) the orthogonal basis The vector's oversampling offset factor v=0.
[0227] Specific steps are as follows:
[0228] Step 1: Obtain transform domain matrix (including time domain matrix);
[0229]Step 2: Take the square root of the sum of the absolute values of each element of the transformation domain matrix with a dimension of 2L×M (Frobenius norm), directly calculate the magnitude of each element or other methods, and then add or In other methods, a transformation domain value matrix with a length of 1×M is obtained.
[0230] Step 3: From the 1×M transformed domain value matrix at position Select consecutive X values and add them or other methods to obtain the value Carrying out the above processing sequentially from the start...
example 2
[0244] In Example 2, the following conditions are met: (1) each layer is independently selected; (2) the transform domain matrix 2L row independently selects the index of the orthogonal basis vector; (3) the oversampling factor O=1; (4) the orthogonal basis The vector's oversampling offset factor v=0.
[0245] Specific steps are as follows:
[0246] Step 1: Obtain transform domain matrix (including time domain matrix);
[0247] Step 2: Take the square root of the sum of the absolute values of each element of the transformation domain matrix with a dimension of 2L×M (Frobenius norm), directly calculate the amplitude of each element or other methods, and obtain a length of 2L×M Transform domain-value matrix of .
[0248] Step 3: Transform the position of the nth row of the domain value matrix from dimension 2L×M Select consecutive X values on n rows and add them or other methods to obtain the value Starting from the starting position of the nth row, the above process...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More