Signal processing unit and method for searching for peaks in a two-dimensional matrix
A signal processing unit and a method for searching for peaks in a two-dimensional matrix of numbers are described. The matrix is analyzed row by row and then column by column. Analyzing a row comprises, for each element of the row, tagging the element in response to determining that the element is a local maximum of the row Analyzing a column comprises determining a bit field associated with the column by determining, for each element of the column, a corresponding bit field element Determining the bit field element comprises: if the element of the column has not been tagged, setting the bit field element to a predefined first value, and, if the element of the column has been tagged, determining whether the element is a local maximum and, in this case, setting the bit field element to a predefined second value different from the first value and, otherwise, setting the bit field element to the first value.
1. A signal processing unit for searching for peaks in a two-dimensional matrix of numbers, by analyzing the matrix row by row and then column by column,
wherein analyzing a row comprises, for each element of the row, tagging the element in response to determining that the element is a local maximum of the row; and
wherein analyzing a column comprises determining a bit field associated with the column by determining, for each element of the column, a corresponding bit field element, wherein determining the bit field element comprises:
if the element of the column has not been tagged, setting the bit field element to a predefined first value, and,
if the element of the column has been tagged, determining whether the element is a local maximum and, in this case, setting the bit field element to a predefined second value different from the first value and, otherwise, setting the bit field element to the first value.
2. The signal processing unit of claim 1 , wherein the first value is zero and the second value is one.
3. The signal processing unit of claim 1 , further arranged to generate an aggregated list, the aggregated list consisting of those bit fields that include at least one bit field element having the second value.
4. The signal processing unit of claim 1 , wherein an element of a row or column is a local maximum of the row or column if it is greater than both its predecessor element and its successor element in the row or column,
or wherein the element of the row or column is a local maximum of the row or column if it is greater than its predecessor element in the row or column and greater than or equal to its successor element in the row or column,
or wherein the element of the row or column is a local maximum of a row or column if it is greater than or equal to its predecessor element in the row or column and greater than its successor element in the row or column.
5. The signal processing unit of claim 1 , wherein a threshold is defined and wherein the element of the row or column is a local maximum of the row or column if it is greater than the threshold, greater than a first one of its predecessor element and its successor element in the row or column, and also greater than or equal to a second one of its predecessor element and its successor element in the row or column.
6. The signal processing unit of claim 1 , wherein the first element of a row or column and the last element of the row or column are considered neighboring elements.
7. The signal processing unit of claim 1 , wherein each of said numbers has a magnitude and any comparison between any two elements of the matrix is a comparison between the magnitude of a first one of said two elements and the magnitude of the second one of said two elements.
8. The signal processing unit of claim 7 , wherein the magnitude 1 z 1 of a real or complex number z is defined by one of the following definitions:
| z |=abs( x ) wherein abs( x )= x if x≥ 0 and abs( x )=− x if x< 0, provided that z=x is real;
| z |=mag( z )=sqrt( z*cc ( z ));
| z |=ssq( z )= z*cc ( z );
| z|=db ( z )=log( z*cc ( z )).
9. A radar device comprising the signal processing unit of claim 1 .
10. A method of searching for peaks in a two-dimensional matrix of numbers, comprising:
operating a signal processing unit to analyze the matrix row by row and then column by column,
wherein analyzing a row comprises, for each element of the row, tagging the element in response to determining that the element is a local maximum of the row; and
wherein analyzing a column comprises determining a bit field associated with the column by determining, for each element of the column, a corresponding bit field element, wherein determining the bit field element comprises:
if the element of the column has not been tagged, setting the bit field element to a predefined first value, and,
if the element of the column has been tagged, determining whether the element is a local maximum and in this case setting the bit field element to a predefined second value different from the first value and, otherwise, setting the bit field element to the first value.
11. The method of claim 10 , further comprising:
generating an aggregated list, the aggregated list consisting of those bit fields that include at least positive bit field element that has been set to the second value.
12. The method of claim 10 , wherein an element of a row or column is a local maximum of the row or column if it is greater than both its predecessor element and its successor element in the row or column,
or wherein the element of the row or column is a local maximum of the row or column if it is greater than its predecessor element in the row or column and greater than or equal to its successor element in the row or column,
or wherein the element of the row or column is a local maximum of a row or column if it is greater than or equal to its predecessor element in the row or column and greater than its successor element in the row or column.
13. The method of claim 10 , wherein a threshold is defined and wherein the element of the row or column is a local maximum of the row or column if it is greater than the threshold, greater than a first one of its predecessor element and its successor element in the row or column, and also greater than or equal to a second one of its predecessor element and its successor element in the row or column.
14. The method of claim 10 , wherein the first element of a row or column has the last element of the row or column as one of its two nearest neighboring elements and wherein the last element of the row or column has the first element of the row or column as one of its two nearest neighboring elements.
15. The method of claim 10 , wherein each of said numbers has a magnitude and any comparison between any two elements of the matrix is a comparison between the magnitude of a first one of said two elements and the magnitude of the second one of said two elements.