Documentation

This is machine translation

Translated by Microsoft
Mouseover text to see original. Click the button below to return to the English verison of the page.

Note: This page has been translated by MathWorks. Please click here
To view all translated materals including this page, select Japan from the country navigator on the bottom of this page.

spsmooth

Spatial smoothing

Syntax

RSM = spsmooth(R,L)
RSM = spsmooth(R,L,'fb')

Description

example

RSM = spsmooth(R,L) computes an averaged spatial covariance matrix, RSM, from the full spatial covariance matrix, R, using spatial smoothing (see Van Trees [1], p. 605). Spatial smoothing creates a smaller averaged covariance matrix over L maximum overlapped subarrays. L is a positive integer less than N. The resulting covariance matrix, RSM, has dimensions (NL+1)-by-(NL+1). Spatial smoothing is useful when two or more signals are correlated.

RSM = spsmooth(R,L,'fb') computes an averaged covariance matrix and at the same time performing forward-backward averaging. This syntax can use any of the input arguments in the previous syntax.

Examples

collapse all

Construct a 10-element half-wavelength-spaced uniform line array receiving two plane waves arriving from 0° and -25° azimuth. Both elevation angles are 0°. Assume the two signals are partially correlated. The SNR for each signal is 5 dB. The noise is spatially and temporally Gaussian white noise. First, create the spatial covariance matrix from the signal and noise. Then, solve for the number of signals, using rootmusicdoa. Next, perform spatial smoothing on the covariance matrix, using spsmooth, and solve for the signal arrival angles again using rootmusicdoa.

Set up the array and signals. Then, generate the spatial covariance matrix for the array from the signals and noise.

N = 10;
d = 0.5;
elementPos = (0:N-1)*d;
angles = [0 -25];
ac = [1 1/5];
scov = ac'*ac;
R = sensorcov(elementPos,angles,db2pow(-5),scov);

Solve for the arrival angles using the original covariance matrix.

Nsig = 2;
doa =  rootmusicdoa(R,Nsig)
doa = 

    0.2755  -49.7762

The solved-for arrival angles are wrong - they do not agree with the known angles of arrival used to create the covariance matrix.

Next, solve for the arrival angles using a smoothed covariance matrix. Perform spatial smoothing to detect L-1 coherent signals. Choose L = 3.

Nsig = 2;
L = 2;
RSM = spsmooth(R,L);
doasm = rootmusicdoa(RSM,Nsig)
doasm = 

  -25.0000   -0.0000

In this case, computed angles do agree with the known angles of arrival.

Input Arguments

collapse all

Spatial covariance matrix, specified as a complex-valued, positive-definite N-by-N matrix. In this matrix, N represents the number of sensor elements.

Example: [ 4.3162, –0.2777 –0.2337i; –0.2777 + 0.2337i , 4.3162]

Data Types: double
Complex Number Support: Yes

Maximum number of overlapped subarrays, specified as a positive integer. The value L must be less than the number of sensors, N.

Example: 2

Data Types: double

Output Arguments

collapse all

Smoothed covariance matrix, returned as a complex-valued, M-by-M matrix. The dimension M is given by M = N–L+1.

References

[1] Van Trees, H.L. Optimum Array Processing. New York, NY: Wiley-Interscience, 2002.

Extended Capabilities

Introduced in R2013a

Was this topic helpful?