Hi everyone,
I have a simulated data set with an antenna array consisting of 8 elements. I simulated an emitting radio source for the azimuth and elevation of 35 deg and 45 deg respectively and a distance of about 60 m. The SNR is around 7 dB. I programmed the MUSIC algorithm in Matlab. The estimation of the azimuth angle is always correct. But I always get the wrong elevation, no matter which scenario I'm testing. It's stuck at around 6-8 deg (see figure).
Is there something i have not considered, maybe some part on the steering vector? Or is the performance of the MUSIC algorithm not more accurate in case of uniform circular arrays? I'm alrady out of ideas and any answers will be gladly accepted. Thanks in advance.
Best regards
Did you use a linear array? In this case, you can estimate only one angle.
Otherwise, please upload your code (a minimalistic example).
In Uniform Circular Array, both angles can be found. Are you stuck at 6-7 degree for any arbitraty angle?
I have found my mistake. Basically, I have set a wrong radius in the steering vector for my circular array. As can be seen in the Figure, the algorithm now properly works as the azimuth (136 deg) and elevation (36 deg) of my signal source is clearly visible in the spectrum.
Now I have 2 further questions:
1) I have simulated a rotating antenna and now I'm trying the get the DOA for a virtual array. The signal's frequency is 1.575.42 GHz and the sampling frequency is 2.046 MHz. The radius of the array is 9 cm and the revolution time is 1 second. My question: how do i set up the covariance matrix of the data samples? What things do I have to consider? Unfortunately i cannot fully comprehend the approaches in diverse papers that deal with this problem.
2) Is it possible to get proper results for the distance in a 3D MUSIC approach? My signal source is around 900 m away from my array. What exactly is the difference between near- and far-field signal sources? is that range already too far to get a proper result?
Sorry for the noobish questions and many thanks to everbody in advance!
Hi,
What is the steering vector used is it r*cos((2*pi*n/N) - azimuth)*sin(elevation) or any other??
For any input elevation my output elevation angle is always fluctuating between -90 degrees and -85 degrees,my elevation scan range is from -90 to 90 in the intervals of 5
Is there any mistake in the delay matrix?
Thanks
- Badges
- Science topic
- Similar topics
- Invertebrate Zoology
- Carcinology
- Crustacea
- Malacostraca
- Brachyura