Research On Parameters Estamation Of Electromagnetic Vector Sensor Arrays Based On USRP

The research of the array signal parameter estimation has been lasted for several decades and so as the revelent algorithms. The most useful algorithms are MUSIC(Multiple Signal Classification algorithm) and ESPRIT(Estimating Signal Parameters via Rotation Invariant subspace algorithm) because of the high performance. However, the scalar sensor can measure only one component of the electromagnetic wave and waste the rest five components. The EMVS(Electromagnetic Vector Sensor) can sense all the six components electromagnetic information of the signal,which means the EMVS array can measure not only spatial information but also polarizing information. So the more efficient EMVS sensor worked with the high performance algorithm is expected to get higher performance include steady detective ability, higher space resolution as well as stronger anti-jamming ability by the researchers, and this has been the interested research direction by these days.Because the special structure increases the process difficulity, there is rarely EMVS antenna for people to use to get the measured signal to study more. And this has been a limitation of the research. The researchers tried several ways to construct an EMVS with the scalar sensor in hand and come up with the algorithms based on MUSIC, but it remains at theoretical stage. A new approach to construct a simplified EMVS and the separated simplified EMVS with several antennas(USRP) and estimate the DOA(Direction of Arrival) and polarization information is introduced in this paper. Besides, estimating the DOA of the source and tracking the direction to a moving source using only one EMVS is investigated, together with the performance simulated with MATLAB.