Abstract
In this paper, a novel direction-of-arrival (DOA) estimation for unknown (anonymous) emitter signal (ES) based on time reversal (TR) and coprime array (CA) is proposed. The resolution and accuracy of DOA estimation are enhanced from two aspects: one is from the view of array arrangement: the new distribution of CA is designed to reduce the holes, increase the degree of freedom (DOF) and apertures by rotating and translating only one subarray, which simplifies the operation. The other one is from the view of the algorithm: a neoteric DOA estimation algorithm with noise suppression based on TR, Capon and adaptive neuro-fuzzy inference system (ANFIS) is proposed for solving the wide sidelobe, multipath effect, low resolution and accuracy produced by conventional algorithms, in particular, those cannot work effectively under the existed hole condition. Furthermore, the resubmitting distorted noise and channel noise are suppressed effectively, which is not taken into considered in the conventional Capon algorithm. Simulation results including the resolution, accuracy, root mean square error (RMSE), Cramér-Rao lower bound (CRLB) and the compared analyses on uniform linear array (ULA), nested array (NA) and minimum redundancy array(MRA) demonstrate the performance advantages of the proposed DOA estimation algorithm even at very low signal-to-noise ratio (SNR) condition.
Highlights
With the continuous appearance of new wireless communication and position systems, locating the emitter signal (ES) plays a more and more important role in public security, fraud detection, and intelligent transportation systems [1,2,3]
For obtaining useful transmission and reflection parts of ES, and forbidding the negative effect of multipath diversity on direction of arrival (DOA) estimation, this paper considers time reversal (TR) as a good candidate, because TR is able to take advantage of multipath which is recognized as clutter or noise and ignored/mitigated in other most DOA estimation algorithms, such as multiple signal classification (MUSIC), decomposition of reverse time operator (DORT), back projection [32,33]
In order to enhance the resolution and accuracy of DOA estimation, the main contributions of this paper can be summarized as the following two points: from the view of array arrangement, we design a virtual large aperture linear array based on coprime array (CA); from the view of algorithm, we proposed a noise suppression DOA estimation algorithm based on TR and Capon
Summary
With the continuous appearance of new wireless communication and position systems, locating the emitter signal (ES) plays a more and more important role in public security, fraud detection, and intelligent transportation systems [1,2,3]. There are many DOA estimation methods springing up, such as estimation of signal parameter via rotational invariance technique (ESPRIT) [8], multiple signal classification (MUSIC) [9] and decomposition of reverse time operator (DORT) [10] These subspace methods present a high complexity due to the fact that they strongly rely on eigenvalues or singular value decomposition for differentiating the signal or noise subspace. Most DOA estimation algorithms attempt to eliminate the effect of multipath using deconvolution with the approximated channel impulse response or channel equalization, which treats multipath as clutter or noise These approaches result in the loss of some useful information on ES, and the resolution and accuracy are limited. DOF Design and Method of Increasing Effective Aperture of Array for DOA Estimation.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
