Abstract
Direction-of-arrival (DOA) estimation plays a vital role in the field of array signal processing. However, the need for heavy computing tasks in most traditional DOA algorithms, e.g., multiple signal classification (MUSIC), makes their engineering practicality significantly compromised in satellite communication systems. The neuroevolution of augmenting topologies (NEAT) can quickly search for appropriate topologies and weights of neural network functions, but its computational complexity is still too high for satellite systems. This paper proposes a modified NEAT architecture featuring a recurrent structure (RNEAT) that only needs a small number of phase components of the received signal covariance matrix as inputs to reduce the complexity and simplify the neural network architecture. The proposed RNEAT incorporated with multiple signal classification (RNEAT-MUSIC) features low complexity to achieve high resolution and low complexity simultaneously. Validation has been done by applying the proposed method in a two-dimensional direction of arrival estimation (2D-DOA) problem. Results show that the proposed RNEAT-MUSIC efficiently restricts the scanning region before forwarding the covariance matrix to the MUSIC stage. Consequently, the computational workload is reduced by 3/4 compared with the traditional 2D-MUSIC algorithm while maintaining satisfactory DOA resolution.
Highlights
O RBITING satellites and other space vehicles have complex trajectories, and ground stations need to acquire their angular positions quickly and accurately
The implementation of neuroevolution presented in this paper relies on the neuroevolution of augmenting topologies (NEAT) algorithm with recurrent links to automate the search for appropriate topologies and weights of neural network function
Algorithm: The proposed RNEAT-multiple signal classification (MUSIC) estimation algorithm Step1: The data covariance matrix R is obtained from the received data of antenna array; Step2: The data covariance matrix is replaced by the sampling covariance matrix R; Step3: The reshaped form of R′′, as shown in Fig. 3, is fed into the improved neuroevolution NEAT algorithm; Step4: The improved neuroevolution NEAT algorithm combined with recurrent structure is used to minimize the restricted scanning region’s neural network and automatically set the weights; Step5: The sub-region is obtained, and transmit the information to the MUSIC algorithm; Step6: Identify the angle in the sub-region corresponding to the maximum points, which are the signal sources incident directions
Summary
O RBITING satellites and other space vehicles have complex trajectories, and ground stations need to acquire their angular positions quickly and accurately. The MUSIC algorithm obtains mutually orthogonal signal subspace and noise subspace by performing eigenvalue decomposition (EVD) on the array output covariance matrix Since the signal subspace and the noise subspace are completely orthogonal under the noisefree model, MUSIC theoretically features super-resolution for arbitrarily close targets It can be applied in a different type of antenna array [19]. A modified NEAT architecture featuring a recurrent structure (RNEAT) is proposed that only needs a small number of phase components of the received signal covariance matrix as inputs to reduce the complexity and simplify the neural network architecture.
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
