| Direction-of-arrival (DOA) estimation of narrowband plane wave signals impinging on a sensor array
has been intensively studied in the past few decades for several military applications. Many high-resolution
algorithms have been developed, e.g. MUSIC, ESPRIT, MODE, for this problem. However, the structure
of incoming signals has not been considered in those well-known traditional subspace methods. Some
works have been proposed to take into account the available information about the incoming signals.
One method has been developed for signals with known waveforms. Constant modulus (CM) signals
have been studied in several other literature, where some algorithms are developed to the case of phase
modulated signals.
Recently, the work by Y. Wang made use of non-circular PAM/BPSK signals by exploiting the complex
conjugate counterpart of the received signals. This algorithm is shown to increase the number of detectable
directions as well as to improve the performance accuracy.
However, in practice, it is more realistic that some users send PAM/BPSK signals while others still send
complex circular signals. In this work, we study this more general case and develop a new MUSIC-based
DOA estimation algorithm. The newly proposed algorithm can improve the DOA estimation accuracy
compared to the conventional MUSIC algorithm. Meanwhile, it allows the maximum number of detectable
signals beyond the traditional MUSIC limit. Moreover, the spatial smoothing technique for dealing with
the correlated sources is modified for our proposed scenario, and the closed form Cramer-Rao bound is
derived. More interestingly, the proposed algorithm is able to discriminate real sources from complex
sources. |