Akifumi Hirata

Fast beamforming of electronically steerable parasitic array radiator antennas: theory and experiment

A low-power consumption, small-size smart antenna, named electronically steerable parasitic array radiator (ESPAR), has been designed. Beamforming is achieved by tuning the load reactances at parasitic elements surrounding the active central element. A fast beamforming algorithm based on simultaneous perturbation stochastic approximation with a maximum cross correlation coefficient criterion is proposed. The simulation and experimental results validate the algorithm. In an environment where the signal-to-interference-ratio is 0 dB, the algorithm converges within 50 iterations and achieves an output signal-to-interference-plus-noise-ratio of 10 dB. With the fast beamforming ability and its low-power consumption attribute, the ESPAR antenna makes the mass deployment of smart antenna technologies practical.

Reactance domain MUSIC algorithm for electronically steerable parasitic array radiator

Conventional adaptive array antenna processing must observe signals on all of the array antenna elements. However, because the low-cost electronically steerable parasitic array radiator (ESPAR) antenna has only a single-port output, none of the signals on the antennas parasitic elements can be observed. A direct application of most of the algorithms for the conventional adaptive array antenna is impractical. In this paper, A technique of estimation of direction-of-arrivals (DoAs) is proposed for the ESPAR antenna. This technique is based on the modified MUltiple SIgnal Classification (MUSIC) algorithm. The correlation matrix used in the MUSIC algorithm is estimated from the signal received through the single-port output of the ESPAR antenna as it switches over a set of antenna patterns. Simulation results show that DoAs can be estimated by the reactance domain MUSIC algorithm for ESPAR antennas. Furthermore, the statistical performance on estimation error variance of the reactance domain MUSIC estimator is analyzed and compared with the Crame/spl acute/r-Rao lower bound. Analytic and empirical results show that high-resolution DoAs estimation can be achieved by using the reactance domain MUSIC algorithm for ESPAR antennas.

Reactance-domain ESPRIT algorithm for a hexagonally shaped seven-element ESPAR antenna

A direction-of-arrival (DoA) method that combines the reactance-domain (RD) technique and the ESPRIT algorithm is proposed for use with the 7-element electronically steerable parasitic array radiator (ESPAR) for the estimation of noncoherent sources. Simulations show that the method could resolve up to three incoming signals with an estimation performance that depends on the signals angle of arrival. Moreover, the method is compared with the Cramer-Rao lower bound (CRB) and the MUSIC asymptotic error variance, both modified for the RD technique. Numerical comparison between this lower bound and the MUSIC algorithm confirmed that the proposed method can achieve the CRB and provide high-precision DoA estimation with a level of performance that is sufficient for many DoA finding applications. The proposed method could be demonstrated by means of experiments on DOA estimation conducted in an anechoic chamber.

Reactance domain MUSIC algorithm for ESPAR antennas

Conventional adaptive array antenna processing must observe signals on all of the array antenna elements. However, because the low-cost electronically steerable parasitic array radiator (ESPAR) antenna has only a single-port output, it is not possible to observe all of the signals on the antenna elements. A direct application of most of the algorithms for the conventional adaptive array antenna is impractical. In this paper, we propose a new technique of estimation for direction-of-arrivals (DoAs) for the ESPAR antenna. This technique is based on a modification of the MUSIC (multiple signal classification) algorithm. An M-dimension vector y of the output of the antenna is obtained by using the angular diversity provided by the ESPAR antenna reactances. Then theM byM correlation matrix of the y vector is employed to estimate the MUSIC spectrum. Simulation results show that DoAs can be estimated by the reactance domain MUSIC algorithm for ESPAR antennas.