Title :
Neurospectral analysis of coaxial fed rectangular patch antenna
Author :
Mishra, R.K. ; Patnaik, A.
Author_Institution :
Electron Sci. Dept., Berhampur Univ., Orissa, India
Abstract :
The importance of the spectral domain approach (SDA) is attributed mostly to its accuracy. Being a full-wave method, it is computer intensive. There are difficulties in numerically integrating the Green´s function due to the presence of poles in the integral path. The artificial neural network technique is being used in the field of microwave during the last five years or so. The use is mostly restricted to optimisation type of problems. This paper tries to extend earlier work of Assailly et al. (1988) and of Mishra and Patnaik (see IEEE MGWL, vol.9, no.9, p.351-53, 1999) on the use of ANN for analysis. Closed form expressions are obtained for the integrals involving the Green´s functions in the SDA. The resonant frequency is found using reverse training. The method is demonstrated by application to a coaxial fed rectangular patch antenna.
Keywords :
Green´s function methods; antenna feeds; electric field integral equations; electrical engineering computing; microstrip antennas; neural nets; resonance; spectral-domain analysis; ANN; EFIE; Green´s functions; artificial neural network; closed form expressions; coaxial fed rectangular patch antenna; electric field integral equation; full-wave method; impedance calculation; neurospectral analysis; optimisation problems; poles; resonant frequency; reverse training; spectral domain approach; Artificial neural networks; Coaxial components; Impedance; Integral equations; Microstrip antennas; NIST; Patch antennas; Resonant frequency; Transfer functions; Transmission line matrix methods;
Conference_Titel :
Antennas and Propagation Society International Symposium, 2000. IEEE
Conference_Location :
Salt Lake City, UT, USA
Print_ISBN :
0-7803-6369-8
DOI :
10.1109/APS.2000.875405