414 IEEE TRANSACTIONS ON ELECTROMAGNETIC COMPATIBILITY, VOL. zyxw 38, NO. 3, AUGUST 1996 ceurate Analysis of T M Horn tennas for Pulse Radiation Kurt L. Shlager, zyxwvutsr Member, IEEE, Glenn S. Smith, Fellow, IEEE, and James G. Maloney, Member, IEEE Abstract-In the past, various approximate theoretical models have been used to analyze TEM horn antennas. Because of the limitations of these approximate models, there has been, to date, only qualitative agreement of measurements for TEM horn antennas with the predictions of the theories. In this paper, the finite-difference time-domain (FDTD) method is used to accurately analyze TEM horn antennas for pulse radiation. First, the metallic triangular-plate TEM horn antenna is considered. Computed results for the reflected voltage in the feeding transmis- sion line and the time-varying radiated electric field are shown to be in very good agreement with new experimental measurements. Graphs of the electric field in the space surrounding the antenna (magnitude of field plotted on a color scale) are used to give physical insight into the process of radiation. Next, the method is used to analyze two TEM horns that were previously designed for pulse radiation. The geometry and electrical properties of these antennas are more complicated than for the metallic, triangular- plate horn. One has shaped metallic plates with a resistive termination at the open end; the other has plates whose resistance varies continuously along their length. The computed results for these antennas are compared with previously made experimental measurements. I. INTRODUCTION OR MANY years, the TEM horn antenna ha,s been used to radiate and receive pulsed signals [1]-[14]. The TEM horn antenna, in its basic form, is shown in Fig. l(a). It consists of two thin metallic plates, each an isosceles triangle of side length zyxwvutsrqp s, with the angle at the apex being a. The planes of these triangles are separated by the angle p, and the antenna is fed with a parallel-wire transmission line connected at the apices. The metallic plates of the horn form a TEM transmission line, whose characteristic impedance zyxwvu Z,, is generally matched to that of the feeding transmission line 2, to reduce reflections. When the dimensions of the aperture of the hom are electrically small, the operation of the hom is fairly simple. On transmission, the horn radiates, on axis, a signal that is Manuscript received March 31, 1995; revised April 1, 1996. This work was supported in part by the Joint Services Electronics Program under Contracts DAAL-03-90-C-0004 and DAAH-04-93-G-0027, and by the Pittsburgh Su- percomputing Center under Grant ECS930005P. K. L. Shlager was with the School of Electrical and Computer Engineering, Georgia Institute of Technology, Atlanta, GA 30332 USA. He is now with the Antenna Systems Department, TRW Electronics Systems and Technology Division, Space and Electronics Group, Redondo Beach, CA 90278 USA. G. S. Smith is with the School of Electrical and Computer Engineering, Georgia Institute of Technology, Atlanta, GA 30332 USA. zyxwvuts J. G. Maloney is with the Signatures Technology Laboratory, Georgia Tech Research Institute, Atlanta, GA 30332 USA. Publisher Item Identifier S 0018-9375(96)06135-2. the first temporal derivative of the incident voltage pulse on the feeding transmission line. On reception, the horn receives a signal that is of the same form as the electric field of the plane-wave pulse normally incident on the aperture. Typical applications for the TEM hom are as the feed antenna for reflectors designed to radiateheceive pulses; as the transmitting and receiving antennas in short-pulse radars, such as those used to detect buried objects (ground penetrating radars): and as the antennas for broad-band sensors used for evaluating electromagnetic compatibility. Even though the TEM horn antenna has been used to radiate and receive pulsed signals for the past 20-25 years, there has been, to date, no full theoretical analysis of this structure. Various approximate analyses have been performed, but they have provided only qualitative agreement with experimental measurements. The simplest analyses assume a field in the aperture of the horn and sometimes in the open spaces along the sides of the horn. This field is usually taken to be the field of the TEM mode that would exist in a horn of infinite length: that is, the higher order modes that accompany the reflection at the aperture are ignored. Huygen’s principle is used with this field to predict the radiated field of the horn [7], [SI. One- dimensional (1-D) analyses approximate the TEM horn by a thin-wire V-antenna and either assume a current on the antenna or solve for the current by a numerical method [9], [lo]. The radiated field is then computed from the current. The reflection coefficient at the aperture of the horn has been estimated by assdming it is the same as that for the open end of a parallel-plate transmission line with plates of infinite width u11. The reflections that occur at the aperture of the horn generally degrade the performance of the antenna, and various schemes have been developed to reduce these reflections. In 197 1, Wohlers introduced the concept of shaping the plates of the TEM horn (varying their width, see Fig. 7) so as to continuously change the characteristic impedance of the equivalent transmission line over the length of the antenna [4]. At the apex, the characteristic impedance is set equal to that of the feeding transmission line (typically 50 R), and at the aperture it is set equal to the wave impedance of free space (c, zyxwvu M 377 R). The idea is that there would be little reflection from the open end of a transmission line whose characteristic impedance is equal to that of free space. In fact, this is not the case, as will be shown in Section IV. Other researchers have perfected the design of Wohlers; an example will be discussed 0018-9375/96$05.00 0 1996 IEEE