IEEE TRANSACTIONS ON ANTENNAS AND PROPAGATION, VOL. 56, NO. 1, JANUARY 2008 245
Talbot Effect Applied to Antennas
M. G. Keller, J. Shaker, and Y. M. M. Antar
Abstract—An experimental investigation is reported on how the Talbot
effect may be exploited to produce a low volume antenna configuration that
is suitable for high effective isotropic radiated power applications. In this
architecture a periodic phase grating is combined with a sparse antenna
array to form a structure that is simpler to fabricate than an equivalent
active antenna array. This configuration is shown to have improved gain
and lower grating lobe levels as compared to the sparse array. Finally, a
simple technique is described to predict the far-field radiation pattern of
this Talbot antenna array.
Index Terms—Antennas, diffraction, interference, Talbot effect.
I. INTRODUCTION
The Talbot effect is an optical phenomenon that was discovered in
1836 by H. F. Talbot [1]. It is a near-field diffraction and interference
phenomenon which occurs when a periodic amplitude or phase grating
is illuminated by a plane wave of electromagnetic energy. Constructive
and destructive interference of the diffracting waves emerging from be-
hind a periodic grating form “bright spot” field pattern zones at frac-
tions of the periodic “Talbot distance” that resemble the grating
in form or in contrast, while in other regions shifted or multiple “bright
spot” image zones appear.
Following the pioneering work performed by Lohmann [2] on op-
tical applications of the Talbot effect, a millimeter-wave Talbot array
illuminator (TAIL) was investigated [3]. The Talbot effect has also been
employed as a means to perform power combining within a planar
waveguide at 8 GHz [4]. In addition, there have been several experi-
mental and theoretical investigations of the free-space Talbot phenom-
enon in the millimeter-wave band which introduced several applica-
tions such as high power transmitting antennas, spatial power com-
bining architectures, oscillators, and frequency doublers [5]–[7]. These
interesting applications involved a quasi-optic approach, whereby the
active array elements are spatially fed, thereby avoiding transmission
line losses. However, this approach leads to structures that are electri-
cally large in the longitudinal direction. This problem can be mitigated
by employing a circuit feeding scheme for the active array elements.
Antennae serve to radiate energy into a region of space in a manner
that is quantified by their radiation patterns. Further, some system ap-
plications may additionally require a certain effective isotropic radiated
power (EIRP). One low-volume approach to achieve the required EIRP
specification has been to employ active antenna arrays. This solution
typically merges densely packed microstrip patch antennas with active
devices such that a high-gain antenna is combined with a large power
output to produce a suitable EIRP. One limitation of this approach is the
requirement for a large number of active radiating elements. This lends
itself to complex design and fabrication issues, as well as heat dissipa-
tion problems that may potentially restrict its EIRP. To overcome some
Manuscript received November 29, 2006; revised May 3, 2007. This work
was carried out at the Communications Research Centre Canada.
M. G. Keller and Y. M. M. Antar are with the Department of Electrical
and Computer Engineering, Royal Military College of Canada, Station Forces
Kingston, ON K7K 7B4, Canada (e-mail: michael.keller@crc.ca; antar-y@
rmc.ca).
J. Shaker is with the Communications Research Centre Canada, Ottawa, ON
K2H 8S2, Canada (e-mail: jafar.shaker@crc.ca).
Digital Object Identifier 10.1109/TAP.2007.913167
Fig. 1. Normalized power distribution (dB) over an surface at
beyond a 2-D phase grating.
of these issues, it would be desirable if a sparse antenna array architec-
ture could be adapted so as to allow for less dense packing of the active
elements, and hence easier thermal considerations. This paper reports
on an experimental and theoretical investigation on how the Talbot ef-
fect may be exploited to produce a low volume antenna configuration
that is suitable for high EIRP applications.
II. TALBOT EFFECT APPLIED TO ANTENNAS
It is worthwhile to briefly review some of the previously obtained
results for when a 2-D phase grating is illuminated by a Gaussian nor-
mally incident wave. For instance, in [3], a nine period by nine period
phase grating, with a phase duty cycle (PDC) equal to 1/3, was formed
by 81 identical unit cells, each of which had three phase shifts elements
of 0 , 120 , and 240 . The grating period was 20 mm. As shown in the
contour plot of Fig. 1, when illuminated by a Gaussian beam, 81 “bright
spots” were generated at the fractional Talbot distance of
beyond the structure, This figure displays the normalized power mea-
sured by a planar near-field measurement system.
A Talbot effect antenna may be formed from the amalgamation of
a periodic phase grating and a periodic antenna array. As shown in
Fig. 2, these two structures are combined to form an antenna based on
the circuit-fed/spatially combined power combiner architecture [8]. In
this configuration, the antenna elements are circuit-fed, and their indi-
vidual outputs are combined in free-space. The underlying foundation
of this concept is the fact that a periodic Talbot phase grating under
plane wave illumination yields a periodic array of “bright spots” at a
fractional Talbot distance [3]. In this particular application, a circuit-fed
array of antenna elements is placed coincident with selected “bright
spot” locations. By reciprocity the phase grating will act to transform
the “bright spots” (now formed by the radiating elements) into a plane
wave. The following sections will describe the design and analysis of
Talbot antenna elements and arrays.
A configuration suitable for high EIRP applications is now intro-
duced. As a corporate-fed microstrip patch antenna array was used to
illuminate the phase grating, its design will be described first. In this
example, passive antenna elements are utilized; however, it would be a
straight forward step to integrate active devices into the antenna array
structure [9].
0018-926X/$25.00 © 2008 IEEE