The authors started investigating the post-wall waveguide in 1997. The post-wall waveguide is realized by making two lines of via-holes in a metal-clad dielectric substrate and metal-plating their walls, as shown in the feed waveguide of the post-wall waveguide-fed parallel-plate slot array antenna in Fig.1. It can be obtained by conventional print-circuit-board fabrication technique at low cost. This paper presents the up-to-date results of the post-wall waveguide slot array antennas.

Fig.1 shows the post-wall waveguide-fed parallel-plate slot array antenna. The feed waveguide is placed at the end of a parallel plate waveguide to excite a plane TEM wave. The antennas are designed to excite a slot array uniformly on the parallel-plate waveguide at 61.25GHz. PTFE substrates are used and the dielectric constant is 2.17. The thickness of the substrate is 1.2mm, which corresponds to 0.36 wavelengths considering the dielectric constant. It should be as large as possible within 0.5 wavelengths in order to reduce the conductor loss. The transmission loss is 0.10dB/cm in the estimation where the dielectric loss tanδ=8.5x10^-4 and the conductivity is 5.8x10⁷(S/m). This level of the loss is confirmed by measurements. The post diameter is 0.5mm and the post spacing is 1.2mm. Fig.2 shows the frequency characteristic of the gain in various sizes of the antennas. The aperture size is 2x2cm to 8x8cm. The peak gain becomes high for a larger aperture. However the bandwidth becomes narrow because the antennas are series-fed arrays. The efficiency keeps high (60-55%) in the range of 21-33dBi gain.

Fig.3 shows a beam-switching antenna by integrating a 16x4-element slot array (145.7mm length) and a 4-way Butler matrix (58.8mm) in a single-layer substrate by the post-wall waveguides at 25.6GHz in order to reduce the connecting loss between them.(All air holes are covered by Aluminum tapes in measurements.) The cross couplers are realized by a short-slot coupler in single layer as well as the hybrids in the Butler matrix. Two air rectangular holes are installed near the side walls in the coupled region of each coupler to enhance the difference in phase constant between the two propagating modes in the coupled region, so that the coupler length is halved in comparison with the conventional coupler using two posts in the coupled region. Four beams cover a 115-degree sectoral area equally in the H-plane as shown in Fig.4.

This paper presents an overview of the wideband phased arrays recently developed by Raytheon under contracts and internal funds for radar and EW applications that require more than 4:1 bandwidths. These arrays are characterized by low profile, light weight, and dual-polarizations with very low cross-pol components. The design principles apply to low and high frequency bands, ranging from UHF (150-600 MHz) to X- and Ku-band (8-18 GHz). Computer modeling results and prototype test performance data will be presented. Specifically, two types of array apertures using different radiating elements will be discussed; namely the low-profile flare dipoles ("bunny-ear") and the continuous long slot apertures. The flare dipole is a variation of the well-known flare notch (Vivaldi) design, but it is shorter, about 1/2 Ć at the high end of the band, as opposed to 2-3 Ć of the conventional Vivaldi element. Also, its cross-pol components are much lower in the diagonal plane, an important and desirable feature of the design for dual-pol and CP systems that demand high isolation between the orthogonal components. The other type of the wideband arrays to be reviewed is the long slot aperture recently developed by Raytheon for radar and EW applications. The new array antenna is intrinsically a robust design characterized by ultrawide bandwidth. Rigorous analysis starting with Green Function for the aperture field distribution has been published, providing parametric curves for practical array designs. The theory is also applicable in general to the "connected" electric dipole arrays, a conjugate configuration of the long slot aperture. Active VSWR and bandwidth of the long slot radiating element, with and without a back plane, will be discussed. Compact, light weight prototype arrays from UHF to L-band with measured data, including array patterns, input match, cross-pol level, array gain and edge effects will be presented. In addition, methods of extending the bandwidth to cover 20:1 band with acceptable loss will be described.

Satellite broadcasting in the 21-GHz band is expected to broadcast a wide-band program such as Super Hi-Vision which has ultra high-definition more than 4000 scanning lines. The 21-GHz band, however, suffers from heavy signal attenuation due to rain at a power level about three times that of the 12-GHz band in terms of decibels.

This paper describes a mitigation technique using onboard imaging reflector antennas that can boost power as a °$B!H(Bboosted beam$B!I(B to offset rain attenuation only in areas with rainfall while keeping a constant level of power in the rest of the service area as a $B!H(Bnationwide beam$B!I(B. The system concept is shown in figure1. The imaging reflector antenna consists of a main reflector of 4m diameter, an imaging arrangement of smaller sub reflector of 0.5m diameter and a small aperture array feed of 188 active array elements with a spacing of 1.8 wavelengths as shown in figure2. As the power amplifier for the active array, the miniature TWT was developed. The cross-sectional dimension of the TWT was 20 x 20mm, which corresponds to approximately 1.5wavelength x 1.5wavelength.

Figure 3 shows the calculated radiation pattern when forming a boosted beam over Tokyo that was about 10 dB greater than the average gain of the nationwide beam over the service area. The maximum gain of the boosted beam was 50.6 dBi, while minimum gain of the nationwide beam was 39.3 dBi.

Introduction

An alternating phase fed single-layer slotted waveguide array [1] has been developed for millimeter wave and microwave application [2]. This antenna consists of a slotted plate and a waveguide base plate joined by a screw which dispenses with electrical contact in the strict sense. The waveguide base and the slot plate are mass produced by die casting and etching respectively. Unfortunately this array has a relatively narrow bandwidth.Fixed Wireless Access systems in the 26 GHz band uses a hundred or so MHz of bandwidth for each carrier. Until now, different arrays were designed for each carrier with different channels, though manufacturing base plates by die casting needs quite large volume orders for cost requirement. On the other hand, the slot plate can be produced cost effectively even in small number. If the waveguide base can be commonly used and only slot plates interchanged for individual frequency, every channel can be covered with reasonable cost.

This paper discusses the possibility of the diversion of the waveguide base plate for frequencies different from the designed ones; multi-frequency operation is theoretically evaluated by FEM full structure simulation of the array.

Structure and analysis of full models

Figure 1 shows the structure of 3 different slot plates for 3 design frequencies (slot plate #1 designed at 24.85GHz, slot plate #2 designed at 25.3GHz and slot plate #3 designed at 25.75GHz) and base plate designed at 25.3GHz. The usual slot array start position line is shown in Figure 1(b). On the other hand, the slot plate #1 and the slot plate #3 have different start position lines for each slot array adjusted by the waveguide length. Each slot plate and waveguide base is joined by a screw. Figure 2 shows the calculated results of directivity gain using HFSSTM (FEM analysis). The antenna gain using the slot plate #2 decreases 0.83dB (at 24.85GHz) and 2.62dB (at 25.75GHz) respectively. By interchanging slot plates #1 or #3, the antenna gain decreases only 0.12dB (at 24.85GHz or at 25.75GHz).

Conclusion

We have proposed an alternating phase fed single-layer slot array with interchangeable slot plates for multi-frequency operation. The characteristics are confirmed by simulation. We will perform an experiment in future.

References

[1] N. Goto, "A waveguide-fed printed antenna," IEICE Technical Report, AP89-3, Apr. 1989.

[2] Y. Kimura, Y. Miura, J. Hirokawa and M. Ando, "A LOW-COST AND COMPACT WIRELESS TERMINAL WITH AN ALTERNATING PHASE FED SINGLE-LAYER WAVEGUIDE ARRAY FOR 26GHz FIXED WIRELESS ACCESS SYSTEMS," JINA, Nov. 2002

Figure 1: Structure of interchangeable slot plates and one waveguide base. Figure 2: Gain characteristics of only slot change model. (Cal.)

An optimum dual band self-diplexed antenna, particularly suitable to accomplish the demanding requirements of the European Navigation System (GALILEO) is proposed.

The antenna relies on the key enabling technologies previously developed for the pioneering GIOVE-A satellite (Galileo In Orbit Validation Element) successfully flown on 28 December 2005. Its excellent operational status and the quality of transmitted signal are a validation of the technology solutions adopted to the antenna. These are based on detachable symmetrical fed self diplexed and stacked patch radiators supported by quartz honeycomb, fed by independent beamforming networks in square coaxial technology embedded in honeycomb structural sandwiches and supported by grounding stubs for thermal design. Current GIOVE-A NAVANT (see Fig 1.a) is a 36 elements dual band transmitting planar array having dual band functionality, high RHCP purity, isoflux beam shape, phase flatness, group delay stability, high gain and high input power.

The new proposed antenna, that still retains the GIOVE-A modular design and enabling technologies, is based on a mixed lattice, specifically optimized for the dual band functionality at minimum hardware, in order to meet the much demanding performance (mainly coverage gain at high frequency band, mass and stiffness) required by Galileo.

The New NAVANT has 28 radiating elements, but only 20 are operative at low and high band respectively, as illustrated in fig 1.b, where the 12 center elements are common to both bands while the outer ring operates at low band and the inner ring at high band.

These two diameters are close to the optimum aperture sampling at low and high band. Under-sampling along phi can be tolerated producing the only effects of slight side-lobes increase out of the earth field of view.

In so doing the antenna can be divided into four identical sectors having only five elements active at any band, allowing a great simplification on the BFN ‘s in terms of losses and overall complexity (less components and avoidance of additional input power sandwiches).

The modular configuration, fully meets the coverage gain, mass and stiffness requirements at minimum hardware and complexity, preserving the key navigation parameters such as phase and group delay stability.

A X-Band electronically scanned active phased array antenna operating in X band has been developed for an airborne SAR/MTI Radar, with an European team from France, Germany, Italy, Netherlands and Spain.

The antenna assembly comprises on its structure: a Radiating Unit consisting of parallel vertical planks (with radiating elements operating in linear polarization, active Transmit-Receive modules, RF distribution for simultaneous SAR and MTI operation, phase shifters and True Time Delay lines for SAR), a Radio Frequency Interface configuring paths and distributing the RF signals for the appropriate applications, a Beam Steering Computer, Power Supplies, a Roll Actuator, a Radome and its fairings, and a Cooling Unit providing a controlled coolant flow to the Radiating Unit.

Extensive simulation has been performed in order to obtain a design optimising the performance in transmission and in reception in SAR mode and in MTI mode, in terms of quality of radiation pattern, radiated power level and best trade-off between noise figure, active channel gain (and G/T) and RF signal distortion (TOI).
RF simulations and re-use or adaptation of available elements have made it possible to design and develop in a reduced time frame the different constitutive parts of the antenna assembly.

The various parts of the active phased array antenna have been separately tested and then assembled in the demonstrator. In the integration process, the panels and then the radiating unit have undergone a calibration process in a near field test range, in order to compensate for the discrepancies between the active paths and obtain a best fit to model the antenna behaviour with modules in saturated state. Radiation patterns for the various modes, in the frequency band and within the angular field-of-view have been obtained through near-field to far-field transform.

After mounting the protective radome, more field measurements have been performed in order to evaluate its RF effects, and through near-field procedures, evaluate the antenna assembly radiated power (EIRP) and quality factor (G/T).

The antenna assembly tests for the various radar modes have then been performed in a compact test range, demonstrating at the same time the reliability and sturdiness of the antenna concept.

Flight demonstration will soon take place on board a Fokker aircraft.

Two reconfigurable antenna frontend configurations of the PAMIR SAR/GMTI imaging radar, an experimental active phased array system with a very high simultaneous bandwidth of 1.82 GHz in X-band are described: an interferometric and a linear aperture configuration. Both configurations use identical subarrays. The broadband beamforming together with azimuth wide scan capabilities (±45°@1.82 GHz bandwidth) requires a switchable true time delay network with time increments equivalent to a fraction of a wavelength. PAMIR serves as an ideal airborne platform for future broadband SAR/GMTI research activities including interferometric image formation.

1. Introduction

The experimental SAR/GMTI system PAMIR (Phased Array Multifunctional Imaging Radar) was developed to investigate sophisticated reconnaissance abilities characterised by high flexibility and multi-mode operation. The tasks of this airborne system are SAR imaging at a very high resolution (10 cm at a range of 30 km and 30 cm at 100 km in the spot light mode) and high resolution ISAR imaging of ground moving targets. Due to the five parallel receiving channels array processing techniques like ground moving target indication (GMTI) via space-time adaptive processing (STAP) and electronic counter-counter measures (ECCM) will be feasible and implemented for wideband application. Further areas of research will be across track interferometric SAR (IfSAR) with high resolution capabilities and bistatic SAR.

2. Mechanical Designs

The reconfigurable antenna aperture chosen for the interferometric investigations consists of three antenna rows with 3 subarrays in each row. The size of the antenna aperture is adapted to the dimensions of the door of the TRANSALL, which serves as carrier platform. The antenna aperture chosen for an optimized SAR/GMTI performance is a linear array with up to 16 subarrays.

3. Design Of The Swithable True Time Delay Network A frequency-independent beam scan with ±45° azimuth coverage without bandwidth related beam squint forces the use of switchable true time delay networks up to the single element, if such high instantaneous bandwidths are envisaged. The demand on low insertion loss with a small ripple vs. frequency and low cost led to the choice of a microstrip solution with switching MMICs minimizing the number of line elements and so the number of discontinuities. The subarray TTD-line network with four switches and 6 delay lines per channel providing 16 different delay states with a time increment of 37 ps is described. An additional TTD network per subarray delays the broadband subarray sum signal resulting in a total time delay of 1700 ps for the the global delay network of the interferometric aperture configuration.

4. Design And Performance Of The Subarray Unit

In order to allow SAR experiments with variable antenna configurations, the subarrays are designed as autonomous replaceable units accommodating the turnable 16 element Vivaldi unit, 16 T/R channels with each having two receiver outputs, an MTI-RF combining net-work, a SAR combining network with the above described TTD-lines in each path, the RF calibration combining network and the control board. The Vivaldi part of each subarray can be turned mechanically over 30° in elevation by an individual step motor to adapt to the operated radar range of up to 100 km.

The paper will focus on the active antennas architectures (operating at X band and C band respectively) and on array-fed reflector based SAR with different concepts for flexibility at feedrarray level.

X-band Active SAR.

Alcatel Alenia Space Italy is currently providing the four active phased array antennas for the Cosmo/Sky-Med Mission, under ASI and Italian MOD contract. The SAR antenna works in X band and provides electronic beam steering in both azimuth and elevation planes, supporting H and V polarizations. The radiating aperture is of about 8 sqm patch array. It includes 1280 T/R modules with independent amplitude and phase control arranged in 40 tiles. Due to the large bandwidth and wide angle steering capability, the antenna is also equipped with variable true time delay lines to stabilize the beam pointing. Two independent high rate production lines for T/R modules and tiles have been implemented in the L’Aquila Plant (Italy). Current production capability amounts to 300 T/R modules and 8 tiles per month.

The proposed modular approach allow to implement different antenna dimensions : reduced scale aperture with different number of tiles can be considered to have antennas reduced in cost and size, without significant changes in electrical and mechanical I/Fs. A new antenna based on a 16 Tiles configuration is under study in this moment for a commercial program. Moreover the tile architecture allows to divide the antenna in separated apertures (two or more) to realize multibeam sensor capabilities.

C-band Active SAR.

In the frame of the Global Monitoring for Environment and Security programme (GMES) ESA is undertaking the development of a European Radar Observatory (Sentinel -1) for the continuation of SAR operational applications. The C-band SAR antenna is an active phased array with wide electronic steering capability along the elevation plane and limited steering angle along the azimuth plane. The radiating aperture is 1.4m x 10 m and includes 320 + 320 single channel T/R modules with independent amplitude and phase control. The SAR antenna operates simultaneously in H and V polarization in receiving mode and in H or V selectable polarization in transmitting mode. The antenna is organized in five mechanical panels, with the lateral wings foldable. Each wing is made up with 2 panels each.

Reflector based SAR.

AAS-I is currently engaged in studies aimed at investigating low cost multifeed concepts for SAR antennas suitable to Scansar, Stripmap and limited MTI functions. The baseline configuration optics consists of a three fold reflector of 1.4 x 5.8 m. Several operation modes are under investigation according to different types of feed clusters and beamforming networks. These are consistent with the realization of a limited number of ( fixed , switchable ) fan beams by multimode sectorial horns in single polarization or a cluster of dual polarization beams by using high efficiency small horns for an efficient segmentation of each sectorial horn. Regarding MTI, the use of a dual mode feed cluster along the azimuth plane for a dual aperture or a dual beam implementation is also under evaluation. Alternative concepts based on multimode apertures baked by VPD/PS in order to implement a limited continuous scan along the range plane is being investigated also.

An improved traveling wave (TW) expansion is utilized in the efficient high frequency (HF) analysis for predicting the collective radiation from large finite planar arrays. The current distribution over the whole array for a given excitation is first obtained from a moment method (MoM) solution of the governing array integral equation. A TW basis set is next used to represent this MoM based array distribution. The number of significant TW basis is less than 1% of the total number of array elements (or the number of array aperture field samples). Such a TW expansion is obtained via a convenient parameter estimation method. The importance of the TWs is that they facilitate an asymptotic HF analysis based on the uniform geometrical theory of diffraction (UTD) (P. Janpugdee, P. H. Pathak et al., IEEE AP-S Int. Symp., Washington, DC, Jul. 2005.), as well as the boundary diffraction wave (BDW) method (P. H. Pathak and P. Janpugdee, General Assembly of URSI, New Delhi, India, Oct. 2005.), respectively, to predict the "collective" radiation (or scattering) from the whole array at once. The collective radiation based on the UTD can be described in terms of just a few diffracted rays arising from the array element truncation boundary and Floquet rays from the array interior. In the case of the BDW, the same radiation is described in terms of the same interior Floquet ray contributions and an integral over the array element truncation boundary to describe the diffraction by the finite array. The BDW is essentially exact and requires a boundary integral that must be evaluated numerically, while the UTD is an asymptotic HF approximation and is essentially in closed form. Thus, the UTD is more efficient but requires ray tracing; on the other hand, the BDW is more robust since it needs no ray tracing from the array boundary. An asymptotic reduction of the BDW leads to the UTD as expected. Most importantly, both the UTD and BDW provide a physical picture for the array radiation (and scattering) mechanisms. In contrast, the conventional array element by element field summation approach is not only highly inefficient for large arrays, but also lacks physical insight. The TWs represent a complex realistic array distribution in terms of a much simpler basis set; the latter is crucial, since the UTD and BDW approach cannot be directly developed for realistic array distributions which generally show highly pronounced ripples near and at the array element truncation boundaries. Only for some hypothetical, smoothly varying and weakly tapered array distributions can the UTD (F. Mariottini, F. Capolino et al., IEEE Trans. Ant. Prop., vol. 53, pp. 608-620, Feb. 2005) and BDW be applied directly with slope diffraction corrections. Numerical results based on the UTD and BDW will be presented for large slot and printed antenna arrays, respectively, to demonstrate the excellent utility and accuracy of the present HF methods.

This paper presents pattern synthesis methods and calibration methods for the conformal array antennas. First, the author shows two types of the pattern synthesis methods, the projection method and the null points adjusting method. These have been developed in order to obtain low sidelobe characteristics for the conformal array antenna, for example, spherical circular array antenna. In the projection method, all element antenna positions are projected on a plane normal to the desired beam direction. Then, excitation amplitude and phase of each element antenna is determined from the desired amplitude distribution and the element spacing density on the projection plane, the element radiation pattern amplitude to the beam direction and the distance between the element antenna and the projection plane. On the other hand, the null points adjusting method so determines the excitation amplitude and phase of each element antenna that the null points of the obtained pattern coincides with them of the desired low sidelobe pattern, for example, the Taylor pattern by using the plane wave synthesis method. Furthermore, the author presents the rotating-element electric-field vector method (REV) method. This is one of the most famous and practical calibration method for phased array antennas and conformal array antennas and is able to measure relative amplitude and phase of an electric field vector radiated from each element antenna including mutual coupling effects among element antennas and environmental conditions around an array antenna by only measuring power of a composite electric field vector radiated from an array antenna The author also presents the modified REV method. This method can measure phase error of each bit of a digital phase shifter by using amplitude and phase of the composite field vector.