Search

There are ever increasing demands for additional capacity in wireless communications to handle voice, data, and wideband Internet applications. These demands are constrained by the bandwidth that was allocated to wireless communications. The spectral efficiencies in present day wireless systems hover around 1 bit/s/Hz. Bell-labs Layered Space-Time (BLAST) is a communication technique for achieving very high spectral efficiencies in highly scattering environments using multiple transmit and receive antennas. A measurement campaign was undertaken to assess the BLAST gains in spectral efficiency in the suburban outdoor environment for stationary subscribers. The measurements employed directive antennas to better control interference from adjacent cells. The measurements were performed over a narrow band at 2.44 GHz with five transmitting and seven receiving antennas, respectively. Extensive calibration methods, assisted by simulations, were developed to assure accurate results for the BLAST capacities of the measured remote subscriber sites. Initial results indicate that BLAST capacities of C/sub B//spl ges/38 bits/s/Hz at 20% of the measured locations and C/sub B//spl ges/24 bits/s/Hz at 50% of these locations are feasible, for reasonable link parameters and negligible interference.

We present the results of propagation measurements at 2.485 GHz for fixed wireless loops. Path loss measurements were performed and characterized at 43 subscriber locations around a base station antenna located on top of Crawford Hill in Holmdel, NJ. This suburban location is characterized by rolling hills, foliage, and terrain blockages. Temporal and horizontal motion path loss fluctuations were found to be uncorrelated, each characterized by a different Ricean distribution. Lower r.m.s. delay spreads were obtained with directive subscriber antennas than with omni-directional antennas. No substantial gain loss (less than 2 dB) of subscribers' directive antennas was observed. The effects of trees, with foliage, surrounding the base station upon the path loss and the ratio of scattered power to specular power are also examined. The distance exponent of path loss versus distance (about 1.5) was observed to be less than free-space. Diffraction loss from hilltop trees, shadowing the base station, are suspected to be the cause. This loss decreases as the remote moves further away and comes out of the shadow. Scattered power from directions other than line-of-sight was observed to be as high as one half of the specular contribution when tree scattering near the base station was significant.

Future cellular and fixed wireless loops (FWL) systems employing highly directive base stations antennas (5 to 8 degrees beamwidth) and moderately directive subscribers' antennas (15 to 25 degrees beamwidth) offer very large capacity due to reduced interference. An important property of the environment of such systems is the interference caused by scattering of the signal from a subscriber transmitter into directions other than the direct line of sight between the subscriber and the base station. In particular, for multibeam base station applications scattering could arrive at the base station in the direction being used by a different beam, resulting in interference that is difficult to reduce by normal nulling techniques. Thus interference can be highly dependent on the incident power density pattern (IPDP) caused by scattering of the signal radiated from the subscriber. We discuss herein the use of a uniformly illuminated array accompanied by electric field deconvolution to measure the crucial IPDP with the same performance as a low sidelobe array of the same size. The mathematical correction technique uses deconvolution of the measured complex electric field pattern with that of the antenna in free space by means of the Fourier Series and limiting the range of Fourier coefficients to those that are not negligible in the free space pattern. Application of the technique to an experimental uniform array with a 2.5 degree azimuthal beamwidth shows the practicality of the deconvolution with real antennas in real environments. The improved resolution and accuracy provided by Taylor weighting versus unweighted deconvolution when trying to measure weak scattered components in the presence of a nearby strong specular component is demonstrated. The IPDP was measured from many sites surrounding a suburban base station. A plot of the cumulative distribution of the ratio of the widely scattered power to that within a prescribed beamwidth summarizes the result of using the deconvolution technique on this experimental data.

An architecture for a combined personal, mobile, and (possibly) loop access microcellular communications system in which calls are digitized and packetized is presented. A network of microcells is utilized to communicate with the various subscribers via radio. A fast handoff/resource assignment process is introduced. It is subscriber-controlled and is jointly performed by the subscribers and base station employing the resource auction multiple access (RAMA) algorithm. In this deterministic algorithm, resource utilization is independent of traffic load. Using selected GSM parameters with RAMA, for the purpose of illustration, it is shown that, in microcells with delays of 1 mu s, 1080 resource assignments per second are feasible: and with delays as high as 45 mu s, 216 assignments per second are feasible. >

Research projects involving fiber-connected microcell base units, measurements and models of propagation in urban microcells, and radio link performance in urban microcells are summarized. Developments in microcell system issues, such as channel assignments and handoff in microcells and architectures for combining microcells with traditional macrocells, are discussed. Research on both indoor radio propagation and wireless indoor systems is reviewed. >

Cumulative distribution estimates are presented for the one-second-average bit-error-rate (BER) which would be experienced on UHF links in a microcellular environment. It is assumed that users (subscribers) are moving at speeds ranging from quasi-stationary to 96 km/h. The authors concentrate on lineal microcells, in which there is always a line-of-sight path between the microcell and subscriber antennas, in addition to road-reflected and/or building-reflected rays. Using extensive UHF data collected in Manhattan, the signal and interference levels are characterized, point-by-point, for a subscriber moving in a lineal microcell. Combining these measurement-based characterizations with a mathematical model of modem performance, the cumulative distribution of the BER is computed. For moving subscribers, simple diversity reception gives large benefits in grade-of-service (percentage of time that the BER is below a specified threshold value) and uplink and downlink transmission qualities tend to be similar, with the latter being somewhat better. Performance estimates based on a multiray model are generally similar to those based on measured data and tend to be conservative. >

A propagation experiment has been designed and conducted at 900 MHz and 11 GHz to characterize microcell channels using various antennas at two distinct frequencies. It is found that propagation in rural areas is dominated by interference between the direct, line-of-sight ray and a specular roadway-reflected ray. In urban areas, the addition of four specular wall-reflected rays adequately represents microcell propagation. The dependence of mean power falloff, measured mean power and calculated power on distance was determined. The lambda /2 scale microvariations of the received power are reduced compared to the variations in present cellular radio systems. For urban sites using omnidirectional base and mobile antennas, the RMS delay spread due to road- and wall-reflected rays was obtained from a six-ray model. Using a 20-dB horn for the mobile antenna can reduce this delay spread. >

A single-mode inline variable optical attenuator utilizing optical fiber tapers is discussed. Optical attenuation control is achieved through axial separation between two tapers with beveled endfaces aligned within a ceramic sleeve. Increased axial separation causes increased coupling loss between the two taper fundamental modes. Insertion loss of 75 dB with continuous 0.05 dB resolution, and reflection return loss of better than 58 dB have been realized. Theoretical studies indicate low polarization and wavelength dependence, and inherent insertion loss of >

A numerical analysis of an infinite phased array of open rectangular waveguides has been made which includes the effects of wall thickness. Two planes of scan, the H and quasi- E planes, have been considered. In these cases, the general vector problem can be expressed in the form of a scalar one-dimensional Fredholm integral equation of the first kind. The approximate fields obtained numerically from the integral equation are used for the evaluation of the input complex reflection coefficient. A variational expression for the reflection coefficient is developed and used for improving the accuracy. Numerical results for the H and quasi- E plane scans are presented as a function of wall thickness and scan angle. Agreement with experimental results is very good.

A series of measurements was conducted in both rural and semiurban areas to measure radio propagation characteristics at microwave and millimeter-wave frequencies in a microcellular mobile radio environment where transmission is essentially line-of-sight (LOS). Measurements were made at 11 GHz with a fixed-site antenna located at the side of the road at heights of up to 30 ft. Data have been taken with both omnidirectional and directional antennas on the mobile, to compare received signal characteristics. It was found that, in rural and semiurban environments, propagation is dominated by interference between the direct ray and a specular, roadway-reflected ray. The average power received does not decay any faster than 1/r/sup 2/ over the measured range of transmitter-to-receiver separation. A wideband communication channel looks feasible, since the dual-ray interference effects and the random components of the received signals can be overcome by simple methods such as higher antenna directivity or space diversity. >

Describes the benefits and outline of resource auction multiple access (RAMA). Discusses RAMA application to extended IS-54 (uplink frequency-time utilisation). The advantages of RAMA are: fast and efficient multi-priority medium access; can easily double system voice spectrum utilization through statistical speech multiplexing; can provide variable bandwidth on demand; collision independent throughout, full (100%) utilization of all communication resources (slots) irrespective of traffic load; flexible control of priority/grade of service - fairness; and can be utilized for speech and data multiplexing.

Resource auction multiple access (RAMA) is a technique for rapidly assigning communication resources (e.g., TDMA time slots) in a wireless environment. This scheme is deterministic in that, barring errors caused by fading and noise, a resource assignment is made in each and every assignment cycle. Previous papers reported on the call assignment capacity of this technique and also demonstrated its capabilities for statistical speech multiplexing. In these studies, we used a selected set of GSM parameters while assuming perfect radio channel operation. In this paper, we extend the previous work in two directions: (1) we examine the effects of channel fading and noise; and (2) we consider both GSM and IS-54 cellular environments. Propagation is modeled, for purposes of this study, as flat Rayleigh fading, with mean propagation loss as a parameter. We focus on the subscriber-to-base link, which is most vulnerable to RAMA errors, and consider several forms of diversity at the base. We also consider two RAMA modes (call assignment only and statistical speech multiplexing). RAMA signaling is two-level amplitude shift keying (ASK). We show that a simple form of two-branch diversity can be very effective. Our final results show the probability of assignment error as a function of mean propagation loss for different diversities, systems, and average transmit powers. In IS-54, for example, simple diversity and an average transmit power of 100 mW yields error probabilities of 58 or less for losses as high as 154 dB. We also show that such error rates are more than adequate for purposes of call assignment. >

Recent breakthroughs in the application of information theory have shown great capacity increase by deploying multiple transmitting and multiple receiving antennas as compared to conventional single antenna systems. This work presents an experimental verification of the theoretical capacity prediction and a summary of measurement results for 16 fixed wireless locations in suburban New Jersey.

There are ever increasing demands, in wireless communications, for additional capacity to handle voice, data and wideband Internet applications. These demands are constrained by the bandwidth that has been allocated to wireless communications. A novel unorthodox concept to substantially increase the spectral efficiency in highly scattering environments, named BLAST (Bell-labs LAyered Space-Time), has been introduced (see Foschini, G.J., Bell Labs Technical Journal, vol.1, no. 2, p.41-59, 1966; Foschini, G.J. and Gans, M.J., Wireless Personal Communications, vol.6, p.311-35, 1998). A measurement campaign was undertaken to assess the BLAST gains in spectral efficiency in the suburban outdoor environment for stationary subscribers. The measurements employed directive antennas for better control of interference from adjacent cells. Extensive calibration methods, assisted by simulations, were developed to assure accurate results for the BLAST capacities of the measured remote subscriber sites.

Future cellular and fixed wireless loops (FWL) systems employing highly directive base stations antennas (5 to 8 degrees beamwidth) and moderately directive subscribers' antennas (15 to 25 degrees beamwidth) offer very large capacity due to reduced interference. An important property of the environment of such systems is the interference caused by scattering of the signal from a subscriber transmitter into directions other than the direct line of sight between the subscriber and the base station. In particular, for multibeam base station applications scattering could arrive at the base station in the direction being used by a different beam, resulting in interference that is difficult to reduce by normal nulling techniques. Thus interference can be highly dependent on the incident power density pattern (IPDP) caused by scattering of the signal radiated from the subscriber. We discuss herein the use of a uniformly illuminated array accompanied by electric field deconvolution to measure the crucial IPDP with the same performance as a low sidelobe array of the same size. The mathematical correction technique uses deconvolution of the measured complex electric field pattern with that of the antenna in free space by means of the Fourier series and limiting the range of Fourier coefficients to those that are not negligible in the free space pattern. Application of the technique to an experimental 25 degree beamwidth uniform array of 320 elements shows the practicality of the deconvolution with real antennas in real environments. The improved resolution and accuracy provided by Taylor weighting versus unweighted deconvolution when trying to measure weak scattered components in the presence of a nearby strong specular component is demonstrated.

Recent breakthroughs in the application of information theory have demonstrated great increases in capacity by deploying multiple transmitting and multiple receiving antennas compared to conventional single antenna systems. An experimental verification of the theoretical capacity prediction is presented. The propagation study was performed in a controlled free-space environment. The near field capacity was measured and modelled. Measurement results match very well with the theory.

There are ever increasing demands, in wireless communications, for additional capacity to handle voice, data and wideband Internet applications. These demands are constrained by the bandwidth that was allocated to wireless communications and practical limits on maximum power and minimum cell size. The spectral efficiencies in present day wireless systems hover around 1 bit/s/Hz. BLAST (Bell-labs LAyered Space-Time) is a communication technique for achieving very high spectral efficiencies in highly scattering environments using multiple transmit and receive antennas. A measurement campaign was undertaken to assess the BLAST gains in spectral efficiency in various environments. The measurements were to be performed over a narrow band at 2.44 GHz with five transmitting and seven receiving antennas, respectively. The effects of thermal noise on the accuracy of measured BLAST capacities are investigated.

The excess capacity of the access mechanism in resource auction multiple access (RAMA) is evaluated for statistical multiplexing of speech. The aim is to exploit this excess capacity to increase the number of subscribers for a given bandwidth. It is shown that, for packet dropping probabilities of 1%, RAMA can have a multiplexing gain as high as 2.56 with cast speech detection and 2-2 with slow speech detection. The simulation results also indicate that RAMA is suitable for cast handoffs and resource allocation with statistical speech multiplexing over areas with a wide variety of cell sizes

The importance of the incident power density pattern (IPDP) for future fixed wireless loop (FWL) systems is discussed. Also an investigation into some of the important S/I implications of the IPDP for proper system capacity simulations is presented.

A novel method is presented to address the need for fast and flexible access, handoff, and resource assignment in future high capacity wireless personal communications systems (PCSs). In one deterministic algorithm, called resource auction multiple access (RAMA), fast handoff, access and resource assignments are incorporated. Resource use and throughput are unaffected by heavy traffic demand as in traditional contention schemes. Using selected GSM parameters with RAMA, it is shown that, in microcells with propagation delays of 1 μs, 2160 resource assignment/s are feasible. With delays of 25 μs, 648 assignment/s are feasible.

A 4-PSK microstrip modulator operating at data rates of up to 800 Mbits/s at 12 GHz is described. The circuit is made of two cells in series. Each cell consists of a 3-dB branch-line hybrid coupler and two BTL p-i-n diodes. One cell provides 0° or 90° phase shifts, and the Other 0° or 180° phase shifts, so that four carrier phase values are obtained by appropriately exciting the two cells. The switching time of each cell is 200 ps. Simultaneous switching of both cells increases the switching time to a maximum value of 400 ps. The phase waveforms are nearly rectangular at the above-mentioned data rates. RF insertion loss is 1 dB ± 0.1dB for the four phase values over the 11.7-12.2 GHz frequency band.

A dually polarized narrow ( ) beam antenna which provides a geosynchronous arc coverage of 60\deg is proposed and analyzed. To track the geosynchronous arc accurately, the properly oriented antenna produces a conically scanned beam by means of a linear array of feed horns with bias cut apertures illuminating a pair of parabolic cylinder reflectors in an imaging arrangement. This design, with reduced size array and singly carved reflectors, is relatively simple to construct. Calculations for a 0.35\deg beamwidth Ku band earth station antenna show a 0.05\deg pointing accuracy with scan loss due to aberrations less than 1 dB and little pattern degradation throughout the scan region. For multiple beam capability, a Rotman lens is proposed and analyzed as a means of exciting the feed array. It allows communication with multiple satellites spaced as close as 1\deg .

The possible existence of blind spots, due to forced aperture resonances, is demonstrated theoretically and experimentally at 14 GHz in rectangular grid arrays of tapered rectangular horn elements with oversized (overmoded) apertures. The dominant contributor to these blind spots is the resonance of the equivalent transverse magnetic ( TM_12 ) waveguide mode. The numerical results based upon our approximate methods of formulation and solution of the array boundary value problem agree well with experiments. A method for eliminating the blind spots by proper taper design is presented and experimentally verified. The analytical methods used in this work could possibly be extended to conical horus and other tapered-structure arrays as well as other periodic grid geometries. New simple and accurate formulas for the zeros of Bessel and Legendre functions and their derivatives are obtained as a result of this analysis.

An extensive CRAY-computer-based simulator using the propagating beam method for the accurate evaluation of the performance of optical fiber up-tapers is discussed. Up-tapers with and without straight tips and with arbitrary index profiles and tapering functions can be evaluated. Results of the evaluation of various 1.5-cm-long optical up-tapers followed by a straight tip and drawn from special preforms with various index profiles are quite encouraging. They show that beam expansion ratios in the range of 5-10 are feasible with a corresponding insertion loss per taper of less than 0.01-0.025 dB. Tapers shorter than 1.5 cm with insertion loss in the same range are also feasible but have slightly smaller beam expansion ratios. >

We have experimentally demonstrated the applicability of optical fiber tapers at 1.3 μm as a simple and practical means of achieving beam expansion in a self-aligned unitary structure. These devices have a standard 8.1-μm core at one end which gradually increases in cross section to the order of 100 μm at the other end. Tapers are envisioned as basic building blocks in a multitude of single-mode optical components. Experiments performed on a batch of eight tapers verified, to experimental accuracy, that no significant amount of mode conversion or beam distortion takes places in the taper. The insertion loss of the taper was found to be under 0.1 dB. The sensitivity of the excess loss between two tapers to lateral and axial displacement is greatly reduced as compared to that between two single-mode fibers. For example, the 0.5-dB loss point of taper coupling corresponds to a 10-μm lateral displacement and a 700-μm axial displacement versus respective 1.6-μm and 36.5-μm displacements for fiber coupling. The increased sensitivity to angular displacement is within practical limits.

Multisatellite communication may require coverage of a segment of the geosynchronous arc as large as 60\deg . A properly oriented ground station antenna can cover the above segment by means of a beam whose axis traces a conical surface (i.e., conically scanned beam). The antenna consists of two linear arrays feeding an imaging reflector arrangement through a polarization diplexer. This diplexer separates the incoming wave into two orthogonal polarizations without loss. In general, the polarization of the wave reflected by the diplexer does not remain the same as that corresponding to the feed array as the beam is scanned (polarization mismatch). This mismatch can cause an appreciable degradation of the antenna gain when the feed array is scanned in certain directions. For example, a ground station located at Seattle, WA could experience 3.6 dB reduction in gain due to polarization mismatch when the antenna beam is scanned along a 60\deg segment of the geosynchronous satellite arc. The above polarization mismatch loss can be substantially reduced by the use of two properly inclined polarization rotators. These devices are sufficiently broadband to cover the entire 12/14 GHz satellite communication band. Design formulas for the inclination angle between the polarization rotators are presented. Applying these formulas we have reduced the above-mentioned polarization mismatch loss to less than 0.01 dB.

The performance of step index up-tapers subjected to certain index perturbations in investigated by computer simulation. Concentric symmetric and asymmetric index profile perturbations as well as localized spherical, called regions bubbles, having a different index are studied. It is found that some low-valued perturbations of small scale size (less than a wavelength) do not alter the optical fiber performance. However, the amount of beam expansion in the up-taper may be ultimately limited by such perturbations. Excessive magnification in the presence of perturbations can cause a distortion of the quasi-Gaussian shape and symmetry of the pure step-index fundamental mode as well as increase the excess loss and degrade the lateral and axial displacement tolerance improvements of the up-tapers. For example, an index variation of Delta n approximately=10/sup -5/ in conjunction with a perturbation scale size of 40 mu m can be detrimental. Bubble-type perturbations could also cause similar problems. A single 12.8- mu m-diameter bubble with a relative index of -0.02, embedded in a 128- mu m-diameter step index core with relative index of 0.5%, produced a 0.3-dB excess loss. >

A quasi-optical and wide-angle (of incidence) polarization rotator is described and analyzed. It consists of three or more parallel, equispaced wire-grid polarizers whose wires are oriented in different directions. With the proper choice of the number of grids, their spacings, and their relative orientations, such a polarization rotator is capable of operating over a broad frequency band and a wide range of angles of incidence. A seven-grid 90\deg polarization rotator was designed to operate over a 42 percent relative bandwidth for a \pm 60\deg range of angles of incidence with less than 0.1 dB transmission loss. Various designs based upon our analysis are given. The impetus for this work is the specific application of this polarizer in a dually polarized multibeam satellite ground station antenna employing a linear array feed. The design of a seven-grid polarization rotator for the ground station antenna is presented.

A method for coupling single-mode fiber to laser diodes with high efficiency and greatly reduced alignment sensitivities is discussed. It consists of a spherical ruby lens and a single-mode optical fiber up-taper. Using a double-channel planar buried heterostructure (DCPBH) laser, more than 60% of the light from the lens is coupled into a single-mode fiber via the taper. The lateral and axial tolerances of the lens-taper arrangement are reduced by factors of 20 and 1000 times, respectively, compared with laser coupling to a lensed single-mode fiber. These relaxed alignment tolerances have the potential to allow the taper to interface with a hermetically-sealed laser-lens package via a window and a connector on the package exterior. >

We investigate the performance characteristics of single-mode optical fiber tapers. These devices have a standard single-mode fiber geometry at one end and gradually increase in cross section so that the size of the core at the other end is comparable or greater to that of a multimode fiber. These tapers effectively expand the single-mode spot size and are envisioned as basic building blocks in a multitude of optical components. Analytical and experimental studies, at \lambda = 0.63 \mu m, show that the dominant mode is preserved while traveling through the taper, from either direction. The excess coupling loss between two tapers is less than 0.1 dB. The sensitivity of the excess loss to lateral and axial displacements for two coupled tapered sections is greatly reduced compared to that between two single-mode fibers. The sensitivity to angular displacement is increased but is within practical limits. For example, for an excess loss of 0.5 dB, the maximum allowed lateral displacement is 3.1 μm for taper coupling, while only 0.73 μm is allowed in the case of fiber coupling. An axial displacement of 291 μm for taper coupling produces 0.5 dB loss while a displacement of only 16.5 μm produces a 0.5 dB loss for fiber coupling. For the same loss, angular displacements of 0.42° for the tapers and 1.77° for the fiber are allowed.

A theory of linear least-mean-squares equalization in digital data communications operating over two coupled linear dispersive channels, with particular application to dually polarized terrestrial radio systems is presented. We jointly optimize transmitter and receiver matrix filters when the inputs are two independent quadrature-amplitude-modulated data signals of fixed total average power. Formulas for minimum total mean-square error, upper bounds on probability of error, matched filter bounds, and the Shannon limit are provided. Using this theory in conjunction with a proposed propagation model for the dually polarized radio channel, we provide estimates of probability distributions of the data rates that can be supported by the optimum receiver structures as well as the jointly optimized transmitter-receiver. The distribution of data rates predicted from the Shannon limit as well as from the matched filter bounds are also given. For the dually polarized radio channel, we find that linear equalization structures can practically eliminate the effects of cross-polarization interference contributed by the antennas and their feeds. At very low outage probabilities, the combined-channels data rate achieved with the jointly optimized structures is about 4 b/s/Hz less than predicted from their matched filter bounds. Also, the combined data rate associated with the Shannon limit is about 3 b/s/Hz higher than predicted by the matched filter bound at very low outage probabilities. Sensitivity analyses reveal that about 1 b/s/Hz can be gained for each 3-dB increase in carrier-to-noise ratio. For each order-of-magnitude decrease in the required error rate, about 0.3 b/s/Hz is given up.

Recent analysis/simulation studies have quantified the multipath outage statistics of digital radio systems using ideal adaptive equalization. In this paper, we consider the use of finite-tap delay line equalizers, with the aim of determining how many taps are needed to approximate ideal performance. To this end, we assume an M -level QAM system using cosine rolloff spectral shaping and an adaptive equalizer with either fractionally spaced or synchronously spaced taps. We invoke a widely used statistical model for the fading channel and computer-simulate thousands of responses from its ensemble. For each trial, we compute a detection signal-to-distortion measure, suitably maximized with respect to the tap gains. We can thereby obtain probability distributions of this measure for specified combinations of system parameters. These distributions, in turn, can be interpreted as outage probabilities (or outage seconds) versus the number of modulation levels. A major finding of this study is that, for the assumed multipath fading model, very few taps (the order of five) are needed to approximate the performance of an ideal infinite-tap equalizer. We also find that a simple, suboptimal form of timing recovery is generally quite adequate, and that fractionally spaced equalizers are more advantageous than synchronously spaced equalizers with the same number of taps. This advantage is minor for rolloff factors of 0.5 and larger but increases dramatically as the rolloff factor approaches zero.

The design and implementation of a high-speed, lowpower, and TTL-compatible driver for a 12 GHz digital p-i-n phase shifter are shown. The switching performance of the phase shifter, combined with its driver in a compact module, is presented. Measurements of 40 phase shifter/driver modules show that the longest time required to change phase states (switching time) is under 8 ns. The maximum driving power consumption is 36 mW at a phase switching repetition rate of 1 MHz. Each module can handle 800 mW of CW RF power. The performance and compactness of the phase shifter/driver make this module especially attractive for satellite and airborne applications.

The feasibility of using an appropriately squinted linear scan in narrow multibeam satellite ground station antennas employing phased arrays is demonstrated. This linear scan has the potential of reducing the complexity of a narrow-beam planar array to that of a linear array. Calculations for such antennas placed at cities throughout the U.S. show tht the peak beam pointing error in covering the 70\deg W to 130\deg W geosynchronous equatorial arc (GEA) is under 5/ 1000th of a degree. Communication at a 300 MBd rate in the 12/14 GHz band can be made feasible, for a grating lobe-free scan and 0.5\deg beamwidth antenna, by using a relatively simple time equalization.

The signal-to-noise ratio (SNR) statistics of a crosspolarized interference cancellation scheme is derived in terms of the statistics of nondispersive fading of the signals in the radio channels. The results are applied to certain digital radio channels subjected to nondispersive fading. It is found that error rates not exceeding 10-6could be maintained during 99 percent of the duration of a 30 dB fade in a 26.4 mi path. The method and data presented support the feasibility of cross polarization interference cancellation. They are applicable to the determination of SNR budget in the transmission system.

In this paper, a method is introduced for determining the practically realizable matching characteristic of a given array radiating element configuration over an assignable scanning region. The matching scheme assumes that a simple impedance change and a reactive element may be introduced in each channel, and \pi -type reactive networks may interconnect adjacent element channels. A computer program has been developed to optimize the network components and evaluate the array match. Numerical results for a square waveguide array are presented. The method developed here provides a practical guideline in the evaluation and development of physically realizable array element configurations. The effects of this matching scheme on the radiation pattern and polarization characteristics of a single excited element in the array are discussed as well.

The analysis of the attenuation due to ohmic losses in periodic linear arrays of metallic cylinders, ribbons, and slots in a metallic ground plane is presented. Calculations indicate that the loss per unit length of the ribbon and cylinder arrays is comparable to that of a standard rectangular waveguide operated in the TE/sub 10/ mode. With a proper choice of parameters, the loss per unit length of the slot array can be brought to within a factor of 2 of that of rectangular waveguides.

In this work, a planar phased array of circular waveguides arranged in an equilateral triangular grid is analyzed. The boundary value problem is first formulated rather generally in terms of a vector two dimensional integral equation for an array of elements that are arranged in a doubly periodic grid along two skewed (nonorthogonal) coordinates. Dielectric plugs, covers, and loading, as well as thin irises at the aperture, are accounted for in the formulation. Numerical solutions are obtained by using the Ritz-Galerkin method to solve the integral equation. Excellent agreement with experimental measurements using a waveguide simulator is observed. The existence of forced surface wave phenomena in equilateral triangular grid arrays and their strong dependence upon the mode of excitation is also demonstrated. These phenomena are shown to exist at isolated points in the scan coordinates. Reflection characteristics as well as the polarization characteristics of the radiation pattern are illustrated at selected planes of scan for both linear and circular polarization excitation.

The use of dielectric materials for the hardening and matching of phased-array antennas in recent years has shown that a more complete understanding of the effects of these materials upon the array performance is necessary. The characteristics of fully loaded, plugged, and sheath covered circular waveguide phase arrays are analyzed and discussed. Numerical solutions of the boundary-value problem are verified by experimental and convergence tests. Particular emphasis is placed on the study of (forced) surface wave resonance effects. Three different cases for surface wave resonances were obtained. These include the case in which surface wave resonances are present in the absence of dielectrics, the case in which they are trapped by the presence of dielectric plugs, as well as the case in which waves are trapped by the presence of a dielectric sheath. The surface wave resonance due to the plug is shown to vanish for certain "bandpass" ranges of plug thickness which repeat periodically for a single trapped waveguide mode. On the other hand, the surface wave trapped in the sheath exhibits no "bandpass" characteristics. Instead, multiple surface wave resonances occur with increasing sheath thickness. Finally, the surface wave resonances observed here appear at isolated points in the scan plane.

The feasibility of correcting phase aberrations of spherical reflector antennas with planar array feeds has been investigated. This type of feed seems to be particularly attractive for applications requiring several closely spaced beams. A synthesis procedure for the array excitation has been developed which minimizes the mean-square error with respect to a prescribed reflector illumination. This method was applied to the analysis of a spherical reflector antenna with a 642\lambda diameter and an effective F number of 0.9.

A systematic method for designing a phased array element that is matched over a wide scan region is presented. Measured coupling coefficients from a small array of trial elements, combined with computer simulation, yield the design parameters of an individual element matching obstacle. These obstacles will optimize the match versus scan characteristics when properly inserted within the element channels. The method presented has been used to optimize the match versus scan characteristic of a circularly polarized coaxial horn type of element over a 15 percent bandwidth. A close agreement between the predicted and measured returned powers has been observed over the frequency band for a broad scan region.

Electromagnetic scattering problems, including waveguide discontinuity, phased array, and scattering (exterior type) problems, are frequently described by integral equations that can be solved by the Ritz-Galerkin or generalized method of moments. Under appropriate conditions, it has been shown that reciprocity and variational properties are, in fact, preserved in the approximate solutions. It is shown here that in the Ritz-Galerkin method, energy is also conserved under certain conditions, even in those scattering problems where reciprocity does not exist. Hence energy conservation cannot serve as a check for accuracy of a numerical solution obtained by the Ritz method or other related methods.

Forced surface wave resonances are shown to occur in dielectric-free 45-degree triangular-grid circular waveguide phased arrays. Similar surface wave phenomena were observed in an identical grid array of rectangular waveguides in the H -plane of scan. In the circular waveguide array, however, the resonance is observed in the E -plane and is shown to be an isolated point. This isolation is further illustrated by observing certain vector symmetries in the radiation patterns.