Phased arrays have a long history in radio astronomy. Large, sparse synthesis arrays have been in use for decades to capture high-resolution images of deep space objects. More recent work has extended the range of applications to other types of arrays, including aperture arrays (AAs) and phased array feeds (PAFs) for multibeam reflector antennas. The extreme sensitivity required for astronomical instrumentation is driving advances in numerical electromagnetic modeling, design optimization of large arrays, low noise amplifiers, minimization of receiver noise, cryogenic PAFs, array calibration, optimal beamforming, interferometric imaging, and array signal processing algorithms for radio-frequency interference mitigation. We give an overview of research progress, current and planned array-based instruments, and open challenges in these areas related to the new generation of sparse arrays, PAFs, and AAs that are in development for astronomical observatories around the world.

A major research emphasis in current radio astronomy instrument development is the use of phased array feeds (PAF) to provide radio telescopes with larger fields of view. Statistically optimal beamformers have been shown to provide high sensitivity for PAFs, however, inability to precisely control beampattern details (even in sidelobes) is problematic. Enforcing beampattern structure with a deterministic beamformer is useful. A hybrid beamforming approach is presented in this paper, which offers the best in sensitivity for a desired amount of beampattern control. Designing any beamformer with simulated calibration data is inadequate for PAF operation and introduces distortion in the final beampattern. A transformation step is required to reduce such effects.

Adaptive interference canceling beamformers are known to suffer from beampattern distortion, particularly when interferers approach the nominal mainlobe. In some applications such as radio astronomy, even small pattern variations can be problematic. We address the issue of computing power spectral density (PSD) estimates at the output of a beamforming array in the presence of strong moving interference. A bias corrected PSD estimation algorithm is introduced which uses subspace projection methods to cancel interference at the beamformer. A correction step eliminates both pattern-distortion-induced PSD bias and spatial response errors over the long-term PSD averaging window.

The LOFAR astronomical array currently under development in the Netherlands produce synthesis images of the most distant (and thus oldest) space objects by observing at unusually low frequencies (30-250 MHz) over a large aperture (100 km) using many antennas (on the order of 10,000). This presents some significant challenges for sensor calibration because at these frequencies Earth's ionosphere acts as a random refractive sheet which over the large aperture induces source direction dependent gain and phase errors that must be estimated and calibrated out. Existing array self calibration algorithms used at higher frequencies by other radio astronomy instruments such as the very large array (VLA) do not address direction dependence and does not work in the LOFAR environment. A new algorithm called "demixed peeling" is presented and assessed as a solution to the direction dependent calibration problem for large distributed sensor arrays

In radio astronomy, cosmic sources are observed which are many orders of magnitude weaker than the telescope system noise level. The necessary sensitivity is achieved by large telescope collecting areas, long integration times, and large bandwidths. In the coming two decades, telescopes are planned which are even one to two orders of magnitude more sensitive than the current generation. Examples are the Low Frequency Array (LOFAR), currently under construction in the Netherlands, and the Square Kilometer Array, for which the envisaged start of construction is in 2012. For this next generation of telescopes, a dynamic range in the sky maps of over 10/sup 6/ is required. In order to reach these numbers, accurate calibration is needed. As these telescopes will observe with relatively large bandwidths, and because of the changing spectrum environment, interference mitigation techniques become increasingly important. Approaches for calibration and interference mitigation are presented, and results from the LOFAR initial phased array test station (ITS) are given.

Radio astronomical observations of important L-band spectral lines must often be made at frequencies allocated to pulsed air surveillance radar in the 1215-1350 MHz band. Such pulsed interference must be dealt with at the Green Bank Telescope (GBT) and other observatories by "blanking" to remove corrupted data. This paper presents a new algorithm which improves aircraft echo detection by using a Kalman tracker to follow the path of a sequence of echoes. This Bayesian method enables more sensitive weak echo detection. Track information is used to form a spatial prior probability distribution for the presence of echoes in the next antenna sweep. A lower detection threshold is used in higher probability regions to pull out low level pulses without increasing the overall probability of false alarm detection.

An algorithm is presented for estimating complex transmit and receive array calibration gains in a MIMO communications system. Calibration values are computed from a series of distinct channel transfer function matrix estimates over a time-varying channel. The new method, called RIMM (robust iterative multiple-channel MUSIC), is an extension of a previous algorithm proposed by B. Friedlander and A.J. Weiss (see Proc. ICASSP, p.2681-4, 1988). Several significant weaknesses in the earlier algorithm are demonstrated, and robust solutions are presented. An auto-calibration experiment is presented using 2.43 GHz indoor channel probing data samples.

Spatial filtering and subspace projection methods have been proposed for removing interference signals at radio astronomy (RA) imaging arrays. Given the fact that RA signal levels are usually below noise levels, and since high gain antennas significantly reduce the interference to noise level at antenna feeds, it is often difficult to form the accurate interference signal parameter estimates. By adding a few (1-3) low gain "auxiliary" antennas to an imaging array, it is possible to overcome this problem. Using low cost auxiliaries with an existing array can improve interference rejection by tens of decibels. New extensions to subspace projection spatial filtering methods are presented, along with simulated results for performance comparison.

Detailed performance assessment of space-time coding algorithms in realistic channels is critically dependent upon accurate knowledge of the wireless channel spatial characteristics. This paper presents an experimental measurement platform capable of providing the narrowband channel transfer matrix for wireless communications scenarios. The system is used to directly measure key multiple-input-multiple-output parameters in an indoor environment at 2.45 GHz. Linear antenna arrays of different sizes and construction with up to ten elements at transmit and receive are utilized in the measurement campaign. This data is analyzed to reveal channel properties such as transfer matrix element statistical distributions and temporal and spatial correlation. Additionally, the impact of parameters such as antenna element polarization, directivity, and array size on channel capacity are highlighted. The paper concludes with a discussion of the relationship between multipath richness and path loss, as well as their joint role in determining channel capacity.

In MIMO wireless communication using space-time coding techniques, it is important to understand channel characteristics so that optimum coding schemes, array configurations, and performance analysis can be developed. Increasingly, understanding of MIMO channels is being aided by direct channel probing experiments to measure transfer matrix, H. An alternate channel representation based on propagating ray parameters has many advantages. We propose an algorithm to compute high precision estimates of channel ray parameters directly from H measurements. Examples are presented where rays are identified, and departure angles, arrival angles, and complex path gains are estimated.

This paper introduces a method for experimentally probing indoor wireless channels to compute MIMO transfer function matrices, H. The data is gathered using rotating transmit and receive antennas to observe the multipath ray structure of the channel. H can be computed for any arbitrary transmit and receive array sizes and geometries using the single probe data set. Estimating H is crucial for space-time coding communications algorithm performance analysis. Example results are presented.

Blind point-source image restoration refers to the problem of high resolution recovery of point-like sources which are blurred by an unknown point spread function (PSF). Applications include astronomical star field localization, magnetoencephalogram brain current imaging, and seismic deconvolution. This paper shows that with suitable constraints the problem can be cast in the language of subspace decomposition as used in blind signal copy algorithms for digital wireless communications. Assuming a separable PSF, we propose a deterministic, non-iterative ESPRIT-like solution to the restoration problem. We next extend the algorithm to non-separable PSFs by approximating them as a series expansion of a few separable components.

This paper presents a new non-iterative, closed-form approximation to the maximum entropy (ME) image restoration method. A fast frequency domain implementation of this closed form approach is developed for the case of circular convolutional blur. This result dramatically reduces the computational demands compared to conventional iterative ME algorithms such as MART. Some limitations and advantages of ME restoration are investigated, including its dismal performance for high resolution restoration of decimated or randomly sampled blurred observations.

Most previously proposed statistical models for the indoor multipath channel include only time of arrival characteristics. However, in order to use statistical models in simulating or analyzing the performance of systems employing spatial diversity combining, information about angle of arrival statistics is also required. Ideally, it would be desirable to characterize the full spare-time nature of the channel. In this paper, a system is described that was used to collect simultaneous time and angle of arrival data at 7 GHz. Data processing methods are outlined, and results obtained from data taken in two different buildings are presented. Based on the results, a model is proposed that employs the clustered "double Poisson" time-of-arrival model proposed by Saleh and Valenzuela (1987). The observed angular distribution is also clustered with uniformly distributed clusters and arrivals within clusters that have a Laplacian distribution.

This paper addresses image and signal processing problems where the result most consistent with prior knowledge is the minimum order, or "maximally sparse" solution. These problems arise in such diverse areas as astronomical star image deblurring, neuromagnetic image reconstruction, seismic deconvolution, and thinned array beamformer design. An optimization theoretic formulation for sparse solutions is presented, and its relationship to the MUSIC algorithm is discussed. Two algorithms for sparse inverse problems are introduced, and examples of their application to beamforming array design and star image deblurring are presented.

This paper presents a new spatial smoothing approach for rank enhancement in sensor array direction finding applications. The method works with coherent scene data when non-uniform array element gains would make conventional smoothing ineffective. Sub-array data is pre-weighted before smoothing to regularize the problem, and so that a consistent phase and gain structure is maintained across the sub-arrays.

The generalized Gaussian Markov random field (GGMRF) is used as an image prior model in MAP restoration of blurred and noise corrupted images. This model is adapted to the characteristics of the true image by jointly estimating the true image and the GGMRF shape parameter, p, from the corrupted observation. A simple estimator for p based on sample kurtosis is introduced. It is shown that the value of p ranges widely when modeling typical images and texture fields. Higher quality restorations can be obtained when the estimated p value is used, rather than commonly used arbitrary choices.

A model adaptive method is proposed for restoring blurred and noise corrupted images. The generalized p-Gaussian family of probability density functions is used as the approximating parametric noise model. Distribution shape parameters are estimated from the image, and the resulting maximum likelihood optimization problem is solved. An iterative algorithm for data-directed restoration is presented and analyzed.

We address the problem of resolving blurred point sources in intensity images. A new approach to image restoration is introduced which is a 2-D generalization of techniques originating from the field of direction of arrival estimation (DOA). In the 2-D frequency domain, algorithms, such as MUSIC, may be adapted to search for these point sources. A generalization of array smoothing based on a regularization operator is introduced for 2-D arrays in order to achieve rank enhancement in the signal space of the covariance matrix.>

A new model adaptive method is presented for restoration of blurred and noise corrupted images by exploiting information available from observed data to choose the appropriate optimization criterion. The derived maximum likelihood solution is based on the l/sub p/ minimization criterion that naturally arises from the adoption of the generalized p-Gaussian family of probability distributions as an additive noise model. A fast and efficient iterative algorithm for this adaptive method is developed and analyzed. Experimental results indicate that this method adapts to the non-Gaussian nature of the noise process and outperforms the least squares method, which lacks the flexibility of the former method.>

This paper addresses the problem of resolving and localizing blurred point sources in intensity images, such as atmospherically blurred star images. It is shown that this problem can be modeled in a form similar to that used for sensor array data in direction of arrival (DOA) estimation. A new superresolving restoration algorithm is described as a generalization of eigenstructure decomposition methods popular in DOA estimation. The algorithm includes a signal subspace rank enhancement technique that works in the presence of non-uniform element weighting due to blur. A theoretical analysis of source localization error is presented, and an expression predicting error performance is derived. Restoration examples are presented illustrating the accuracy of the error analysis and demonstrating algorithm performance.>

The problem of removing blur from, or sharpening, astronomical star field intensity images is discussed. An approach to image restoration that recovers image detail using a constrained optimization theoretic approach is introduced. Ideal star images may be modeled as a few point sources in a uniform background. It is argued that a direct measure of image sparseness is the appropriate optimization criterion for deconvolving the image blurring function. A sparseness criterion based on the l(p) is presented, and candidate algorithms for solving the ensuing nonlinear constrained optimization problem are presented and reviewed. Synthetic and actual star image reconstruction examples are presented to demonstrate the method's superior performance as compared with several image deconvolution methods.

The problem of removing blur from, or sharpening, astronomical star field intensity images is addressed. A new image restoration algorithm is introduced which recovers image detail using constrained optimization theoretic approach. Ideal star images may be modeled as a few point sources in a uniform background. It is therefore argued that a direct measure of image sparseness is the appropriate optimization criterion for deconvolving the image blurring function. A sparseness criterion based on the l/sub p/ quasinorm is presented, and an algorithm for sparse reconstruction is described. Synthetic and actual star image reconstruction examples are presented which demonstrate the algorithm's superior performance as compared with the CLEAN algorithm, a standard star field deconvolution method.>