The authors present a covariance differencing algorithm for bearing estimation in situations where the noise covariance matrix is unknown. Conventional covariance differencing methods acquire a set of vectors which are orthogonal to the direction vectors by obtaining an eigenvalue decomposition of the difference of the covariance matrices of two measurements of the array. Eigendecomposition algorithms, however, involve a considerable computational burden. The authors also consider a procedure which does not require eigenvalue decomposition and is thus computationally more efficient. Results of simulation studies are included to show that the proposed approach performs nearly as well as the more conventional covariance differencing techniques in terms of signal resolution and estimation error.>

In recent years a new, and very powerful technique for parameter estimation - the eigenstructure, or signal subspace method - has been developed. Eigenstructure algorithms are closely related to Pisarenko's method for estimating the frequencies of sinusoids in white Gaussian noise. In theory they yield asymptotically unbiased estimates of arbitrarily close parameters, independent of the signal-to-noise ratio (SNR). Although signal subspace methods have proven to be powerful tools, they are not without drawbacks. An important weakness of all signal subspace algorithmis their need to know the noise covariance explicitly. The important problem of developing signal subspace based procedures for signals in noise fields with unknown covariance has not been satisfactorily addressed. It is our intent to propose a solution to the problem of direction-of-arrival (DOA) estimation for a broad class of unknown noise fields. We will then briefly discuss other important estimation problems for which modified versions of this procedure can be applied.

Through both theoretical developement and simulation the method we have put forward has been shown to yield improved performance over the conventional spatial smoothing algorithm. Our discussion has demonstrated that for a given array length the proposed method can be used to increase the number of coherent signals that can be resolved and thus effectively increase array aperture. Our simulation results underscore this fact. As we will see this increase in aperture is obtained, to some extent, at the expense of robustness.

The paper discusses some results of application of a stochastic realization algorithm for developing an ARMA type predictor for a class of speech signals. It is shown that a minimal order adaptive predictor can be obtained for the given speech data using a finite window version of the realization algorithm. The utility of the predictor is first established by processing synthetic signals generated by an ARMA

In this paper the problem of impulsive noise suppression in analog communication systems is considered. The problem of demodulation of amplitude and phase modulated signals in the presence of impulsive noise is formulated by introducing a binary random process to indicate the presence of impulsive noise. It is shown that sub-optimal receiver structures can be derived using this model and these are similar to some experimental receivers reported earlier in the literature. The receivers derived are non-linear in nature and their working is demonstrated through digital computer simulations.

The problem of designing optimum receivers with a finite built-in delay, for a class of stochastic signals, is considered. The approach used here is based on the so-called fixed-lag smoothing techniques. Specific cases of amplitude and phase modulation are considered in detail to illustrate the type of optimal (suboptimal in the nonlinear case) receivers which are obtained using this approach. The performance of these finite-delay demodulators is compared with that obtainable using the ideal infinite-delay and the common zero-lag demodulators. It is found that introduction of a nominal delay results in a significant improvement in performance over the zero-lag demodulators, of course, at the cost of increased complexity.

This correspondence examines the possibility of recursive prediction for the daily maximum temperature based on a state variable model. It is shown that past data can be processed for identifying a suitable noisy state model for this process. Once the model is selected the ``corrector-predictor'' algorithm of Kalman is readily applied for predicting the daily maximum temperature for a chosen lead time. The method is illustrated by predicting the daily maximum temperature of Delhi, India.