Arnt-Børre Salberg

Stochastic multipulse-PAM: a subspace modulation technique with diversity

In this paper, we introduce a non-conventional communication technique appropriately named Stochastic Multipulse-PAM (SM-PAM). SM-PAM systems operate by transmission of stochastic processes residing in low-dimensional subspaces. The transmitted signal is thus inherently stochastic, which is beneficial from a low-probability of intercept transmission point of view. The underlying structure of the transmission is such that the information is actually coded by the subspaces themselves, rather than by fixed coordinates in a preassigned basis, as is the case for conventional communication. We introduce a Fourier-based version of SM-PAM, and show that intersymbol interference (ISI) may readily be mitigated by employing a zero-forcing strategy and the insertion of a cyclic prefix in the transmitted vector. A substantial part of this paper is devoted to the derivation of matched subspace detectors for the transmitted SM-PAM waveform, for various classes of interference. We carry out a performance evaluation of SM-PAM, which shows that a proper energy scaling of the transmitted stochastic symbol vector has positive effects on the performance of the system. Various interesting scenarios are demonstrated by numerical simulations. The proposed multidimensional, stochastic, broadband, non-coherent communication technique buys security and diversity at the expense of spectral efficiency. We believe that SM-PAM may be important for non-standard applications where high data rates are not mandatory, but where security and interference rejection is imperative.

A subspace theory for differential chaos-shift keying

In this brief, we introduce a subspace theory for differential chaos-shift keying (DCSK) systems. We show that DCSK systems operate by transmission of chaotic signals residing in a low-dimensional subspace. The subspace formalism of DCSK schemes leads to the derivation of useful subspace detectors that can be applied to decode the DCSK signal for various types of channels. Closed form expressions for the bit error rate is derived for an M-ary FM-version of DCSK, under the assumption of orthogonal subspace generating vectors. Numerical simulations demonstrate that the proposed subspace detector in general outperforms the conventional correlation detector for DCSK

Secure digital communications by means of stochastic process shift keying

A digital communications keying concept that applies stochastic processes rather than deterministic waveforms is introduced and demonstrated. The technique implies a higher degree of security that conventional digital communications systems, reducing the risk of eavesdropping. The resulting receivers are non-coherent, and simpler than those of spread spectrum and chaotic encoders.

Robust Multidimensional Matched Subspace Classifiers Based on Weighted Least-Squares

We propose and design two classes of robust subspace classifiers for classification of multidimensional signals. Our classifiers are based on robust M-estimators and the least-median-of-squares principle, and we show that they may be unified as iterated reweighted oblique subspace classifiers. The performance of the proposed classifiers are demonstrated by two examples: noncoherent detection of space-time frequency-shift keying signals, and shape classification of partially occluded two-dimensional (2-D)_ objects. In both cases, the proposed robust subspace classifiers outperform the conventional subspace classifiers

Subspace detectors for stochastic process shift keying

We present a new digital modulation technique that introduces covertness in digital communications in a simple fashion. The basic principle is to transmit realizations of a stochastic process in such a manner that the transmitted waveform appears noiselike. The transmitted waveform is expressed in a subspace formalism, allowing for an elegant geometrical interpretation of the waveform, and a simple and accurate subspace detector for the receiver. The effect of intersymbol interference (ISI) is also studied, and a simple zero-forcing subspace detector is suggested. The technique is demonstrated by numerical simulations, and it shows that our simple subspace detectors yield reliable high-quality receivers.

A novel modulation method for secure digital communications

We present a new digital modulation technique that introduces covertness in digital communications. The basic principle is to transmit realizations of a stochastic process in such a manner that the transmitted waveform appears noise-like. In this paper, we have chosen to express the transmitted waveform in a subspace formalism. This allows for an elegant geometrical interpretation of the waveform, and it naturally suggests a simple and accurate matched subspace detector for the receiver. The technique is demonstrated by numerical simulations, and a comparison with an optimal Neyman-Pearson detector shows that our simple subspace detector yields a high-quality and reliable receiver for the modulated signal.We present a new digital modulation technique that introduces security in digital communications. The basic principle is to transmit realizations of a stochastic process in such a manner that the transmitted waveform appears noiselike. In this paper, we have chosen to express the transmitted waveform in a subspace formalism. This allows for an elegant geometrical interpretation of the waveform, and it naturally suggests a simple and accurate matched subspace detector for the receiver. The technique is demonstrated by numerical simulations, and a comparison with an optimal Neyman-Pearson detector shows that our simple subspace detector yields a high-quality and reliable receiver for the modulated signal.

Object segmentation and feature estimation using shadows

We introduce an automatic underwater video-based system for estimating features such as length and 3-D velocity of fish from a single-view camera system. Artificial light provides a high contrast object shadow, which is exploited both in the segmentation process, and to achieve 3-D inference about the fish. The proposed video segmentation method is demonstrated successfully on video-recordings of Greenland halibut.

Shape classification of partially occluded objects using subspace detectors

In this paper we consider shape classification of partially occluded objects. We model the occlusion as non-Gaussian noise, and apply robust subspace detectors in the classification module. We show that the robust subspace detectors can be formulated as a weighted subspace detector, and the elements in the boundary vector will be weighted before they are matched. The part of the boundary vector that corresponds to the occluded part of the object, will be suppressed by the weight vector, and hence have a reduced effect on the classification performance. The detectors are demonstrated on fish species classification applications.

Robust subspace detectors based on weighted least-squares

In this paper, we propose and design robust subspace detectors for classification of multidimensional subspace signals. Using the principle of M-estimators and least-median-of-squares (LMedS), we formulate the robust subspace detectors as weighted subspace detectors, where we weigh the rows of the measurement matrix prior to the signal matching. The detectors are demonstrated numerically by communication signals transmitted over an unknown frequency selective channel in impulsive noise, and shape classification of partially occluded two-dimensional objects. In both cases, the proposed robust subspace detectors outperform the classical subspace detector.

Digital communication using low-rank noise processes: subspace detectors

We propose a new digital modulation technique that yields a natural resistance against hostile-detection. The transmitted waveform is constructed of orthogonal low-rank stochastic processes, allowing for an elegant geometrical interpretation of the waveform. Channel effects are considered,, and we propose a channel-interference eliminating subspace detector. The technique is demonstrated by numerical simulations which shows that our subspace detectors yield high-quality and reliable receivers.