Worst-case Error Probability Research Articles

Kernel Robust Hypothesis Testing

The problem of robust hypothesis testing is studied, where under the null and the alternative hypotheses, the data-generating distributions are assumed to be in some uncertainty sets, and the goal is to design a test that performs well under the worst-case distributions over the uncertainty sets. In this paper, uncertainty sets are constructed in a data-driven manner using kernel method, i.e., they are centered around empirical distributions of training samples from the null and alternative hypotheses, respectively; and are constrained via the distance between kernel mean embeddings of distributions in the reproducing kernel Hilbert space, i.e., maximum mean discrepancy (MMD). The Bayesian setting and the Neyman-Pearson setting are investigated. For the Bayesian setting where the goal is to minimize the worst-case error probability, an optimal test is firstly obtained when the alphabet is finite. When the alphabet is infinite, a tractable approximation is proposed to quantify the worst-case average error probability, and a kernel smoothing method is further applied to design test that generalizes to unseen samples. A direct robust kernel test is also proposed and proved to be exponentially consistent. For the Neyman-Pearson setting, where the goal is to minimize the worst-case probability of miss detection subject to a constraint on the worst-case probability of false alarm, an efficient robust kernel test is proposed and is shown to be asymptotically optimal. Numerical results are provided to demonstrate the performance of the proposed robust tests.

Fingerprinting With Equiangular Tight Frames

Digital fingerprinting is a framework for marking media files, such as images, music, or movies, with user-specific signatures to deter illegal distribution. Multiple users can collude to produce a forgery that can potentially overcome a fingerprinting system. This paper proposes an equiangular tight frame fingerprint design which is robust to such collusion attacks. We motivate this design by considering digital fingerprinting in terms of compressed sensing. The attack is modeled as linear averaging of multiple marked copies before adding a Gaussian noise vector. The content owner can then determine guilt by exploiting correlation between each user's fingerprint and the forged copy. The worst case error probability of this detection scheme is analyzed and bounded. Simulation results demonstrate that the average-case performance is similar to the performance of orthogonal and simplex fingerprint designs, while accommodating several times as many users.

Robust MIMO Precoding for Several Classes of Channel Uncertainty

The full potential of multi-input multi-output (MIMO) communication systems relies on exploiting channel state information at the transmitter (CSIT), which is, however, often subject to some uncertainty. In this paper, following the worst-case robust philosophy, we consider a robust MIMO precoding design with deterministic imperfect CSIT, formulated as a maximin problem, to maximize the worst-case received signal-to-noise ratio or minimize the worst-case error probability. Given different types of imperfect CSIT in practice, a unified framework is lacking in the literature to tackle various channel uncertainty. In this paper, we address this open problem by considering several classes of uncertainty sets that include most deterministic imperfect CSIT as special cases. We show that, for general convex uncertainty sets, the robust precoder, as the solution to the maximin problem, can be efficiently computed by solving a single convex optimization problem. Furthermore, when it comes to unitarily-invariant convex uncertainty sets, we prove the optimality of a channel-diagonalizing structure and simplify the complex-matrix problem to a real-vector power allocation problem, which is then analytically solved in a waterfilling manner. Finally, for uncertainty sets defined by a generic matrix norm, called the Schatten norm, we provide a fully closed-form solution to the robust precoding design, based on which the robustness of beamforming and uniform-power transmission is investigated.

Optimal fingerprinting strategies with one-sided error

Fingerprinting enables two parties to infer whether the messages they hold are the same or different when the cost of communication is high: each message is associated with a smaller fingerprint and comparisons between messages are made in terms of their fingerprints alone. In the simultaneous message passing model, it is known that fingerprints composed of quantum information can be made exponentially smaller than those composed of classical information. For small message lengths, we present constructions of optimal classical fingerprinting strategies with one-sided error, in both the one-way and simultaneous message passing models, and provide bounds on the worst-case error probability with the help of extremal set theory. The performance of these protocols is then compared to that for quantum fingerprinting strategies constructed from spherical codes, equiangular tight frames and mutually unbiased bases.

Classical and quantum fingerprinting with shared randomness and one-sided error

Within the simultaneous message passing model of communication complexity, under a public-coin assumption, we derive the minimum achievable worst-case error probability of a classical fingerprinting protocol with one-sided error. We then present entanglement-assisted quantum fingerprinting protocols attaining worst-case error probabilities that breach this bound.

An Extended Quadratic Frobenius Primality Test with Average Case Error Estimates

We present an Extended Quadratic Frobenius Primality Test (EQFT), which is related to the Miller-Rabin test and the Quadratic Frobenius test (QFT) by Grantham. EQFT is well-suited for generating large, random prime numbers since on a random input number, it takes time about equivalent to 2 Miller-Rabin tests, but has much smaller error probability. EQFT extends QFT by verifying additional algebraic properties related to the existence of elements of order 3 and 4. We obtain a simple closed expression that upper bounds the probability of acceptance for any input number. This in turn allows us to give strong bounds on the average-case behaviour of the test: consider the algorithm that repeatedly chooses random odd k bit numbers, subjects them to t iterations of our test and outputs the first one found that passes all tests. We obtain numeric upper bounds for the error probability of this algorithm as well as a general closed expression bounding the error. For instance, it is at most 2^{-143} for k=500, t=2 . Compared to earlier similar results for the Miller-Rabin test, the results indicates that our test in the average case has the effect of 9 Miller-Rabin tests, while only taking time equivalent to about 2 such tests. We also give bounds for the error in case a prime is sought by incremental search from a random starting point. While EQFT is slower than the average case on a small set of inputs, we present a variant that is always fast, i.e. takes time about 2 Miller-Rabin tests. The variant has slightly larger worst case error probability than EQFT, but still improves on previous proposed tests.

A spread-spectrum communication system protecting information data from interception

This correspondence tries to construct a direct-sequence spread-spectrum multiple-access communication system, so that the difficulty of wiretapping from insiders' and outsiders' attacks increases and the worst case error probability decreases, as much as possible. Bent sequences with optimal periodic correlation properties are assigned to users, as spreading sequences corresponding to secret keys.

Worst-case error probability of a spread-spectrum system in energy-limited interference

We consider a communication channel corrupted by thermal noise and by an unknown and arbitrary interference of bounded energy. For this channel, we derive a simple upper bound to the worst-case error probability suffered by a direct sequence (DS) communication system with error-correction coding, pseudorandom interleaving, and a correlation receiver. This bound is exponentially tight as the block length of the error correcting code becomes large. Numerical examples are given that illustrate the dependence of the bound on the choice of error correcting code, the type of interleaving used, and the relative energy of the Gaussian noise and arbitrary interference.

Worst case additive noise for binary-input channels and zero-threshold detection under constraints of power and divergence

Additive-noise channels with binary inputs and zero-threshold detection are considered. We study worst case noise under the criterion of maximum error probability with constraints on both power and divergence with respect to a given symmetric nominal noise distribution. Particular attention is focused on the cases of a) Gaussian nominal distributions and b) asymptotic increase in worst case error probability when the divergence tolerance tends to zero.

On nonlinear direct-sequence detectors in arbitrary power-limited interference

Communication over a waveform channel corrupted by additive white Gaussian noise and by an unknown and arbitrary interference of bounded power is considered. For this channel, an upper bound is presented for the worst-case error probability of a communication system comprising a direct-sequence spread spectrum modulator and a nonlinear correlation receiver. It is shown that this bound is exponentially tight as the number of chips used in the modulator becomes large. This bound is evaluated for several detector nonlinearities. Numerical examples and comparisons to the performance of a pure Gaussian noise channel are also given.

Error probability for FH/MDPSK in multitone jamming, fast Rician fading, and Gaussian noise

A method is presented for the calculation of M-ary DPSK error probability in partial-band multitone jamming and white gaussian noise. Both unfaded and faded signals and jamming tones may be analyzed by the method. The fading considered is frequency-flat, time selective, Rician fading. The results have application in mobile, slow frequency-hopped, military, satellite systems. The method uses the Pawula-Rice-Roberts (1982) method for calculating the probability distribution function of the phase angle between two vectors when they are perturbed by noise. The worst case error probability in partial-band jamming with fading is efficiently computed using Houston's (1975) approach. A fundamental parameter, independent of the actual jammer power being used on the system, is identified. The effect of the ratio of the direct signal energy to the faded (indirect) signal energy is shown by example, as is the effect of fading bandwidth and Doppler shift. For practical values of the parameters, jammer fading was not found to be significant. Results for M=2, 4, and 8 are given. >

On the optimality of direct sequence for arbitrary interference rejection

Communication over a waveform channel corrupted by additive white Gaussian noise, and by an unknown and arbitrary interfering signal of bounded power is considered. For this channel, the authors derive an upper bound to the worst case error probability of direct-sequence spread spectrum communication with a correlation receiver, and also a lower bound applicable to any binary signaling technique and any receiver. By comparing these two bounds, it is shown that, if a small error probability is required, then no other binary signaling scheme or receiver can substantially improve upon the performance of direct-sequence with a correlation receiver for the same power and bandwidth. >

An asymptotically optimal random modem and detector for robust communication

Coherent communication over a waveform channel corrupted by thermal noise and by an unknown and arbitrary interfering signal of bounded power is considered. For a fixed encoder, a random modulator/demodulator (modem) and detector are derived. They asymptotically minimize the worst-case error probability as the blocklength of the encoder becomes large. This optimal modem is independent of the encoder, and the optimal detector is the standard correlation receiver. A simple upper bound to the performance of any encoder when used with the optimal modem and detector is presented. These results provide a benchmark with which the performance of spread-spectrum modems and robust detection rules can be compared.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Coded Error Probability for DS Spread-Spectrum Systems with Periodic Pulsed Multiple-Tone Interference

The coded error probability for direct sequence (DS) BPSK or QPSK spread-spectrum systems, used with or without interleaving, and operating in the presence of pulsed multiple-tone interference, is investigated. We consider the worst-case channel error probability under conditions of pulsed multiple-tone jamming for the various systems, and present coded error probability as a function of interleaving delay for (n, k) block codes with hard-decision decoding. It is shown that when the maximum level of continuous multiple-tone interference does not exceed the level of the desired signal, the duty cycle of the corresponding pulsed interference which yields the worst-case error probability is usually small, and that the smaller the duty factor of the jamming, the more considerable is the performance improvement due to the use of interleaving in conjunction with the coding.