PROBLEMS OF INFORMATION TRANSMISSION
A translation of Problemy Peredachi Informatsii

CONTENTS                  

On Synthesis of Asymptotically Nonredundant Self-correcting Arrangements of Unreliable Functional Elements
S. I. Ortyukov
pp. 247–251

Abstract—The article describes a method of synthesizing self-correcting arrangements of unreliable functional elements. It is assumed that malfunctions occur in all elements independently with probability $\varepsilon$. It is shown that the operation of an arrangement with error probability of order $\varepsilon$ can be achieved asymptotically without increasing the Shannon function when the increase in the mean number of malfunctions in the arrangement is slight.

Limits of Attainable Accuracy for Transmission of a Parameter over a White Gaussian Channel
M. V. Burnashev
pp. 251–260

Abstract—Assume that an unknown parameter $\theta\in[0,1]$, while a modulating signal $S_t(\theta)$ is transmitted over a channel with white Gaussian noise $$ dX_t=S_t(\theta)\,dt+dW_t,\quad t\in T,\quad \theta\in\Theta=[0,1], $$ where $S_t(\theta)$ satisfies only the energy constraint $$ \int\limits_0^1 \|S_t(\theta)\|^2\,d\theta\le A. $$ New upper and lower bounds are obtained for the minimum possible mean $\alpha$-power risk $e_\alpha(A)=\inf\limits_{S,\hat{\theta}}e_\alpha(S,\hat{\theta})$. In particular, for $\alpha\ge 3$ the exponents of the upper and lower bounds coincide.

Estimating Using Statistics That Assume a Finite Number of Values
A. M. Samarov
pp. 260–265

Abstract—The authors take up the problem of estimating an unknown parameter $\theta$ by using statistics that can assume not more that $L$ different values. A lower bound is obtained for the mean-square estimation accuracy as the number of observations $n\to\infty$ and $L\to\infty$, and the estimate is constructed whose risk coincides with this bound in the principal terms.

Estimation of Signal Parameter in Gaussian Noise
Yu. A. Kutoyants
pp. 266–271

Abstract—The author investigates the asymptotic properties of maximum-likelihood estimates and Bayesian estimates of a univariate parameter when a continuous signal is transmitted over a channel with arbitrary Gaussian noise. In terms of the norms of a signal and its derivative with respect to a parameter in Hilbert space, for which the correlation function of the noise acts as a reproducing kernel, conditions are found that guarantee that these estimates will be asymptotically normal, consistent, and asymptotically effective. Transmission of a harmonic signal over a channel with stationary noise is considered as an example.

On Hypothesis Discrimination under Limited Observation Resources and Limited Information
N. G. Gatkin and Yu. L. Daletskii
pp. 272–277

Abstract—The article considers the problem of (statistically) optimal signal detection in random noise by using polynomial functionals. It is shown that the problem can be reduced to a system of equations that utilizes complete information on the random functions under consideration. This system is investigated under certain conditions; in particular, a linearized system corresponding to small signals is obtained, this being now determined only by a finite array of moments of the random functions in question.

Optimum Signal Filtering in Asynchronous Address Communication System with TFSK and Internal Noise
M. V. Chernyakov
pp. 277–283

Abstract—Markov theory of optimum nonlinear filtering in the Gaussian approximation is used to solve the problem of optimum signal filtering in asynchronous address communication systems with TFSK and internal noise. The useful signal is considered in conjunction with discrete pulse noise that approximates internal noise, and stationary white noise. Messages to be transmitted are represented by components of two-dimensional continuous Markov processes.

Linear Block Codes That Correct Multiple Bursts and Independent Errors
S. I. Zhegalov
pp. 283–288

Abstract—The article examines linear block codes over the field $GF(2^m)$ that correct multiple bursts for the case in which there are independent errors. A decoding algorithm for these codes is given.

Codes That Correct Triple Defects in Memory
I. B. Belov and A. M. Shashin
pp. 289–291

Abstract—A class of codes is proposed for correcting distortions that arise in storage devices with three defective cells. It is assumed that the location of the defective cells and the nature of the defect are known upon read-in but are unknown upon readout. The redundancy of the proposed codes is lower than that of available codes provided that the code length is large enough.

Quantity of Information Transmitted over a Quantum Channel with Various Dequantization Methods
A. G. Vantsyan and E. V. Astashkina
pp. 291–296

Abstract—Expressions are obtained for the Shannon quantity of information regarding the state of a quantum field at the receiver aperture, corresponding to different modes of input-signal processing. In the particular case involving reception of two signals without thermal noise, the results are compared quantitatively.

State Assignment for a Finite Automaton with Simplification of Logic Converter
V. V. Sapozhnikov and Vl. V. Sapozhnikov
pp. 296–303

Abstract—State assignment algorithms are proposed for an asynchronous finite automaton (AFA) that yields a simplified structure for the logic converter by reducing the functional dependence among internal variables. The algorithms are based on the Liu–Sagalovich method [C.N. Liu, J. Assoc. Comput. Mach., 10, No. 2, 209–216 (1963), Yu.L. Sagalovich, Probl. Peredachi Inf., 3, No. 2, 73–85 (1967)].

Use of Sequential Decoding in Recognition of Continuous Speech
K. Sh. Zigangirov and V. N. Sorokin
pp. 303–308

Abstract—The use of a sequential-decoding algorithm with return for recognition of continuous speech is considered.

Flow of Serviced Requests in Infinite-Channel Queueing Systems in a Transient Mode
M. Ya. Postan
pp. 309–313

Abstract—The conditions are determined under which the outgoing flow in a transient mode is a nonstationary Poisson flow, for infinite-channel systems with a nonstationary nonunary Poisson incoming flow of requests and a servicing time whose distribution is arbitrary or follows a nonstationary exponential law. Explicit forms are found for the generating functions of the joint probability distribution of the number of requests being serviced and of the number of serviced requests. Expressions are computed for the leading functions of the outgoing flows.

On Some Queueing Systems of Finite Capacity
P. P. Bocharov and V. A. Naumov
pp. 314–320

Abstract—Queueing systems of limited capacity that can be described by a Markov chain with semi-Markov control are considered. It is assumed that the distribution functions of the intervals between transitions of the control semi-Markov process have rational Laplace–Stieltjes transforms. A method of computing the stationary characteristics of such systems is proposed. The results make it possible to unify algorithms and computational programs for the stationary characteristics of certain queueing systems.

On Multistage One-Shot Batch Switching Arrangements
V. A. Garmash and L. A. Shor
pp. 320–323

Abstract—It is shown that Ofman arrangements [Tr. Mosk. Mat. Ob-va, 14, 186–199 (1965)] can be used for one-shot batch switching without inversions; a connection-establishment algorithm is proposed. Multistage one-shot batch switching arrangements are set up that have loop connections and binary commutators, each having two states.

INDEX
pp. 325–331