Probabilistic Modeling of Digital Detectors with Correlated Input Effects

• Main Authors: L. O. Ivannikova, V. N. Zhurakovskiy
• Format: Article
• Language: Russian
• Published: MGTU im. N.È. Baumana 2017-01-01
• Series: Nauka i Obrazovanie
• Subjects: correlation; detector; simulation and probabilistic modeling; transients; false operation; filter
• Online Access: http://technomag.edu.ru/jour/article/view/1161

<p>Most modern radio systems use digital signal processing. To debug algorithms of digital signal processing, most often is used simulation modelling, which has a number of significant drawbacks, since most modern programmes that generate random sequences of numbers begin to repeat previously obtained combinations of values with a large number of measurements. In addition, simulation modelling does not allow us to obtain very small probabilities of detector states. Consequently, the problem of mathematical modelling of digital devices remains relevant. Therefore, to study the operation of digital detectors when there are correlated input effects there is an article’s proposal to apply a probabilistic simulation method, using the Markov chain apparatus, based on a recurrent procedure and a concept of the extended digital automaton.</p><p>The simulation modelling to debug digital detector operation is as follows: a generated correlated signal comes to the threshold device to give at the output a sequence of binary signals arriving at the counter. Since in real conditions a digital detector has to have high accuracy, an error emerges when debugging based on simulation modelling. The article shows that with a probabilistic modelling method it is necessary to use the multidimensional Markov chains and the extended digital automaton to reach the higher accuracy. The article also compares the results obtained by various types of mathematical modelling to analyse the effect of the input signal correlation on the probability of detector false response.</p><p>The analysis has shown that probabilistic modelling in debugging the digital detector operation with a correlated input signal of high accuracy corresponds to the simulation model with the same specified parameters. With increasing input signal correlation there is a significantly raising probability of false response. Thus, it is fair to say that Markov chains-based probabilistic simulation can be used to study operation of radar signal processors.</p>