Skip to main content
main-content
Top

About this book

This book constitutes the refereed post-conference proceedings of the 23rd International Conference on Distributed and Computer and Communication Networks, DCCN 2020, held in Moscow, Russia, in September 2020.

The 54 revised full papers and 1 revised short paper were carefully reviewed and selected from 167 submissions. The papers cover the following topics: computer and communication networks; analytical modeling of distributed systems; and distributed systems applications.

Table of Contents

Frontmatter

Computer and Communication Networks

Frontmatter

Power Domain NOMA Without SIC in Downlink CSS-Based LoRa Networks

Low Power Wide Area Network, such as LoRa is one of the main building blocks of the Internet of Things. One of the main issues is to scale up the number of devices and one strong limitation comes from the downlink communication. In fact, the access point is constrained by the duty cycle, therefore it cannot address a large number of devices. We propose a superposition scheme to transmit multiple packets to multiple devices in the same frequency, time slot, and spreading factor. This scheme is applied to the specific physical layer proposed by LoRa, based on the chirp spread spectrum. Our proposal includes the power allocation scheme and the decoding technique that are very specific to this physical layer and show a significant performance improvement, increasing the number of devices that can be connected at least by ten compared to the classical LoRa-like system.

Angesom Ataklity Tesfay, Eric Pierre Simon, Ido Nevat, Laurent Clavier

Simulation-Based Analysis of Mobility Models for Wireless UAV-to-X Networks

Recently, the use of air base stations located on unmanned aerial vehicles (UAVs) has attracted great attention. Static deployment of a sufficient number of UAVs allows uniform wireless coverage in the demanded areas, where the existing cellular infrastructure has white spots or insufficient capacity. However, UAVs mobility may be required for applications, where UAVs are used to provide communications for mobile groups of users (e.g., massive sport or community events like marathon or music festival) or for patrolling tasks with relaxed requirements for data transmission delays (for example, when collecting information from a large number of mMTC sensors). In such tasks, the movement of UAVs can significantly increase the efficiency of the system, since in this case the coverage of the area can be provided by a smaller number of UAVs following the dynamics of ground users. Nowadays, more and more often the question arises about the mobile communications availability in a remote area, for example, during public events or search operations. The lack of on-demand connectivity with sufficient quality in such areas is unacceptable in modern conditions. Therefore, the study of the behavior of a dynamic UAV network is necessary for decision-making operation in such scenarios. The main contribution of this work is making the user mobility model more human-alike according to the real scenarios. The paper considers two models of UAVs movement, the effectiveness of which is estimated from the point of view of the coverage probability and average fade duration of the signal.

Vladislav Prosvirov, Viktoriia Khalina, Ekaterina Lisovskaya, Yuliya Gaidamaka, Jiri Pokorny, Jiri Hosek, Konstantin Samouylov

Structures and Deployments of a Flying Network Using Tethered Multicopters for Emergencies

In recent years, the interest of tethered UAVs high-altitude platforms has been widely constantly increasing in many fields. The long-time operating possibility is one of the main advantages of tethered unmanned high-altitude platforms compared to autonomous UAVs. In the paper, a flying network for emergencies using tethered multicopters is proposed. The combination of tethered unmanned high-altitude platforms and groups of UAVs in flying network for emergencies is expected to enhance the effectiveness of search and rescue operation in the wilderness as well as after natural disasters.

Truong Duy Dinh, Vladimir Vishnevsky, Andrey Larionov, Anastasia Vybornova, Ruslan Kirichek

Multipath Redundant Network Protocol Without Delivery Guarantee

Timely and faultless packets delivery problem in real-time systems is described in the paper. New protocol based on multipath redundant transmissions with high probability of faultless and timely delivery is presented. This approach is based on UDP protocol and using redundant transmissions via multipath reserve channels between a client and a server. The multiplicative criteria based on faultless and timely packets delivery probability and average delivery time reserve relatively delivery time restriction defined in the real-time computer system is used for efficiency evaluation for time sensitive systems in this paper. Faultless and timely delivery packet probability and faultless probability were used as a criterion for developed protocol efficiency analyzing in systems without strong delivery packet time limitations or packet delivery time insensitive systems. Effective using areas of developed protocol are described in the paper. Developed protocol provides faultless and timely delivery packets improvements for different type of systems (real-time and time insensitive). The efficiency of redundant multipath transmissions is analyzed and researched using obtained results from experiments with simulation models developed in OMNeT++ environment. This paper can be useful for network engineers who develop new transport or application layer protocols to provide reliable network transmissions in computer networks. This paper can be considered as a theoretical base for developing new multipath redundant protocol implementation and using it in real computer networks.

Ilya Noskov, Vladimir Bogatyrev

Modelling Multi-connectivity in 5G NR Systems with Mixed Unicast and Multicast Traffic

3GPP New Radio (NR) radio access technology operating in millimeter wave (mmWave) frequency band is considered as key enabler for Fifth-generation (5G) mobile system. Despite the enormous available bandwidth potential, mmWave signal transmissions suffer from fundamental technical challenges like severe path loss, sensitivity to blockage, directivity, and narrow beamwidth, due to its short wavelengths. To address the problem of quality degradation due to the line-of-sight (LoS) blockage by various objects in the channel, 3GPP is currently working on multi-connectivity (MC) mechanisms that allow a user to remain connected to several mmWave access points simultaneously as well as switch between them in case its active connection drops. In this paper, exploiting the methods of stochastic geometry and queuing theory we propose a model of 5G NR base station (BS) serving a mixture of unicast and multicast traffic. MC techniques is proposed to be used for cell-edge users. The proposed model is validated against computer simulations in terms of session drop probabilities and system resource utilization metrics. Our findings are illustrated with a numerical example.

Vitalii Beschastnyi, Daria Ostrikova, Sergey Melnikov, Yuliya Gaidamaka

IoT Traffic Prediction with Neural Networks Learning Based on SDN Infrastructure

In recent time, there are more achievements in technologies for 5G/MT-2020 networks the international research area have. Any way, one of the main important task in this area is the IoT traffic recognition and prediction. Currently, researchers face a new challenge to their experience, talents and desire to reach new heights in information and communication technologies. The new challenge is IMT-2030 networking technologies and services. In this case, the question with effective IoT traffic prediction methods is still relevant during transition to the next IMT-2030 network and services. IoT it is the ubiquitous conception, on which the new IMT-2030 services also based. For example, Tactile Internet, part of the solutions in digital avatars, and others. There more, these algorithms have to be more efficient and fastly in work with huge data capacity, which characterized the different services of Internet of Things. Recently, there are Machine Learning and Big Data algorithms are took this place of new algorithms for efficient and complex algorithms. In this paper, we implement IoT traffic prediction approaches using single step ahead and multi-step ahead prediction with NARX neural network. As a data we used the metadata of flows which were received through the northern interface. The prediction accuracy has been evaluated using three neural network traing algorithms: Traincgf, Traincgp, Trainlm, with MSE as performance function in term of using mean absolute percent of error (MAPE) as prediction accuracy measure IoT.

Artem Volkov, Ali R. Abdellah, Ammar Muthanna, Maria Makolkina, Alexander Paramonov, Andrey Koucheryavy

Transmission Latency Analysis in Cloud-RAN

Cloud-based Radio Access Network (C-RAN) is a centralized cloud computing architecture for radio access networks (RANs) that provides large-scale deployment, joint support for radio technologies, and real-time virtualization capabilities. By moving signal processing functions to a data center, C-RAN significantly reduces power consumption and deployment cost. The architecture of the cloud radio access network consists of three main components: a pool of base-band units (BBU pool), remote radio heads (RRHs), and a transport network. In C-RAN, base stations are replaced by remote radio heads: data blocks are digitized, transmitted through the fiber-optical infrastructure, and remotely processed in BBU pool. One of the main issues is to control the round-trip delay between the remote radio heads and the BBU pool. In the paper, we describe a C-RAN in terms of queuing network and accurately evaluate all delay components. Besides, we analyze the required computational resources of the BBU pool required to satisfy the strict round-trip delay budget in C-RAN.

Eduard Sopin, Anatoly Botvinko, Alexandra Darmolad, Dinara Bixalina, Anastasia Daraseliya

Resource Queuing System with Preemptive Priority for Performance Analysis of 5G NR Systems

One of the ways to enable smooth coexistence of ultra reliable low latency communication (URRLC) and enhances mobile broadband (eMBB) services at the air interface of perspective 5G New Radio (NR) technology is to utilize preemptive priority service. In this paper, we provide approximate analysis of the queuing system with random resource requirements, two types of customers and preemptive priority service procedure. The distinctive feature of the systems – the random resource requirements – allows to capture the essentials of 5G NR radio interface but inherently increases the complexity of analysis. We present the main performance metrics of interest including session drop probability and system resource utilization as well as assess their accuracy by comparing with computer simulations. The developed model is not inherently limited to URLLC and eMBB coexistence and can be utilized in performance evaluation of 5G NR systems with priority-based service discipline at the air interface, e.g., in context of network slicing. Among other conclusions we explicitly show that both session drop and interruption probabilities of low priority traffic heavily depend not only on the intensity of high priority traffic but on stochastic characteristics of the resource request distribution.

Eduard Sopin, Vyacheslav Begishev, Dmitri Moltchanov, Andrey Samuylov

Redundant Servicing of a Flow of Heterogeneous Requests Critical to the Total Waiting Time During the Multi-path Passage of a Sequence of Info-Communication Nodes

The possibilities are investigated and analytical models of redundant multiway servicing of a heterogeneous request flow with their replication rate depending on the maximum permissible waiting time for replicas accumulated in the queues of nodes that make up the path for real-time information and communication systems are proposed. Two options are considered for redundant servicing of a heterogeneous flow during the sequential passage of copies of requests through parallel-connected nodes grouped in groups. For the first option, when generating a request, a certain number of copies are created, for each of which a path is predefined as a sequence of nodes of different groups involved in servicing this copy. For the second option, the paths are formed dynamically at each stage, and a copy of the request, executed first at some stage of the sequential passage of groups of redundant nodes, is transferred for redundant service to the next group of nodes. At various stages of service, the redundancy ratio can vary.

V. A. Bogatyrev, A. V. Bogatyrev, S. V. Bogatyrev

Agriculture Management Based on LoRa Edge Computing System

Internet of Things (IoT) technologies represent the future challenges of computing and communications. They can also be useful to improve traditional farming practices worldwide. Since the areas where agricultural land is located in remote places, there is a need for new technologies. These technologies must be suitable and reliable for communication over long distances and, at the same time, consume little energy. In particular, one of these relatively new technologies is the LoRa communication protocol, which uses long waves to work over long distances. This is extremely useful in agriculture, where the communicating areas are broad fields of crops and greenhouses. This study developed a greenhouse monitoring system based on LoRa technology, designed to work over long distances. The edge computing paradigms with a machine learning mechanism are proposed to analyze and control the state of the greenhouse, and in particular, to reduce the mount of data transmitted to the server.

Fatkhullokhodzha Sharofidinov, Mohammed Saleh Ali Muthanna, Van Dai Pham, Abdukodir Khakimov, Ammar Muthanna, Konstantin Samouylov

Dynamic Algorithm for Building Future Networks Based on Intelligent Core Network

6G/IMT-2030 is designed to provide users with innovative speeds of terabit per second, which are proposed to be achieved using a number of advanced technologies, such as Mobile Edge Computing (MEC), Internet of Things (IoT), millimeter wave (mmWave), new radio and software defined networking. It is necessary to solve several important aspects in order to satisfy Quality of Service (QoS), first of all, to ensure network coverage density even in sparsely populated areas. In this paper we proposed software defined network based mobile edge computing dynamic algorithm for improving network performance. In addition, this algorithm can help the service provided to adapt with a required load on the radio links. Furthermore, local content caching and Local Internet Breakout (LIB) can be utilized to reduce the transport network requirements. Finally, the proposed algorithm is analyzed using some use cases and we developed testbed to emulate operator infrastructure.

Abdukodir Khakimov, Ammar Muthanna, Ibrahim A. Elgendy, Konstantin Samouylov

Methods and Models for Using Heterogeneous Gateways in the Mesh LPWANs

Nowadays, a group of low-power wide-area networks (LPWANs) is one of the options providing the communication infrastructure for the Internet of Things applications. Most of LPWANs use a star topology model between end-nodes and a base station or a gateway. However, the mesh topology has shown the mobility and the dynamic in the network deployment. This article considers a question for constructing mesh LPWAN networks. In this article, the authors investigated existing standards of the developing gateways in the LPWAN networks, creating a structure of the mesh LPWAN with a conversion structure – heterogeneous gateway. We used this structure to create a simulation model of this network. The simulation model was mapped with a model based on the more traditional star topology with a simple gateway and an edge server. The results of the simulation can be used to design new mesh LPWAN networks.

Viacheslav Kulik, Van Dai Pham, Ruslan Kirichek

Research on Using the AODV Protocol for a LoRa Mesh Network

In this paper, we consider the LoRa technology to expand sensor network coverage in smart sustainable cities. A model of a LoRa mesh network is proposed using the AODV protocol in packet routing. With a simulation model developed based on OMNET++, a series of computer experiments was carried out with changing various parameters. In the experiments results, the end-to-end delay and packet loss ratio were analyzed in the dependence on the number of nodes and packet size in the network. The simulation results show that the latency is relatively high in the LoRa mesh network, but it might be accepted for some applications.

Van Dai Pham, Duc Tran Le, Ruslan Kirichek, Alexander Shestakov

On the Algebraic Theory of Loop Free Routing

Validation models not only provide a better understanding of the system, but can also help in improving the reliability and robustness of the design. EIGRP metric can be modeled algebraically using bi-semigroups and semirings in case the cost function is homomorphic [1]. EIGRP uses DUAL algorithm which is the basis for loop-free distance vector routing with non lexical metric. DUAL was validated using the classical shortest path problem [13, 20]. However, it was shown that DUAL does not perform as expected in the absence of monotonicity [1, 23]. This article approaches loop free routing from an algebraic perspective. Conditions for loop free routing and the relations between them were presented algebraically and proved correct. Then, we investigate loop free routing in the presence and the absence of monotonicity. Our goal is to provide theory for loop free routing with an arbitrary metric.

Hussein Khayou, Margarita A. Rudenkova, Leonid I. Abrosimov

Queueing System with Two Unreliable Servers and Backup Server as a Model of Hybrid Communication System

In this paper, we analyze a queueing system with two main unreliable servers and backup reliable server. The input flow is defined by the BMAP (Batch Markovian Arrival Process). Heterogeneous breakdowns arrive to the main servers according to a MMAP (Marked Markovian Arrival Process). Service times and repair times have PH (Phase type) distribution. The queueing system under consideration is an adequate model of operation of hybrid communication systems which combine the use of Free Space Optics and radio technologies. We derive a condition for the stable operation of the system, compute its stationary distribution and the key performance measures. Illustrative numerical examples give some insight into the behavior of the system.

Valentina Klimenok, Alexander Dudin, Vladimir Vishnevsky

Flexible Random Early Detection Algorithm for Queue Management in Routers

With recent advancements in communication networks, congestion control remains a research focus. Active Queue Management (AQM) schemes are normally used to manage congestion in routers. Random Early Detection (RED) is the most popular AQM scheme. However, RED lacks self-adaptation mechanism and it is sensitive to parameter settings. Many enhancements of RED were proposed and are yet to provide stable performance under different traffic load situations. In this paper, AQM scheme called Flexible Random Early Detection (FXRED) is proposed. Unlike other RED’s enhancements with static drop patterns, FXRED recognizes the state of the current network’s traffic load and auto tune its drop pattern suitable to the observed load situation in order to maintain stable and better performance. Results of the experiments conducted have shown that regardless of traffic load’s fluctuation, FXRED provides optimal performance and efficiently manages the queue.

Aminu Adamu, Vsevolod Shorgin, Sergey Melnikov, Yuliya Gaidamaka

Architecture and Functionality of the Collective Operations Subnet of the Angara Interconnect

The Angara interconnect developed by JSC “NICEVT” is designed to connect the nodes of supercomputers and computing clusters. The paper describes the main architectural solutions, algorithms and functionality of the collective operations subnet of the Angara interconnect and presents the forecast of its characteristics based on the simulation modeling and actual operation. The proposed solutions allow bringing the time complexity of the collective operations execution to the theoretical limit for the kD-torus topology network.

Alexey Simonov, Oleg Brekhov

The Model of WBAN Data Acquisition Network Based on UFP

The paper determines the general model of an unmanned flying platforms based network for collecting information from wearable wireless body area networks. There are considered several possible approaches to its implementation. A list of tasks to be solved and their features in the framework of building a network are given. There are proposed variants of optimal network radio technologies and topologies, antenna devices and types of unmanned flying platforms to create the acquisition network. Variants of implementation of the interaction protocol and methods of organizing secure data transmission, taking into account the peculiarities of the problem being solved, are recommended.

S. Vladimirov, V. Vishnevsky, A. Larionov, R. Kirichek

Development and Investigation of Model Network IMT2020 with the Use of MEC and Voice Assistant Technologies

Humanity is at a new stage in the development of information and computer technologies. The digitalization of most areas of activity creates a contradiction. On the one hand, to start the operation of the device, you must promptly enter information. Only the information entered by the user, allows the software agent to perform tasks. On the other hand, the technological diversity of modern devices requires new complex skills from the user. Consequently, the search for a solution to the problem of the quick introduction of information by a person without special and long-term preparation becomes relevant. The technology of voice input of information makes it possible to solve this problem. It underlies the work of virtual voice assistants, the use of which is constantly growing. The paper developed a simulation model using voice assistants (VA) based on software-defined (SDN) networks and mobile edge computing (MEC) technology. The analysis of the scope of voice assistants. An experimental study is given of the influence of the dependence of the execution time of processes and the total delay on the method of processing speech information of devices. The dependence of the total delay on the type of traffic and computing device for the final processing of packets’.

Maria Makolkina, Nikolay Shypota, Andrey Koucheryavy

On Overall Measure of Non-classicality of N-level Quantum System and Its Universality in the Large N Limit

In this report we are aiming at introducing a global measure of non-classicality of the state space of N-level quantum systems and estimating it in the limit of large N. For this purpose we employ the Wigner function negativity as a non-classicality criteria. Thus, the specific volume of the support of negative values of Wigner function is treated as a measure of non-classicality of an individual state. Assuming that the states of an N-level quantum system are distributed by Hilbert-Schmidt measure (Hilbert-Schmidt ensemble), we define the global measure as the average non-classicality of the individual states over the Hilbert-Schmidt ensemble. We present the numerical estimate of this quantity as a result of random generation of states, and prove a proposition claiming its exact value in the limit of $$N\rightarrow \infty $$ N → ∞ .

V. Abgaryan, A. Khvedelidze, I. Rogojin

Analytical Modeling of Distributed Systems

Frontmatter

Queues with Markovian Arrivals, Phase Type Services, Breakdowns, and Repairs

In service sectors, the server interruptions occur naturally and are studied using queueing models with interruptions. Such models have been studied extensively in the literature. The server interruptions may occur due to external events such as shocks and negative customers, or due to internal events which are dictated by a timer or a clock which when expired may result in an interruption. In this paper, we analyze models of MAP/PH/1-type with phase type breakdowns and phase type repairs analytically. The breakdowns are modeled using internal clocks. The study of such breakdowns in the context of MAP/PH/c-type models is carried out via simulation. Some illustrative examples are discussed.

Srinivas R. Chakravarthy, Rakesh Kumar Meena, Alka Choudhary

Statistical Analysis of the End-to-End Delay of Packet Transfers in a Peer-to-Peer Network

The paper is devoted to the statistical analysis of the end-to-end (E2E) delay of packet transfers between source and destination nodes in a peer-to-peer (P2P) overlay network. We focus on the identification of the E2E delay and the longest per-hop delay distributions and the stochastic dependence of the associated random process. The E2E delay is determined by the sum of a random number of dependent per-hop (p-h) delays along the links of a considered overlay path and the longest per-hop delay by their maximum. We propose to use the sum of the p-h delays to get a distribution of the maximum which is motivated by the available statistical data of the E2E delays. Based on recent analytic results derived from extreme-value theory we show that such sums and maxima corresponding to different paths may have the same tail and extremal indexes. These indexes determine the heaviness of the distribution tail and the dependence of extremes. Using the extremal index we identify limit distributions of the maxima of the E2E delays and the maxima of the p-h delays at a path among all source-destination paths. Considering real-time applications with stringent E2E-delay constraints, the distributions are used to identify quality-of-service (QoS) metrics of a P2P model like the packet missing probability and the corresponding playback delay as well as the equivalent capacity of a transport channel.

Natalia M. Markovich, Udo R. Krieger

Multidimensional Central Limit Theorem of the Multiclass M/M/1/1 Retrial Queue

In this paper, we consider the multiclass M/M/1/1 retrial queueing system. Customers of each class arrive from outside the system according to a Poisson process. The service times of customers are assumed to be exponentially distributed with the parameter corresponding to the type of the customer. If the server is busy incoming customers join the orbit according to their type and make a delay for an exponentially distributed time. Equations for the characteristic function of the multi-dimensional probability distribution of the numbers of customers in the orbits are obtained. These equations are investigated by method of asymptotic analysis under the long delay condition of customers in the orbits. It is shown that the probability distribution can be approximated by a multi-dimensional Gaussian distribution. Equations are obtained for finding the parameters of this probability distribution.

Anatoly Nazarov, Tuan Phung-Duc, Yana Izmailova

A Retrial Queueing System in Which Server Searches to Accumulate Customers for Optimal Bulk Serving

In this paper, we introduce the concept of orbital search as a means for accumulating customers in a bulk service retrial queueing system. In this model, customers enter the service facility from a finite buffer. If the buffer is full at the time of arrival of a customer, it enters an orbit from where retrials for entering the buffer has been made. Here search is done as a means for accumulating customers in the buffer so that optimum level of bulk service can be provided. A service policy considered here is (a, b) bulk service policy with search. Under this policy, search will be initiated when the number of customers in the buffer reaches a and it has been continued until either the search has been done for a random duration of time or the number of customers in the buffer increases to b where $$a<b$$ a < b . Steady state analysis of the model has been done and some important measures of performance has been evaluated. We analyze the model numerically as well as graphically.

Varghese C. Joshua, Ambily P. Mathew, Achyutha Krishnamoorthy

Approximate Analysis of the Queuing System with Heterogeneous Servers and N-Policy

In this paper, we propose an approximate method to investigate the Markovian queuing system with two separate pools of heterogeneous servers (HS) under N-policy. It is assumed that fast servers (F-servers) remain awake all the time while slow servers (S-servers) will go to sleep independently when number of calls in the buffer less than some threshold. At the end epoch of a sleep period, if the number of the calls gathered in the system buffer reaches or exceeds a given threshold, the corresponding S-server will wake up independently; otherwise, the S-server will begin another sleep period. An approximate method is applied under the condition that the sleep rates is essentially less than both arrival intensity of calls and their service intensity. The joint probability distribution of the number of calls in system and number of busy S-servers is determined by simple computational procedures. Illustrative numerical examples show the high accuracy of the proposed approximate method.

Agassi Melikov, Sevinc Aliyeva, Mammed Shahmaliyev

Analysis of a Resource-Based Queue with the Parallel Service and Renewal Arrivals

Classical queueing theory is often not suitable to model modern computer and communication systems, in which the service itself can require random amounts of multiple resources. For instance, this is true for distributed computation and wireless devices connected through different access technologies.To model such systems we propose a resource queueing system with customer duplication, in which the service time and the amount of requested resources in each block are independent random variables. In more detail, we assume that customers arrive according to a general renewal process and, taking advantage of the dynamic screening and the asymptotic analysis methods, we derive a Gaussian approximation for the stationary probability distribution of the occupied resources in the system blocks. Finally, simulation experiments point out the applicability region (in terms of arrival rate) of the proposed approximation.

Ekaterina Lisovskaya, Ekaterina Pankratova, Svetlana Moiseeva, Michele Pagano

Two-Phase Resource Queueing System with Requests Duplication and Renewal Arrival Process

In this paper, we analyze a two-phase resource queueing system with duplication at the second phase under the assumption that customers enter the system according to a renewal process and take up random resource amounts that do not affect the time of their service (for video-conference or streaming traffic). We apply the dynamic screening method and the asymptotic analysis method to obtain an approximation for the stationary probability distribution of the total amount of occupied resources in the system under increasing arrival rate. In more detail, we show that the three-dimensional probability distribution of the total resource amounts on the system blocks is three-dimensional Gaussian, obtain its parameters (means vector and covariance matrix) and verify the correctness of the asymptotic using discrete-event simulation.

Anastasia Galileyskaya, Ekaterina Lisovskaya, Michele Pagano, Svetlana Moiseeva

Deep Neural Networks for Emotion Recognition

The paper investigates the problem of recognizing human emotions by voice using deep learning methods. Deep convolutional neural networks and recurrent neural networks with bidirectional LSTM memory cell were used as models of deep neural networks. On their basis, an ensemble of neural networks is proposed. We carried out computer experiments on using the constructed neural networks and popular machine learning algorithms for recognizing emotions in human speech contained in the RAVDESS audio record database. The computational results showed a higher efficiency of neural network models compared to machine learning algorithms. Accuracy estimates for individual emotions obtained using neural networks were 80%. The directions of further research in the field of recognition of human emotions are proposed.

Eugene Yu. Shchetinin, Leonid A. Sevastianov, Dmitry S. Kulyabov, Edik A. Ayrjan, Anastasia V. Demidova

A Simulation Approach to Reliability Assessment of a Redundant System with Arbitrary Input Distributions

With the rapid development and spread of computer networks and information technologies, researchers are faced with new complex challenges of both applied and theoretical nature in investigating the reliability and availability of communication networks and data transmission systems. In the current paper, we perform the system-level reliability analysis for a redundant system with arbitrary distributions of uptime and repair time of its elements using a simulation approach. Also, we obtained the values of the relative recovery speed at which the desired level of reliability is achieved, presented dependency plots of the probability of system uptime and plots of the uniform difference of the obtained simulation results against the relative speed of recovery; also plots of the empirical distribution function $$F^*(x)$$ F ∗ ( x ) and reliability function $$R^*(x)$$ R ∗ ( x ) relative to the reliability assessment. Software implementation of simulation algorithms was carried out on the basis of the R language.

H. G. K. Houankpo, D. V. Kozyrev, E. Nibasumba, M. N. B. Mouale, I. A. Sergeeva

Problem of Overbooking for a Case of a Random Environment Existence

An overbooking supposes that a booking of some product or service exceed given possibilities. It takes into consideration that a part of the booking will be cancelled. This situation is considered following to example of aviation ticket booking. It is supposed that an external random environment exists. The environment is described as a continuous-time finite irreducible Markov chain. A demand on the booking depends on the state of the random environment. An using of economical criterion supposes a consideration of such indices as costs of engaged seats of the aircraft and the penalty for the refusal of passenger with sold tickets. This criterion is optimized by means of dynamic programming method. A numerical example is considered.

Alexander Andronov, Iakov Dalinger, Diana Santalova

Optimization of Signals Processing in Nodes of Sensor Network with Energy Harvesting and Expenditure for Admission and Transmission

Operation of a sensor node of a wireless sensor network with energy harvesting is described by the single-server queue. Customers and energy units arrive according to the Markov arrival processes (MAP) and are stored in the corresponding buffers. Service of a customer is possible only in presence of an energy unit. In contrast to previously investigated in the literature models, we assume that, besides the use of one energy unit for service of any customer, one more unit is expended at the moment of a customer arrival if the customer is accepted to the system. To optimize operation of the system, a parametric strategy of admission control is used. The goal of control is to minimize the risk of the server starvation in case of too strict control and the risk of wasting the energy due to acceptance of too many customers that finally will not receive a service (due to the lack of energy or impatience) in case of too liberal control. Under the fixed value of control parameter, the behavior of the system is described by the six-dimensional Markov chain. The generator of this Markov chain is obtained. Expressions for computation of the key performance indicators of the system are presented. Numerical results illustrating the effectiveness of the proposed control strategy are presented.

Sergey Dudin, Olga Dudina, Alexander Dudin, Chesoong Kim

The Analysis of Resource Sharing for Heterogenous Traffic Streams over 3GPP LTE with NB-IoT Functionality

The main feature in the development of Internet of Things (IoT) applications is the necessity of conjoint servicing of heterogenous data streams over existent network infrastructure. This trend has been recognized and supported by 3GPP with introducing of NarrowBand IoT (NB-IoT) technology, which allows to use the same resource by 3GPP LTE high-end equipment and NB-IoT low-end devices. The need of sharing the limited amount of available resource efficiently emphasizes the importance of theoretical study of formulated problem. The model of resource allocation and sharing for conjoint servicing of real time video traffic of surveillance cameras and NB-IoT data traffic of smart meters and actuators over LTE cell facilities is constructed. In the model the access control is used to create the conditions for differentiated servicing of coming sessions. All random variables used in the model have exponential distribution with corresponding mean values but the obtained results are valid for models with arbitrary distribution of service times. Using the model the main performance measures of interest are given with help of values of probabilities of model’s stationary states. The recursive algorithm of performance measures estimation is suggested. The model and derived algorithms can be used for study the scenarios of resource sharing between heterogenous data streams over 3GPP LTE with NB-IoT functionality.

Sergey N. Stepanov, Mikhail S. Stepanov, Umer Andrabi, Juvent Ndayikunda

Performance Measures of Emergency Services in Case of Overload

The mathematical model of public-safety answering points (PSAP) functioning is constructed and analyzed. In the model the usage of interactive voice response (IVR) and the possibility of call repetition in case of blocking or unsuccessful waiting are taken into account. Algorithm of characteristics estimation based on truncation of used infinite space of states and solving the system of state equations is suggested. Relative error of characteristics calculation caused by truncation is found. The first two or three terms in the asymptotic expansion of basic stationary performance measures into a power series of the intensity of primary calls as it tends to infinity are derived. Approximate algorithm of performance measures estimation is constructed. The usage of the model for elimination of negative effects of PSAP overload is considered.

Sergey N. Stepanov, Mikhail S. Stepanov, Maxim O. Shishkin

Evaluation and Prediction of an Optimal Control in a Processor Sharing Queueing System with Heterogeneous Servers

In this paper we study the problem of optimal controlling in a processor sharing (PS) M/M/2 queueing system with heterogeneous servers. The servers differ in the service intensities, operating and usage costs. The objective is to find the optimal policy to allocate the customers either to an idle or partially loaded server, or to the queue at each arrival and service completion epoch to minimize the long-run average cost per unit of time. We handle this optimization problem as Markov decision problem and study numerically structural properties of the optimal control policy. Using a policy-iteration algorithm we show that this policy for the current model is of threshold type. In this case the faster server handles customers with maximum capacity, while the number of simultaneously serviced customers at the slower server can be increased only when the number of waiting customers exceeds a certain threshold level. The data-sets generated by classic methodology of analyzing the controlled queues are used to explore predictions for optimal thresholds through artificial neural networks. The presented theoretical results are accompanied by heuristic solution and numerical examples.

Dmitry Efrosinin, Vladimir Rykov, Natalia Stepanova

On Exponential Convergence of Dynamic Queueing Network and Its Applications

This paper is a continuation of previous research in ergodicity of some models for unreliable networks. The set of random graphs and the sequence of matrixes describing the failure and recovery process has been used instead of the fixed graph for network structure. The main results about an ergodicity and bounds for rate of convergence to stationary distribution are formulated under more general assumptions on intensity rates.

Elmira Yu. Kalimulina

Leader Nodes in Communities for Information Spreading

The paper is devoted to the effective information spreading in random complex networks. Our objective is to elect leader nodes or communities of the network, which may spread the content among all nodes faster. We consider a well-known SPREAD algorithm by Mosk-Aoyama and Shah (2006), which provides the spreading and the growth of the node set possessing the information. Assuming that all nodes have asynchronous clocks, the next node is chosen uniformly among nodes of the network by the global clock tick according to a Poisson process. The extremal index measures the clustering tendency of high threshold exceedances. The node extremal index shows the ability to attract highly ranked nodes in the node orbit. Considering a closeness centrality as a measure of a node’s leadership, we find the relation between its extremal index and the minimal spreading time.

Natalia M. Markovich, Maxim S. Ryzhov

Sensitivity Analysis of a k-out-of-n:F System Characteristics to Shapes of Input Distribution

The problem of sensitivity of a redundant system’s probability characteristics to shapes of the input distributions is considered. In some previous works, closed-form representations have been obtained for stationary characteristics of hot-standby redundant systems with exponential lifetime distribution of their elements and general distribution of their repair time. In the current paper we carry out the sensitivity analysis of a k-out-of $$-n $$ - n :F system with the help of a simulation approach. Comparison of analytic and simulation results is presented.

V. V. Rykov, N. M. Ivanova, D. V. Kozyrev

Prioritized Service of URLLC Traffic in Industrial Deployments of 5G NR Systems

The simultaneous support of enhanced mobile broadband (eMBB) and ultra-reliable low latency (URLLC) traffic types at the air interface in upcoming 5G New Radio systems is a challenging problem requiring new connection admission control and scheduling strategies. To enable this coexistence while still maintaining the prescribed quality-of-service guarantees the state of the art solutions utilize non-orthogonal multiple access and traffic isolation with explicit resource reservation. In this paper, we study an explicit prioritization of URLLC traffic over other services. Using the tools of queuing theory we mathematically characterize and investigate several techniques for priority-based resource allocation. Our results demonstrate that preemptive priority service is a viable option to fulfill strict delay and loss guarantees at the NR air interface. We also show that elasticity of lower priority eMBB service allows for additional capacity gains in terms of the eMBB session drop probability during the service.

Ekaterina Markova, Dmitri Moltchanov, Rustam Pirmagomedov, Daria Ivanova, Yevgeni Koucheryavy, Konstantin Samouylov

Milestone Developments in Quantum Information and No-Go Theorems

In this article we present milestone developments in the theory and application of quantum information from historical perspectives. The domain of quantum information is very promising to develop quantum computer, quantum communication and varieties of other applications of quantum technologies. We also give the light on experimental manifestations of major theoretical developments. In addition, we present important no-go theorems frequently used in quantum information along with ideas of their respective mathematical proofs.

K. K. Sharma, V. P. Gerdt, P. V. Gerdt

Practical Application of the Multi-model Approach in the Study of Complex Systems

Different kinds of models are used to study various natural and technical phenomena. Usually, the researcher is limited to using a certain kind of model approach, not using others (or even not realizing the existence of other model approaches). The authors believe that a complete study of a certain phenomenon should cover several model approaches. The paper describes several model approaches which we used in the study of the random early detection algorithm for active queue management. Both the model approaches themselves and their implementation and the results obtained are described.

Anna V. Korolkova, Dmitry S. Kulyabov, Michal Hnatič

Three Approaches in the Study of Recurrent Markovian and Semi-Markovian Processes

We present three classical methods in the study of dynamic and stationary characteristic of processes of Markovian or Semi-Markovian type which possess points of regeneration. Our focus is on the stationary distributions and conditions of its existence and use.The first approach is based on detailed probability analysis of time dependent passages between the states of the process at a given moment. We call this approach Kolmogorov approach.The second approach uses the probability meaning of Laplace-Stieltjes transformation and of the probability generating functions/ Some additional arteficial excrement construction is used to show how derive direct relationships between these functions and how to find them explicitly.The third approach obtains relationships between the stationary characteristics of the process by use of so called “equations of equilibrium”. The input flow in each state must be equal to the respective output flow from that state. In such a way no accumulations should happen on each of that states when process gets its equilibrium.In all the illustrations of the these approaches we analyze a dynamic Marshal-Olkin reliability model with dependent components functioning in parallel. Results on this example are new.

Boyan Dimitrov, Vladimir Rykov, Sahib Esa

The Remaining Busy Time in a Retrial System with Unreliable Servers

In this paper, we consider a multiserver retrial queuing system with unreliable servers class-dependent retrial rates and N classes of customers following Poisson input processes. We analyze the distribution of the stationary generalized remaining service time which includes all unavailable periods (setup times) occurring during service of the customer. During service of a class-i customer, the interruptions occur according to the i-dependent Poisson process and the following i-dependent random setup time of the server. We consider two following disciplines caused by the service interruptions: preemptive repeat different and preemptive resume. Using coupling method and regenerative approach, we derive the stationary distribution of the generalized remaining service time in an arbitrary server. For each class i, this distribution is expressed as a convolution of the corresponding original service times and setup times, and in general is available in the terms of the Laplace-Stieltjes transform allowing to calculate the moments of the target distribution. Some numerical examples are included as well.

Evsey Morozov, Taisia Morozova

Diffusion Approximation for Multiserver Retrial Queue with Two-Way Communication

In this paper, we consider a multiserver retrial queue with two-way communication. Incoming calls arrive according to the stationary Poisson process and occupy the servers. Durations of incoming calls have an exponential distribution. If all the servers are busy upon arrival, the incoming call joins the orbit. The time spent by the call in the orbit is an exponential random variable. Idle servers also make outgoing calls whose durations follow an exponential distribution. We derive the diffusion limits of the number of calls in the orbit and the approximation of its stationary probability distribution.

Anatoly Nazarov, Tuan Phung-Duc, Svetlana Paul, Olga Lizyura

On a Single Server Queueing Inventory System

In this paper, we consider a queueing inventory model. Customers arriving to a single server queueing system in which, server uses some commodities from the inventory to fulfill the service. Service of a customer is initiated only when the server is free and the commodities for the service are available in the inventory. Otherwise, the arriving customer joins the queue. The inventory level is managed with (s,S) policy. Lead time is assumed to be positive. Steady state analysis of the model is performed. Various performance measures are estimated. A numerical example is provided.

K. A. K. AL Maqbali, V. C. Joshua, A. Krishnamoorthy

Rare-Event Simulation for the Hitting Time of Gaussian Processes

In reliability theory and network performance analysis a relevant role is played by the time needed to reach a given threshold, known in probability theory as hitting time. Although such issue has been widely investigated, closed-form results are available only for independent increments of the input process. Hence, in this paper we focus on the estimation of the upper tail of the hitting time distribution for general Gaussian processes by means of discrete-event simulation. Indeed, Gaussian processes often arise as a powerful modelling tool in many real-life systems and suitable ad-hoc techniques have developed for their analysis and simulation. Since the event of interest becomes rare as the threshold increases, a variant of Conditional Monte Carlo, based on the bridge process, is introduced and the explicit expression of the estimator is derived. Finally, simulation results highlight the unbiasedness and effectiveness (in terms of relative error) of the proposed approach.

Oleg Lukashenko, Michele Pagano

A Queueing Inventory System with Two Channels of Service

We consider a queueing inventory model with positive service time and two service channels. Channel I is a single server facility and channel II is a bulk service facility. There are two types of customers, type-I and type-II. The same type of commodity is served to both types of customers. Channel I provides service to type-I customers and channel II provides service to type-II customers. Service is initiated only if inventory is available. Bulk service is initiated at the end of a random clock or by the accumulation of N type-II customers. The inventory replenishment follows the (s, S) policy with positive leadtime. The service time follows phase type distribution. Steady state analysis of the model is performed. Some performance measures are evaluated.

Nisha Mathew, V. C. Joshua, A. Krishnamoorthy

The Analytical Model of Six-Dimensional Linear Dynamic Systems with Arbitrary Piecewise-Constant Parameters

In this paper the $$6 \times 6$$ 6 × 6 fundamental matrix of a linear six - dimensional system with arbitrary piecewise-constant parameters is obtained in the analytical form in elementary functions for the first time. To write this matrix the new sign-function $$F_q^{(i,j)}$$ F q ( i , j ) is introduced and described here. This function determines the order of interaction of eigen-oscillations of intervals with constant parameters in a resulting equivalent oscillation. The fundamentally new concept of equivalent oscillations for six-dimensional systems is introduced for the first time too. The stability problem of these systems is considered and the stability conditions are obtained in the analytical form in original parameters of a linear six-dimensional system. The results obtained allow solving a number of problems in mechanics, electrical engineering, communication systems, optoelectronics, automatic control theory and information theory. It can be, for example, the problems of determining the stability of dynamical systems, the problems of synthesizing dynamical systems of various nature, etc.

K. A. Vytovtov, E. A. Barabanova, V. M. Vishnevsky, I.Yu. Kvyatkovskaya

Evaluation of the End-to-End Delay of a Multiphase Queuing System Using Artificial Neural Networks

The article discusses a multiphase queuing system with a recurrent incoming flow, storage units of unlimited capacity, and recurrent service times at its nodes. To analyze and evaluate the main parameters of the performance of this network, a new approach is presented, which consists in combining simulation with data mining methods. This approach is universal in nature, and its area of application is not limited to studies of tandem networks only, but, on the contrary, extends to any other networks of arbitrary topology, as well as to queuing systems with various combinations of its main components. To assess the quality of the characteristics obtained, namely the average end-to-end delay, a comparative analysis is carried out with the known approximate results obtained in earlier works.

A. V. Gorbunova, V. M. Vishnevsky, A. A. Larionov

Model of Navigation and Control System of an Airborne Mobile Station

In this paper the theoretical foundation of the navigation system for airborne mobile communication objects is proposed. This system is served for detection of a positioning location, a space shift, speed of an air mobile station based on a drone. The working frequency of the radio-frequency navigation system is chosen to be 2.4 GHz. First of all the new approach of positioning location control and the method of space coordinate calculation are proposed by authors. To design such the system the helix antenna is proposed. The main parameters of it are calculated and the numerical simulation is carried out. The analytical method of velocity and acceleration determination for airborne mobile communication objects is developed too.

V. M. Vishnevsky, K. A. Vytovtov, E. A. Barabanova

Modeling D2D-Enhanced IoT Connectivity: An Approach Through the Simplified Analytical Framework

In the light of the proliferation of the Internet of Things (IoT), the device-to-device (D2D) communication is becoming a promising technology and a key enabler for enhancing the energy efficiency of the wireless network environment and reducing the traffic latency between user equipments (UEs) within their communication range. This papers considers one simplified analytical framework for modeling of communication offloading scenario within the D2D communications underlying cellular network, in which a UE generates a session toward another UE in the same cell. The model aims to improve the system capacity and energy-efficiency by offloading cellular traffic onto D2D communications, when the source and destination UEs are proximate enough to satisfy their QoS requirements.

Tatiana Milovanova, Dmitry Kozyrev

Optimization of SPTA Acquisition for a Distributed Communication Network of Weather Stations

We consider a topology of a distributed communication network of weather stations and its hardware configuration. The operability criteria for the considered network are formulated. The theoretical foundations for optimizing the acquisition of spare parts, tools and accessories (SPTA) for distributed systems are considered. An algorithm for SPTA optimization for distributed systems has been developed. Optimization of the SPTA acquisition for the hardware components of the distributed communication network of 24 weather stations was carried out. The acquisition options are proposed, taking into account various criteria.

Evgeny Golovinov, Dmitry Aminev, Sergey Tatunov, Sergey Polesskiy, Dmitry Kozyrev

Distributed Systems Applications

Frontmatter

Crystal-Ball and Magic Wand Combined: Predicting Situations and Making Them Happen

The Internet of Things (IoT) envisions an ecosystem in which everyday objects are enhanced with sensing, computation, and communication capabilities. These ‘smart’ devices (i.e., IoT devices) can sense and collect considerable amounts of data and share it with each other via the Internet. This paper proposes an IoT middleware platform enhanced with context- and situation-prediction capability, called Context-Prediction-as-a-Service (CPaaS). CPaaS offers real-time context prediction capabilities to a variety of IoT applications as a service and enables more effective decision support using relevant validated dependable information. A number of use cases where CPaaS can be deployed are also discussed.

Arkady Zaslavsky, Ali Hassani, Pari Delir Haghighi, Antonio Robles-Kelly, Panos K. Chrysanthis

Architectural ML Framework for IoT Services Delivery Based on Microservices

The Internet of Things (IoT) is the interconnection of devices and services that allows free data flow. Managing and analyzing this data is the actual added value that IoT is beneficial for. Machine learning plays an increasingly important role in performing data analysis in IoT solutions. This paper presents an architectural framework with machine learning solutions implemented as a service in the microservice group. This architectural framework for IoT services delivery is designed following the Agile methodology. The requirements for the software architecture and expected functionalities of the system are defined. The microservices collection is explained by providing a separate description for every service. Machine learning (ML) analytics on IoT (as the processing paradigm for intelligently handling the IoT data) is represented as a part of the microservice platform. Several strategic advantages of the proposed microservice-based IoT architecture over others are discussed together with implementation issues.

Kristina Dineva, Tatiana Atanasova

Applying Machine Learning to Data from a Structured Database in a Research Institute to Support Decision Making

The article deals with the problems of creating an information base for accumulating data on the scientific activities of employees in the research organization, to study these data as the basis of scientometrics indicators and to provide tools for making strategical decisions at research institutions. The study is based on Weka – open source machine learning environment that gives several mechanisms for finding consistent patterns in the data.

Nina Bakanova, Tatiana Atanasova, Arsenii Bakanov

Generation of Metadata for Network Control

In this paper we propose a method of ensuring network information security by controlling network connections using metadata. The metadata contains information about admissible task interactions and application locations in a corporate network.Metadata is formed based on a mathematical model of an information technology. Information technology models are represented in the form of directed acyclic graphs. This paper describes hierarchical decompositions of directed acyclic graphs. Hierarchical decomposition makes it possible to optimally form blocks of information transformations for their placement on hosts of distributed information and computing system (DICS). Protocols of metadata usage are briefly discussed. The problem of transforming directed acyclic graphs into metadata is solved. Security risks introduced by user interaction with information technology process are considered.

Alexander Grusho, Nick Grusho, Michael Zabezhailo, Elena Timonina

Backmatter

Additional information

Premium Partner

    Image Credits