Distributed Computer and

Communication Networks: 22nd

International Conference, DCCN 2019,
Moscow, Russia, September 23–27,
2019, Revised Selected Papers Vladimir
M. Vishnevskiy
Visit to download the full and correct content document:
https://textbookfull.com/product/distributed-computer-and-communication-networks-2
2nd-international-conference-dccn-2019-moscow-russia-september-23-27-2019-revis
ed-selected-papers-vladimir-m-vishnevskiy/
More products digital (pdf, epub, mobi) instant
download maybe you interests ...

Distributed Computer and Communication Networks: 22nd

International Conference, DCCN 2019, Moscow, Russia,
September 23–27, 2019, Revised Selected Papers Vladimir
M. Vishnevskiy
https://textbookfull.com/product/distributed-computer-and-
communication-networks-22nd-international-conference-
dccn-2019-moscow-russia-september-23-27-2019-revised-selected-
papers-vladimir-m-vishnevskiy/

Distributed Computer and Communication Networks 19th

International Conference DCCN 2016 Moscow Russia
November 21 25 2016 Revised Selected Papers 1st Edition
Vladimir M. Vishnevskiy
https://textbookfull.com/product/distributed-computer-and-
communication-networks-19th-international-conference-
dccn-2016-moscow-russia-november-21-25-2016-revised-selected-
papers-1st-edition-vladimir-m-vishnevskiy/

Distributed Computer and Communication Networks 17th

International Conference DCCN 2013 Moscow Russia
October 7 10 2013 Revised Selected Papers 1st Edition
Vladimir Vishnevsky
https://textbookfull.com/product/distributed-computer-and-
communication-networks-17th-international-conference-
dccn-2013-moscow-russia-october-7-10-2013-revised-selected-
papers-1st-edition-vladimir-vishnevsky/

Distributed Computer and Communication Networks 18th

International Conference DCCN 2015 Moscow Russia
October 19 22 2015 Revised Selected Papers 1st Edition
Vladimir Vishnevsky
https://textbookfull.com/product/distributed-computer-and-
communication-networks-18th-international-conference-
dccn-2015-moscow-russia-october-19-22-2015-revised-selected-
Supercomputing 6th Russian Supercomputing Days RuSCDays
2020 Moscow Russia September 21 22 2020 Revised
Selected Papers Vladimir Voevodin

https://textbookfull.com/product/supercomputing-6th-russian-
supercomputing-days-ruscdays-2020-moscow-russia-
september-21-22-2020-revised-selected-papers-vladimir-voevodin/

Analysis of Images Social Networks and Texts 8th

International Conference AIST 2019 Kazan Russia July 17
19 2019 Revised Selected Papers Wil M. P. Van Der Aalst

https://textbookfull.com/product/analysis-of-images-social-
networks-and-texts-8th-international-conference-aist-2019-kazan-
russia-july-17-19-2019-revised-selected-papers-wil-m-p-van-der-
aalst/

Supercomputing Second Russian Supercomputing Days

RuSCDays 2016 Moscow Russia September 26 27 2016
Revised Selected Papers 1st Edition Vladimir Voevodin

https://textbookfull.com/product/supercomputing-second-russian-
supercomputing-days-ruscdays-2016-moscow-russia-
september-26-27-2016-revised-selected-papers-1st-edition-
vladimir-voevodin/

Cognitive Cities Second International Conference IC3

2019 Kyoto Japan September 3 6 2019 Revised Selected
Papers Jian Shen

https://textbookfull.com/product/cognitive-cities-second-
international-conference-ic3-2019-kyoto-japan-
september-3-6-2019-revised-selected-papers-jian-shen/

Mathematical Optimization Theory and Operations

Research 18th International Conference MOTOR 2019
Ekaterinburg Russia July 8 12 2019 Revised Selected
Papers Igor Bykadorov
https://textbookfull.com/product/mathematical-optimization-
theory-and-operations-research-18th-international-conference-
motor-2019-ekaterinburg-russia-july-8-12-2019-revised-selected-
 Vladimir M. Vishnevskiy
Konstantin E. Samouylov
Dmitry V. Kozyrev (Eds.)
LNCS 11965

Distributed Computer
and Communication Networks
22nd International Conference, DCCN 2019
Moscow, Russia, September 23–27, 2019
Revised Selected Papers
Lecture Notes in Computer Science 11965

Founding Editors
Gerhard Goos
Karlsruhe Institute of Technology, Karlsruhe, Germany
Juris Hartmanis
Cornell University, Ithaca, NY, USA

Editorial Board Members

Elisa Bertino
Purdue University, West Lafayette, IN, USA
Wen Gao
Peking University, Beijing, China
Bernhard Steffen
TU Dortmund University, Dortmund, Germany
Gerhard Woeginger
RWTH Aachen, Aachen, Germany
Moti Yung
Columbia University, New York, NY, USA
More information about this series at http://www.springer.com/series/7411
Vladimir M. Vishnevskiy •

Konstantin E. Samouylov •

Dmitry V. Kozyrev (Eds.)

Distributed Computer
and Communication Networks
22nd International Conference, DCCN 2019
Moscow, Russia, September 23–27, 2019
Revised Selected Papers

123
Editors
Vladimir M. Vishnevskiy Konstantin E. Samouylov
V.A. Trapeznikov Institute of Control Peoples’ Friendship University of Russia
Sciences of Russian Academy of Sciences Moscow, Russia
Moscow, Russia
Dmitry V. Kozyrev
V.A. Trapeznikov Institute of Control
Sciences of Russian Academy of Sciences
Moscow, Russia
Peoples’ Friendship University of Russia
Moscow, Russia

ISSN 0302-9743 ISSN 1611-3349 (electronic)

Lecture Notes in Computer Science
ISBN 978-3-030-36613-1 ISBN 978-3-030-36614-8 (eBook)
https://doi.org/10.1007/978-3-030-36614-8
LNCS Sublibrary: SL5 – Computer Communication Networks and Telecommunications

© Springer Nature Switzerland AG 2019

This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the
material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation,
broadcasting, reproduction on microfilms or in any other physical way, and transmission or information
storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now
known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication
does not imply, even in the absence of a specific statement, that such names are exempt from the relevant
protective laws and regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in this book are
believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors
give a warranty, expressed or implied, with respect to the material contained herein or for any errors or
omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in
published maps and institutional affiliations.

This Springer imprint is published by the registered company Springer Nature Switzerland AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
 Preface

This volume contains a collection of revised selected full-text papers presented at the
22nd International Conference on Distributed Computer and Communication Networks
(DCCN 2019), held in Moscow, Russia, September 23–27, 2019.
DCCN 2019 is an IEEE (Region 8 + Russia Section) technically cosponsored
international conference. It is a continuation of traditional international conferences
of the DCCN series, which took place in Sofia, Bulgaria (1995, 2005, 2006, 2008,
2009, 2014); Tel Aviv, Israel (1996, 1997, 1999, 2001); and Moscow, Russia (1998,
2000, 2003, 2007, 2010, 2011, 2013, 2015, 2016, 2017, 2018) in the last 22 years. The
main idea of the conference is to provide a platform and forum for researchers and
developers from academia and industry from various countries working in the area
of theory and applications of distributed computer and communication networks,
mathematical modeling, methods of control, and optimization of distributed systems,
by offering them a unique opportunity to share their views as well as discuss the
perspective developments and pursue collaboration in this area. The content of this
volume is related to the following subjects:
1. Communication networks algorithms and protocols
2. Wireless and mobile networks
3. Computer and telecommunication networks control and management
4. Performance analysis, QoS/QoE evaluation, and network efficiency
5. Analytical modeling and simulation of communication systems
6. Evolution of wireless networks toward 5G
7. Centimeter- and millimeter-wave radio technologies
8. Internet of Things and Fog Computing
9. Probabilistic and statistical models in information systems
10. Queuing theory and reliability theory applications
11. High-altitude telecommunications platforms
The DCCN 2019 conference gathered 174 submissions from authors from 26 dif-
ferent countries. From these, 132 high quality papers in English were accepted and
presented during the conference. The current volume contains 46 extended papers
which were recommended by session chairs and selected by the Program Committee
for the Springer post-proceedings.
All the papers selected for the post-proceedings volume are given in the form
presented by the authors. These papers are of interest to everyone working in the field
of computer and communication networks.
vi Preface

We thank all the authors for their interest in DCCN, the members of the Program
Committee for their contributions, and the reviewers for their peer-reviewing efforts.

September 2019 Vladimir Vishnevskiy

Konstantin Samouylov
 Organization

DCCN 2019 was jointly organized by the Russian Academy of Sciences (RAS), the
V.A. Trapeznikov Institute of Control Sciences of RAS (ICS RAS), the Peoples’
Friendship University of Russia (RUDN University), the National Research Tomsk
State University, and the Institute of Information and Communication Technologies of
Bulgarian Academy of Sciences (IICT BAS).

International Program Committee

V. M. Vishnevskiy (Chair) ICS RAS, Russia
K. E. Samouylov (Co-chair) RUDN University, Russia
Ye. A. Koucheryavy Tampere University of Technology, Finland
(Co-chair)
S. M. Abramov Program Systems Institute of RAS, Russia
S. D. Andreev Tampere University of Technology, Finland
A. M. Andronov Riga Technical University, Latvia
N. Balakrishnan McMaster University, Canada
A. S. Bugaev Moscow Institute of Physics and Technology, Russia
S. R. Chakravarthy Kettering University, USA
T. Czachorski Institute of Computer Science of Polish Academy
of Sciences, Poland
A. N. Dudin Belarusian State University, Belarus
D. Deng National Changhua University of Education, Taiwan
A. V. Dvorkovich Moscow Institute of Physics and Technology, Russia
Yu. V. Gaidamaka RUDN University, Russia
P. Gaj Silesian University of Technology, Poland
D. Grace York University, UK
Yu. V. Gulyaev Kotelnikov Institute of Radio-engineering
and Electronics of RAS, Russia
J. Hosek Brno University of Technology, Czech Republic
V. C. Joshua CMS College, India
H. Karatza Aristotle University of Thessaloniki, Greece
N. Kolev University of São Paulo, Brazil
J. Kolodziej Cracow University of Technology, Poland
G. Kotsis Johannes Kepler University Linz, Austria
T. Kozlova Madsen Aalborg University, Denmark
U. Krieger University of Bamberg, Germany
A. Krishnamoorthy Cochin University of Science and Technology, India
A. E. Koucheryavy Bonch-Bruevich Saint-Petersburg State University
of Telecommunications, Russia
Ye. A. Koucheryavy Tampere University of Technology, Finland
viii Organization

N. A. Kuznetsov Moscow Institute of Physics and Technology, Russia

L. Lakatos Budapest University, Hungary
E. Levner Holon Institute of Technology, Israel
S. D. Margenov Institute of Information and Communication
Technologies of Bulgarian Academy of Sciences,
Bulgaria
N. Markovich ICS RAS, Russia
A. Melikov Institute of Cybernetics of the Azerbaijan National
Academy of Sciences, Azerbaijan
G. K. Miscoi Academy of Sciences of Moldova, Moldavia
E. V. Morozov Institute of Applied Mathematical Research
of the Karelian Research Centre RAS, Russia
V. A. Naumov Service Innovation Research Institute (PIKE), Finland
A. A. Nazarov Tomsk State University, Russia
I. V. Nikiforov Université de Technologie de Troyes, France
P. Nikitin University of Washington, USA
S. A. Nikitov Institute of Radio-engineering and Electronics
of RAS, Russia
D. A. Novikov ICS RAS, Russia
M. Pagano Pisa University, Italy
E. Petersons Riga Technical University, Latvia
V. V. Rykov Gubkin Russian State University of Oil and Gas,
Russia
L. A. Sevastianov RUDN University, Russia
M. A. Sneps-Sneppe Ventspils University College, Latvia
P. Stanchev Kettering University, USA
S. N. Stepanov Moscow Technical University of Communication
and Informatics, Russia
S. P. Suschenko Tomsk State University, Russia
J. Sztrik University of Debrecen, Hungary
H. Tijms Vrije Universiteit Amsterdam, The Netherlands
S. N. Vasiliev ICS RAS, Russia
M. Xie City University of Hong Kong, Hong Kong, China
Yu. P. Zaychenko Kyiv Polytechnic Institute, Ukraine

Organizing Committee
V. M. Vishnevskiy (Chair) ICS RAS, Russia
K. E. Samouylov RUDN University, Russia
(Vice Chair)
D. V. Kozyrev RUDN University and ICS RAS, Russia
A. A. Larionov ICS RAS, Russia
S. N. Kupriyakhina ICS RAS, Russia
S. P. Moiseeva Tomsk State University, Russia
T. Atanasova IIICT BAS, Bulgaria
I. A. Gudkova RUDN University, Russia
 Organization ix

S. I. Salpagarov RUDN University

D. Yu. Ostrikova RUDN University

Organizers and Partners

Organizers
Russian Academy of Sciences
V.A. Trapeznikov Institute of Control Sciences of RAS
RUDN University
National Research Tomsk State University
Institute of Information and Communication Technologies of Bulgarian Academy of
Sciences
Research and Development Company “Information and Networking Technologies”

Support
Information support is provided by the IEEE (Region 8 + Russia Section) and the
Russian Academy of Sciences. The conference has been organized with the support
of the “RUDN University Program 5-100.”
 Contents

Computer and Communication Networks

5G New Radio System Performance Analysis Using Limited Resource

Queuing Systems with Varying Requirements. . . . . . . . . . . . . . . . . . . . . . . 3
Valeriy Naumov, Vitalii Beschastnyi, Daria Ostrikova,
and Yuliya Gaidamaka

On the Performance of LoRaWAN in Smart City: End-Device Design

and Communication Coverage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
Dmitry Poluektov, Michail Polovov, Petr Kharin, Martin Stusek,
Krystof Zeman, Pavel Masek, Irina Gudkova, Jiri Hosek,
and Konstantin Samouylov

Adaptive Cyclic Polling Systems: Analysis and Application

to the Broadband Wireless Networks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
V. M. Vishnevsky, O. V. Semenova, D. T. Bui, and Alexander Sokolov

Multichannel Diffusion Approximation Models for the Evaluation

of Multichannel Communication Networks . . . . . . . . . . . . . . . . . . . . . . . . . 43
Tadeusz Czachórski, Godlove Suila Kuaban, and Tomasz Nycz

Model and Algorithm of Next Generation Optical Switching Systems

Based on 8
8 Elements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
E. Barabanova, K. Vytovtov, and V. Podlazov

Characterizing the Degree of LTE Involvement in Supporting

Session Continuity in Street Deployment of NR Systems . . . . . . . . . . . . . . . 71
Vyacheslav Begishev, Andrey Samuylov, Dmitri Moltchanov,
and Konstantin Samouylov

Dolph-Chebyshev and Barcilon-Temes Window Functions Modification . . . . 84

V. P. Dvorkovich and A. V. Dvorkovich

Principles of Building a Power Transmission System for Tethered

Unmanned Telecommunication Platforms . . . . . . . . . . . . . . . . . . . . . . . . . . 94
V. M. Vishnevsky, B. N. Tereschenko, D. A. Tumchenok,
A. M. Shirvanyan, and Alexander Sokolov
xii Contents

On the Stability of D2D Connection with the Use of Kinetic Equation

for SIR Empirical Distribution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111
Yurii Orlov, Anastasia Ivchenko, Natalia Podzharaya,
Anastasiia Sochenkova, Vsevolod Shorgin, Aliaksandr Birukou,
Yuliya Gaidamaka, and Konstantin Samouylov

The Use of Asymmetric Numeral Systems Entropy Encoding

in Video Compression . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125
Fedor Konstantinov, Gennady Gryzov, and Kirill Bystrov

Statistical Model of Computing Experiment on Digital Color Correction . . . . 140

E. V. Borevich, S. V. Mescheryakov, and V. E. Yanchus

Synthesis of High-Performance Window Functions Using Minimization

of Difference Between Its Waveform and Spectrum. . . . . . . . . . . . . . . . . . . 151
V. P. Dvorkovich and A. V. Dvorkovich

Analytical Modeling of Distributed Systems

Optimal Control by the Queue with Rate and Quality of Service Depending
on the Amount of Harvested Energy as a Model of the Node of Wireless
Sensor Network . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165
Alexander Dudin, Chesoong Kim, and Sergey Dudin

On Optimal Control Policy of MAP(t)/M/2 Queueing System

with Heterogeneous Servers and Periodic Arrival Process. . . . . . . . . . . . . . . 179
Dmitry Efrosinin and Natalia Stepanova

Estimation of the Parameters of Continuous-Time Finite Markov Chain . . . . . 195

Alexander Andronov, Irina Jackiva (Yatskiv), and Diana Santalova

Asymptotic-Diffusion Analysis for Retrial Queue with Batch Poisson

Input and Multiple Types of Outgoing Calls. . . . . . . . . . . . . . . . . . . . . . . . 207
Anatoly Nazarov, Tuan Phung-Duc, Svetlana Paul, and Olga Lizura

A Multistage Queueing Model with Priority for Customers Become Fit . . . . . 223
Dhanya Babu, V. C. Joshua, and A. Krishnamoorthy

Renewal Redundant Systems Under the Marshall-Olkin

Failure Model. Sensitivity Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 234
Vladimir Rykov and Boyan Dimitrov

Unreliable Queueing System with Threshold Strategy of the Backup

Server Connection. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 249
Valentina Klimenok, Chesoong Kim, V. M. Vishnevsky,
and Alexander Dudin
 Contents xiii

Detection and Detectability of Changes in a Multi-parameter

Exponential Distribution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 263
Igor Nikiforov

Distribution Parameters Estimation in Recurrent Synchronous

Generalized Doubly Stochastic Flow of the Second Order . . . . . . . . . . . . . . 276
Lyudmila Nezhel’skaya, Michele Pagano, and Ekaterina Sidorova

Queue with Partially Ignored Interruption in Markovian Environment . . . . . . 289

A. Krishnamoorthy and S. Jaya

Modeling and Reliability Analysis of a Redundant Transport System

in a Markovian Environment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 302
Udo R. Krieger and Natalia Markovich

Heterogeneous Queueing System MAP/GIðnÞ /1 with Random

Customers’ Capacities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 315
Ekaterina Lisovskaya, Ekaterina Pankratova, Yuliya Gaidamaka,
Svetlana Moiseeva, and Michele Pagano

Cluster Modeling of Lindley Process with Application to Queuing . . . . . . . . 330

Natalia Markovich and Rostislav Razumchik

Discrete-Time Insurance Models. Optimization of Their Performance

by Reinsurance and Bank Loans . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 342
Ekaterina V. Bulinskaya

Hidden Markov Model of Information System with Component-Wise

Storage Devices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 354
Yuriy E. Obzherin, Stanislav M. Sidorov, and Mikhail M. Nikitin

Statistical Distributions of Partial Correlators of Network Traffic

Aggregated Packets for Distinguishing DDoS Attacks . . . . . . . . . . . . . . . . . 365
Andrey Evgenievich Krasnov and Dmitrii Nikolaevich Nikol’skii

Approximate Product Form Solution for Performance Analysis

of Wireless Network with Dynamic Power Control Policy . . . . . . . . . . . . . . 379
Yves Adou, Ekaterina Markova, and Irina Gudkova

The Modeling of Call Center Functioning in Case of Overload . . . . . . . . . . . 391

Sergey N. Stepanov, Mikhail S. Stepanov, and Hanna M. Zhurko

Evaluation of Information Transmission Resource While Processing

Heterogeneous Traffic in Data Networks . . . . . . . . . . . . . . . . . . . . . . . . . . 407
Veronika M. Antonova, Natalia A. Grechishkina,
Ludmila Yu. Zhilyakova, and Nickolay A. Kuznetsov
xiv Contents

On Failure Rate Comparison of Finite Multiserver Systems . . . . . . . . . . . . . 419

Evsey Morozov, Irina Peshkova, and Alexander Rumyantsev

Queue with Retrial Group for Modeling Best Effort Traffic with Minimum
Bit Rate Guarantee Transmission Under Network Slicing . . . . . . . . . . . . . . . 432
Ekaterina Markova, Yves Adou, Daria Ivanova, Anastasia Golskaia,
and Konstantin Samouylov

Reliability Model of a Homogeneous Warm-Standby Data Transmission

System with General Repair Time Distribution . . . . . . . . . . . . . . . . . . . . . . 443
H. G. K. Houankpo and Dmitry Kozyrev

Distributed Systems Applications

Methodology for Data Processing in Modular IoT System . . . . . . . . . . . . . . 457

Kristina Dineva and Tatiana Atanasova

Effect of Heterogeneous Traffic on Quality of Service in 5G Network . . . . . . 469

Omar Abdulkareem Mahmood, Abdukodir Khakimov,
Ammar Muthanna, and Alexander Paramonov

Flying Ad-Hoc Network for Emergency Based on IEEE 802.11p

Multichannel MAC Protocol. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 479
Truong Duy Dinh, Duc Tran Le, Thi Thu Thao Tran,
and Ruslan Kirichek

Minimizing the IoT System Delay with the Edge Gateways . . . . . . . . . . . . . 495
Van Dai Pham, Trung Hoang, Ruslan Kirichek, Maria Makolkina,
and Andrey Koucheryavy

States with Minimum Dispersion of Observables in Kuryshkin-Wodkiewicz

Quantum Mechanics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 508
A. V. Zorin, L. A. Sevastianov, and N. P. Tretyakov

Multi-criteria Method for Calculating a SPTA Package

for a Mobile Communications Vehicle . . . . . . . . . . . . . . . . . . . . . . . . . . . . 520
Bogdan Pankovsky and Sergey Polesskiy

Leaky Modes in Laser-Printed Integrated Optical Structures . . . . . . . . . . . . . 534

A. A. Egorov, D. V. Divakov, K. P. Lovetskiy, A. L. Sevastianov,
and L. A. Sevastianov

Information Flow Control on the Basis of Meta Data. . . . . . . . . . . . . . . . . . 548

Alexander Grusho, Nick Grusho, and Elena Timonina
 Contents xv

Application of Machine Learning Algorithms to Handle Missing Values

in Precipitation Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 563
Andrey Gorshenin, Mariia Lebedeva, Svetlana Lukina,
and Alina Yakovleva

The Bans in Finite Probability Spaces and the Problem of Small Samples . . . 578
Alexander Grusho, Nick Grusho, and Elena Timonina

Reliability Evaluation of a Distributed Communication Network

of Weather Stations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 591
Dmitry Aminev, Evgeny Golovinov, Dmitry Kozyrev, Andrey Larionov,
and Alexander Sokolov

Large-Scale Centralized Scheduling of Short-Range Wireless Links. . . . . . . . 607

Alexander Pyattaev and Mikhail Gerasimenko

Author Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 617

Computer and Communication
Networks
 5G New Radio System Performance
Analysis Using Limited Resource Queuing
Systems with Varying Requirements

Valeriy Naumov1 , Vitalii Beschastnyi2 , Daria Ostrikova2 ,

and Yuliya Gaidamaka2,3(B)
1
Service Innovations Research Institute, 8 Annankatu Street, Helsinki 00120, Finland
valeriy.naumov@pf.fi
2
Peoples’ Friendship University of Russia (RUDN University),
6 Miklukho-Maklaya Street, Moscow 117198, Russian Federation
{beschastnyy-va,ostrikova-dyu,gaydamaka-yuv}@rudn.ru
3
Federal Research Center “Computer Science and Control” of the Russian Academy
of Sciences (FRC CSC RAS), 44-2 Vavilov Street, Moscow 119333, Russian Federation

Abstract. Prospective 5G New Radio (NR) systems offer unprece-

dented capacity boost by the ultradense deployments of small cells
operating at mmWave frequencies, with massive available bandwidths.
They will facilitate the provisioning of exceptionally demanding mission-
critical and resource-hungry applications that are envisaged to utilize the
5G communications infrastructure. In this work, we provide an analyt-
ical framework for 5G NR system analysis in terms of queuing theory.
We consider a multiservice queuing system with a limited resource with
customers that demand varying amount of resource within their service
time. Such an approach provides more accurate performance evaluation
compared to conventional multiservice models. For the considered model,
we propose a method that allows to calculate the stationary probability
distribution to the specified accuracy. Our findings are illustrated with
a numerical example.

Keywords: 5G NR · Blocking probability · Loss system · Limited

resource · Random requirements

1 Introduction
5G NR systems bring a set of new unique challenges to systems designers includ-
ing much higher propagation losses compared to microwave communications,
blockage of propagation paths by small dynamic objects in the channel, the
need for efficient electronic beamstreering mechanisms, etc. In real-life outdoor
The publication has been prepared with the support of the “RUDN University Pro-
gram 5-100” (D.Yu. Ostrikova, visualization). The reported study was partially funded
by RFBR, projects Nos. 18-07-00576 (Yu.V. Gaidamaka, methodology and project
administration) and 18-37-00380 (V.A. Beschastnyi, numerical analysis).
c Springer Nature Switzerland AG 2019
V. M. Vishnevskiy et al. (Eds.): DCCN 2019, LNCS 11965, pp. 3–14, 2019.
https://doi.org/10.1007/978-3-030-36614-8_1
4 V. Naumov et al.

deployments 5G NR systems mostly suffer from mobile obstacles such as humans

and cars, which are often termed “blockers” [1,2].
Depending on the propagation environment and the distance to NR base
station (BS), a user equipment (UE) experiencing such type of blockage may
either enter outage conditions or lower its modulation and coding scheme such
that block error probability at the air interface is satisfied [3]. To target out-
age situations 3GPP has recently proposed multi-connectivity operation, where
several simultaneously active links for adjacent NR BSs are maintained and the
connection is transferred between them in case of blockage events [4,5]. When no
outage conditions are experiences by UE, the service may continue at the cur-
rent BS. However, to support the required rate at the air interface more physical
resources are needed. When this surplus of resources is not available an ongoing
session is dropped or its rate needs to be reduced.
Resource multiserver loss systems are widely used in analysis of communi-
cation systems [6,7]. In typical communication scenarios resource is allocated
once the session is initiated and the amount of the occupied resource is not
changed within the session lifetime. Meanwhile, there are many scenarios that
entail resource demand variation, particularly in wireless systems where demand
depends on channel quality which is heavily affected by mobility of objects. How-
ever, loss systems with varying requirements have not been well studied so far.
In this paper, we analyze a multiserver loss system with varying requirements,
derive formulations for the stationary probability distribution, and propose an
efficient accurate method that allows to calculate it. Finally, we apply our ana-
lytical results to the performance analysis of an 5G NR system. The rest of the
paper is organized as follows. First, in Sect. 2 we describe our system model of
an NR BS, then in Sect. 3 we introduce the analytical framework can be used as
a baseline for building analytical models for 5G NR networks, and the accurate
method. Numerical results are presented in Sect. 4. Conclusions are drawn in the
last section.

2 System Model
We consider the scenario with 5G NR Base Station (BS) deployment serving
point-to-point (unicast) sessions, see Fig. 1. Each BS has a circularly-shaped
coverage area of radius RC estimated using the mmWave propagation model,
the set of MCS [8], and network topology. Users are assumed to be randomly
distributed according to Poisson Point Process (PPP) with parameter ρ. So
the intensity of user requests for service is a Poisson process with parameter
λ = Λρπ(dE 2
LoS ) , where Λ is the parameter of exponentially distributed intervals
between two consecutive requests from a single UE and ρ is the density of users.
To process a user request BS allocates radio frequency resource of the size that
is generally a random variable and determined by the UE location. In accordance
with [9], the mmWave path loss LdB for UEs in LoS and nLoS conditions is given
by:
 5G NR System Performance Analysis Using Limited Resource Loss Systems 5

NR BS
O
UE with LoS

I UE without LoS
−1
RI Blocker

−1
nLoS area

Trajectory

Fig. 1. 5G NR Multiconnectivity deployment.


32.4 + 21 log(x) + 20 log fc , non-blocked,
LdB (x) =
47.4 + 21 log(x) + 20 log fc , blocked,
where fc is operational frequency measured in MHz, and x is the distance
between BS and UE. From these equations we derive maximum distances dE nLoS
and dELoS at which a UE can establish a session in LoS and nLoS states respec-
tively by setting value of LdB threshold as the worst possible Signal-to-Noise
Ratio (SNR) given by modulation and coding scheme (MCS) for mmWave prop-
agation model [3,10], as shown in (1) and (2):
⎛   ⎞ γ1
PT · 10
10GT +GR
dE
LoS =
⎝ √ ⎠ , (1)
10
10NO W +S · 102log10 fc +3,24

⎛   ⎞ γ1
PT · 10
10GT +GR
dE
nLoS =
⎝ √ ⎠ , (2)
10
10NO W +S · 102log10 fc +4,74

where PT is the transmit power, GT - BS antenna gain, GR - UE antenna gain,

No - the Johnson-Nyquist noise at one Hz, S - SNR value in dB, W - the
bandwidth assigned to the BS, and γ - path loss exponent.
From resource demand point of view, we divide the coverage area into two
subareas: the circularly-shaped “inner zone” with radius RI = dE nLoS ; and the
annulus-shaped “outer zone” with width RO = RC − RI , where RO is defined
by the radius of the coverage area RC . The coverage area is limited by either
dE
nLoS , or Inter-Site Distance (ISD), which is the distance between two adjacent
BSs. Thus, RC = min(dE LoS , ISD − dnLoS ), as there is no need to extend the
E

coverage to the inner zone of adjacent BS, where sessions can be maintained
regardless of LoS conditions.
6 V. Naumov et al.

LoS towards a NR BS may by blocked by mobile obstacles. In our model

we consider the most common type of blockers which is humans that have their
blocker radius rB . Movement of blockers can be modeled with two exponentially
distributed random variables with parameters θLoS and θnLoS that represent
blockage duration (nLoS state intensity) and time intervals between consecutive
blockages (LoS state intensity) [6].
Figure 2 illustrates resource allocation process in system time where event
aIn describes establishment of the n-session from inner zone, aO
n - establishment
of a new n-session from outer zone, dIn - end of the n-session service from inner
zone, τnnLoS - the moment when the UE with n-session enters nLoS conditions,
τnLoS - the moment when the UE enters LoS conditions, C = sWA - the number
of RBs in the pool of resources, and sA - the service unit (size of an RB in
frequency domain).

Fig. 2. Resource allocation in system time.

On UE request for new session initiation, NR BS first process the channel

quality data and defines a proper MCS. Then, in accordance with selected MCS
and requested service type it calculates the required number of RBs r. If r is less
or equal to the number of unallocated RBs from the whole pool of resources, the
UE request is accepted and the demanded resource is allocated starting from the
next time slot. If r is greater than the number of unallocated RBs, BS rejects
the request and the session is considered to be dropped.
In our model the resource demands are calculated as mean demands for
the corresponding zones and LoS conditions. Particularly, we divide a zone into
adjacent annuluses: one for each scheme from the set of MCS. Then, using the
spectral efficiently parameter we quantify the demand for each annulus and
finally get the weighted demand in accordance with the area of the annulus.
Demand of the blocked session from outer zone equals to 0 since the session is
 5G NR System Performance Analysis Using Limited Resource Loss Systems 7

handed over to an adjacent NR BS and do not require resource from the origin
BS until the end of blockage.
Table 1 describes the mapping between CQIs and spectral efficiency for 3GPP
NR systems. This approach follows [6], where authors compute mean demand
for the model without multiconnectivity feature.

Table 1. CQI, MCS and Spectral efficiency mapping.

CQI MCS Spectral efficiency SNR in dB RBs for 50 Mbit/s

1 QPSK, 78/1024 0,15237 −9,478 228
2 QPSK, 120/1024 0,2344 −6,658 149
3 QPSK, 193/1024 0,377 −4,098 93
4 QPSK, 308/1024 0,6016 −1,798 58
5 QPSK, 449/1024 0,877 0,399 40
6 QPSK, 602/1024 1,1758 2,424 30
7 16QAM, 378/1024 1,4766 4,489 24
8 16QAM, 490/1024 1,9141 6,367 19
9 16QAM, 616/1024 2,4063 8,456 15
10 16QAM, 466/1024 2,7305 10,266 13
11 16QAM, 567/1024 3,3223 12,218 11
12 16QAM, 666/1024 3,9023 14,122 9
13 16QAM, 772/1024 4,5234 15,849 8
14 16QAM, 873/1024 5,1152 17,786 7
15 16QAM, 948/1024 5,5547 19,809 7

LoS from a UE towards its NR BS may be blocked while serving an active

UE’s session. In this case the BS recalculates the new demand for the worsened
channel quality. If the number of additionally required RBs is less or equal to
the number of unallocated RBs, BS proceeds serving the session. Otherwise, the
session is dropped.
In our model we consider the appearance of LoS blockers for users through-
out the whole coverage area [11]. On LoS blocker appearance between UE and
BS at the distance less then dE nLoS the session can be still maintained by allo-
cating greater amount of resource to compensate the worsened channel quality,
thus we define the radius of “inner zone” RI = dE nLoS . The radius of “outer
zone” RO , that defines the distance from BS to the edge of the coverage area, is
directly connected to the notion of multiconnectivity. UE from outer zone can-
not maintain a session through the closest BS without LoS towards it. However,
the session may be handed over to another BS with LoS condition. Therefore,
radius of outer zone determines ISD that is dISD = RO + RI . If the distance
between UE and BS is greater than RC , user enters outage condition and cannot
be provided with a service.
8 V. Naumov et al.

In the outer zone when the LoS to UE is blocked, channel quality decreases
to such an extent that it is no longer worth or possible to maintain connection
between UE and current BS, and the session is handed over to the adjacent
BS. As UE is always served at BS with better channel quality, after handover it
will demand greater amount of resource than it was previously allocated by the
origin BS. On re-establishment of LoS towards origin BS, the session is handed
over back with the former demand. However, at this moment it may appear that
origin BS does not have available resource to resume session, in this case it is
lost.

3 Analytical Model
3.1 Queuing System Formalization

Consider a multiserver loss system with N ≤ ∞ servers and a Poisson arrival

flow with the rate λ [10]. Assume customer service time be independent of the
arrival process, independent of each other, and have a phase-type cumulative
distribution function (CDF) B(x) = 1 − aexM u, – where a is a row vector
of length n represents the probability distribution of initial phase, u - all-ones
column vector of length n, and M = [μij ] is non-degenerate square matrix of
order n that represents service inter-phase transition rates.
In our model customer demands for resource may vary within the service
time and depend on the service phase. On transition to the j-th service phase,
the customer demands for rj resource units. If at the moment of arrival or service
phase transition the amount of the demanded resource exceeds the amount of
available resource, the customer is considered to be lost. Let us denote R =
n
j=1 rj the total amount of resource.
The system state at the moment t is described by a Markovian process X(t) =
(X1 (t), ..., Xn (t)), where Xj (t) - is the number of customers being served at phase
j. The process X(t) has the finite state space (3).

⎧ ⎫
⎨ 
n 
n ⎬
X (R) = (k1 , ..., kn )|kj ≥ 0, j = 1, ..., n, kj ≤ N, kj rj ≤ R . (3)
⎩ ⎭
j=1 j=1

n
We denote μi0 = − j=1 μij as service termination rate at i-phase, and
μi = −μii - departure rates from i phase. Then, the system of equilibrium
equations (SEE) for the process X(t) can be presented in the form (4) where
I(S) is the indicator function of statement S, r = (r1 , ..., rn ) and µ = (μ1 , ..., μn )
are column vectors, ei is row vector of zeros with figure one at the i-th position.
 5G NR System Performance Analysis Using Limited Resource Loss Systems 9
 

n
λI (ku < N ) I(kr + ri ≤ R)ai + kµ p(k)
i=1

n 
n
=λ I(ki > 0)p(k − ei )ai + I(ku < N ) p(k + ei )(ki + 1)μi0
i=1 i=1

n 
n
(4)
+I(ku < N ) I(kr + rj > R)p(k + ei )(ki + 1)μij
i=1 j=1
j=i

n 
n
+ I(kj > 0)I(kr + rj ≤ R)p(k + ei − ej )(ki + 1)μij , k ∈ X (R).
i=1 j=1
j=i

3.2 Computation Approach

We propose an approach of the loss system analysis with limited resources as an

iterative method based on Gauss–Seidel scheme. At the initiation stage we set
uniform distribution as the first approximation:
1
p(k) = , k ∈ X (R).
|X (N, R)|
Then we calculate consecutive approximations of the  probabilitydistribution
pm , where m is the counting number of iteration, until pm − pm−1  ≤ , where
is the required computational accuracy, performing the following steps:

1. Set C := 0;
2. Calculate new values of p using (5);
3. Normalize p(k): p(k) = C1 p(k), k ∈ X (R).

1
p(k) =  n

λI (ku < N ) I(kr + ri ≤ R)ai + kµ
i=1
 n 
n
× λ I(ki > 0)p(k − ei )ai + I(ku < N ) p(k + ei )(ki + 1)μi0
i=1 i=1

n 
n (5)
+I(ku < N ) I(kr + rj > R)p(k + ei )(ki + 1)μij
i=1 j=1
j=i

n 
n 
+ I(kj > 0)I(kr + rj ≤ R)p(k + ei − ej )(ki + 1)μij , k ∈ X (R).
i=1 j=1
j=i
10 V. Naumov et al.

4 Numerical Results
In this Section we concentrate on session drop probabilities (6) and mean NR
BS resource utilization (7) for a fixed and variable ISD.

 
B = ab, b = (b1 , ..., bn ), bi = p (k) + p (k). (6)
k∈X (R): j∈{1,...,n}\i,k∈X (R):
kr+ri >R kr−rj +ri >R


U= kr · p(k). (7)
k∈X

The default system parameters [6,12] are summarized in Table 2.

Table 2. Input data

Notation Description Values

fc Operational frequency 28 GHz
W Bandwidth 1 GHz
sA Service unit 1.44 MHz
rB Blocker radius 0.4 m
PT Transmit power 0.2 W
γ Path loss exponent 2.1
GT BS antenna gain 5.57 dBi
GR UE antenna gain 2.58 dBi
No W Johnson-Nyquist noise, −84 dBi
RO Outer zone width 200–800 m
RI Inner zone radius 200 m
dISD Inter site distance 400–1200 m
v Service Data Rate 50 Mbps
r1 nLoS session demand in outer zone 16 RBs
r2 LoS session demand in inner zone 158 RBs
r3 LoS session demand in outer zone 31 RBs
r4 nLoS Session demand in inner zone 0 RBs
Λ−1 Session mean inter-arrival time 1000 s
−1
μ Mean service time 30 s
−1
θnLoS Mean blockage time 2.94 s

For the considered scenario the transition matrix has the form (8) and vector
RI2 RI2
a = ( (RI +R O ) 2 , 0, 1 − (R +R )2 , 0) which represents initial phase probability
I O
distribution. Vector a has four components, where a1 and a3 are equal to the ratio
of inner and outer zone areas to the coverage area respectively, while a2 = a4 = 0
 5G NR System Performance Analysis Using Limited Resource Loss Systems 11

which follows our assumption about impossibility to establish a session in nLoS

conditions.
A customer service may be transited between the first and the second phases,
as well as between the third and the fourth phases. The first phase corresponds
to sessions from inner zone in LoS conditions, while the second - to sessions
from inner zone in nLoS conditions. Similarly, the third and the fourth phases
correspond to sessions from outer zone in LoS/nLoS conditions.

⎛ ⎞
−(θLoS
I
+ μ) I
θLoS 0 0
⎜ θI −(θnLoS
I
+ μ) 0 0 ⎟
M=⎜
⎝
nLoS ⎟.
⎠ (8)
0 0 −(θLoS + μ)
O O
θLoS
0 0 O
θnLoS −(θnLoS + μ)
O

Figure 3 illustrates the session drop probabilities for two defined zones and
the cumulative drop probabilities as a function of density of users ρ. Drop prob-
ability in inner zone is significantly lower as the mean demand of I-session is less
compared to the outer zone.

0.03
Inner
Outer
0.025 Overall
Session drop probability

0.02

0.015

0.01

0.005

0
0.05 0.06 0.07 0.08 0.09 0.1 0.11 0.12 0.13 0.14 0.15
Density of users, units/m 2

Fig. 3. Session drop probabilities on arrival.

Figure 4 presents the utilization fraction for the two defined zones and the
aggregate utilization fraction for LoS and nLoS sessions as a function of density
of users ρ. It should be noted that when there is a lack for resources with the load
increase, the nLoS session in inner zone suffer most due to their huge demands
giving the resource away for sessions with lower demands.
Optimization of ISD is one of the most challenging problems for mmWawe
systems [6]. By increasing the density of BSs, i.e. shortening the ISD, it is possible
12 V. Naumov et al.

80
UTILO
UTILLoSI
70
UTILnLoSI
UTIL
60
Utilization fraction

50

40

30

20

10

0
0.05 0.06 0.07 0.08 0.09 0.1 0.11 0.12 0.13 0.14 0.15
Density of users, units/m 2

Fig. 4. The mean NR BS resource utilization.

to lower the drop probabilities and improve the quality of experience. However, it
requires massive number of expensive equipment to be installed and may not be
commercially effective. To address this problem, our modeling approach allows
for analyzing session drop probabilities and resource utilization fraction as a
function of ISD, as shown in Figs. 5 and 6.

0.01
Session drop probability

0.0001

ISD = 400m
1x10-6
ISD = 600m
ISD = 800m
ISD = 1000m
1x10-8
ISD = 1200m

1x10-10

1x10-12

0.01 0.011 0.012 0.013 0.014 0.015 0.016 0.017 0.018 0.019 0.02
Density of users, units/m 2

Fig. 5. Session drop probabilities as a function of ISD.

 5G NR System Performance Analysis Using Limited Resource Loss Systems 13

100
ISD = 400m ISD = 800m ISD = 1200m
90 ISD = 600m ISD = 1000m

80

70
Utilization fraction

60

50

40

30

20

10

0
0.01 0.011 0.012 0.013 0.014 0.015 0.016 0.017 0.018 0.019 0.02
Density of users, units/m 2

Fig. 6. The mean NR BS resource utilization as a function of ISD.

Both figures show that quality of service and NR BS load is severely under-
mined in case ISD is greater than 800m. When ISD approaches 1000m, the
BS covers the number of users that it is not capable to provide with sufficient
resources, resulting in session drop probability higher than 10− 5. Thus, knowing
the average expected density of users in a specified area, it is possible to predict
the cost-effective deployment scheme of NR BSs that will allow to fulfill user
and system needs.

5 Conclusion

In this paper, we considered the radio resource allocation process by an 5G NR

Base Station. We proposed an analytical model in terms of multiserver queuing
system with limited resources and varying resource demands. We also proposed
an approximate method for calculation of stationary probability distribution that
allows for estimation of user- and system-centric performance metrics that can
be further used in performance optimization of practical NR BS deployments,
which is highly sought as deploying dense sites increases capital and operating
expenditure (CAPEX and OPEX) for operators, thus increasing cost for end-
users.
Our future work is to introduce overlapping resource requirements [13] to
our model which will allow for modeling multicast services as they are widely
used in conventional wired and wireless networks significantly improving resource
utilization in presence of users interested in the same content.
14 V. Naumov et al.

References
1. IEEE Standard: 802.11ad-2012: Enhancements for very high throughput in the 60
GHz band. https://ieeexplore.ieee.org/document/6392842/
2. 3GPP: 5G; study on channel model for frequencies from 0.5 to 100 GHz (Release
14), 3GPP TR 38.901, May 2017
3. 3GPP: NR: physical channels and modulation (Release 15), 3GPP TR 38.211,
December 2017
4. 3GPP: NR: multi-connectivity; Overall description (Release 15), 3GPP TS 37.34
V15.2.0, June 2018
5. Drago, M., Azzino, T., Polese, M., Stefanović, Č., Zorzi, M.: Reliable video stream-
ing over mmWave with multi connectivity and network coding. In: International
Conference on Computing, Networking and Communications (ICNC), pp. 508–512
(2018)
6. Kovalchukov, R., et al.: Improved session continuity in 5G NR with joint use of
multi-connectivity and guard bandwidth. In: 2018 IEEE Global Communications
Conference (GLOBECOM), pp. 1–7 (2018)
7. Basharin, G., Gaidamaka, Y., Samouylov, K.: Mathematical theory of teletraffic
and its application to the analysis of multiservice communication of next generation
networks. Autom. Control Comput. Sci. 47, 62–69 (2013)
8. Petrov, V., et al.: Dynamic multi-connectivity performance in ultra-dense urban
mmWave deployments. IEEE J. Sel. Areas Commun. 35(9), 2038–2055 (2017)
9. Naumov, V., Samouylov, K., Yarkina, N., Sopin, E., Andreev, S., Samuylov, A.:
LTE performance analysis using queuing systems with finite resources and random
requirements. In: 7th International Congress on Ultra Modern Telecommunications
and Control Systems and Workshops (ICUMT), pp. 100–103 (2015)
10. Polese, M., Zorzi, M.: Impact of channel models on the end-to-end performance
of Mmwave cellular networks. In: IEEE 19th International Workshop on Signal
Processing Advances in Wireless Communications (SPAWC), pp. 1–5 (2018)
11. Venugopal, K., Valenti, M.C., Heath Jr., R.W.: Analysis of millimeter wave net-
worked wearables in crowded environments. In: 2015 49th Asilomar Conference on
Signals, Systems and Computers, pp. 872–876 (2015)
12. Mezzavilla, M., et al.: End-to-end simulation of 5G mmWave networks. IEEE
Comm. Surv. Tutorials 20(3), 2237–2263 (2018)
13. Samouylov, K., Gaidamaka, Y.: Analysis of loss systems with overlapping resource
requirements. Stat. Pap. 59(4), 1463–1470 (2018)
 On the Performance of LoRaWAN
in Smart City: End-Device Design
and Communication Coverage

Dmitry Poluektov1(B) , Michail Polovov1 , Petr Kharin1 , Martin Stusek2 ,

Krystof Zeman2 , Pavel Masek2 , Irina Gudkova1,2,3 , Jiri Hosek2 ,
and Konstantin Samouylov1,2,3
1
Applied Mathematics & Communications Technology Institute,
Peoples’ Friendship University of Russia (RUDN University),
Moscow, Russian Federation
poluektov dmitri@mail.ru
2
Department of Telecommunication, Brno University of Technology,
Brno, Czech Republic
3
Institute of Informatics Problems, Federal Research Center Computer Science
and Control of the Russian Academy of Sciences, Moscow, Russian Federation

Abstract. Expected communication scenarios within the emerging

landscape of Internet of Things (IoT) bring the growth of smart devices
connected in the communication network. The communication tech-
nologies of greatest interest for IoT are known as Low-Power Wide-
Area Networks (LPWANs). Today, there are LPWA technologies (Sig-
fox, Long-Range Wide Area Network (LoRaWAN), and Narrowband IoT
(NB-IoT)) capable to provide energy efficient communication as well as
extended communication coverage. This paper provides an analysis of
LPWA technologies and describes experimental evaluation of LoRaWAN
technology in real conditions. The LoRaWAN technology provides over
150 dB Maximum Coupling Loss (MCL), which together with maxi-
mum transmission power (TX) 14 dBm and spreading factor 7 results in
theoretical communication distance in units of kilometers. The obtained
results from field-deployment in the city of Brno, Czech Republic confirm
the initial expectations as it was possible to establish reliable commu-
nication between low-end LoRaWAN device and LoRaWAN gateway on
the distance up to 6 km.

The described research was supported by the National Sustainability Program

under grant LO1401. The publication has been prepared with the support of
the “RUDN University Program 5-100” (recipients Irina Gudkova, Dmitry Poluek-
tov, Petr Kharin). The reported study was funded by RFBR, project numbers
18-00-01555(18-00-01685), 19-07-00933 (recipient Konstantin Samouylov). This arti-
cle is based as well upon support of international mobility project MeMoV, No.
CZ.02.2.69/0.0/0.0/16 027/00083710 funded by European Union, Ministry of Educa-
tion, Youth and Sports, Czech Republic and Brno University of Technology. For the
research, infrastructure of the SIX Center (Czech Republic) as well as the 5G Lab
RUDN (Russia) was used.
c Springer Nature Switzerland AG 2019
V. M. Vishnevskiy et al. (Eds.): DCCN 2019, LNCS 11965, pp. 15–29, 2019.
https://doi.org/10.1007/978-3-030-36614-8_2
Another random document with
no related content on Scribd:
den, and cut her loose; so we dragged the ship away around South
of the den, taking us all the forenoon, and we could see the lions on
the rocks sunning themselves and probably talking over in lion
language what they would do to us if we got fresh, and every little
while they would cough like a case of pneumonia, and it made my
hair raise, but Pa was so cool he had to turn his collar up.
After a few hours we got the ship in the right place, about a quarter
of a mile south of the den, and Pa got the cowboy ready with his
lasso, and the German ready to yell murder in his language, and the
negro ready to throw overboard for the lions to eat, and Pa said,
“Turn her loose,” and we let go of the rope, and the ship sailed right
straight for the den, and we all climbed upon a big rock to watch the
proceedings. It was the most exciting moment of my life, except the
time the fat woman in the circus sat down in Pa’s lap, and crushed
him beyond recognition and they had to scrape him up with case
knives.
There was Pa at the wheel, his eyes staring ahead at the lions, all of
the lion family having come out of the den to see the air ship, and the
dog lion, the head of the household waving his tail and making the
air fairly tremble with his roaring.
Pretty soon the airship was right over the den, the lasso was thrown
over the dog lion’s neck, and drawn tight, and he coughed and
strangled like a negro being lynched, and then he turned tail and ran
down into the den in the rocks, with all the other lions after him,
dragging the ship back into the entrance of the den, and closing the
hole completely, and we all rushed up and tied the rope to trees, so
the gas bag was right over the hole, tight as a drum, and Pa got
down off the frame, and as Mr. Hagenbach came up in a perspiration
Pa said, “There’s your lions, about a dozen of them captured down in
that hole; help yourselves,” and Pa sat down on the ground like a
man who had conquered the world, and was waiting for the
applause. Mr. Hagenbach said that was all right, so far as it had
gone, but what he wanted was lions in cages, ready to ship to
Germany, and not down in a hole in the ground that might be as
deep as a copper mine, with no elevator to bring the lions to the
surface. “Well,” said Pa, as he lit a cigar, “there’s a perfectly good
dog Numidian lion, with a black mane, on the end of that lasso, and
all you got to do is to pull him up, just as you would a muscalonge on
a line, and when he comes to the surface after I have finished my
cigar, I will hog tie him and have him ready for shipment quicker’n a
wink,” and Pa yawned, as though capturing wild lions was as easy
for him as catching mice in a trap.

“There’s Your Lions, About a Dozen Captured Down in That

Hole; Help Yourselves,” Said Pa.
So the crowd all got hold of the lasso and began to pull up, and of all
the snarling and howling you ever heard, that beat the band. The old
lion seemed to catch on to everything coming up, and all the other
lions roared until the rocks on which we stood fairly trembled like
there was an earthquake, but the old dog kept coming and I felt as
though something terrible was going to happen, and I began to get
farther away. Pa knocked the ashes off his cigar and asked the
cowboy how much more rope there was left, and was told about ten
feet, so he told them to let up a minute until the driver drove the cage
up to a point on the rock not far from where the lion would come out,
and when the cage was ready and the door open, so the lion could
see a goat tied in the cage eating hay, Pa said to the men to give a
few more jerks, and, by Gosh, pretty soon the lion’s head and neck
came out of the hole, and he was the maddest looking animal I ever
saw, and the men looked scared.
The lion was bracing with his front feet, and using all kinds of
language, but Pa was the coolest man in the bunch. “Now, let him
rest a minute,” says Pa, “but hold the line taut,” and Pa took out a
bag of tobacco and a piece of paper and rolled a cigarette, and lit it,
and we all looked at Pa in admiration for his nerve.
After puffing his cigarette a little, and looking to see if the cage was
entirely right, he ordered the men who were not pulling on the rope
to line up in two lines from the hole to the cage, like the honorary pall
bearers at a funeral, and told them not to move until the lion was in
the cage, and when they were all in place, Pa said, “Now jerk his
head plumb off,” and the crowd pulled and the lion came out of the
hole mad and frothing at the mouth. Pa stepped one side and gave
the lion a swift kick in the ham, and the king of beasts put his tail
between his legs and started for the hearse cage, and Pa said, “Get
in there, you measly cur dog,” and Pa followed him, kicking him
every jump, until the big lion rushed into the cage and laid down, so
completely conquered that he bellowed pitifully when the goat butted
him off of the hay, and Pa closed the door and locked it and turned to
Mr. Hagenbach and asked, “How many of these vermin do you
want?” and he said, “Now that we were about it we had better get the
whole bunch.” Pa said “all right, he was there after lions, and he
wanted to get the limit,” so they signalled camp for some more
cages, and Pa said we had better have lunch right there on the rock
beside the airship in the shade, while he prepared to catch the rest
of the lions.

“Get in There, You Measly Cur Dog,” Said Pa, Kicking the
Big Lion at Every Jump.

Pa was attaching a long rubber hose to the gas bag, and as he got it
fastened and reeled about fifty feet of the hose down in the hole, Mr.
Hagenbach said, “Say, old man, I don’t want to kick on any of your
new inventions, but what are you going to do now?” and Pa said, as
he turned a faucet in the gas bag and let the gas into the hose,
“Didn’t you ever drown gophers out of a hole by pouring water in,
until the gophers came to the top strangling, and you put them in a
shot bag and let them chew your fingers? Well, I am going to drown
out big gophers with gas, and in about fifteen minutes after we have
had lunch, you will see the dammest procession of sneezing lions
come up out of that hole that ever were in captivity, and I want all of
you brave ducks to hold the bags over the hole, and when you get a
lion in a bag tie the bag and roll the beast over the rock, see?”
Well, they got the gunny sacks ready, and after we had our lunch
and the gas was filling the hole good and plenty, there was a lot of
sneezing and roaring down the hole, and Pa said the medicine was
working all right, and pretty soon Pa turned off the gas and
unscrewed the hose, and loosened the ropes on the air ship so she
sailed off across the veldt for a block or so, and then the trouble
began.
First a big she lion came up with a mess of cubs, and they held the
bag all right, but she went right through it like a bullet through
cheese, and then there was an explosion away down in the bowels
of the earth, from the toe nails of some unmanicured lion striking fire
on a flint stone, and fire began to pour out of the hole, and about
nine singed lions of all sizes came up out of the hole scared to
death, and the smell of burned hair was awful.
The lions began to cuff the men and they stampeded down the
rocks, leaving Pa and two or three of us alone. Pa and I seized a
couple of the baby lions and started to run for camp, and the lions
took after us and chased us awhile, until Pa got out of wind, when
we climbed trees with the cubs, and the lions rolled in the grass to
put out the fire, and then they took to the jungle, and Pa said when
Roosevelt got to Africa and shot a few singed lions, he would think it
was a new kind of beast.
We got back to camp with the two cubs, and called the roll to see
who was missing, and we found the natives had packed up and
moved away, claiming that the old man was a devil who had
produced a burning mountain, and the whole country would be
devastated.
We sent all our animals to the coast to be shipped to Berlin and
moved our camp up to the jungle, about fifty miles, where there is a
new tribe of natives, and where it is said the country is inhabited with
gorillas.
Pa says he is going to move a cage into the gorilla country, and call
the gorillas around him, learn their language, get their confidence,
and eventually reform them and bring them to realize that they are
endowed with certain inalienable rights, and teach them white man’s
customs, and Pa will do it or die trying, but I don’t like the idea, as it
seems dangerous to Pa. Say, those gorillas are bigger than John L.
Sullivan, and they hug like bears. Gee, but I want to see gorillas
hanging by their tails on trees, and Pa says I may go with him.
 CHAPTER XVI.
The Boy’s Pa Shows Bravery in the Jungles in Africa—Four Gorillas
Chase Pa—The Boy and His Pa Don’t Sleep Much at Night—The Boy
Discovers a Marsh Full of Wild Buffaloes.

I do not know whether Pa is an expert in hypnotism or what it is, but

he certainly delivers the goods when he goes after a wild animal in
the jungles of Africa, and he shows bravery at times that astonishes
everybody, but he admits that he is a coward at heart, and would run
if anybody pulled a gun on him, and I guess he would, but you turn
him loose in a wild animal congress and he will be speaker and
make the whole bunch get on their knees.
I was scared when Pa wanted to have a cage with iron bars hauled
into the jungle where the gorillas live, and insisted that he be left
there alone for two days, with rations to last a week, as he said he
expected to have some gorilla boarders to feed, but Mr. Hagenbach
let Pa have his way, and the cage was hauled about eight miles into
the black wilderness, with great trees and vines and suckes and
gorillas all around him, but Pa insisted on having a phonograph full
of jig tunes, and when we got the cage located and Pa in it and were
ready to leave, I cried, and the whole crowd felt as though we would
never see Pa alive again, and it was a sad parting.
When we left Pa he was cooking some bacon on an oil stove in the
cage and frying some eggs for his dinner, and as we took the trail
back to camp, in silence, we could smell the bacon frying, and when
we got a mile or so away we heard music and stopped to listen and
could plainly hear the phonograph playing “There will be a hot time,”
and Mr. Hagenbach said it reminded him of a dirge.
It was a long two days before we could go back and find Pa’s
remains, but the second day we hiked out through the jungle and
into the woods. Pa had told us that when we came after him to come
quiet and not disturb the menagerie, so when we got near the place
where we left Pa we slowed down and crept up silently and peeked
through the bushes and a sight met our eyes that scared me.
There were four big gorillas and several little ones around the cage,
and some were gnawing ham bones and others were eating dog
biscuits, but it was so silent in the cage that I thought Pa had been
killed and that the gorillas were eating him, so I yelled, “Pa, are you
all right?” and he answered back, “You bet your sweet life I am all
right,” and then we prepared to go the cage, when Pa said for us to
climb trees, and just then the gorillas started for us with their teeth
gleaming, and we all shinned up the trees around the cage, and we
had front seats at the biggest show on earth. Pa told us that the
gorillas that treed us were afraid we were going to harm him, and
they proposed to protect him.
He said he had been feeding the animals for two days and had got
their confidence so he could make them understand what he wanted
them to do.
“Now watch ’em dance when I turn on the music,” and then Pa gave
them the “Merry Widow” waltz, and by gosh if a big gorilla didn’t put
his arm around his wife, or some other gorilla’s wife, and dance
barefooted right there in front of the cage, and all the rest joined in,
and the baby gorillas rolled over on the ground and laughed like
hyenas. Pa stopped the music and called one big gorilla Rastus and
told him to sit down in the cactus, and the others did the same, and
Pa repeated an old democratic speech of his, and they clapped their
hands just like a caucus. “Well, what do you know about that
already,” said Mr. Hagenbach, and then he asked Pa how he was
going to capture them.
 Pa Stopped the Music and Repeated an Old Democratic
Speech of His and They Acted Just Like a Caucus.

Pa said he had them in the cage several times and let them out, and
when we got ready to go to camp all he had to do was to let the
phonograph play “Supper is now ready in the dining car,” and they
would come in and he would slip out and lock the door and we could
haul the cage to camp.
 All He Had to Do Was to Play “Supper is Now Ready in the
Dining Car” on the Phonograph.

Well, you ought to have seen my old gentleman call the whole bunch
of gorillas into the cage and feed them and see them act like a lot of
boys in camp, reaching for potatoes and bacon and wiping their lips
on their hairy arms, but none of them asked for napkins or finger
bowls. When the food was all gone they began to kick like boarders
at a second-class boarding house, and then Pa slipped out of the
door and locked it, and we came down out of the trees and
surrounded the cage, and Pa acted as barker and told us the names
he had given to the gorillas.
Pa brought the phonograph out of the cage and set it going and the
gorillas began to dance. Mr. Hagenbach was so pleased that he
fairly hugged Pa, and we got ready to haul the cage to camp.
Pa always makes some mistake before he has a proposition well in
hand, and he did this time, of course. As we were about to start the
gorilla Rastus, who had become Pa’s chum, looked at Pa so pitiful
that Pa said he guessed he would let Rastus out and he and Rastus
would walk along ahead and get the brush out of the road, so he
opened the door of the cage and beckoned to Rastus, and the big
gorilla came out with his oldest boy, and Pa and the two of them took
hold of hands and started on ahead, and we started to haul the
wagon by drag ropes, when the worst possible thing happened.
Rastus reached in Pa’s pistol pocket, where Pa had just put a large
plug of tobacco after he had bit off a piece, and Rastus thought
because Pa ate the tobacco he could, so he bit off about half of the
plug and ate it and gave his half-grown boy the rest of it, and that
was eaten by the boy. Pa tried to take it away from them, but it was
too late, and they were both mad at Pa for trying to beat them out of
their dessert.
It was not long before Rastus turned pale around the mouth, but his
face was so covered with hair that you couldn’t tell exactly how sick
he was; though, when he put both hands on his stomach, gave a yell
and turned some somersaults, we knew he was a pretty sick gorilla,
and his boy rolled over and clawed his stomach and had a fit.
Rastus had the most pained and revengeful look on his face I ever
saw, and he looked at Pa as though he was to blame.
Pa had one of the men get the medicine chest, and Pa fixed two
seidlitz in a tin cup, but before he could put in the water Rastus had
swallowed the powder from the white and blue paper and reached
for a wash basin of water, and before Pa could prevent Rastus from
drinking it on top of those powders, he had swallowed every drop of
the water, and the commotion inside of him must have been awful,
for he frothed at the mouth and the bubbles came out of his nose,
and he rolled over and yelled like a man with gout, and he seemed to
swell up, and Pa looked on as though he had a case on his hands
that he couldn’t diagnose, while Rastus’ boy just laid on the ground
and rolled his eyes as though he were saying his “Now I lay me,” and
Mr. Hagenbach said to Pa he guessed he had broke up the show,
and Pa said, “Never you mind, I will pull them both through all right.”
Finally the siedlitz powder fiz had all got out of Rastus’ system and
he seemed to be thinking deeply for a moment, and then he got off
his haunches and looked steadily into Pa’s eyes for a minute, and
then he took Pa by one hand and his boy with the other and started
right off through the jungle, Pa pulling back and yelling to us to
rescue him from the gorilla kidnapers, but Rastus walked fast and
before he had got ought of sight he had picked his sick boy up and
carried him under his arm and both were groaning, and he held on to
Pa’s hand and went so fast that Pa’s feet only hit the high places.
The gorillas in the cage looked at them disappear and tried to get out
of the cage to go along, but they couldn’t get out.
Finally Mr. Hagenbach said me, “Hennery, I guess your Pa has got
what is coming to him this time. Rastus will probably drag your Pa up
a tree and eat him when his appetite comes back, but we can’t help
him, so we better haul the cage and the gorillas that have not had
any tobacco to camp, and in a day or two we will all come out here
and find your father’s bones and bury them.”
And then we all went to camp, and the poor gorillas just remained
listlessly in the cage, mourning as though they knew Rastus and his
boy were dead. We fed them everything we could spare, but they
would not eat, and by watching them we found there was a case of
jealousy in the cage, as two male gorillas seemed to be stuck on a
young female, and they were scrapping all the time.
Gee, but we needed Pa worse than ever to settle the gorilla dispute,
but we all felt that Pa was not of this earth any more, and the camp
took on an air of mournfulness, and they all wanted to adopt me,
’cause I was alone in the world. There was not much sleep in camp
that night, and the next day we were going out with guns to find Pa’s
remains and shoot Rastus, but a little after daylight we heard the
night watchman say to the cook, who was building a fire, “Look
who’s here, and what do you know about that,” and he called the
whole camp up, and we looked out across the veldt and there came
Pa astradle of a Zebra, with Rastus’ boy up behind him and Rastus
thoroughly subdued, leading the Zebra with a hay rope Pa had
twisted out of grass.
The whole camp came to attention and Pa scratched a match on
Rastus’ hair and lighted a cigarette, and when he got near enough
he said: “Slept in the crotch of a tree all night. Gave Rastus and his
boy a drink of whiskey out of my flask and cured them of the tobacco
sickness, had some mangoes for breakfast, sent Rastus to catch a
Zebra, and here we are ready for coffee and pancakes.”
Pa got off his zebra, opened the door of the cage and pointed to it,
and Rastus and his boy got in, and Pa kicked Rastus right where the
hair was worn off sitting down, and Rastus looked at Pa as though
that was all right and he deserved it. Then Pa closed the door,
washed his hands and sat down to breakfast, and when Mr.
Hagenbach said, “Old man, you have got Barnum and Forepaugh
skinned a mile,” Pa said, “O that is nothing; I have located a marsh
full of wild Buffaloes, and we will go out there and get a drove of
them in a few days.
“They are the ugliest and fightingest animals in the world, but I will
halter break some of them and ride them without any saddle.” Mr.
Hagenbach said he believed it, and Pa said, “Hennery, one spell I
thought you would be an orphan, but whiskey saved you. When they
got a big drink of whiskey they began to laugh, and then fell on my
neck and cried, just like a white man when he is too drunk to fight.
Well, I am going to take a nap,” and Pa laid down on a bale of hay
and slept all day, and the crowd talked about what a hero he was.
 CHAPTER XVII.
The Boy’s Experience With an African Buffalo—The Boy’s Pa Shoots
Roman Candles to Scare the Buffaloes—The Boy’s Pa Tames the
Wild Animals.

When Pa told us that he had located a place where we could get all
the wild African buffalo that we wanted, I thought of the pictures I
had seen of the killing of buffaloes in America, where all the buffalo
hunter had to do was to ride a horse after a herd of the animals, that
couldn’t run faster than a yoke of oxen, pick out a big bull and ride
alongside of him and fire bullets into his vital parts at about ten feet
range, until his liver was filled full of holes and he had the nose
bleed, and when he fell down from loss of blood, dismount and skin
him for a lap robe. The American buffalo would always run away and
the hunter could kill him if he had cartridges enough, and never be in
any more danger than a farmer milking a cow.
I thought we would have about the same kind of experience with
African buffalo, only we intended to lasso them and bring them to
camp alive for the show business, but instead of the African buffalo
running away from you, he runs at you on sight and tries to gouge
out your inside works with his horns, and paws you with his hoofs,
and when he gets you down he kneels down on you and runs horns
all through your system and rolls over on your body like a setter dog
rolling on an old dead fish.
The African buffalo certainly has a grouch, as though he had
indigestion from eating cactus thorns, and when he sees a man his
eyes blaze with fire and he gets as crazy as an anarchist and seems
to combine in his make-up the habits of the hyena, the tiger, the
man-eating shark and the Texas rattlesnake.
I wouldn’t want such an animal for a pet, but Pa said the way to get
buffaloes was to go after them and never let up until you had them
under your control. So we started out under Pa’s lead to capture
African buffalo, and while the returns are not all in of the dead and
wounded, we know that our expedition is pretty near used up.
These African buffaloes live in a marsh, where the grass and cane
grows high above them, and the only way you can tell where they
are is to watch the birds flying around and alighting on the backs of
the animals to eat wood ticks and gnats. The marsh is so thick with
weeds that a man cannot go into it, so we planned to start the airship
on the windward side of the marsh, after lining up the whole force of
helpers, negroes and white men, and building a corral of timber on
the lee side of the marsh. Pa and the cowboy and I went in the
airship, with those honk-honk horns they have on automobiles, and
those megaphones that are used at football games, and Pa had a
bunch of Roman candles to scare the buffaloes.
When the fence was done, which fifty men had worked on for a
week, it run in the shape of a triangle or a fish net, with a big corral at
the middle. Mr. Hagenbach sent up a rocket to notify Pa that he was
ready to have him scare the buffaloes out of the marsh, down the
fence into the corral.
Pa had the gas bag all full, a mile across the marsh, tied to a tree
with a slip noose, so when we all got set he could pull a string and
untie the slip noose.
Well, everything worked bully, and when Pa tied her loose we went
up into the air about fifty feet, and Pa steered the thing up and down
the marsh like a pointer dog ranging a field for chickens.
It was the greatest sight I ever witnessed, seeing more than two
hundred buffalo heads raise up out of the tall grass and watch the
airship, looking as savage as lions eating raw meat.
First they never moved at all, but we began to blow the honk horns,
and then we yelled through the megaphones to “get out of there, you
sawed off short horns,” and then they began to move away from the
airship across the marsh, and we followed until they began to get
into a herd, nearly on the other side of the marsh, but they only
walked fast, splashing through the mud.
When we got almost across the marsh Pa said now was the time to
fire the Roman candles, so we each lit our candle, and the fire and
smoke and the fire balls fairly scorched the hair of the buffaloes in
the rear of the herd, and in a jiffy the whole herd stampeded out of
the marsh right toward the fence, bellowing in African language,
scared half to death, the first instance on record that an African
buffalo was afraid of anything on earth.
We followed them until they got to the fence, but only about one
hundred got into the corral, the others going around the fence and
chasing the keepers into the jungle and hooking the negroes in the
pants, and some of the negroes are running yet, and will, no doubt,
come out at Cairo, Egypt.
 Some of Those Negroes Are Running Yet, and Will No
Doubt Come Out at Cairo, Egypt.

Mr. Hagenbach and the white men got up in trees and watched Pa
and the airship, and when we got where the fence narrowed up at
the corral Pa let the airship come down to the ground and anchored
it to a stump and yelled for the boss of the expedition and the men to
come down out of the trees and help capture some of the best
specimens; so they came down and tore out the wings of the fence
and placed them across so we had the buffaloes in a pen, and then
Mr. Hagenbach, who had been getting a little jealous of Pa, came up
to him and shook his hand and told him he was a wonder in the
capturing of wild animals, and Pa said don’t mention it, and Pa took
the makings and made himself a cigarette and smoked up, and Mr.
Hagenbach asked Pa how we were going to get the buffaloes out of
the corral, ’cause they were fighting each other in the far end of the
pen, and Pa said you just wait, and he sent for the cages, enough to
hold about ten of the buffaloes, and we let the gas out of the airship
and went into camp right there, and Pa bossed things for about two
days, until the buffaloes got good and hungry, and then we backed
the cages up to an opening in the fence and put hay in the far end of
the cages, and the herd began to take notice.
We wanted the big bulls and some cows, and nature helped us on
the bulls, ’cause they fought the weaker ones away from the cages,
and walked right up the incline into the cages, and Pa went in and
locked the doors, and when we got the cages full of bulls and started
to haul the cages to camp by the aid of some of the negroes who
had returned alive, by jingo, the cows followed the cages with the
bulls in, and you couldn’t drive them away.
We loaded the gas bag on to a sort of stone boat, and Pa rigged up
a couple of ox yokes and in some way hypnotized a few cow
buffaloes so he could drive them, and they hauled the stone boat
with the airship to camp, and we got there almost as soon as the
cages did, and Pa was smoking as contented as though he was
walking on Broadway, and with an ox gad he would larrup the oxen
and say, “Haw Buck,” like a farmer driving oxen to plow a field.
Pa got his wild oxen so tame before we got to camp that they would
eat hay out of his hand, and when we rounded up in our permanent
camp and looked over our stock and killed some of the buffaloes that
had followed the cages, for meat for the negroes, and lit some sky
rockets and fired them at the balance of the herd to drive them away
from camp, the negroes, who had always had a horror of meeting
wild buffaloes, thought Pa was a superior being to be able to tame a
whole herd of the most savage animals, and they got down on their
knees and placed their faces in the dust in front of Pa and
worshipped him, and they wouldn’t get up off the ground until Pa had
gone around and put his feet on the necks of all the negroes in token
that he acknowledged himself to be their king and protector, and the
wives of the negroes all threw their arms around Pa and hugged him
until he got tired, and he said he had rather fight buffaloes than be
hugged by half naked negro women that hadn’t had a bath since
Stanley discovered them, but Pa appreciated the honor, and Mr.
Hagenbach said Pa was the greatest man in the world.

Pa Had to Put His Foot on Their Necks and Acknowledge

Himself Their King and Protector.

The next day we shipped the buffaloes to the coast and had them
sent to Berlin, and when we got the mail from headquarters there
was an order for a lot more tigers, so I suppose we will be tigering as
soon as the open season is on.
The idea is that we must get all the animals we can this year, for it is
rumored that Roosevelt is coming to Africa next year to shoot big
game, and all of us feel that wild animals will be scarce after he has
devastated Africa.
We got short of Salt Pork, and some time ago Pa salted down some
sides of rhinoceros, and yesterday was the day to open the barrel.
Pa showed the cooks how to fry rhinoceros pork, and I tell you it
made you hungry to smell rhinoceros frying, and with boiled potatoes
and ostrich eggs and milk gravy, made from elephant milk, we lived
high, but the next day an epidemic broke out, and they laid it to Pa’s
rhinoceros pork dinner, but Pa says any man who eats eight or nine
fried ostrich eggs is liable to indigestion.
Gee, but this is a great country to enjoy an outing in.