Proceedings of Sixth International Congress On Information and Communication Technology ICICT 2021 London Volume 2 1st Edition Xin-She Yang
Proceedings of Sixth International Congress On Information and Communication Technology ICICT 2021 London Volume 2 1st Edition Xin-She Yang
Proceedings of Sixth International Congress On Information and Communication Technology ICICT 2021 London Volume 2 1st Edition Xin-She Yang
com
https://ebookmeta.com/product/proceedings-of-sixth-
international-congress-on-information-and-communication-
technology-icict-2021-london-volume-2-1st-edition-xin-she-
yang/
OR CLICK BUTTON
DOWLOAD NOW
https://ebookmeta.com/product/the-proceedings-of-the-2021-asia-
pacific-international-symposium-on-aerospace-technology-
apisat-2021-volume-1-1st-edition-sangchul-lee/
https://ebookmeta.com/product/international-conference-on-
cognitive-based-information-processing-and-applications-
cipa-2021-volume-2-1st-edition-bernard-j-jansen/
https://ebookmeta.com/product/proceedings-of-the-third-
international-conference-on-information-management-and-machine-
intelligence-icimmi-2021-dinesh-goyal/
https://ebookmeta.com/product/information-communication-and-
computing-technology-6th-international-conference-
icicct-2021-new-delhi-india-may-8-2021-revised-selected-papers-
Proceedings of International Conference on Data Science
and Applications : ICDSA 2021, Volume 1 1st Edition
Mukesh Saraswat
https://ebookmeta.com/product/proceedings-of-international-
conference-on-data-science-and-applications-
icdsa-2021-volume-1-1st-edition-mukesh-saraswat/
https://ebookmeta.com/product/international-conference-on-
innovative-computing-and-communications-proceedings-of-
icicc-2021-volume-3-1st-edition-ashish-khanna/
https://ebookmeta.com/product/information-and-communication-
technologies-in-tourism-2021-proceedings-of-the-
enter-2021-etourism-conference-january-19-22-2021-1st-edition-
wolfgang-worndl/
https://ebookmeta.com/product/applied-optimization-and-swarm-
intelligence-springer-tracts-in-nature-inspired-computing-eneko-
osaba-editor-xin-she-yang-editor/
Lecture Notes in Networks and Systems 236
Xin-She Yang
Simon Sherratt
Nilanjan Dey
Amit Joshi Editors
Proceedings of Sixth
International
Congress
on Information and
Communication
Technology
ICICT 2021, London, Volume 2
Lecture Notes in Networks and Systems
Volume 236
Series Editor
Janusz Kacprzyk, Systems Research Institute, Polish Academy of Sciences,
Warsaw, Poland
Advisory Editors
Fernando Gomide, Department of Computer Engineering and Automation—DCA,
School of Electrical and Computer Engineering—FEEC, University of Campinas—
UNICAMP, São Paulo, Brazil
Okyay Kaynak, Department of Electrical and Electronic Engineering,
Bogazici University, Istanbul, Turkey
Derong Liu, Department of Electrical and Computer Engineering, University of
Illinois at Chicago, Chicago, USA
Institute of Automation, Chinese Academy of Sciences, Beijing, China
Witold Pedrycz, Department of Electrical and Computer Engineering, University of
Alberta, Alberta, Canada
Systems Research Institute, Polish Academy of Sciences, Warsaw, Poland
Marios M. Polycarpou, Department of Electrical and Computer Engineering,
KIOS Research Center for Intelligent Systems and Networks, University of Cyprus,
Nicosia, Cyprus
Imre J. Rudas, Óbuda University, Budapest, Hungary
Jun Wang, Department of Computer Science, City University of Hong Kong,
Kowloon, Hong Kong
The series “Lecture Notes in Networks and Systems” publishes the latest
developments in Networks and Systems—quickly, informally and with high quality.
Original research reported in proceedings and post-proceedings represents the core
of LNNS.
Volumes published in LNNS embrace all aspects and subfields of, as well as new
challenges in, Networks and Systems.
The series contains proceedings and edited volumes in systems and networks,
spanning the areas of Cyber-Physical Systems, Autonomous Systems, Sensor
Networks, Control Systems, Energy Systems, Automotive Systems, Biological
Systems, Vehicular Networking and Connected Vehicles, Aerospace Systems,
Automation, Manufacturing, Smart Grids, Nonlinear Systems, Power Systems,
Robotics, Social Systems, Economic Systems and other. Of particular value to both
the contributors and the readership are the short publication timeframe and the
world-wide distribution and exposure which enable both a wide and rapid
dissemination of research output.
The series covers the theory, applications, and perspectives on the state of the art
and future developments relevant to systems and networks, decision making, control,
complex processes and related areas, as embedded in the fields of interdisciplinary
and applied sciences, engineering, computer science, physics, economics, social, and
life sciences, as well as the paradigms and methodologies behind them.
Indexed by SCOPUS, INSPEC, WTI Frankfurt eG, zbMATH, SCImago.
All books published in the series are submitted for consideration in Web of Science.
Proceedings of Sixth
International Congress
on Information
and Communication
Technology
ICICT 2021, London, Volume 2
Editors
Xin-She Yang Simon Sherratt
Middlesex University University of Reading
London, UK Reading, UK
© The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature
Singapore Pte Ltd. 2022
This work is subject to copyright. All rights are solely and exclusively licensed 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 Singapore Pte Ltd.
The registered company address is: 152 Beach Road, #21-01/04 Gateway East, Singapore 189721,
Singapore
Preface
v
vi Preface
valuable efforts and support. We are also thankful to the sponsors, press, print, and
electronic media for their excellent coverage of this conference.
vii
viii Contents
Xin-She Yang obtained his D.Phil. in Applied Mathematics from the University
of Oxford and subsequently worked at the Cambridge University and the National
Physical Laboratory (UK) as Senior Research Scientist. He is currently Reader in
Modelling and Optimization at Middlesex University London and Adjunct Professor
at Reykjavik University (Iceland). He is also Elected Bye-Fellow at the Cambridge
University and IEEE CIS Chair for the Task Force on Business Intelligence and
Knowledge Management. He was included in the “2016 Thomson Reuters Highly
Cited Researchers” list.
Amit Joshi is Director of the Global Knowledge Research Foundation and the Inter-
national Chair of InterYIT at the International Federation of Information Processing
(IFIP, Austria). He has edited more than 40 books for Springer, ACM and other
reputed publishers. He has also organized more than 50 national and international
xv
xvi Editors and Contributors
conferences and workshops in association with the ACM, Springer and IEEE in, e.g.
India, Thailand and the UK.
Contributors
Ashhadul Islam ICT Division, College of Science and Engineering, Hamad Bin
Khalifa University, Doha, Qatar
Fakir M. Amirul Islam Faculty of Health, Arts and Design, Swinburne University
of Technology, Hawthorn, VIC, Australia
Ngomane Issah Department of Computer Science, University of Limpopo,
Menkweng, South Africa
Urszula Jagodzińska-Szymańska PIT-RADWAR S.A., Warsaw, Poland
Łukasz Jeleń Department of Computer Engineering, Wrocław University of
Science and Technology, Wrocław, Poland
Fernando Jiménez Department of Information and Communication Engineering,
University of Murcia, Murcia, Spain
Juan Jose Fallas-Monge Escuela de Matemática, Instituto Tecnológico de Costa
Rica, Cartago, Costa Rica
Akira Kamiya Nagoya Institute of Technology, Aichi, Japan
D. G. T. L. Karunathilaka Faculty of Engineering, University of Jaffna,
Kilinochchi, Sri Lanka
Swetlana Kedich Emperor Alexander I St. Petersburg State Transport University,
St. Petersburg, Russia
Jagdish Kene Shri Ramdeobaba College of Engineering and Management, Nagpur,
India
Hitendra Shankarrao Khairnar Research Scholar PICT, Cummins College of
Engineering, Pune, India
Zakir Khan Time Research & Innovation (Tri), Southampton, UK;
Khilgaon, Dhaka, Bangladesh
Alexander A. Kharlamov Institute of Higher Nervous Activity and Neurophysi-
ology, RAS, Moscow, RF, Russia;
Moscow State Linguistic University, Moscow, RF, Russia;
Higher School of Economics, Moscow, RF, Russia
Nwe Ni Khin Yangon Technological University, Computer Engineering and Infor-
mation Technology, Yangon, Republic of the Union of Myanmar
Takumi Kobayashi Shimane University, Matsue, Shimane, Japan
Panagiotis Kofinas Department of Biomedical Engineering, University of West
Attica, Athens, Greece
Boris Kontsevoi Intetics Inc., Naples, FL, USA
Mridula Korde Shri Ramdeobaba College of Engineering and Management,
Nagpur, India
xxii Editors and Contributors
Atiq Ur Rehman ICT Division, College of Science and Engineering, Hamad Bin
Khalifa University, Doha, Qatar
J. A. van Biljon School of Computing, UNISA, Pretoria, South Africa
Alta Van der Merwe University of Pretoria, Pretoria, South Africa
John Andrew van der Poll Graduate School of Business Leadership (SBL),
University of South Africa, Midrand, South Africa
C. J. van Staden School of Computing, UNISA, Pretoria, South Africa
Paul F. M. J. Verschure Synthetic Perceptive Emotive Cognitive Systems
(SPECS) Lab, Institute for Bioengineering of Catalonia (IBEC), Barcelona
Institute of Science and Technology (BIST), Barcelona, Spain;
Institució Catalana de Recerca i Estudis Avançats, Barcelona, Spain
Naumov Victor Institute of State and Law, Russian Academy of Sciences, Saint
Petersburg, Russia
Carlota Villasante Marcos Ericsson España SA, Madrid, Spain
Thanh-Hai Vo Industrial University of Ho Chi Minh city, Ho Chi Minh city,
Vietnam
Matthias Volk Otto-von-Guericke University Magdeburg, Magdeburg, Germany
Dražen Vuk Mobilisis d.o.o, Varaždin, Jalkovec, Croatia
Franz-Josef Wache University of Applied Science Aachen, Aachen, Germany
Jing Wan Guangdong University of Foreign Studies, Guangzhou, China
Hiroshi Watanabe Graduate School of Fundamental Science and Engineering,
Waseda University, Tokyo, Japan
Hanbin Wu Jiangxi University of Traditional Chinese Medicine, Nanchang,
Jiangxi, China
Jie Xue University of California, Santa Barbara, CA, USA
Runze Yan University of Virginia, Charlottesville, VA, USA
William Emmanuel S. Yu Ateneo de Manila University, Quezon City, Philippines
Olaf Zukunft Hamburg University of Applied Sciences, Department of Infor-
matics, Hamburg, Germany
Highly Efficient Stochastic Approaches
for Computation of Multiple Integrals
for European Options
Abstract In this work we investigate advanced stochastic methods for solving a spe-
cific multidimensional problems related to computation of European style options
in computational finance. Recently stochastic methods have become very impor-
tant tool for high performance computing of very high dimensional problems in
computational finance. The advantages and disadvantages of several highly efficient
stochastic methods connected to European options evaluation will be analyzed. For
the first time multidimensional integrals up to 100 dimensions related to European
options will be computed with highly efficient lattice rules.
1 Introduction
Recently Monte Carlo (MC) and quasi-Monte Carlo (QMC) approaches are estab-
lished as a very attractive and necessary computational tools in finance [11]. The field
of computational finance is more complicated with increasing number of applications
[2]. The option pricing is a key problem in financial markets [5, 6, 12] and especially
V. Todorov · S. Poryazov
Institute of Mathematics and Informatics, Bulgarian Academy of Sciences,
Department of Information Modeling, Acad. Georgi Bonchev Str., Block 8, 1113 Sofia, Bulgaria
e-mail: [email protected]
V. Todorov (B) · I. Dimov
Institute of Information and Communication Technologies, Bulgarian Academy of Sciences,
Department of Parallel Algorithms, Acad. G. Bonchev Str., Block 25 A, 1113 Sofia, Bulgaria
e-mail: [email protected]; [email protected]
I. Dimov
e-mail: [email protected]
S. Apostolov
Faculty of Mathematics and Informatics, Sofia University, 5 James Bourchier Blvd.,
1164 Sofia, Bulgaria
© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2022 1
X.-S. Yang et al. (eds.), Proceedings of Sixth International Congress on Information
and Communication Technology, Lecture Notes in Networks and Systems 236,
https://doi.org/10.1007/978-981-16-2380-6_1
2 V. Todorov et al.
difficult when the dimension of the problem goes higher [1]. MC and QMC methods
are appropriate for solving multidimensional problems [7] and are used not only for
option pricing [9], but also in other problems in computational finance [4, 13]. The
basic definitions that we are using are taken from [11].
Let’s deal with a European call option [11] whose payoff depends on k > 1 assets
with prices Si , i = 1, ..., k.. Following [11] we assume that at expiry time T , and
risk-free interest rate r , the payoff is given by h(S1 , . . . , Sk ), S is the value at expiry
of the i-th asset. Then for the option value V :
∞ ∞
h(S1 , . . . , Sk )
...
S1 . . . Sk
0 0
exp −0.5α −1 α dS1 . . . dSk ,
−1
αi = σi (T − t)1/2 ln(Si /Si ) − (r − σi2 /2)(T − t) .
According to [11] the most important case in recent models is when the payoff
function is the exponent function.
We will now give a brief explanation which demonstrates the strength of the MC
and QMC approach [7]. According to [7], a time of order 1093 s will be necessary for
computation of the integral with the deterministic approach, and 1 year has 31536 ×
103 s.
According to [7] a time of
will be necessary in order to evaluate the multidimensional integral with the same
accuracy. We summarize that in the case of 100-dimensional integral it is 5 × 1091
times faster than the deterministic one. That motivates our study on the new highly
efficient stochastic approaches for the problem under consideration.
Highly Efficient Stochastic Approaches for Computation … 3
log s N
D ∗N = O .
N
⎛ ⎛ ⎞⎞
5
5
exp ⎝ 0.5ai xi2 ⎝2 + sin x j ⎠⎠ ≈ 2.923651, (5)
i=1 j=1, j =i
[0,1]5
20
exp xi ≈ 1.00000949634. (7)
i=1
[0,1]20
We also have done high performance computing with our methods for the first
time on a 100-dimensional integral:
100
I100 = exp xi , (8)
i=1
[0,1]100
using the exponential function in Taylor series and integrating (x1 · · · x100 )n we
receive
100
exp xi =
[0,1]100 i=1
∞
1
= =100 F100 (1, · · · , 1; 2, · · · , 2; 1).
n=0
(n + 1)100 n!
∞
(a1 )n · · · (a p )n x n
p Fq (a1 , · · · , a p ; b1 , · · · , bq ; x) = ,
n=0
(b1 )n · · · (bq )n n!
50
I50 = exp xi . (9)
i=1
[0,1]50
The results are given in the Tables including the relative error (RE) of the MC and
QMC method that has been used, the CPU-time (T) in seconds and the samples (#).
We will make a high performance computation, including the Optimized lattice rule
(OP), the Fibonacci based rule (FI), the Adaptive approach (AD) [8] and the Sobol
quasi-random sequence (SO) [3].
For the 3-dimensional integral, for the number of samples Generalized Fibonacci
numbers of the corresponding dimensionality, the best relative error is produced by
the optimized lattice algorithm OP—see Table 1, but for a preliminary given time in
seconds the optimized method OPT and the Fibonacci latice rule FI gives results of
the same order—see Table 2. For the 5-dimensional integral again the best approach
is OPT method, for N = 440096 it gives relative error of 8.16e − 7—see Table
3, while for 20s again FI method gives results of the same order as the optimized
method—see Table 4. For the 8-dimensional integral the Adaptive approach, the
Sobol QMC algorithm, and the Fibonacci approach produce relative error of the
same order—see Table 5, but for a preliminary given time in seconds, Fibonacci
approach is better than both Sobol QMC and Adaptive approach—see Table 6. For
the 20-dimensional integral Sobol QMC approach is better than both Fibonacci and
Adaptive approach—see Table 7 and Adaptive approach requires very huge amount
of time—near one hour for number of samples N = 524888 due to the division of
the subareas in the description of the algorithm. Thats why we omit this algorithm for
the 50- and 100-dimensional integrals. For 20s for 20-dimensional integral the best
result is produced again by the optimized lattice rule—1.23e − 8 in Table 8. For the
50-dimensional integral Fibonacci approach is worse than Sobol approach by at least
1 order—see Table 9, but for a preliminary given time in seconds Sobol QMC and
Fibonacci approach give relative errors of the same order—see Table 10. It is worth
mentioning that the Sobol approach requires more amount of time due to generation
of the sequence, while Fibonacci lattice rules and Optimized approach are more
faster and computationally efficient algorithms. For the 100-dimensional integral the
best result is produced by the optimized lattice approach—it gives 4.78e − 6 for
number of samples N = 220 —see Table 11 and for 100s it produces a relative error
of 8.16e − 7 which is very high accuracy and with 3–4 orders better than the other
stochastic approaches. So we demonstrate here the advantages of the new lattice
method and its capability to achieve very high accuracy for less than a minute on a
laptop with a quad-core CPU (Table 12).
6 V. Todorov et al.
5 Conclusion
Acknowledgements Venelin Todorov is supported by the Bulgarian National Science Fund under
Project DN 12/5-2017 “Efficient Stochastic Methods and Algorithms for Large-Scale Problems”
and by the National Scientific Program “Information and Communication Technologies for a Single
Highly Efficient Stochastic Approaches for Computation … 9
Digital Market in Science, Education and Security (ICT in SES)”, contract No DO1-205/23.11.2018,
financed by the Ministry of Education and Science in Bulgaria. Stoyan Apostolov is supported by
the Bulgarian National Science Fund under Young Scientists Project KP-06-M32/2 - 17.12.2019
“Advanced Stochastic and Deterministic Approaches for Large-Scale Problems of Computational
Mathematics”.
References
Abstract Cognitive Radio Networks equipped with dynamic spectrum access are
envisioned to address spectrum scarcity by allowing secondary users (SU) to utilise
vacant spectrum bands opportunistically. The SUs utilise cooperative spectrum
sensing (CSS) to make accurate spectrum access decision and to avoid interference.
Unfortunately, malicious users can cooperate with SUs and share false observations
leading to inaccurate spectrum access decision. The Spectrum Sensing Data Falsi-
fication (SSDF) attack is caused by malicious users. In this study, we investigated
the SSDF attack and a dynamic defence mechanism called the reputation and q-out-
of-m rule scheme designed to address the effects of SSDF attack. The scheme was
implemented in cognitive radio ad hoc networks. The fusion node was not consid-
ered. The success, missed detection, and false alarm probabilities were considered
as evaluation metrics and the MATLAB simulations.
1 Introduction
© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2022 11
X.-S. Yang et al. (eds.), Proceedings of Sixth International Congress on Information
and Communication Technology, Lecture Notes in Networks and Systems 236,
https://doi.org/10.1007/978-981-16-2380-6_2
12 V. Mthulisi et al.
2 Related Work
The SSDF attack can cause DoS to SUs or interference to PUs. The authors in [23]
proposed a scheme to counter the SSDF attack in CRAHN. The scheme implements
a q-out-of-m rule with a modified z-test. Chen et al. [24] implemented a scheme that
mitigates the SSDF attack in a distributed CRN environment called destiny-based
scheme (DBS). This scheme incorporated CSS where SUs share their sensing reports.
However, the hit and run attack was not considered.
Pongaliur et al. [25] proposed a distributed scheme to counter the SSDF attack
known as the multi-fusion-based distributed spectrum sensing (MFDSS). The scheme
implemented the modified z-test to combat extreme outliers and the reputation-based
system for the final decision-making. The authors in [26] proposed a reputation-based
system that clustered the SUs based on their history and initial reputation. However,
the study did not consider the unintentionally misbehaving SUs and the hit and run
attacks.
The work in [27] used the suspicious level to address the SSDF attack. SUs that
were deemed suspicious were isolated. The reports from SUs with trustworthy history
were included in decision-making. The advantage of this study was that it restored the
reports of unintentionally misbehaving SUs. Ye et al. [28] investigated and isolated
the SSDF attack by implementing a statistical consensus-based algorithm in CRAHN.
The scheme also isolated reports from unintentionally misbehaving SUs.
3 Network Model
We considered two types of users, SUs and PUs. The SUs cooperate in both sensing
and sharing sensing data. However, cooperative sensing can be compromised by
malicious nodes, which report incorrect spectrum observations during CSS. We study
three kinds of malicious nodes; the always yes, always no, and the hit and run. The
Spectrum Sensing Data Falsification Attack … 13
always yes report that a vacant spectrum is occupied. The always no report that an
occupied spectrum is vacant. The hit and run malicious node alters its reports to
avoid detection [29, 30]. We also studied two types of SUs, the legitimate SUs, and
the unintentionally misbehaving SUs. The legitimate SUs make use of the cognitive
capabilities and the unintentionally misbehaving SUs report incorrect sensing data
due to hidden terminal problem, signal fading, and multipath fading [31]. The SUs
incorporate energy detection in sensing the vacant channels. They also determine the
off periods of the spectrum by detecting the signal strength of the PUs as shown in
Eqs. (1–8).
where H0 denotes that the PU signal is absent and H1 denotes that the PU is present.
N being the number of SUs, xi (t) is the ith sample of the received signal, si (t) is
the PU signal, and ci (t) is the channel gain, while ai (t) denotes the additive white
Gaussian noise (AWGN). The energy E of a given signal xi (t) is:
∞
E= |xi (t)|∧ 2dt (3)
−∞
∞ ∞
|xi (t) | dt =
2
|xiπ ( f )|∧ 2dt (4)
−∞ −∞
where
∞
xiπ = e−2πi dt (5)
−∞
n n! n Φ n−Φ
= = P(Φ) = s f , Φ = 0, 1 . . . , n and 0 ≤ p ≤ 1.
Φ Φ!(n − Φ)! Φ
(9)
With mean
n n
n! n Φ n−Φ
E(Φ) = s Φ f n−Φ = Φ s f = ns. (10)
Φ=1
(n − Φ)!Φ! Φ=1
Φ
And variance
After the SUs have computed their binary decisions, they share observations as
depicted in Fig. 1. After sensing the spectrum, the SUs report their spectrum obser-
vations to their neighbouring nodes. Unfortunately, the malicious nodes MU 1 and
MU 2 interfere with the normal operations of the network for either greedy reasons
or to monopolise the band or to mislead the SUs and cause DoS or interference. The
goal of the SSDF attack can be modelled as follows:
Reputation-based system
The reputation-based system considered the trustworthiness of the SUs by evaluating
the history of their reports. SUs with reputation values that were above the TV of
0.6 were considered as malicious and their reports were isolated otherwise they are
included in the final decision. When the SUs with TVs below 0.6 are selected, the
q-out-of-m is implemented. In the q-out-of-m rule, 60% of the nodes with good
Spectrum Sensing Data Falsification Attack … 15
Fig. 1 Cooperative
spectrum sensing [22]
reputation are selected and the final decision is informed by q, which is either 0 or 1.
The reputation-based system and the q-out-of-m rule are shown in algorithms 1 and
2, respectively.
Algorithm 1 [30]
where m is the assessor node which performs the data fusion. The variable i denotes
the neighbour SU and di(t) is the status of the spectrum band. Si(t) is the value
of the neighbour report, and gm(t) is the final decision at device m. and r mi is the
current reputation of the device i at device m.
Algorithm 2 [31]
16 V. Mthulisi et al.
The q-out-of-m rule randomly polls 60% of the nodes with good reputation to
be considered in decision-making. If majority report 1, then the final transmission
decision is that the band is occupied, otherwise 0, it’s not occupied.
MATLAB R2015a was used to implement and simulate the scheme. The false alarm
probability and missed detection probability were set to 0.1 with SNR to 10 dB.
Energy detection was used as the sensing technique with a detection TV of 0. Different
network scenarios ranging from N = 10, 50,100,150–250 nodes and the percentage
of malicious nodes ranged from 10, 20, 40, 50–60% on each network size were
considered. The SUs sensed the spectrum band then computed their binary obser-
vations and shared their observations. We evaluated our scheme and compared its
performance with the MFDSS [25] and the DBS [24] schemes.
Figures 2, 3, and 4 depict the reputation-based system results. It shows the nodes
IDs in a network with 50 nodes and their reputation values. Figure 3 shows the
nodes that were isolated because their reputation values were above the TV value
of 0.6. The inputs of these nodes were not considered in the final decision-making.
Figure 4 shows the results of the nodes that had reputation values that were below
the TV. These nodes were selected and their reputation values were included in the
q-out-of-m rule phase.
In Fig. 5, the scheme’s success probability was evaluated. We varied the number of
nodes from (N = 10 to N = 250) with 10% of the nodes being MUs. We investigated
the hit and run (H n R), the always yes (Ay), and the always no attacks (An). We also
evaluated the impact of the unintentionally misbehaving nodes (U m). In N = 10,
Other documents randomly have
different content
Mitään sanomatta tarttui Jaakko Nikin kaulukseen ja työnsi hänet
ovesta ulos.
Jaakko meni kamariin. Siellä oli Emmi. Katsoi pitkään niinkuin olisi
sanonut: »Siinä sen näet.» Jaakon täytyi paeta sitä katsetta, niin
koski se häneen. Ja istuessaan pienen vierashuoneen keinutuolissa
hän tunsi olevansa kuin yksin koko talossa.
XVI.
Hän oli nyt köyhä, mutta oliko se hänen oma syynsä? Eivätkö nuo
lukuisat iilimadot, liikemiehet ja kauppiaat, olleet häntä kynineet
puhtaaksi niinkuin kaikkia muitakin. Hän oli yrittänyt rehellisesti
työskennellä välitystoimistossaan, mutta toiset, voimakkaammat,
löivät häneltä tien tukkoon. Nyt hän oli työtön, joita nykyinen
yhteiskuntajärjestelmä synnytti lukemattomia. Villen ajatukset
jatkuivat. Hänelle tuli polttava halu työskennellä nykyisen
vallankumouksellisen työväen parissa sen aatteita seuraten ja
tulkiten toisille tietämättömille. Hän oli mielestään kehittynyt
muutamassa hetkessä täydellisesti demokratiaa ymmärtäväksi ja oli
valmis sitä toisille opettamaan. Itse hän ei kyennyt käsittämään, että
hänen ajatuksensa, kerran irti päästyään ja saatuaan sysäyksen
ahdistavasta puutteesta, kääntyivät pian anarkokommunistiseen
suuntaan. Ei pilkahdustakaan edes hetkellisestä häpeän tunteesta,
että oli hylännyt maan ja omaa laiskuuttaan joutunut rappiolle.
Hänhän oli muka rehellisesti koettanut talohuijarina pyrkiä eteenpäin.
— Sinä ainakin.
— Mutta minä tahdon elää reilusti, niinkuin tähänkin asti. Jos sinä
et saa rahaa, niin kyllä minä…
— Miten?
Illoin työn loputtua ei tehnyt mieli olla asunnossa, jossa vaimo aina
valitti ja lapset pyysivät leipää. Ne eivät ymmärtäneet, että sitä piti
haukata vain suurimpaan nälkäänsä. Ja kun tätä ei jaksanut kuulla ja
katsella, tuli aina kävelleeksi työväen kahvilaan. Siellä sai istua
ainakin siltä rauhassa ja sitäpaitsi sai siellä aina kahvitilkan
hiukovaan nälkäänsä. Joskus tarjosivat toverit, toisinaan osti itsekin.
— Se on alkanut jo. Mutta miksi sen piti näin alkaa? Miksi ei kaikki
omat miehet ryssiä lyömään ja maata puhdistamaan vuosisataisesta
saastasta. Punakaarti, mikä se oli? Joukko hulikaanien harhaan
johtamia nälkäisiä työmiehiä, oman maan kansalaisia.
— Minähän olen aina tehnyt heille oikeutta eikä minun siis tarvitse
pelätä.
Nikki Purola istuu tupansa penkillä ja silmää silloin tällöin ulos, eikö
sieltä jo alkaisi sanantuojaa näkyä. Lyhyt talvinen päivä on jo
lopussa, ja kuu nousee metsän takaa vetäen valojuovan hämärän
tuvan lattialle.
Nikki kiroaa.