Professional Documents
Culture Documents
Textbook Information and Communication Technologies of Ecuador Tic Ec Miguel Botto Tobar Ebook All Chapter PDF
Textbook Information and Communication Technologies of Ecuador Tic Ec Miguel Botto Tobar Ebook All Chapter PDF
https://textbookfull.com/product/advances-in-information-and-
communication-technologies-processing-and-control-in-information-
and-communication-systems-mykhailo-ilchenko/
https://textbookfull.com/product/ethics-of-information-and-
communication-technologies-1st-edition-adriano-fabris-auth/
https://textbookfull.com/product/reclaiming-information-and-
communication-technologies-for-development-first-edition-unwin/
Communication: Innovation & Quality Miguel Túñez-López
https://textbookfull.com/product/communication-innovation-
quality-miguel-tunez-lopez/
https://textbookfull.com/product/cityscape-in-the-era-of-
information-and-communication-technologies-1st-edition-agata-
bonenberg-auth/
https://textbookfull.com/product/integrating-information-and-
communication-technologies-in-english-for-specific-purposes-1st-
edition-rosa-munoz-luna/
https://textbookfull.com/product/information-and-communication-
technologies-in-organizations-and-society-past-present-and-
future-issues-1st-edition-francesca-ricciardi/
https://textbookfull.com/product/advances-in-information-and-
communication-networks-proceedings-of-the-2018-future-of-
information-and-communication-conference-ficc-vol-1-kohei-arai/
Advances in Intelligent Systems and Computing 884
Miguel Botto-Tobar
Lida Barba-Maggi
Javier González-Huerta
Patricio Villacrés-Cevallos
Omar S. Gómez
María I. Uvidia-Fassler Editors
Information and
Communication
Technologies of
Ecuador (TIC.EC)
Advances in Intelligent Systems and Computing
Volume 884
Series editor
Janusz Kacprzyk, Polish Academy of Sciences, Warsaw, Poland
e-mail: kacprzyk@ibspan.waw.pl
The series “Advances in Intelligent Systems and Computing” contains publications on theory,
applications, and design methods of Intelligent Systems and Intelligent Computing. Virtually all
disciplines such as engineering, natural sciences, computer and information science, ICT, economics,
business, e-commerce, environment, healthcare, life science are covered. The list of topics spans all the
areas of modern intelligent systems and computing such as: computational intelligence, soft computing
including neural networks, fuzzy systems, evolutionary computing and the fusion of these paradigms,
social intelligence, ambient intelligence, computational neuroscience, artificial life, virtual worlds and
society, cognitive science and systems, Perception and Vision, DNA and immune based systems,
self-organizing and adaptive systems, e-Learning and teaching, human-centered and human-centric
computing, recommender systems, intelligent control, robotics and mechatronics including
human-machine teaming, knowledge-based paradigms, learning paradigms, machine ethics, intelligent
data analysis, knowledge management, intelligent agents, intelligent decision making and support,
intelligent network security, trust management, interactive entertainment, Web intelligence and multimedia.
The publications within “Advances in Intelligent Systems and Computing” are primarily proceedings
of important conferences, symposia and congresses. They cover significant recent developments in the
field, both of a foundational and applicable character. An important characteristic feature of the series is
the short publication time and world-wide distribution. This permits a rapid and broad dissemination of
research results.
Advisory Board
Chairman
Nikhil R. Pal, Indian Statistical Institute, Kolkata, India
e-mail: nikhil@isical.ac.in
Members
Rafael Bello Perez, Universidad Central “Marta Abreu” de Las Villas, Santa Clara, Cuba
e-mail: rbellop@uclv.edu.cu
Emilio S. Corchado, University of Salamanca, Salamanca, Spain
e-mail: escorchado@usal.es
Hani Hagras, University of Essex, Colchester, UK
e-mail: hani@essex.ac.uk
László T. Kóczy, Széchenyi István University, Győr, Hungary
e-mail: koczy@sze.hu
Vladik Kreinovich, University of Texas at El Paso, El Paso, USA
e-mail: vladik@utep.edu
Chin-Teng Lin, National Chiao Tung University, Hsinchu, Taiwan
e-mail: ctlin@mail.nctu.edu.tw
Jie Lu, University of Technology, Sydney, Australia
e-mail: Jie.Lu@uts.edu.au
Patricia Melin, Tijuana Institute of Technology, Tijuana, Mexico
e-mail: epmelin@hafsamx.org
Nadia Nedjah, State University of Rio de Janeiro, Rio de Janeiro, Brazil
e-mail: nadia@eng.uerj.br
Ngoc Thanh Nguyen, Wroclaw University of Technology, Wroclaw, Poland
e-mail: Ngoc-Thanh.Nguyen@pwr.edu.pl
Jun Wang, The Chinese University of Hong Kong, Shatin, Hong Kong
e-mail: jwang@mae.cuhk.edu.hk
Editors
Information
and Communication
Technologies of Ecuador
(TIC.EC)
123
Editors
Miguel Botto-Tobar Patricio Villacrés-Cevallos
Department of Mathematics and Computer Facultad de Ingeniería
Science Universidad Nacional de Chimborazo
Eindhoven University of Technology Riobamba, Ecuador
Eindhoven, Noord-Brabant, The Netherlands
Omar S. Gómez
Lida Barba-Maggi Escuela Superior Politécnica de Chimborazo
Facultad de Ingeniería Riobamba, Ecuador
Universidad Nacional de Chimborazo
Riobamba, Ecuador María I. Uvidia-Fassler
Facultad de Ingeniería
Javier González-Huerta Universidad Nacional de Chimborazo
Department of Software Engineering Riobamba, Ecuador
Blekinge Tekniska Högskola
Karlskrona, Blekinge Län, Sweden
This Springer imprint is published by the registered company Springer Nature Switzerland AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
Preface
v
vi Preface
We would like to express our sincere gratitude to the invited speakers for their
inspirational talks, to the authors for submitting their work to this conference, and
the reviewers for sharing their experience during the selection process.
Honorary Committee
Patricio Villacrés-Cevallos
vii
viii Organization
Program Committee
Sponsoring Institutions
CEDIA
https://www.cedia.edu.ec/es/
Contents
Communication Networks
Millimeter-Wave Channel Estimation Using Coalitional Game . . . . . . . 3
Pablo Palacios, José Julio Freire, and Milton Román-Cañizáres
Resource Allocation in WDM vs. Flex-Grid Networks:
Use Case in CEDIA Optical Backbone Network . . . . . . . . . . . . . . . . . . 18
Rubén Rumipamba-Zambrano, Luis Vargas, Claudio Chacón,
Flavio Rodríguez, and Juan Pablo Carvallo
NFC-Based Payment System Using Smartphones
for Public Transport Service . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
Diego Veloz-Cherrez and Jaime Suárez
An Open Source Synchronous and Asynchronous Approach
for Database Replication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
Marcos Orellana Cordero, Gerardo Orellana Cordero,
and Esteban Crespo Martinez
Forensics Analysis on Mobile Devices:A Systematic Mapping Study . . . 57
Jessica Camacho, Karina Campos, Priscila Cedillo, Bryan Coronel,
and Alexandra Bermeo
Software Engineering
Analytic Hierarchy Process of Selection in Version Control Systems:
Applied to Software Development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
Javier Vargas, Franklin Mayorga, David Guevara, and Edison Álvarez
Reliability and Validity of Postural Evaluations with Kinect v2 Sensor
Ergonomic Evaluation System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86
Christian Mariño, Rafael Santana, Javier Vargas, Luis Morales,
and Lorena Cisneros
xi
xii Contents
Computer Sciences
A Text Mining Approach to Discover Real-Time Transit Events
from Twitter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155
Belén Arias Zhañay, Gerardo Orellana Cordero, Marcos Orellana Cordero,
and María-Inés Acosta Urigüen
Automatic Microstructural Classification with Convolutional
Neural Network . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 170
Guachi Lorena, Guachi Robinson, Perri Stefania, Corsonello Pasquale,
Bini Fabiano, and Marinozzi Franco
Clustering Algorithm Optimization Applied to Metagenomics
Using Big Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 182
Julián Vanegas and Isis Bonet
Intelligent System of Squat Analysis Exercise to Prevent
Back Injuries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193
Paul D. Rosero-Montalvo, Anderson Dibujes, Carlos Vásquez-Ayala,
Ana Umaquinga-Criollo, Jaime R. Michilena, Luis Suaréz, Stefany Flores,
and Daniel Jaramillo
Architecture
Multifunctional Exoskeletal Orthosis for Hand Rehabilitation
Based on Virtual Reality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 209
Patricio D. Cartagena, José E. Naranjo, Lenin F. Saltos, Carlos A. Garcia,
and Marcelo V. Garcia
Contents xiii
IT Management
Importance of ICT’s Use in Business Management and Its
Contribution to the Improvement of University Processes . . . . . . . . . . . 241
Johanna Rosalí Reyes Reinoso and Deisy Carolina Castillo Castillo
ICT and Business Inclusion in the Southern Communities of the City
of Bogotá – Colombia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 253
Camilo José Peña Lapeira and Cliden Amanda Pereira Bolaños
Edition, Publication and Visualization of Geoservices
Using Open-Source Tools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 266
Pablo Landeta, Jorge Vásquez, Xavier Rea, and Iván García-Santillán
Web Technologies
LOD-GF: An Integral Linked Open Data Generation Framework . . . . 283
Víctor Saquicela, José Segarra, José Ortiz, Andrés Tello,
Mauricio Espinoza, Lucía Lupercio, and Boris Villazón-Terrazas
Semantic Architecture for the Extraction, Storage, Processing
and Visualization of Internet Sources Through the Use of Scrapy
and Crawler Techniques . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 301
Ramiro Leonardo Ramírez-Coronel, Ana Cristina Cárdenas,
María-Belén Mora-Arciniegas, and Gladys-Alicia Tenesaca-Luna
Use of Apache Flume in the Big Data Environment for Processing
and Evaluation of the Data Quality of the Twitter Social Network . . . . 314
Gladys-Alicia Tenesaca-Luna, Diego Imba, María-Belén Mora-Arciniegas,
Verónica Segarra-Faggioni, and Ramiro Leonardo Ramírez-Coronel
ICT in Education
Sophomore Students’ Acceptance of Social Media for Managing
Georeferenced Data in a Socially-Enhanced Collaborative
Learning Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 329
Erika Lozada-Martínez, Félix Fernández-Peña, and Pilar Urrutia-Urrutia
xiv Contents
1 Introduction
Large antenna arrays (i.e. Massive MIMO) at both sides eNodeB (eNB) and
UE is a promising technology to achieve high-throughput services [1]. Large
antenna arrays at the same time deal with the high path-loss in millimeter
frequencies. By the other hand, channel states information (CSI) in terms of
channel matrix or beam alignment are needed at the eNB to point the beams in
the UE direction. Both strategies are usually acquired by a training sequence [2].
The sequence is used to measure every beamformer and combiner to estimate
the pair of beams that are closer to the desired angles, but this exhaustive-search
method need a large number of measurements to estimate the best beam-pair.
Additionally, this fact could leads to lower channel rate. In vehicular or train
scenarios where due to the UE speed, the channel coherence time becomes shorter
and the training period could occupies all the coherence time, leaving no time
for data transmission [3].
c Springer Nature Switzerland AG 2019
M. Botto-Tobar et al. (Eds.): TICEC 2018, AISC 884, pp. 3–17, 2019.
https://doi.org/10.1007/978-3-030-02828-2_1
4 P. Palacios et al.
2 Related Work
about channel estimation for static and high speed environments have been done
for single carrier and multi-carrier systems in micro-wave frequencies, such that
this is a practical assumption, this solution applies virtual channel decomposition
[14] to the lower frequency channel in order to extract the dominant paths, and
reminding the fact that due to higher path loss and shorter wavelength, high
frequency systems are expected to use a larger antenna array than sub-6 Ghz
[15], this implies a narrower beam at mm-wave system.
Notation: A is a set, |A| is the cardinality of set A. lower-case a is a scalar,
a is a vector, A is a matrix. AT , AH denote the transpose and Hermitian of
matrix A.
3 System Model
We consider a mm-wave MIMO downlink system with uniform linear arrays
(ULAs) conformed by Nt transmitter antennas in the UE and Nr receiver anten-
nas in the eNB, as show in Fig. 1. We consider that both the transmitter and the
receiver have only one RF chain, hence, only analog beamforming/combining
can be applied.
We use f, and q to denote the beamformer and combiner vector, respectively:
1 2π
f = √ [1, ..., ej(Nr −1) λ d cos φ ]T , (1)
Nt
where φ ∈ [−π/2, π/2], is a quantized angle of departure, besides f has constant
modulus entries and only phase can varying, in similar fashion the combiner:
1 2π
q = √ [1, ..., ej(Nr −1) λ d cos θ ]T , (2)
Nr
where θ ∈ [−π/2, π/2], is a quantized angle of arrival, the AoAs and AoDs can
be taken following regular or non regular sampling strategies, the detail about
how we choose this angles is discussed later in Sect. 3.2. Then considering a
narrowband channel model H ∈ CNr ×NT , the received signal in the eNB can be
modeled as:
√
y= ρqH Hfx + qH v, (3)
√
where ρ is the average transmit power in the training phase, x is the training
symbol, and v is the vector of i.i.d. ∼ CN (0, σ02 I) noise.
Fig. 2. Illustration of virtual channel matrix for: (a) Sub-6 Ghz with Nr = Nt = 16, (b)
Mmwave with Nr = Nt = 64.
where L is the number of paths, αl represents the complex path gain of the
l -th propagation path, θl ∈ [−π/2, π/2] and φl ∈ [−π/2, π/2] denote the AoA
and AoD of the L-th path at transmitter and receiver, respectively. The vectors
at (·) and ar (·) denote the array response vectors for transmitting and receiving
antenna arrays. The array response vector of ar (θl ) is given by:
1 2π 2π
ar [θl ] = √ [1, ej λ d cos θl , ..., ej(Nr −1) λ d cos θl ]T , (5)
Nr
where λ is the transmission wave length and d is the antenna spacing.√Further-
more the array response vector in (5) has a unit norm and the factor Nt Nr in
(4) reflects this normalization. The array response vector at (φl ), can be written
in a similar fashion.
where every beamforming vector has the same form of the array response vector,
such that:
The angles are chosen as in [14] such that every beam has same magnitude but
different width, that is to say narrower at broadside direction and broader at
endfire direction, if we set the inter-antenna spacing as λ/2, the positive and
negative angles of departure are given by:
2m Nt
φ̃(+)
m = arcsin( ), m ∈ {1, ..., } (10)
Nt 2
2m −Nt
φ̃(−) = arcsin( ), m ∈ {−1, ..., } (11)
m
Nt 2
The AoAs design follow the same rule than AoDs.
sub-6 Ghz, reducing the searching space to those overlapped mm-wave angles,
then the candidate beam list provided by the spatial information from the sub-
6 Ghz channel, is stored in the set S. At this point the task is select the pair of
virtual angles in the mm-wave virtual representation that provide higher gain in
order in the receiver in order to estimate the mm-wave channel.
Considering the set S is carrying the bunch of possibles arrival and departure
virtual angles θ̄p and φ̄r respectively, where θ̄p , p = 1, 2, ..., P ; P < Nr and
φ̄r , r = 1, 2, ..., Nt ; R < Nt decreasing the searching space and training overhead
to P × R. Then using the set of arrival virtual angles to construct the combiner
at the receiver i.e. Q = [q1 (θ̄1 ), q2 (θ̄2 ), ..., qP (θ̄P )]. The departure virtual angles
are feedback to the transmitter in order to build its beamformer F, where log2 P
bits are needed, that could be sent using the sub-6 Ghz channel. The new received
signal in mm-wave systems is
√
Y = ρQH S HFS + V, (13)
1.2
0.8
0.6
0.4
0.2
0
5 10 15 20 25 30 35 40
Length of signal x
We apply the linear LS estimator to calculate the complex path value associated
to the given AoA and AoD, given by
z = (xxH )−1 xyH (16)
√
where z = ρqH î
Hfî , that carries the path gain estimation. About the vector
training symbol length the Fig. 3 shows the variability of the path gain estimated
αl according to the signal length, we can observe there is a convergence with a
length further than 20.
Additionally we focus on analyze the rate between the eNB and an UE in
downlink transmission. Then the rate R is affected by changing the channel
coherence time due to MS velocity, as
To − Tτ ρqH Hfî fH
î
HH qî
RSingle = log2 (1 + î ) (17)
To σo2
where the pre log-factor takes account the training overhead necessary to esti-
mates the channel, here To = λB2vo , λ, Bo , vo , are mm-wave carrier wavelength,
o
5 Multicell Analysis
From now, considering an Orthogonal Frequency Division Multiple Access
(OFDMA) multicell system as is shown in Fig. 4. We assume in every cell there
is a small cell microwave station and millimeter-wave small station located in
the same position. Also, we assume a microwave channel estimation have been
already done in every small cell microwave station, such that a CE training-based
method is applied in every millimeter-wave small station. We will analyze the
performance of this mm-wave CE proposed method. Therefore, we focus on this
frequency band. Additionally we assume no interference among the subchannels.
Let’s consider the Fig. 4 where the UE1 is located in the eNB2 coverage area
border, such that this user must deal with handoff management and interfer-
ence from neighboring cells. In order to overcome these problems, we propose a
method based on cooperative model using coalitional games between the con-
cerned eNBs. The goals of this section is to increase the channel rate in the
transmission data stage.
with each other via a wire backhaul, e.g., fiber, providing reliable links for traffic
control. Each MMWSC i ∈ N works over the same set of channels, that is the
available spectrum, i.e., is the frequency resource shared between all MMWSCs.
Therefore the MMWSCs should share these bandwidth allocated in an oppor-
tunistic way in order to avoid handoff recurrent, besides diminishing the inter-
ference, leading to an improve in the channel rate.
on subchannel l, Hi,l is the massive MIMO channel between the MMWSC i and
the UE on the l-th subchannel, qi,l is the combining vector in the UE pointing
to MMWSC i on subchannel l and IS denotes the power interference suffered by
an UE from neighboring MMWSCs, as
IS = ρj,l qH
j,l Hj,l fj,l fj,l Hj,l qj,l ,
H H
(20)
j∈N ,j=i
The term IS can significantly reduce the rates achieved. Specifically, depend-
ing on the signal to noise and interference ratio (SNIR) feedback from the UE,
the MMWCs can decide to form cooperative groups called coalitions, in order
to overcome the interference between neighboring cells and the handoff manage-
ment. Under this coalitional game approach, the MMWCs are modeled as players
that access the spectrum, avoiding coalitions among them by jointly scheduling
their transmissions, we state the following definition for a coalition:
Definition 1. A coalition S ⊆ N is a non−empty subset of N in which players
inside the set access the spectrum via a coordinated manner.
Therefore we consider that if a coalition S is formed, the transmissions inside S
are managed by a local scheduler as in [20], using Time Division multiple access
(TDMA) mode, such that the subchannels are split in time slots allocated for
every MMWSC. As a result no more than one MMWSC will access the every
channel in each time slot, mitigating the interference inside the coalition S.
Although coordination can help to increase the channel rate by decreasing
the interference, it also incurs in a coordination cost. Here we consider this coor-
dination cost in terms of transmit power. Thus, the power spent by a MMWSC
i to reach the farthest MMWSC ĵ in a coalition S is ρĵ,i . Then the power cost
needed to form a coalition S is:
ρS = ρĵ,i (21)
i∈S
In addition, we define a maximum tolerable power cost ρlim for every coalition
S as in [21]. By considering the coalition cost in this way we take account the
spatial distribution of the MMSCs and the coalition size.
We also assume that v(∅) = 0. Thus, the definition above imposes a dependence
on the coalitional structure N when evaluating the value of S ⊆ N , i.e. to the
players N \S as well. By the other hand the TU property implies that the total
utility represented by a real number, (in our case the channel rate) can be divided
in any manner between the coalition members.
Clearly, the cooperation model can be stated as a game in partition form
where the MMWSC are the players, since the channel rate of a coalition is
affected by the interference from others players, thus there is a dependence
between the coalition and the players which do not belong to the coalition
(coalitional structure). Therefore the utility achieved by the coalition S can
be expressed in terms of the channel rate as:
ρi,l qH H H
i,l Hi,l fi,l fi,l Hi,l qi,l
U(S, GN ) = αil log2 (1 + ), (22)
i∈S l∈Γ
σo2 + ÎS
where αil ∈ [0, 1] denotes the fraction of time duration during which MMWSC i
transmits on the subchannel l to the UE. In the non-cooperative scenario, i.e.,
FDD transmission mode, each transmission occupies a whole subchannel, hence
αil = 1. In addition IŜ denotes the co-tier interference suffered by the UE served
by MMWSC i on subchannel l from players j ∈ N \S as follows:
IŜ = ρj,l qH
j,l Hj,l fj,l fj,l Hj,l qj,l ,
H H
(23)
j∈GN \S, j=i
when the interfering MMWSCs are closely located in a way that ρS ≤ ρlim , thus
coalition with many members is unlikely to happen, in this sense we are not focus
on analyze the stability of the grand coalition. Finding an optimal coalitional
structure for games in partition form have been studied in, [22] and [23], we
will apply the concept of recursive core as in [23]. The concept of recursive core
studies the behaviour of dynamics coalition formation but also considering the
interference from neighboring MMWSCs. The detail about how the recursive
core works is provided in [23].
In the algorithm proposed (Algorithm 1), at the first step an UE is sens-
ing interference which will trigger the coalitional game. Secondly to resolve the
coalitional game, i.e., achieving the recursive core, we propose three phases: envi-
ronment sensing, coalition formation and scheduling transmission. First of all,
the network is partitioned in N single coalitions, this is the non-cooperative
case. Then by discovering neighbors stage, the MMWSCs can create a list of
existing neighbors in the network, once each MMWSC has a neighboring list,
they can start a recursive coalition formation to find a recursive core. Here
every MMWSC establishes negotiations with the discovered neighbors, to iden-
tify potential partners for cooperation, this information is exchanged by using
the wire reliable control channel. Then the cooperation cost for every coalition
is calculates as in Eq. 21, and the potential payoff is computed as in Eq. 25 for
every member of a coalition. To reach the recursive core, each MMWSC joins
to the coalition which provides the highest revenue, i.e., payoff. Then, once the
coalition is formed, coalition-level scheduling occurs in each coalition.
6 Numerical Results
In this section, simulations are carried out to evaluate the performance for the
proposed channel estimation strategies. We assess the channel estimation error,
+the effect of SNR and speed on the method’s performance of the proposed
methods, additionally the computation time is analyzed.+
The sub-6 Ghz channel works at f6G = 3 GHz and mm-wave channel at
fmm = 28 GHz, bandwidth Bo = 10 MHz alike than mm-wave case, the distance
between the UE and eNB is set at 50m, the path loss exponent at sub-6 Ghz
is equal to 2 while for mm-wave has been set to 3. The angles of arrival and
departure for both environments are limited at [− π2 , π2 ), the antennas number at
sub-6 Ghz and mm-wave will be changing according to every experiment at the
UE and eNB, the inter antenna element distance at both cases is half-wavelength.
Additionally the signal-to-noise ratio is set as SN R = Pσ2o .
o
To assess the estimation error we express the mm-wave virtual channel Hvirt
as sparse [16], such that Hvirt = JTLNr Λ JLNt where Λ is an L × L diagonal
matrix and the L are the entries different than zero of Hvirt , the binary matrices
JTLNr , JLNt are L × Nr and L × Nt selection matrices, generated by keeping L
rows of Nr × Nr and Nt × Nr identity matrices respectively. Therefore we can
compute the mean square error as MSE = E{ Hvirt − H̃virt 2F / Hvirt 2F }.
14 P. Palacios et al.
end if
Step 3: High Speed mmWave Communications
The last experiment takes account the system complexity between the refined
CE method and FCE detailed at [17] in term of their computation time, the
complexity cost are O(Nr T + LNt T + LT ) and O(Nr Nt L + Nr L) respectively.
All simulations are conducted at Matlab R2015a by the Intel Core i5 CPU, in
Fig. 6(a) the Nr is set to 16, while Nt is increasing, then in Fig. 6(b) theNt is set
to 16 and Nt is changing. For the sake of fairness we comparative both method
Fig. 6. Computation time of CE based on training and FCE, varying: (a) Nr and
(b) Nt.
16 P. Palacios et al.
7 Conclusions
References
1. Wang J, Zhu H, Gomes NJ (2012) Distributed antenna systems for mobile com-
munications in high speed trains. IEEE J Sel Area Commun 30(4):675–683
2. Hassibi B, Hochwald BM (2003) How much training is needed in multiple-antenna
wireless links? IEEE Trans Inf Theory 49(4):951–963
3. Heath J, Robert W (2013) What is the role of MIMO in future cellular networks:
massive? Presentation delivered at IEEE international conference on communica-
tions (ICC)
4. Maschietti F, Gesbert D, Kerret P, Wymeersch H (2013) Robust location-aided
beam alignment in millimeter wave massive MIMO. In: GLOBECOM 2017-2017
IEEE global communications conference, pp 1–6
5. Hur S, Kim T, Love DJ, Krogmeier JV, Thomas TA, Ghosh A (2013) Millimeter
wave beamforming for wireless backhaul and access in small cell networks. IEEE
Trans Commun 61(10):4391–4403
6. Alkhateeb A, El Ayach O, Leus G, Robert RW (2014) Channel estimation and
hybrid precoding for millimeter wave cellular systems. IEEE J Sel Top Signal
Process 8(5):831–846
7. Alkhateeb A, Ayach OE, Leus G, Heath RW (2013) Hybrid precoding for mil-
limeter wave cellular systems with partial channel knowledge. In: Proceedings of
information theory and applications workshop (ITA), February 2013, p 15
8. Ayach OE, Rajagopal S, Surra SA, Pi Z, Heath RW (2013) Spatially sparse pre-
coding in millimeter wave MIMO Systems. In: Proceedings of information theory
and applications workshop (ITA), February 2013, p 15
9. Wang Y, Venugopal K, Andreas F, Heath RW Jr (2017) Analysis of urban mil-
limeter wave microcellular networks. IEEE Trans Veh Technol 66(10):8964–8978
Another random document with
no related content on Scribd:
demoniacal laughter. Yet I could not pause to remove those grating
shoes of toothed steel. Every second even might be precious now.
I drew near the turn, the revolver thrust forward in readiness for
instant action.
I reached it, and there just beyond, a dark figure was standing,
framed in a blaze of light.
It was Milton Rhodes.
He turned his head, and I saw a smile move athwart his features.
"Well, we've found it, Bill!" said he.
I was drawing near to him.
"That scream?" I said. "Who gave that terrible scream?"
"Terrible? It didn't sound terrible to me," smiled Milton Rhodes. "Fact
is, Bill, I'd like to hear it again."
"What on earth are you talking about?"
"'Tis so."
"Who was it? Or what was it?"
"Why, the angel herself!" he told me.
"Where is she now?"
"Gone, Bill; she's gone. When she saw me, she fetched up, gave
that scream, then turned and vanished—around that next turn."
"What is she like, Milton?"
"I wish that I could tell you! But how can a man describe Venus? I
know one thing, Bill: if all the daughters of Drome are as fair as this
one that I saw, I know where all the movie queens of the future are
coming from."
I looked at him, and I laughed.
"Wait till you see her, Bill. Complexion like alabaster, white as
Rainier's purest snow! And hair! Oh, that hair, Bill! Like ten billion
dollars' worth of spun gold!"
"Gosh."
"Wait till you see her," said Milton.
"And the demon?" I queried.
"I didn't see any demon, Bill."
There was silence for a little space.
"Then," I said, "the whole thing is true, after all."
"You mean what grandfather Scranton set down in his journal, and
the rest of it?"
I nodded.
"I never doubted that, Bill."
"At times," I told him, "I didn't doubt it. Then, again, it all seemed so
wild and unearthly that I didn't know what to think."
"I think," he said with a wan smile, "that you know what to think now
—now when you are standing in this very way to Drome, whatever
Drome may be."
"Yes. And yet the thing is so strange. Think of it. A world of which
men have never dreamed, save in the wildest romance! An
underground world. Subterranean ways, subterranean cities, men
and women there—"
"Cavernicolous Aphrodites!" said Milton Rhodes.
"And all down there in eternal darkness!" I exclaimed. "Why, the
thing is incredible. No wonder that I sometimes find myself
wondering if I am not in a dream."
Said Milton Rhodes:
"But come, Bill," he added. "Don't let this a priori stuff bowl you over.
In the first place, it isn't dark down there—when, that is, you get
down far enough."
"In Heaven's name, how do you know that?"
"Why, for one thing, if this subterranean world was one of unbroken
darkness, the angel (and the demon) would be blind, like those poor
fishes in the Mammoth Cave. But she is no more blind than you or I.
Ergo, if for no other reason, we shall find light down there."
"Of course, they have artificial light, or—"
"I don't mean that. If there had not been some other illumination, this
strange race (of whose very existence Science has never even
dreamed) would have ceased to exist long ago—if, indeed, it ever
could have begun."
"But no gleam of sunlight can ever find its way down to that world."
"It never can, of course. But there are other sources of light—
nebulas and comets in the heavens, for example, and auroras,
phosphorus and fire-flies here on earth. The phenomena of
phosphorescence are by no means so rare as might be imagined.
Why, as Nichol showed, though any man who uses his eyes can see
it himself, there is light inherent even in clouds."
I have the professor's book before me as I write—J. P. Nichol, LL.D.,
Professor of Astronomy in the University of Glasgow—and here are
his own words:
"Whatever their origin, they [the auroras] show the existence of
causes in virtue of whose energy the upper strata of our atmosphere
become self-luminous, sometimes in a high degree; for, in northern
regions, our travelers have read by their brilliance. But the Aurora is
not the only phenomenon which indicates the existence of a power in
the matter of our globe to emit light. One fact, that must have been
often noticed, forcibly impresses me with the conviction, that here,
through what seems common, truths of much import will yet be
reached. In the dead of night, when the sky is clear, and one is
admiring the brilliancy of the stars, hanging over a perfectly obscured
earth, a cloud, well known to observing astronomers, will at times
begin to form, and it then spreads with astonishing rapidity over the
whole heavens. The light of the stars being thus utterly shut out, one
might suppose that surrounding objects would, if possible, become
more indistinct: but no! what was formerly invisible can now be
clearly seen; not because of lights from the earth being reflected
back from the cloud—for very often there are none—but in virtue of
the light of the cloud itself, which, however faint, is yet a similitude of
the dazzling shell of the Sun."
After mentioning the phosphorescence of the dark hemisphere of
Venus and the belief that something similar has been seen on the
unillumined surface of our satellite, he continues thus:
"But the circumstance most remarkably corroborative of the
mysterious truth to which these indications point, is the appearance
of our midnight luminary during a total eclipse. By theory, she ought
to disappear utterly from the heavens. She should vanish, and the
sky seem as if no Moon were in being; but, on the contrary, and even
when she passes the very centre of the Earth's shadow, she seems
a huge disc of bronze, in which the chief spots can easily be
described by a telescope."
And that, remember, when the moon is in the utter blackness of the
earth's shadow. Of course, another explanation has been advanced;
but it does not take the professor long to dispose of that.
"It has been put forth in explanation," he says, "that a portion of the
rays of the Sun must be reflected by our atmosphere and bent
toward the eclipsed disc, from which again they are reflected to the
Earth—thus giving the Moon that bronze color; but, the instant the
hypothesis is tested by calculation, we discern its utter inefficiency.
Nor is there any tenable conclusion save this:—That the matter both
of Sun and planets is capable, in certain circumstances whose exact
conditions are not known, of evolving the energy we term light."
All this, and more, Rhodes explained to me, succinctly but clearly.
"Oh, we'll find light, Bill," said he.
All the same this subterranean world for which we were bound
presented some unpleasant possibilities, in addition, that is, to those
concomitant to its being a habitat of demons—and Heaven only
knew what besides.
"And then there is the air," I said. "As we descend, it will become
denser and denser, until at last we will be able to use these ice-picks
on it."
Rhodes, who was removing his creepers, laughed.
"We will have to make a vertical descent of three and one half miles
below the level of the sea—a vertical descent of near five miles from
this spot where we stand, Bill—before we reach a pressure of even
two atmospheres."
"The density then increases rapidly, doesn't it?"
"Oh, yes. Three and a half miles more, and we are under a pressure
of four atmospheres, or about sixty pounds to the square inch. Three
and a half miles farther down, or ten and one half miles in all below
the level of the sea, and we have a pressure upon us of eight
atmospheres. Fourteen miles, and it will be sixteen atmospheres. At
thirty-five miles the air will have the density of water, at forty-eight
miles it will be as dense as mercury, and at fifty miles we shall have
it as dense as gold."
"That will do!" I told him. "You know that we can never get down that
far."
"I have no idea how far we can go down, Bill."
"You know that we could never stand such pressures as those."
"I know that. But, as a matter of fact, I don't know what the pressures
are at those depths. Nor does any other man know. What I said a
moment ago is, of course, according to the law; but there is
something wrong with the law, founded upon that of Mariotte—as
any physicist will tell you."
"What's wrong with it?"
"At any rate, the law breaks down as one goes upward, and I have
no doubt that it will be found to do so as one descends below the
level of the sea. If the densities of the atmosphere decrease in a
geometrical ratio as the distances from sea-level increase in an
arithmetical ratio, then, at a distance of only one hundred miles up,
we should have virtually a perfect vacuum. The rarity there would be
absolutely inconceivable. For the atmospheric density at that height
would be only one billioneth of what it is at the earth's surface."
"And what is the real density there?"
"No man knows or can know," replied Rhodes, "until he goes up
there to see. But meteors, rendered incandescent by the resistance
they encounter, show that a state of things exists at that high altitude
very different from the one that would be found there if our formulae
were correct and our theories were valid. And so, I have no doubt,
we shall find it down in Drome.
"Formulae are very well in their place," he went on, "but we should
never forget, Bill, that they are often builded on mere assumption
and that a theory is only a theory until experiment (or experience)
has shown us that it is a fact. And that reminds me: do you know
what Percival Lowell says about formulae?"
I said that I didn't.
"'Formulae,' says the great astronomer, 'are the anaesthetics of
thought.'
"I commend that very highly," Milton Rhodes added, "to our fiction
editors and our writers of short stories."
"But—"
"But me no buts, Bill," said Milton. "And what do your scientists know
about the interior of this old earth that we inhabit, anyway? Forsooth,
but very little, Billy, me lad. Why they don't even know what a
volcano is. One can't make a journey into the interior of the earth on
a scratch-pad and a lead-pencil, or, if he does, we may be pardoned
if we do not give implicit credence to all that he chooses to tell us
when he comes back. For instance, one of these armchair
Columbuses (he made the journey in a machine called d2y by dx2
and came out in China) says that he found the interior in a state of
igneous fluidity. And another? Why, he tells us that the whole earth is
as rigid as steel, that it is solid to the very core."
"It seems," said I, "to be a case of
"'Great contest follows, and much learned dust
Involves the combatants; each claiming truth,
And truth disclaiming both.'"
"The truth, in this case, is not yet known, of course," replied Milton
Rhodes, "though I trust that you and I, Bill, are fated to learn it—
some of it, I should say."
He smiled a queer, wan smile.
"Whether we are fated, also, to reveal it to the world, to our world—
well, as for that, quién sabe?" he said.
"Then," I remarked, my fingers busy removing my ice-creepers,
"what we read about the state of things in the interior of the earth—
the temperature, the pressure, the density—then all that is pure
theory?"
"Of course. How could it be anything else? All theory, save, that is,
the mean density of the earth. And that mean density gives us
something to think about, for it is just a little more than twice that of
the surface materials. With all this enormous pressure that we hear
so much about and the resultant increase of density with depth, the
weight of the earth certainly ought to be more than only five and one
half times that of a globe of equal size composed of nothing but
water."[6]
"Kind of queer, all right," was my comment.
"It is queer, all right—as the old lady said when she kissed the cow.
However, as old Dante has it, 'Son! our time asks thrifter using.'"
As the last words left his lips, I straightened up, the toothed shoes in
my hand; and, as I did so, I started and cried:
"Hear that?"
Rhodes made no answer. For some moments we stood there in
breathless expectation; but that low mysterious sound did not come
again.
I said:
"What was that?"
"I wish that I knew, Bill. It was faint, it was—well, rather strange."
"It was more than that," I told him. "It seemed to me to be hollow—
like the sound of some great door suddenly closing."
My companion looked at me rather quickly.
"Think so, Bill?" he said. "I thought 'twas like the sound of something
falling."
There was a pause, one of many moments, during which pause we
stood listening and waiting; but the gallery remained as silent as
though it had never known the tread of any living thing.
"Well, Bill," said Milton Rhodes suddenly, "we shall never learn what
Drome means if we stay in this spot. As for the creepers, I am going
to leave mine here."
The place where he put them, a jutting piece of rock, was a
conspicuous one; no one passing along the tunnel could possibly fail
to notice the objects resting there. Mine I placed beside them,
wondering as I did so if I should ever see this spot again.
Milton then wrote a short note, which recorded little more than our
names, the date of our great discovery and that we were going
farther. This, carefully folded, he placed beside the creepers and put
a rock-fragment upon it. I wondered as I watched him whose would
be the eyes that would discover it. Some inhabitant of this
underground world, of course, and to such a one the record would
be so much Greek. 'Twas utterly unlikely that any one from the world
which we were leaving would ever see that record.
"And now, Bill," said Milton Rhodes, "down we go!"
And the next moment we were going—had begun our descent into
this most mysterious and dreadful place.
Chapter 16
"ARE WE ENTERING DANTE'S INFERNO
ITSELF?"
When Scranton came with his weird story of Old He, I was, I
confess, not a little puzzled by his and Milton's reference to the
extraordinary scientific possibilities that it presented. At first I could
not imagine what on earth they meant. But I saw all those
possibilities very clearly now, and a thousand more I imagined. I
knew a wild joy, exultation, and yet at the same time the wonder and
the mystery of it all made me humble and sober of spirit. I admit, too,
that a fear—a fear for which I can find no adequate name—had laid
its palsied and cold fingers upon me.
In a few moments we reached that spot where the angel had
vanished. There we paused in curiosity, looking about; but nothing
was to be seen. The gallery—which from this point swung sharp to
the right and went down at a rather steep angle—was as silent as
some interstellar void.
"Bill," smiled Milton Rhodes, "he is idle who might be better
employed."
And he started on, or, rather, down. A hundred feet, however—we
were now under the glacier—and he halted, turned his light full upon
the left-hand wall, pointed and said:
"There you are, Bill—the writing on the wall."
I pressed to his side and stood staring. The rock there was as
smooth, almost, as a blackboard; and upon it, traced in white chalk,
were three inscriptions, with what we took to be names appended to
them. That on the right was clearly a very recent one—had been
placed there doubtless, at the most but a few days since, by that
"cavernicolous Venus" that Milton Rhodes had seen for so fleeting a
moment.
It was Milton's opinion that the characters were alphabetical ones,
though at first I was at a loss to understand how they could be
anything to him but an utter mystery. The letters were formed by
straight lines only. The simplest character was exactly like a plain
capital T, with, that is, the vertical line somewhat elongated. And it
was made to perform the office of another letter by the simple
expedient of standing it upon its head. The number of cross-lines
increased up to six, three at the top and three at the bottom; and in
one or two characters there were two vertical lines, placed close
together.
"Evidently," observed Milton Rhodes, "this alphabet was constructed
on strictly scientific principles."
For a space we stood there looking, wondering what was recorded in
that writing so strange and yet, after all, so very and beautifully
simple. Then Milton proceeded to place another record there, and,
as he wrote, he hummed: