ArticlePDF Available

Abstract and Figures

The increasing urbanization level of the world population has driven the development of a smart city geographic system, conceived as a fully connected wide area characterized by the presence of a multitude of smart devices, sensors, and processing nodes aimed at distributing intelligence into the city. At the same time, the pervasiveness of wireless technologies has led to the presence of heterogeneous networks, operating simultaneously in the same city area. One of the main challenges in this context is to provide sustainable solutions able to jointly optimize the data transfer, exploiting heterogeneous networks, and the data processing, exploiting heterogeneous devices, for managing smart city applications for citizens' communities. In this article, the UMCC framework is developed, introducing a mobile cloud computing model describing the flows of data and operations taking place in the smart city. In particular, we focus on the proposal of a unified offloading mechanism where communication and computing resources are jointly managed, allowing load balancing among the different entities in the environment, delegating both communication and computation tasks in order to satisfy the smart city application requirements. This allows us to cope with the limited battery power and computation capacity of smart mobile devices and plays a key role in a smart environment where wireless communication is of utmost relevance, particularly in the mobility and traffic control domains.
Content may be subject to copyright.
A Unified Urban Mobile Cloud Computing
Offloading Mechanism for Smart Cities
Daniela Mazza, Daniele Tarchi, Senior Member, IEEE, and
Giovanni E. Corazza, Senior Member, IEEE,
Department of Electrical, Electronic and Information Engineering
University of Bologna
40136 Bologna, Italy
Abstract
The increasing urbanization level of the world population has driven the development of a Smart City
geographic system, conceived as a fully connected wide area characterized by the presence of a multitude
of smart devices, sensors and processing nodes aimed at distributing intelligence into the city. At the
same time, the pervasiveness of wireless technologies has led to the presence of heterogeneous networks,
operating simultaneously in the same city area. One of the main challenges in this context is to provide
sustainable solutions able to jointly optimize the data transfer, exploiting heterogeneous networks, and
the data processing, exploiting heterogeneous devices, for managing Smart City applications for the
citizens community. In this paper, the Urban Mobile Cloud Computing (UMCC) framework is developed,
introducing a mobile cloud computing model describing the flows of data and operations taking place
in the Smart City. In particular, we focus on the proposal of a unified offloading mechanism where
communication and computing resources are jointly managed allowing a load balancing among the
different entities in the environment, delegating both communication and computation tasks in order
to satisfy the Smart City application requirements. This allows to cope with the limited battery power
and computation capacity of the Smart Mobile Devices (SMDs), and plays a key role in a smart
environment where wireless communication is of utmost relevance, particularly in mobility and traffic
control domains.
Index Terms
Smart City, Mobile Cloud Computing, HetNets, Offloading mechanisms, QoS management.
1
I. INTRODUCTION
According to the World Urbanization Prospect1published by the United Nations, more
than half of the population is living nowadays in urban areas, and about 70% will be city
persons by 2050. At the same time with urbanization, an extraordinary phenomenon concerning
the Information and Communication Technology (ICT) is happening: according to the Visual
Networking Index2, the number of connected devices in mobility has overtaken the number of
people in the world, and by 2018 it will be over 10 billion, including Machine to Machine
(M2M) modules in the Internet of Things. Mobile data traffic is expected to increase about 11
times in the next five years.
Urbanization and ICT expansions are finding a relevant convergence point in the Smart
City concept, the icon of sustainable and livable city, projecting the ubiquitous and pervasive
computing paradigms to urban spaces, focusing on developing city network infrastructures,
optimizing traffic and transportation flows, lowering energy consumption and offering innovative
services. It is through ICT that Smart Cities are truly turning smart [1], in particular exploiting
smart mobile devices in a Mobile Cloud Computing (MCC) context [2]. However, the huge
amount of data generated in a Smart City environment could be overwhelming, due to the rising
and diversified QoS requirements of the city services in relation to the computation time and the
energy consumed by the devices. In order to face the explosion of Big Data to be stored and
elaborated in a Smart City, mobile devices need to be supported by cloud and fog computing
structures [3], allowing an optimized load-sharing in the network for both data storage and
processing features.
For this reason, a new urban framework, named Urban MCC (UMCC), is developed herein.
While in [4]–[6] specific solutions were introduced and analyzed, here the full system view
is provided with requirements and optimization framework. UMCC can be thought of as the
technological nervous system, allowing the networks and information flows of the city to enjoy
a better urban way of life. UMCC is composed by different network and computing elements,
having heterogeneous requirements and capabilities. Within it, the offloading process emerges
as the opportune method for balancing the workload in a twofold way: on one hand, network-
offloading [7] distributes data traffic among the different wireless access technologies within the
1http://esa.un.org/unpd/wup/
2http://www.cisco.com/c/en/us/solutions/service-provider/visual-networking-index-vni/index.html
2
Heterogeneous Network (HetNet) environment. On the other hand, computation-offloading, or
cyberforaging [8], delegates computing functions to the cloud. In this context, a novel unified
offloading mechanism can be envisaged. By means of the UMCC framework, data can be stored
and processed by resource-rich devices using a dynamic cell association for delegating workload,
thus shortening execution time, extending battery life and exploiting the possibility to preserve
data in the cloud. The proposed framework implements a unified offloading mechanism that
allows to optimize the system, by offloading both communication and computing tasks in order
to satisfy the Smart City application requirements.
The unified offloading operation, within the UMCC framework, can be driven by a purposely
defined utility function where throughput, energy efficiency, latency and computing performance
are taken into account. Several works have already analyzed the characteristics of MCC
offloading. In Tab. I the strengths and weaknesses with respect to UMCC of some of the most
important works are summarized.
[TABLE 1 about here.]
The rest of the paper is organized as follows. In the Section II, the main requirements of a
Smart City environment are introduced, by focusing on some specific applications. In Section III,
the proposed UMCC framework is introduced by focusing on the main constitutive entities, while
in Section IV, the offloading mechanism taking advantage of the UMCC framework is discussed.
Finally, in Section V, the conclusions are drawn.
II. REQUIREMENTS OF SMART CITY APPLICATIONS
There are many taxonomies trying to define Smart City key areas, where social aims, care
for environment, and economic issues are related and interconnected. The European Research
Cluster on the Internet of Things (IERC) has identified in [9] a list of applications in different
Internet of Things (IoT) domains, including the Smart City domain. Moreover, the Net!Works
ETP has issued a white paper [10] aiming to identify the major topics of Smart Cities that
will influence the ICT environment. Furthermore, a relevant document aiming to categorize
and define the different applications has been released by the European Telecommunications
Standards Institute (ETSI), where several application types have been specified focusing on their
bandwidth requirements [11].
3
Taking into account all the relevant aforementioned essays, we selected some important Smart
City applications in order to identify their requirements and, then, to leverage the UMCC.
Each application can be defined through the services provided to the citizens, concerning the
requirements in terms of:
Latency: the amount of time required by a certain application between the event happens
and the event is acquired by the system;
Energy Consumption: the energy consumed for executing a certain application locally or
remotely;
Throughput: the amount of bandwidth required by a specific application to be reliably
executed in the Smart City environment;
Computing: the amount of computing processes requested by a certain application;
Exchanged data: the amount of input, output and code information to be transferred by
means of the wireless network;
Storage: the amount of storage space required for storing the sensed data and/or the
processing application;
Users: the number of users for achieving a reliable service.
The Quality of Service (QoS) of a certain application can be seen as a function, where
each requirement plays a role less or more important depending on the application type. In
the following, we list some of the most influential Smart City applications, by highlighting their
technological requirements and characteristics, while, in Tab. II, the considered application types
and the significance of their requirements are summarized.
[TABLE 2 about here.]
a) Mobility: All the components in an intelligent transportation system could be connected
to improve transportation safety, relieve traffic congestion, reduce air pollution and enhance
comfort of driving. The necessary throughput, the computational load and the amount of data
to exchange are high, whereas we can think the storage as a secondary requirement, unless for
security recording.
b) Healthcare: Intelligent and connected medical devices, monitoring physical activity and
providing efficient therapy management by using patients’ personal devices, could be connected
to medical archives and provide information for medical diagnosis. In this case, there are
4
relatively low requirements regarding energy consumption, throughput and number of users,
whereas the requirements in terms of latency, computation, exchanged data and storage are high.
c) Disaster Recovery: In a disaster relief scenario people are facing with the destruction of
the infrastructures and local citizens are asked to use their mobile phones to photograph the site.
In this case there are relatively low requirements regarding throughput, whereas it is important
to have a quick response and to save the energy of the devices.
d) Energy: Energy saving can take advantage from the cloud basically thanks to smart
grid systems, aimed to transform the behavior of individuals and communities towards a more
efficient and greener use of electric power.
e) Waste Management: Automatically generated schedules and optimized routes which take
into account an extensive set of parameters could be planned not only looking at the current
situation, but also considering the future outlook. We can expect non-restrictive requirements of
latency and throughput, whereas resource-poor devices have to be taken into consideration. The
requirements related to data to be exchanged, load of computation, storage and number of users
are not critical.
f) Tourism: Augmented reality and social networks are the characteristics of applications
that more take advantage from the cloud, that becomes also useful for mobile users sharing
photos and video clips, tagging their friends in popular social networks. We can expect
not-restrictive requirements of latency and throughput, whereas resource-poor devices have
to be taken into consideration. There are a great amount of data to be exchanged, load of
computation and storage and number of users are variable.
By comparing the above described applications, it is possible to highlight that a Smart City
scenario is composed of several heterogeneous services with different requirements. However,
it is possible to note that most of them require a high computational complexity and a very
high amount of data to be exchanged in order to be executed. Moreover, it should be noticed,
that in a Smart City scenario multiple services co-exist, increasing even more the system
requirements. This is at the base of the proposed UMCC architecture that, taking benefit from a
joint distributed computing and communication infrastructure, can be implemented through the
use of heterogeneous cloud computing and wireless networks.
5
III. UMCC FRAMEWORK
UMCC sprang from the MCC, that is gaining an increasing interest in the recent years,
due to the possibility of exploiting both cloud computing and mobile devices for enabling a
distributed cloud infrastructure [2]: on one hand, the cloud computing idea has been introduced
as a mean for allowing a remote computation, storage and management of information, and,
on the other hand, the mobility skill allows to gain by the most modern smart devices and
broadband connections for creating a distributed and flexible virtual environment. At the same
time, the recent advances in the wireless technologies are defining a novel pervasive scenario
where several heterogeneous wireless networks interact among them, giving the users the ability
to select the best radio access among those in a certain area. As a consequence, the development
of UMCC is introduced, gaining from both computing and wireless communication technologies.
In the following the three pillars at the basis of the proposed UMCC framework are discussed.
A. Smart Mobile Devices (SMDs)
By analyzing the technology systems underlying a smart city framework, mobile devices can
be considered in a three-fold way:
Sensors: They can acquire different types of data regarding the users and the environment,
transmitting a large amount of information to the cloud in real time, by means of wireless
communication systems.
Nodes: They can form distributed mobile clouds where the neighboring mobile devices are
merged for resource sharing, becoming integral part of the network.
Outputs: They can make the citizens aware of results and able to decide consequently, or
become actuators without need of human intervention.
To perform this triple role, mobile devices have to become part of an infrastructure that is
constituted by different cloud topologies and, at the same time, have to exploit heterogeneous
wireless link technologies, allowing to address the different requirements of a smart city scenario.
This infrastructure starts from the concept of MCC, where the cloud works as a powerful
complement to resource-constrained mobile devices.
B. Cloud Topologies
In relation to the previously described Smart Mobile Device (SMD) roles, we take into account
various cloud topologies. This is a different categorization with respect to the classical as a
6
Service taxonomy used for cloud computing, i.e., Software-as-a-Service (SaaS), Platform-as-a-
Service (PaaS) and Infrastructure-as-a-Service (IaaS). It looks on the different interaction among
the nodes that form the cloud, instead of the services provided by the cloud itself, so we can
distinguish among centralized cloud, cloudlet, distributed mobile cloud and a combination of
them, as shown in Fig. 1:
Centralized Cloud: A centralized cloud provides the citizens to interact remotely, e.g., for
accessing to open data delivered by the public administrations. It refers to the presence
of a remote cloud computing infrastructure having a huge amount of storage space and
computing power, virtually infinite, offering the major advantage of the elasticity of resource
provisioning.
Cloudlet: Cloudlets are fixed small cloud infrastructures installed between the mobile
devices and the centralized cloud, limiting their exploitation to the users in a specific area.
Their introduction allows to decrease the latency of the access to cloud services by reducing
the transfer distance at the cost of using smaller and less powerful cloud devices.
Distributed Mobile Cloud: A third configuration can address the issue of non persistent
connectivity, whereas both the previous concepts must assume a durable state of connection.
In a distributed mobile cloud the neighboring mobile devices are pooled together for resource
sharing [12].
The proposed UMCC framework foresees the joint exploitation of the aforementioned topologies.
[Fig. 1 about here.]
C. Heterogeneous Access Technologies
One of the most actual trend in wireless networks is the presence of a heterogeneous
access platform allowing to several types of devices with multiple network interfaces to select
among them the most suitable. Such a forthcoming scenario, introducing a higher degree of
pervasiveness, allows, especially in a Smart City scenario, to enable the access of a multitude of
different devices, from the high-end broadband user devices to the narrowband M2M devices.
Such network deployment, comprised of a mix of low-power nodes underlying the conventional
homogeneous macrocell network, by deploying additional small cells within the local-area range
and bringing the network closer to users, can significantly boost the overall network capacity
through a better spatial resource reuse. Inspired by the attractive features and potential advantages
7
of HetNets, their development have gained much momentum in the wireless industry and research
communities during the past few years towards the 5G concepts.
D. Towards a unified offloading mechanism
The UMCC approach foresees the definition of a scenario where smart city applications can
exploit jointly the three cloud topologies, as shown in Fig. 2, by distributing and performing
among the different parts composing the framework, and heterogeneous wireless network access
technologies, deployed in the urban area. The application requested by a specific SMD, named
as the Requesting SMD (RSMD), is partitioned and distributed among the different clouds using
the available access networks or computed locally (Fig. 2).
[Fig. 2 about here.]
The main issue is that, for transferring data from the requesting mobile device to the
selected cloud topology, a certain time is required. This mostly depends on some communication
parameters of the selected access network, such as the end-to-end throughput, the amount of
users, the QoS management of a certain transmission technology between the user device and
each type of cloud processing unit. Moreover, the access networks themselves could be already
used by SMDs belonging to the smart city scenario, as well as other devices using the wireless
infrastructures. This involves the necessity of designing a proper offloading method that by
modeling both computing and communication resources as a single unique resource allows to
distribute the computing/communication load in a fair way among the different clouds and access
networks.
Hence, when a RSMD needs to select the cloud infrastructures to be used for computing the
smart city application, two main elements have to be taken into account:
the processing and storage devices - smart mobiles, per se or together forming distributed
mobile clouds, and cloud servers, constituting the cloudlets and the centralized cloud;
the wireless transmission equipment, - different access networks entailing diverse transmis-
sion speeds in relation to their own channel capacity and to the number of linked devices.
In Fig. 2, the UMCC framework is sketched by representing the functional flows of the
architecture. Whenever a smart city application has to be performed, the citizen within the
UMCC can select among different MCC infrastructures, aiming to respect the requirements of
8
the specific application depending on their features. The distribution depends on the application
requirements, and the UMCC features.
Computation, storage, and transmission features: The features of the selected processing
and storage devices, considered per se or in a group forming cloud/cloudlets, are:
Processing Speed: the speed of a device or a group of devices for processing the applications;
Storage Capacity: the amount of storage space provided by a device or a group of devices.
At the same time, the features of the transmission equipment to be taken into account are:
Channel Capacity: The nominal bandwidth of a certain communication technology that can
be accessed by a certain device;
Priority/QoS management: The ability of a certain communication technology to manage
different QoS and/or priority levels;
Communication interfaces: The number of communication interfaces of each device, that
impacts on the possibility of selecting among the available HetNets.
IV. UMCC OFFLOADING MODEL
Let’s focus on one RSMD running an application App, defined through the number of operation
to be executed, O, the amount of data to be exchanged, D, and the amount of data to be stored, S.
An application can be seen as a smart city service, that can be executed either locally or remotely
by exploiting the cloud infrastructures. Furthermore, each application has many requirements
regarding the QoS levels. Among others, the most important are:
the maximum accepted latency TApp, intended as the interval between a task of the
application is requested and its results are acquired,
the minimum level of energy consumption EApp, that the RSMD necessarily uses for
performing the application itself,
the throughput ηApp, intended as the minimum bandwidth that the application needs for
being performed.
The first acting entity in the system is the RSMD, characterized by certain features that are
involved in the offloading operation: the power to compute applications locally, Pl, the power
used for transferring data towards clouds, Ptr, the power for idling during the computation in
the cloud, Pid, the computing speed to perform locally the computation, fl, and its storage
availability, Hl. Furthermore, also the time-varying position of the device plays an important
role in the system interactions.
9
The different types of clouds considered in the Smart City are characterized by their computing
speed to perform the computation, i.e., fcc for the centralized cloud and fcl for cloudlets.
Additionally, while the storage availability of the centralized cloud can be considered infinite,
therefore not constraining in the interaction, the storage availability Hcl of each cloudlet has to
be taken in consideration.
The distributed cloud is a set of SMDs, each characterized by its specific features in the same
way of the RSMD, even if the role played by the SMDs is not a request but a provision of
service. Furthermore, we are considering the system from the point of view of the RSMD. Thus,
the involved features are: connectivity, computation and storage for the data exchange, i.e. the
computing speed fMD, the storage availability HMD, the position posMD(x, y), the throughput
ηMD, the number of devices that can be connected to each SMD nMD, and their coverage range
rMD.
While the connection to the cloudlets can be made only through the unique Access Point
(AP) that can be considered built-in in each cloudlet, and the connection to the SMDs of the
distributed cloud can be made directly, the nodes of the HetNet offer different alternatives to
connect towards the centralized cloud. For each involved node it is possible to define the position
of the node posNod(x, y), the end-to-end throughput in bit per second between the user and the
exploited node ηNod, the number of devices available to connect nNod, and the range of availability
of the node rNod.
Tab. III summarizes the entities and the characteristics above described. They are in a certain
relationship due to some physical and logical bounds that are derived from the following
considerations.
[TABLE 3 about here.]
In order to distribute the computing and communication loads among the different elements,
the system has to evaluate which HetNet nodes, cloudlets, and SMDs are available. On one
hand, there are Mavailable HetNet nodes Nod for the communication offloading towards the
centralized cloud, and Ncloudlets Ccl and KSMDs, able to offer computation offloading
capabilities to the RSMD. On the other hand the system has to distribute, by means of all
these entities, different percentages αiof operations O,βiof data D, and γiof memory S, to
all the available nodes, cloudlets and devices.
The requirements related to the applications, and the associated QoS, can be respected by
10
optimizing the application partitioning and node/cloud association based on the features of the
processing and storage devices and of the transmission devices introduced in Section III-D;
this corresponds to design a unified offloading mechanism, that, by taking into account both
computing and communication resources and their relationships, as listed in Tab. III, as a whole,
can distribute the loads to the different devices of the environment.
In this context a utility function aiming to optimize the application dependent QoS can
be introduced, acting as input for the offloading procedure by selecting the best cloud and
communication infrastructures, as shown in Fig. 3. The model constraints are derived from the
observation that the sum of the offloaded fractions must be equal to 1, thus the optimization
problem becomes:
max
αXiXi {wEf(ERSMD (αXi, βXi)) + wTf(TRSMD (αXi, βXi)) + wηf(ηRSMD (αXi, βXi))}(1a)
s.t. α0+
M
X
i=1
αHNi +
N
X
i=1
αCLi +
K
X
i=1
αMDi = 1 (1b)
M
X
i=1
βHNi +
N
X
i=1
βCLi +
K
X
i=1
βMDi = 1 (1c)
The above equation corresponds to maximize a utility function defined as a weighted sum of
the functions related to the energy consumed, the time spent and the throughput achieved by
the RSMD, with constraints the amount of operations and data to be shared among the different
entities. By doing this the system performs a unified offloading mechanism by considering
jointly the communication and computing resources. In particular, the overall throughput can be
evaluated as the sum of the throughput values ηXi achievable through each node of the scenario.
The throughput ηXi is related to the number of SMDs nXi connected to the ith node and the
channel capacity BWXi of the ith node, and can be expressed by resorting to the Shannon
Formula. With respect to the latency, it can be evaluated as the sum of the local computing, the
data transfer time toward and from the cloud/cloudlets, and the idle time during the offloaded
computation. With respect to the consumed energy, it can be derived from the latency, as the
weighted sum of each latency components by the power consumed in each state.
In Fig. 3, the functional blocks of the UMCC offloading mechanism, based on a utility
function optimization, are represented. On one hand, the smart city applications define specific
requirements, while the cloud topologies in a certain scenario set their features. The utility
function aims at selecting those cloud topologies and access networks that allow to respect the
requirements by setting an optimized distribution of the application itself. The optimization of
11
the partition and the node association will impact again on the UMCC features to be used by
the other applications.
[Fig. 3 about here.]
The maximization of the introduced utility function could be a nontrivial optimization problem,
depending on the considered number of applications and devices acting in the selected scenario.
To this aim different methods to find an optimal, or sub-optimal solution, of the objective function
can be employed.
a) A Greedy approach: If the offloading operation is advantageous with respect to the local
computation, the cell association scheme allows to select the ‘best’ node from the list of those
available; such list can be completed by each SMD that sort each possible access node based on
a self calculated objective function [4]. If the offloading cost is lower than the cost for the local
computation, the SMD will connect to the node which minimizes the cost function, otherwise
it will compute locally the application.
b) A cluster based approach: The idea is to divide the urban area in subareas having
range r; each SMD can share resources only with the other SMDs, cloudlets, and HetNet access
points placed in the same subarea. This approach, even if sub-optimal, can simplify the problem
by reducing the amount of concurrent devices that are involved in the offloading; in [5] a
cluster based optimization model is proposed, where the cluster size plays a significant role for
optimizing the problem while keeping low the complexity.
c) Biased Randomization: A different approach can be obtained by resorting to probabilistic
algorithms based on biased-randomization techniques [6]. In this problem setting, the most
promising node concerning the potential increase in system efficiency has to be selected. The
biased-randomization techniques work by introducing a biased or oriented random effect on the
possible solutions of a problem, allowing to choose the best solution from a set of possible
alternatives that are close to the global optimal. In [6] a biased randomization algorithm is
proposed, allowing to approach the optimal solution by gaining from a heuristic algorithm,
hence keeping the complexity low while approaching to the optimal solution. In [6], it is also
possible to note that such an approach is feasible from the implementation point of view allowing
to have a quasi-optimal solution in a reduced amount of time.
12
V. CONCLUSIONS
In this paper we developed the UMCC framework, a concept that supports the smart city vision
for the optimization of the QoS of various types of smart city applications. By exploiting the
heterogeneous types of applications and devices, typical of a Smart City environment, and from
the heterogeneous computing and communication infrastructure that composes the technological
nervous system of the Smart City, the proposed UMCC framework allows to optimize the
system performance by respecting the application requirements by performing a suitable partial
offloading mechanism. The performance shown in specific applications allows to be optimistic
about the UMCC practical effectiveness.
REFERENCES
[1] M. Dohler, C. Ratti, J. Paraszczak, and G. Falconer, “Smart cities [guest editorial],” IEEE Commun. Mag., vol. 51, no. 6,
pp. 70–71, Jun. 2013.
[2] H. T. Dinh, C. Lee, D. Niyato, and P. Wang, “A survey of mobile cloud computing: architecture, applications, and
approaches,” Wireless Communications and Mobile Computing, vol. 13, no. 18, pp. 1587–1611, Dec. 2013.
[3] F. Bonomi, R. Milito, P. Natarajan, and J. Zhu, “Fog computing: A platform for internet of things and analytics,” in Big
Data and Internet of Things: A Roadmap for Smart Environments, ser. Studies in Computational Intelligence, N. Bessis
and C. Dobre, Eds. Springer International Publishing, 2014, vol. 546, pp. 169–186.
[4] D. Mazza, D. Tarchi, and G. E. Corazza, “A user-satisfaction based offloading technique for smart city applications,” in
Proc. of IEEE Globecom 2014, Austin, TX, USA, Dec 2014.
[5] D. Mazza, D. Tarchi, and G. E. Corazza, “A cluster based computation offloading technique for mobile cloud computing
in smart cities,” in Proc. of IEEE ICC 2016, Kuala Lumpur, Malaysia, May 2016.
[6] D. Mazza, A. Pag`
es-Bernaus, D. Tarchi, A. A. Juan, and G. E. Corazza, “Supporting mobile cloud computing in smart
cities via randomized algorithms,” IEEE Systems Journal, 2016, accepted for publication.
[7] A. Aijaz, H. Aghvami, and M. Amani, “A survey on mobile data offloading: technical and business perspectives,” IEEE
Wireless Commun., vol. 20, no. 2, pp. 104–112, Apr. 2013.
[8] X. Ma, Y. Zhao, L. Zhang, H. Wang, and L. Peng, “When mobile terminals meet the cloud: computation offloading as the
bridge,” IEEE Netw., vol. 27, no. 5, pp. 28–33, Sep./Oct. 2013.
[9] O. Vermesan and P. Friess, Internet of Things - From Research and Innovation to Market Deployment. Aalborg, DK:
River Publishers, 2014.
[10] L. M. Correia and K. Wunstel, “Smart cities application and requirements,” White Paper, Net!Works European Technology
Platform, May 2011.
[11] Electromagnetic compatibility and Radio spectrum Matters (ERM); System Reference document (SRdoc): Spectrum
Requirements for Short Range Device, Metropolitan Mesh Machine Networks (M3N) and Smart Metering (SM) applications,
ETSI Std. TR 103 055, 2011.
[12] D. Huang, T. Xing, and H. Wu, “Mobile cloud computing service models: a user-centric approach,IEEE Network, vol. 27,
no. 5, pp. 6–11, Sep. 2013.
13
[13] Y. Xu and A. Helal, “Scalable cloud-sensor architecture for the internet of things,” IEEE Internet of Things Journal, vol. 3,
no. 3, pp. 285–298, Jun. 2016.
[14] F. H. Bijarbooneh, W. Du, E. C. H. Ngai, X. Fu, and J. Liu, “Cloud-assisted data fusion and sensor selection for internet
of things,” IEEE Internet of Things Journal, vol. 3, no. 3, pp. 257–268, Jun. 2016.
[15] M. Jo, T. Maksymyuk, B. Strykhalyuk, and C. H. Cho, “Device-to-device-based heterogeneous radio access network
architecture for mobile cloud computing,” IEEE Wireless Communications, vol. 22, no. 3, pp. 50–58, Jun. 2015.
Daniela Mazza received the masters degrees in electronic engineering and communication science and the Ph.D. degree in
electronics, telecommunications, and information technologies engineering from the University of Bologna, Bologna, Italy.
She is an expert on system optimization in public administration organizational contexts at Emilia Romagna Local Government,
Italy. Her research interests include the services to the citizens, including solution for distributed multimedia system and services
management in a context of smart city. She possesses more than 15 years of experience in the local government organizations
and has worked as a Project Manager for technological packaging industries since 1991.
Daniele Tarchi (S’99, M’05, SM’12) received the M.Sc. degree in telecommunications engineering and the Ph.D. degree
in informatics and telecommunications engineering from the University of Florence, Florence, Italy, in 2000 and 2004,
respectively. He is currently an Assistant professor at the University of Bologna, Bologna, Italy. His research interests include the
telecommunication area, with particular interests to resource allocation and link adaptation algorithms in wireless and satellite
networks. He has been involved in several national projects as well as European projects and has been active in several industry
funded projects. Dr. Tarchi is an Editorial Board Member for the IEE Communications and for the IEEE Transactions on Vehicular
Technology, and has been an Editorial Board Member for the Wiley Wireless Communication and Mobile Computing, Hindawi
Journal of Engineering, and the Scientific World Journal and has served as an Associate Editor for the IEEE Transactions on
Wireless Communications. He was Symposium Chair at IEEE Wireless Communications and Networking Conference 2011 and
at IEEE Globecom 2014, and Workshop Co-Chair at the IEEE International Conference on Communications 2015.
14
Giovanni Emanuele Corazza (M’92, SM’07) is currently a Full Professor with the Alma Mater Studiorum, University of
Bologna, Bologna, Italy, a Member of the Alma Mater Board of Directors, the Founder of the Marconi Institute for Creativity
(2011), a Member of the Marconi Society Board of Directors, a Member of the Board of the 5G Infrastructure Association, the
Vice-Chairman of the NetWorld2020 European Technology Platform, and the founder of the Mavigex S.r.l. spin-off company. He
was Head of the Department of Electronics, Computer Science and Systems (DEIS) in 20092012, the Chairman of the School
for Telecommunications in 20002003, the Chairman of the Advanced Satellite Mobile Systems Task Force (ASMS TF), the
Founder and Chairman of the Integral Satcom Initiative, a European technology platform devoted to satellite communications.
He has authored or co-authored more than 260 papers. His research interests include wireless and satellite communications,
mobile radio channel characterization, Internet of Things, navigation and positioning, estimation and synchronization, spread
spectrum and multicarrier transmission, and scientific creative thinking. Prof. Corazza served as an Editor for Communication
Theory and Spread Spectrum for the IEEE Transactions on Communications in 19972012. He was the recipient of the Marconi
International Fellowship Young Scientist Award in 1995, the IEEE 2009 Satellite Communications Distinguished Service Award,
the 2013 Newcom# Best Paper Award, the 2002 IEEE Vehicular Technology Society Best System Paper Award, the Best Paper
Award of the IEEE International Symposium on Spread Spectrum Techniques and Application (ISSSTA) 1998, at the IEEE
International Conference on Telecommunication 2001, and at the 2nd International Symposium on. Wireless Communication
Systems 2005. He was the General Chairman of the IEEE ISSSTA 2008, ASMS 20042012 Conferences, and MIC Conference
2013.
15
LIST OF FIGURES
1 Cloud topologies in the UMCC framework: centralized cloud, cloudlet and dis-
tributed mobile cloud. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
2 The process of distributing and performing the application among different parts of
the UMCC. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
3 The utility function acts for distributing and performing the application in different
parts of the Urban MCC. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
FIGURES 16
Fig. 1. Cloud topologies in the UMCC framework: centralized cloud, cloudlet and distributed mobile cloud.
FIGURES 17
Fig. 2. The process of distributing and performing the application among different parts of the UMCC.
FIGURES 18
Fig. 3. The utility function acts for distributing and performing the application in different parts of the Urban MCC.
FIGURES 19
LIST OF TABLES
I State of the Art summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
II Summary of Smart City applications and Requirements . . . . . . . . . . . . . . . 21
III Summary of entities and relations in the UMCC - Involved features and requirements 22
TABLES 20
TABLE I
STATE OF THE ART SUMMARY
Reference Objective Strengths Weaknesses w.r.t UMCC
[13] Cloud-Edge-Beneath (CEB) architecture Scalable ecosystem useful for
Smart City’s massive scale of
devices
Mostly focused on architec-
tural aspects
[14] Cloud Assisted Data Fusion Efficient selection of nodes
with respect to link quality Mostly focused on data-
collection
[15] Device To Device Based architecture Adds D2D communication to
cloud, with increased traffic
capacity
Mostly focused on the global
traffic increase
[12] Mobile as a Representer (MaaR) User-centric characterization
using proactive behaviour Mostly focused on an holistic
perspective
TABLES 21
TABLE II
SUMMARY OF SMART CITY APPLICATIONS AND REQUIREMENTS
Requirements
Application latency energy throughput computing exchanged data storage users
Mobility restrictive variable restrictive high high variable high
Healthcare restrictive non-restrictive non-restrictive high high high low
Disaster Recovery restrictive restrictive non-restrictive high high high variable
Energy non-restrictive non-restrictive non-restrictive high high high high
Waste Management non-restrictive restrictive non-restrictive low low low low
Tourism non-restrictive restrictive non-restrictive high high high variable
TABLES 22
TABLE III
SUMMARY OF ENTITIES AND RELATIONS IN THE UMCC - INVOLVED FEATURES AND REQUIREMENTS
Entity Connectivity Storage Throughput Energy Time latency
App =App(O, D, S, TApp, EApp, ηApp)-S ηApp O,D,EApp O,D,TApp
Dev =Dev(Pl, Ptr, Pid, fl, Hl, posdev(x, y)) posdev(x, y)Hl-Pl,Ptr,Pid,flfl
Ccc =Ccc(fcc)- - - fcc fcc
Ccl =Ccl(fcl, Hcl, poscl(x, y), ηcl, ncl, rcl)poscl(x, y),ncl,rcl Hcl ηcl fcl,ηcl ηcl,fcl
MD =MD(fMD, HMD,posMD(x, y), ηMD, nMD, rMD)posMD(x, y),nMD,rMD HMD ηMD fMD,ηMD ηMD,fMD
Nod =Nod(posNod(x, y), ηNod, nNod, rNod)posNod(x, y),nNod,rNod -ηNod ηNod ηNod
... Differ-ent types of services are evaluated in terms of latency, energy consumption, throughput, computational requirement, volume of data to be transferred, and storage requirement. QoS can be characterized as a function of above parameters [149]. ...
... Cloud computing is facilitating in the domains of smart citizen services [150], mobility, healthcare, disaster management, tourism, energy [151], waste management [149] , education, municipal services, governance, smart buildings and homes [152], security, and transport route planning [148] to deliver quality services to the residents. ...
Article
Full-text available
The concept of smart city evolved with the integration of information and communication technology (ICT) in various sub-systems and processes in urban environment. The development of the smart cities is the best possible solution to major urban issues. It contributes towards economic and social development of the residents. It aims to provide the cordial environment in the domains of healthcare, education, transportation, power generation and dissipation, security, living, industry, etc., to the inhabitants to make their lives comfortable. Sustainability of these services is another major objective in a smart city framework. Along with the true realization of the idea of a smart city, advanced computational and communication technologies are contributing hugely towards its sustainable development. Communication technologies act as backbone to ensure connectivity at the various levels in a smart city framework. Novel smart city solutions for different application domains are designed and deployed by the industry using advanced computational technologies like IoT, Artificial Intelligence, Blockchain, Big Data and Cloud Computing. In this work, authors discuss the concept of smart city, its architecture and sustainability. Different operational domains in a smart city ecosystem are elaborated. The cyber physical aspect of the smart cities is discussed in brief. The role of various computational and communication technologies in the sustainable development of smart cities is presented. Limiting factors in the deployment of various advanced technologies in different smart city domains are highlighted. Security issues associated with the technological sustainable development of different smart city services along with existing solutions are discussed. The article is concluded by highlighting the future research directions.
... Digital technologies, in particular those that refer to the notion of smart and sustainable city (D'Amico et al., 2020b) as sensors (Arkian et al., 2017;D'Amico et al., 2020b), real-time monitoring stations, (Fernández-Ares et al., 2017;Goncalves et al., 2020;Tekouabou et al., 2020), actuators (Fatimah et al., 2020b), digital cameras (Kurilkin et al., 2015;Liu and Xu, 2020), real-time tracking systems (Uhlemann et al., 2017;Rathore et al., 2017;Pioppi et al., 2020), big data analysis systems (Al Nuaimi et al., 2015;Bibri, 2019), artificial intelligence , cloud computing (Mazza et al., 2017;Baucas and Spachos, 2020), smart grids (Daissaoui et al., 2020;Kalluri et al., 2020), Internet-of-Things (IoT) infrastructure (Rathore et al., 2017;Wirtz et al., 2018;D'Amico et al., 2021b), augmented reality (Piao et al., 2020), blockchain (Marsal-Llacuna, 2018, 5G wireless communication (Gozalvez, 2017), social media platforms and other information and communication technologies (ICTs) (Bibri et al., 2017) have been described as the driving force of the digitalisation of the circularity of urban metabolism. ...
Article
Digitalisation of urban metabolism circularity provides policymakers, urban managers, planners and administrators with a useful tool for identifying, controlling and evaluating a wide range of data concerning the flows of social, environmental and economic resources. This approach is based on the crucial role of fixed and mobile digital infrastructures such as real-time monitoring stations, GPS tracking sensors, augmented reality, virtual sharing platforms, social media dashboards, smart grids, and the like in the development and strengthening of the quality and efficiency of the circularity of resources. For these reasons, the integration of digital technologies in mobility, waste, water and wastewater management, energy efficiency, safety, and so on, represents a crucial aspect for cities involved in the circularity of their urban metabolism. Through a systematic literature review and case study approaches, the analysis disclose a wide-range of initiatives adopted by several European circular cities that optimise the circularity of urban metabolic flows, and contributes to the efforts in increasing understanding and awareness of the digitalisation driven by the urban metabolism circularity.
... The order required to achieve the same performance is less than that of the FIR filter. However, the FIR filter is simple to calculate without feedback, and Able to achieve linear phase [12,13]. The frequency of the ECG signal is 0~100 Hz, and 90% of the energy is concentrated within 35 Hz, so a FIR digital low-pass filter is designed here with a sampling rate of 200 Hz and a 50 Hz notch function. ...
Article
Full-text available
The application background of physiological signal research focuses on education and research in the fields of life sciences, clinical diagnosis and monitoring applications, and is an important content in the field of biomedical engineering. The development and use of intelligent physiological signal acquisition systems have also become the current life science research One of the hot spots. The acquisition and processing of physiological and medical signals is an important technology in biomedical engineering, which provides a necessary benchmark for the research of sports physiology and medicine. This article aims to provide some ideas and directions for the research on collecting physiological signal data of athletes under the cloud computing environment and the application of Internet of Things technology.and mainly introduces the design of the intelligent acquisition system of physiological signal data of athletes under cloud computing. This paper proposes an intelligent acquisition method for athletes’ physiological signal data under cloud computing, including the design of power frequency filter and the energy model of the TDMA protocol system, which is used to conduct research experiments on the design of an athlete’s physiological signal data intelligent acquisition system under cloud computing. The experimental results of this article show that the average collection accuracy rate of the system is 97.01%, the test of the intelligent acquisition system shows that the stability of the system is very high, and the accuracy rate is high, which can better collect the physiological signal data of athletes.
... When the output error exceeds the allowable range and the number of repetitions is less than the upper limit, the error will be returned; otherwise, the error will be returned. When the exit error is within the allowable range or the number of repetitions is equal to the upper limit, insert the next sample and start training a new sample [12]. ...
Article
Full-text available
With the continuous development of network technology and the continuous expansion of network scale, the security of the network has suffered more threats, and the attacks facing them have become more and more extensive. The frequent occurrence of network security incidents has caused huge losses. Facing an increasingly severe situation, it is necessary to adopt various network security technologies to solve the problem. Intrusion detection technology can detect internal and external network attacks, respond before the intrusion occurs, and send out alarm information for timely and effective processing. This article mainly introduces the research of cloud computing intrusion detection technology based on BP neural network (BP-NN), and intends to provide ideas and directions for the development of cloud computing intrusion detection technology based on BP-NN. This paper proposes research methods of cloud computing intrusion detection technology based on BP-NN, including BP-NN algorithm, neural network cloud computing intrusion detection technology and artificial bee colony optimization algorithm, which are used to conduct cloud computing intrusion detection technology experiment based on BP-NN; Proposed an artificial bee colony optimization neural network algorithm; designed a cloud computing intrusion detection system based on BP-NN. Experimental result shows that the average detection rate of the ABC-BP network algorithm is 92.67 %, which can effectively distinguish normal data from abnormal data.
... Several scholars highlighted the fundamental role of the integration of advanced urban technologies, such as ICTs [43], sensors [11], IoT [22], big data analytics [102], cloud computing [103], artificial intelligence (AI) [25], augmented reality (AR) [104], realtime monitoring stations [105], digital cameras [106], actuators [107], real-time tracking systems [108], smart grids [109], Blockchain [110], 5G wireless communication [111], and social media platforms [112] with the physical infrastructures that characterize the urban context, such as roads, highways, airports, ports, bridges, rivers, lakes, buildings, public transport stops, vehicles, parks, and so on, to improve the efficiency and digitization of the urban metabolism. Hence, in light of the challenges related to the growing datafication and urbanization, the urban metabolism narrative has shifted from a linear approach (source-consumption-waste), characterized by the theory of the abundance of resources always available to models of smart urban metabolism [3], where real-time and digital monitoring, conservation, sharing, and reformulation of materials and products already in circulation is guaranteed and maximized to reduce the extraction of raw materials and the accumulation of waste. ...
Article
Full-text available
Digital technologies engaged in urban metabolism for efficiency provide policymakers, urban managers, and planners with useful instruments to collect, monitor, analyze, and evaluate the circularity of environmental, social, and economic resources to improve their effectiveness and quality. At present, the digital technology-based approach is strategic for circular cities engaged in the development of smart and sustainable actions in the fields of mobility, energy, environment, waste, telecommunications, and security. Through the ‘Circular Resource Efficiency Management Framework’ developed by the European Commission, this paper generates insights into the digitalization practices of the circularity of urban metabolism by analyzing the initiatives implemented by the municipalities of Kaunas, Flanders region, Porto, Prato, The Hague, and Oslo, which constitute the Partnership on Circular Economy (PCE) of the Urban Agenda of the European Union. The results of the analysis provide a wide range of practices such as real-time monitoring stations for water and energy consumption, digital cameras for controlling vehicle flows, web platforms for sharing goods and services, and tracking sensors for public transport, which aim to optimize the efficiency of the circularity of urban metabolic flows. This study increases the understanding and awareness of digital technologies in this paradigm shift.
... Mazza et al. introduced a mobile cloud computing model to describe the data flow and operations that occurred in smart cities and intelligent parks. In the meantime, they proposed a unified offloading mechanism in which communication and computing resources were jointly managed, allowing load balancing between different entities in the environment and delegating communication and computing tasks simultaneously, thereby satisfying the application requirements of smart cities and parks [15]. Hossain et al. discovered that although traditional cloud computing methods could provide the largest computing and storage facilities to support data processing, the latency was high. ...
Article
Full-text available
Today, intelligence in all walks of life is developing at an unexpectedly fast speed. The complexity of the Internet of Things (IoT) big data system of intelligent parks is analyzed to unify the information transmission of various industries, such as smart transportation, smart library, and smart medicine, thereby diminishing information islands. The traditional IoT systems are analyzed; on this basis, a relay node is added to the transmission path of the data information, and an intelligent park IoT big data system is constructed based on relay cooperation with a total of three hops. Finally, the IoT big data system is simulated and tested to verify its complexity. Results of energy efficiency analysis suggest that when the power dividing factor is 0.5, 0.1, and 0.9, the energy efficiency of the IoT big data system first increases and then decreases as α0 increases, where the maximum value appears when α0 is about 7 J. Results of outage probability analysis demonstrate that the system’s simulation result is basically the same as that of the theoretical result. Under the same environment, the more hop paths the system has, the more the number of relays is; moreover, the larger the fading index m, the better the system performance, and the lower the outage possibility. Results of transmission accuracy analysis reveal that the IoT big data system can provide a result that is the closest to the actual result when the successful data transmission probability is 100%, and the parameter λ values are between 0.01 and 0.05; in the meantime, the delay of successful data transmission is reduced gradually. In summary, the wireless relay cooperation transmission technology can reduce the outage probability and data transmission delay probability of the IoT big data system in the intelligent park by adding the multihop path, thereby improving the system performance. The above results can provide an experimental basis for exploring the complexity of IoT systems in intelligent parks. 1. Introduction Technologies such as the Internet of Things (IoT), cloud computing, big data, and mobile IoT have gradually been employed in various industries as science and technology advance forward. The economic level of the entire human society has been greatly improved. While comprehensively promoting the integration and innovation of a new generation of information and communication technology and urban development, the IoT composed of massive sensors is continuously collecting variously structured and unstructured data day and night. The amount of data in various industries, including surveillance video data, geographic information, traffic data, population data, and security and environmental monitoring data, is undergoing explosive growth. Under the trend of intelligent development in various industries, the phenomenon of “information islands” gets increasingly prominent, protruding the importance of parks [1, 2]. Therefore, how to interconnect and share information in such parks has become the research focus of scholars and experts worldwide. An intelligent park is the vital link of smart city development; its architecture and development models are the epitomai of a smart city [3]. Compared with cities, intelligent parks often have smaller spatial granularity and more specific functions and development goals. Thus, it is easier for them to implement smart city designs and constructions, playing a demonstrative and leading role in smart city construction [4]. Like smart cities, the management and service methods of intelligent parks should be based on the demands of people and meet the needs of the public, society, and government as much as possible. Intelligent parks are the outcomes of the in-depth development of the “Internet plus Technology,” which will have a huge and far-reaching impact on the planning and management of the parks, public services, production methods, people’s livelihood, and market operations [5]. Today, when the Fifth-Generation (5G) communication technology is about to become popular, the situation where “everyone is connected” and “everything is connected” has been formed. Moreover, IoT has become the core symbol of smart cities and intelligent parks in the “Internet +” era as it gradually penetrates various industries. According to statistics, in 2020, the total number of IoT connections worldwide approached 30 billion, and the market size reached 1.7 trillion US dollars; meanwhile, nearly 55% of IoT achievements were concentrated in business fields, such as smart manufacturing, smart home, smart cities, and intelligent parks [6, 7]. Only adopting IoT cannot contribute to intelligent park construction. Instead, this process also requires big data analysis and cloud computing technologies. The big data analysis technology uses its various data to serve the parks and provides big data platforms and tools for enterprises in the parks [8, 9]. Cloud computing can gather massive amounts of data on the cloud servers, analyze and sort out the parks’ data, and utilize these data for optimizing the parks’ management and operation. Cloud computing manifests as a data center in constructing intelligent parks, which is the core of the information system of smart cities and intelligent parks [10]. Therefore, it is vital to collect and process data and information while constructing intelligent parks. In summary, the research purposes are to further develop the intelligent parks and increase the degree of information transmission and sharing. Innovatively, the wireless relay cooperation transmission technology is added to the traditional IoT systems, and an IoT big data system for intelligent parks is constructed based on relay cooperation. Besides, the complexity of the system’s performance is verified through simulation, which can provide a reference basis for the development of intelligent parks in the future. 2. Related work 2.1. The Application Trend of IoT in Intelligent Parks Today, as communication technology advances at an unexpectedly fast speed, IoT technology has been gradually accepted in all walks of life. Scholars worldwide have done a lot of works on IoT technology. Rathore et al. proposed an IoT-based big data analysis next-generation super park planning system. They proposed a complete system, including various intelligent systems based on IoT, such as smart homes, Internet of Vehicles, weather and water systems, smart parking lots, and monitoring objects, for data generation. The final simulation results confirmed that the proposed super park planning system could provide higher efficiency and scalability [11]. Bresciani et al. modeled the data of multiple IoT smart city project alliances and found that an enterprise’s development was closely connected to the rapid development of urbanization. They also emphasized that knowledge management capabilities indirectly improved the flexibility of alliances through the information and communication technology capabilities of enterprises. They recommended that managers of multinational enterprises design knowledge management tools and develop new information and communication technology skills [12]. Qian et al. applied IoT technology to promote the construction of infrastructure in intelligent parks so that the economy could grow sustainably and people’s livelihood could be significantly improved [13]. Watson et al. analyzed and estimated the big data-driven decision-making process in the knowledge-based city economy by comprehensively analyzing the existing achievements and basis of IoT intelligent parks. Consequently, they found that despite the fact that IoT had made a great contribution to intelligent parks, the development needs of intelligent parks could not be met yet [14]. 2.2. The Application Status of Cloud Computing in Intelligent Parks Cloud computing acts as the data center in smart cities and intelligent parks, which is the core of their information systems. Hence, many scholars have researched cloud computing technology. Mazza et al. introduced a mobile cloud computing model to describe the data flow and operations that occurred in smart cities and intelligent parks. In the meantime, they proposed a unified offloading mechanism in which communication and computing resources were jointly managed, allowing load balancing between different entities in the environment and delegating communication and computing tasks simultaneously, thereby satisfying the application requirements of smart cities and parks [15]. Hossain et al. discovered that although traditional cloud computing methods could provide the largest computing and storage facilities to support data processing, the latency was high. Thus, they introduced edge computing into the construction of smart cities and intelligent parks. Finally, they found that processing raw IoT data on edge devices was effective in terms of latency and provided context awareness for smart city decision-makers in a seamless manner [16]. Giannakoulias researched the data security issues in the cloud computing environment and introduced some of the most important security threats in cloud computing, as well as key recommendations on how to deal with these threats, namely, security standards and certifications, service provider auditing, security API, transport layer protection, identity verification and encryption key management, and cloud service agreement [17]. Javadzadeh and Rahmani believed that the technology used to implement smart cities was usually based on cloud computing; however, this technology was accompanied by unreliable delays, lack of mobility support, and location awareness. To further develop another path, they applied fog computing to smart cities and parks to explore their research trends and development directions [18]. 2.3. The Application Status of Big Data in Intelligent Parks The rise of big data provides powerful and efficient solutions for IoT and various fields; moreover, its applications are very broad. People’s lives are undergoing tremendous changes under the influence of communication, network, and computer technology. After reviewing the development trend of industrial communication and IoT, Wollschlaeger et al. applied the 5G telecommunication network to IoT, which greatly improved the efficiency and data processing rate of the IoT systems [19]. Gai et al. proposed a dynamic privacy protection model under the premise that data transmission had great security risks. They also developed an Android application to evaluate the effectiveness of the model and, finally, protected the security of data privacy [20]. Zhang et al. put forward a new network paradigm called CIoT, including the generation of big data perception, efficient computing, and storage at the edge of CIoT, which could be further integrated into deep learning and data analysis to improve the operating efficiency of the system [21]. With the rapid advancement of big data technology, Wang et al. (2021) established a data management system to link international academic research and city-level management policies. Furthermore, they increased the number of uses based on the practicality of the collected data to enhance smart cities and the development efficiency of the parks [22]. In summary, although various new technologies such as cloud computing, IoT, big data, and artificial intelligence are widely accepted in various industries, they do not have information interoperability, resulting in information islands. Therefore, adding a wireless relay cooperation transmission system to the traditional IoT system can provide a basis for the information transmission of intelligent parks, which is of great value to the development of the economy and society. 3. Big Data System of IoT in Smart Park 3.1. Status Quo of Intelligent Park Construction Regarding the actual situation at the current stage, the construction of smart cities and intelligent parks worldwide is in the exploratory and experimental stage. The development of smart cities is a solid foundation for the construction of intelligent parks, and the construction and development of smart cities drive the construction of intelligent parks. According to statistics, there are more than 100 smart cities worldwide, most of which are located in Asia and Europe [23]. In China, the construction of parks shows obvious clustering characteristics, and the current development pattern presents a pattern of “the eastern coastal cluster, the central riverside linkage, and the western characteristic development.” According to statistics, as of the end of 2019, more than 15,000 parks of various types have been built across the country, and GDP accounts for about 30% of the overall economy. Industrial parks have become an important force in promoting high-quality economic development in China [24]. The development trend of the intelligent park in China is shown in Figure 1 [22].
... The upstream fused data toward the cloud data centres need to be fairly distributed among the edge servers [5,6]. In the existing literature [7][8][9], the fused data streams are offloaded to the nearest edge servers. However, this approach is not efficient as some of these servers may overload quickly in comparison to others that remain underutilized. ...
Article
In the densely populated Internet of Things (IoT) applications, sensing range of the nodes might overlap frequently. In these applications, the nodes gather highly correlated and redundant data in their vicinity. Processing these data depletes the energy of nodes and their upstream transmission towards remote datacentres, in the fog infrastructure, may result in an unbalanced load at the network gateways and edge servers. Due to heterogeneity of edge servers, few of them might be overwhelmed while others may remain less-utilized. As a result, time-critical and delay-sensitive applications may experience excessive delays, packet loss, and degradation in their Quality of Service (QoS). To ensure QoS of IoT applications, in this paper, we eliminate correlation in the gathered data via a lightweight data fusion approach. The buffer of each node is partitioned into strata that broadcast only non-correlated data to edge servers via the network gateways. Furthermore, we propose a dynamic service migration technique to reconfigure the load across various edge servers. We assume this as an optimization problem and use two meta-heuristic algorithms, along with a migration approach, to maintain an optimal Gateway-Edge configuration in the network. These algorithms monitor the load at each server, and once it surpasses a threshold value (which is dynamically computed with a simple machine learning method), an exhaustive search is performed for an optimal and balanced periodic reconfiguration. The experimental results of our approach justify its efficiency for large-scale and densely populated IoT applications.
... Some use cases of Mobile Cloud Computing found in the literature are mobile gaming, mobile commerce, mobile learning [30], mobile social media [87], multimedia sharing, and location-based services [117]. Mobility, disaster recovery, waste management, tourism, and healthcare by combining the user physical activity with medical archives would enable smart and connected medical devices to present information for medical diagnosis [75]. ...
Article
Full-text available
The edge computing (EC) paradigm brings computation and storage to the edge of the network where data is both consumed and produced. This variation is necessary to cope with the increasing amount of network-connected devices and data transmitted, that the launch of the new 5G networks will expand. The aim is to avoid the high latency and traffic bottlenecks associated with the use of Cloud Computing in networks where several devices both access and generate high volumes of data. EC also improves network support for mobility, security, and privacy. This paper provides a discussion around EC and summarized the definition and fundamental properties of the EC architectures proposed in the literature (Multi-access Edge Computing, Fog Computing, Cloudlet Computing, and Mobile Cloud Computing). Subsequently, this paper examines significant use cases for each EC architecture and debates some promising future research directions.
Chapter
Full-text available
Medical imaging diagnosis is the most assisted method to help physicians diagnose patient diseases using different imaging test modalities. In fact, Deep learning aims to simulate human cognitive functions. It providing a paradigm shift in the field of medical imaging, due to the expanding availability of medical imaging data and to the advancing deep learning techniques. In effect, deep learning algorithms have become the approach of choice for medical imaging, from image acquisition to image retrieval, from segmentation to disease prediction. In our paper, we present a review that focuses on exploring the application of deep learning in medical imaging from different perspectives.
Chapter
Smart City concept is an emerging vision of cities with wide connectivity, systems inter-operability, sustainable solutions and prevalent security. Wireless Sensor Networks (WSNs) have shown themselves to be very useful in the process of the realization of this paradigm. Data collection is the first step on which the whole system is based. The new services and innovative applications offered by the Smart City are based on the extracted information from the data collected by sensors which are widely deployed and integrated in many smart devices such as smart phones and connected cars. In this paper, we introduce a WSNs information life cycle in Smart Cities encompassing six phases, namely; deployment, management, collect and transfer, use and share, storage, and analyses and plan. We specify for each phase the technologies involved. Then, we discuss the challenges of integration.
Conference Paper
Full-text available
The evolutionary trends of the Information Society have lead to the definition of the so-called Smart City, an environment characterized by the interaction of several heterogeneous technologies aiming at providing unprecedented services and efficiency. The Smart City is one of the most challenging scenarios for the Internet of Things (IoT) applicability. Among several things composing a Smart City, the computing infrastructures are responsible for giving a distributed elaboration intelligence to the environment. This paper takes into consideration the presence of the Mobile Cloud Computing (MCC), where different types of Cloud Computing Infrastructures (CCIs) are available through different wireless connections. MCC can be exploited by Smart Mobile Devices (SMDs) for offloading applications towards powerful remote servers and other SMDs pooled together, shortening execution time and extending battery life. Aim of this paper is to propose a cluster based computation offloading technique able to work in a distributed environment, in order to better satisfy the Quality of Service (QoS) requirements of the SMD users, by minimizing a cost function taking into account a trade-off between SMD’s energy consumption and execution time
Article
Full-text available
Smart cities represent rich and dynamic environments in which a multitude of smart mobile devices (SMDs) interact among them by sharing data. SMDs require from fast access to online services, but they offer limited computing capabilities and battery lifetime. SMDs make frequent use of computation offloading, delegating computing-intensive tasks to the cloud instead of performing them locally. In such a large-scale and dynamic environment, there might be thousands of SMDs simultaneously executing processes and, therefore, competing for the allotment of remote resources. This arises the need for a smart allocation of these resources. Accordingly, this paper proposes a biased-randomized algorithm to support efficient and fast link selection. This algorithm is able to provide “real-time” near-optimal solutions that outperform solutions obtained through existing greedy heuristics. Furthermore, it overcomes the responsiveness limitations of exact optimization methods.
Article
Full-text available
The Internet of Things (IoT) is connecting people and smart devices on a scale that was once unimaginable. One major challenge for the IoT is to handle vast amount of sensing data generated from the smart devices that are resource-limited and subject to missing data due to link or node failures. By exploring cloud computing with the IoT, we present a cloud-based solution that takes into account the link quality and spatio-temporal correlation of data to minimize energy consumption by selecting sensors for sampling and relaying data. We propose a multiphase adaptive sensing algorithm with belief propagation (BP) protocol (ASBP), which can provide high data quality and reduce energy consumption by turning on only a small number of nodes in the network. We formulate the sensor selection problem and solve it using both constraint programming (CP) and greedy search. We then use our message passing algorithm (BP) for performing inference to reconstruct the missing sensing data. ASBP is evaluated based on the data collected from real sensors. The results show that while maintaining a satisfactory level of data quality and prediction accuracy, ASBP can provide load balancing among sensors successfully and preserves 80% more energy compared with the case where all sensor nodes are actively involved.
Article
Full-text available
The emerging heterogeneous mobile network architecture is designed for an increasing amount of traffic, quality requirements, and new mobile cloud computing demands. This article proposes a hierarchical cloud computing architecture to enhance performance by adding a mobile dynamic cloud formed by powerful mobile devices to a traditional general static cloud. A mobile dynamic cloud is based on heterogeneous wireless architecture where device-to-device communication is used for data transmission between user devices. The main advantage of the proposed architecture is an increase in overall capacity of a mobile network through improved channel utilization and traffic offloading from Long Term Evolution-Advanced to device-to-device communication links. Simulations show that the proposed architecture increases the capacity of a mobile network by up to 10 percent depending on the conditions and amount of offloaded data. The offloading probability is also evaluated by taking into consideration the number of devices in the cloudlet and the content matching values. We have gained insight into how content similarity affects offloading probability much more than the number of devices in a cloudlet.
Conference Paper
Full-text available
The Smart cities applications are gaining an increas-ing interest among administrations, citizens and technologists for their suitability in managing the everyday life. One of the major challenges is regarding the possibility of managing in an efficient way the presence of multiple applications in a Wireless Heterogeneous Network (HetNet) environment, alongside the presence of a Mobile Cloud Computing (MCC) infrastructure. In this context we propose a utility function model derived from the economic world aiming to measure the Quality of Service (QoS), in order to choose the best access point in a HetNet to offload part of an application on the MCC, aiming to save energy for the Smart Mobile Devices (SMDs) and to reduce computational time. We distinguish three different types of application, considering different offloading percentage of computation and analyzing how the cell association algorithm allows energy saving and shortens computation time. The results show that when the network is overloaded, the proposed utility function allows to respect the target values by achieving higher throughput values, and reducing the energy consumption and the computational time.
Chapter
Full-text available
Internet of Things (IoT) brings more than an explosive proliferation of endpoints. It is disruptive in several ways. In this chapter we examine those disrup- tions, and propose a hierarchical distributed architecture that extends from the edge of the network to the core nicknamed Fog Computing. In particular, we pay attention to a new dimension that IoT adds to Big Data and Analytics: a massively distributed number of sources at the edge.
Article
Full-text available
It is through information and communications technologies (ICT) that smart cities are truly turning "smart." This is facilitated by means of services that use, among others, networked sensors and actuators deployed in the city, allowing the monitoring of the urban environment in real time, to react just in time if needed and to establish automated control processes with less or even without human intervention.
Article
Full-text available
Over the last few years, data traffic over cellular networks has seen an exponential rise, primarily due to the explosion of smartphones, tablets, and laptops. This increase in data traffic on cellular networks has caused an immediate need for offloading traffic for optimum performance of both voice and data services. As a result, different innovative solutions have emerged to manage data traffic. Some of the key technologies include Wi-Fi, femtocells, and IP flow mobility. The growth of data traffic is also creating challenges for the backhaul of cellular networks; therefore, solutions such as core network offloading and media optimization are also gaining popularity. This article aims to provide a survey of mobile data offloading technologies including insights from the business perspective as well.
Article
Recent advances in the Internet of Things (IoT) and pervasive and ubiquitous computing provide a glimpse into the future of our planet and reveal exciting visions of many smart things: smart cities, smart homes, smart cars, and other smart spaces such as malls, workplaces, hotels, schools, and much more. Driven by a technological revolution offering 'low-power many things and wireless almost everything,' we could, in only a decade, envision and prototype impressive smart space systems that improve quality of life, enhance awareness of resources and the environment, and enrich users' experience. However, prototyping is one thing and actual large-scale deployments are another. The massive scale of sensors and devices that will be deployed in smart cities of the future will be challenging. Without an ecosystem and a scalable architecture in place, it will be extremely difficult to manage or program such an expanding and massive IoT. In this paper, we introduce the Cloud-Edge-Beneath (CEB) architecture and present its salient scalability features. We also present a validation study based on an event-driven programming model demonstrating CEB's scaling behavior in face of IoT expansion and under dynamically increasing loads.