IEEE Electrification Magazine - June 2018 - 28

a significant role in improving the performance of city operations. in particular, big data analytics manage to translate
the collected raw data into actionable intelligence that facilitates real-time decision making in smart city operations.
accordingly, big data analytics support city authorities' goal
of boosting the efficiency, economics, reliability, resilience,
and sustainability of smart city operations.
Cloud computing allows data to be retrieved and processed in real time, which offers a convenient way to perform big data analytics. in essence, cloud computing
represents a group of networked computers and servers
that are easily accessible over the internet. Cloud computing thus provides pathways for dealing with large volumes
of data in cost-effective manners. existing iot-driven
cloud computing platforms include amazon web services,
General electric predix, Google Cloudplatform, azure iot
suite, and salesforce iot Cloud.
however, several limitations block the widespread
deployment of cloud computing applications. in addition
to concerns of bandwidth limitations and processing
delays, one major challenge is tied to heterogeneous data
sources included in iot technologies. For example, some
iot devices generate analog data with proprietary timing
and structural characteristics, which necessitate additional protocol translations before any data transmission to a
cloud. More importantly, privacy leakage is another common concern when applying cloud computing together
with iot technologies. For example, smart electricity
meters may reveal citizens' sensitive personal information
(e.g., domestic energy usage) that prefers local data processing rather than cloud-based services.
edge computing (or fog computing), acting as an extension of cloud computing to the utmost edge, is a promising solution for analyzing localized data with maximized
resolution and minimized latency. hence, it is considered
as an alternative to keep data storage, processing, and
analysis in a more localized manner at edges of the cloud.
similar to cloud computing, edge computing utilizes
semantic intelligence (natural language processing and
machine learning) and computational intelligence (i.e.,
advanced mathematics). but different from cloud computing, edge computing enables more efficient data analysis
by employing local computing resources.
the combination of cloud computing and edge computing merges the merits of both technologies for city
authorities to make data-driven decisions in a rapid and
proactive manner. therefore, city authorities should utilize
flexible data flow management mechanisms to customize
the use of iot technologies for specific applications, with a
tradeoff between the reliability and efficiency of city operations and the privacy and security of citizens.

Hierarchical Control and Management Framework for
the Implementation of Smart City Solutions
the hierarchical control and management framework is
proposed for merging the merits of both technology-based

I EEE E l e c t r i f i c a t i on M a gaz ine / j un e 2018

centralized (i.e., top-down) approaches and human-based
distributed (i.e., bottom-up) approaches for making urban
infrastructures smart and secure. in Figure 19, the proposed framework is divided into three functional hierarchies with distinct requirements on operation timescales
and communication bandwidth.
xx
Field device level. sensors and actuators are networked
with iot technologies, which promote more efficient
integrations of emerging context-aware sensors and
devices, ultimately improving the situational awareness in city operations. sensors embedded into the collection of smart city infrastructures collect the
monitoring and measuring information in real time,
and report it to the smart city control center via a wired
or wireless telecommunication infrastructure.
xx
Area control level. local controllers realize data capture,
processing, storage, and analytics at distributed points
across the city, which reduces architectural complexity and boosts scalability. with the utilization of edge
computing technologies, local controllers increase
their responsiveness to real-time and context-critical
information.
xx
Control center level. the control center helps discover
and maintain a registry of data sources and their characteristics, e.g., periodicity, liveliness, and quality, and
make them available for subsequent analyses. application-specific analytics based on cloud computing
can also be deployed in the control center for facilitating pattern mining and predictive analytics.
the proposed framework is expected to ensure data
integrity, localization, and confidentiality, while enhancing availability, flexibility, and elasticity to meet targeted
performance requirements (e.g., transmission rate and
delay). on the one hand, the framework features a hybrid
communication strategy that collects data from dispersed
field devices through short-range wireless technologies
and delivers real-time data to the smart city control center (i.e., cloud-based) via long-range backhauling communication technologies. on the other hand, the proposed
framework strikes a balance between immediacy and
depth of insight for data processing, which analyzes most
time-sensitive data at the local level and sends selected
data to the smart city control center for more extensive
processing and longer-term storage. hence, the proposed
framework acts as a unified platform for orchestrating all
parts of the city to work together, which ensures the collection of smart city infrastructures so that they can work
in harmony rather than running as separate entities.
since the proposed framework promotes synergies
between iCts and ots, it is expected to not only boost
efficiency, reliability, and resilience of smart city operations but also enhance the quality of life, well-being, and
safety of citizens.
when the framework is friendly to prevalent applications of iCts like iot technologies, it is also compatible
with emerging innovations in iCts like software-defined

Table of Contents for the Digital Edition of IEEE Electrification Magazine - June 2018