RESEARCH OpenAccess Networkingarchitecturesandprotocols ...

Journal of Internet Servicesand Applications

Jawhar et al Journal of Internet Services and Applications (2018) 926 httpsdoiorg101186s13174-018-0097-0

RESEARCH Open Access

Networking architectures and protocolsfor smart city systemsImad Jawhar1 Nader Mohamed2 and Jameela Al-Jaroodi3

Abstract

The smart city model is used by many organizations for large cities around the world to significantly enhance andimprove the quality of life of the inhabitants improve the utilization of city resources and reduce operational costsThis model includes various heterogeneous technologies such as Cyber-Physical Systems (CPS) Internet of Things(IoT) Wireless Sensor Networks (WSNs) Cloud Computing and Unmanned Aerial Vehicles (UAVs) However in order toreach these important objectives efficient networking and communication protocols are needed to provide thenecessary coordination and control of the various system components In this paper we identify the networkingcharacteristics and requirements of smart city applications and identify the networking protocols that can be used tosupport the various data traffic flows that are needed between the different components In addition we provideillustrations of networking architectures of selected smart city systems which include smart grid smart home energymanagement smart water UAV and commercial aircraft safety and pipeline monitoring and control systems

Keywords Smart city Cyber-physical systems (CPS) Networking architectures Unmanned aerial vehicle (UAV)Wireless sensor networks (WSNs)

1 IntroductionA number of large cities around the world are inves-tigating applying the smart city model to heighten theliving quality of their inhabitants and enhance the uti-lization of the city infrastructure and resource s Var-ious advanced technologies and techniques supportingsuch models provide smart services to improve the per-formance and operations in healthcare transportationenergy education and many other fields At the sametime these services reduce operational costs and resourceconsumption in smart cities Examples of these technolo-gies are Wireless Sensor Networks (WSNs) the Internetof Things (IoT) Cyber-Physical Systems (CPS) roboticsUnmanned Aerial Vehicles (UAVs) fog computing cloudcomputing and big data analytics Utilizing these tech-nologies provides many advantages and services for smartcities WSNs are used to provide real-time monitoringof the conditions of smart city resources and infrastruc-tures [1] The IoT facilitates the integration of the physicalobjects in a city network [2] CPS are used to provide

Correspondence ihjawhargmailcom imadjawharmuedulb1Al Maaref University Old Airport Avenue PO Box 25-5078 Beirut LebanonFull list of author information is available at the end of the article

useful interactions between the cyber world and the phys-ical world in smart cities [3] Robotics and UAVs are usedto provide automation and offer useful services for smartcities [4] Such services include enhancement delivery ofservices environmental monitoring traffic monitoringsecurity and safety controls and telecommunication ser-vices [5] Fog computing is used to provide low latencysupport location awareness better mobility support andstreaming and real-time support for smart city applica-tions [6] Cloud computing provides a scalable and costeffective computation and data storage platform to sup-port smart city applications [7] Big data analytics is usedto provide intelligent and optimized short and long termdecisions based on collected data to enhance smart cityservices [8]These advanced technologies are used to implement

a number of smart city services [9ndash11] Examples ofthese smart services are intelligent transportation ser-vices that can be used to enhance route planning andcongestion avoidance in city streets provide intelligenttraffic light controls and parking services enhance vehicu-lar safety and enable self-driving cars Other examples aresmart energy services that provide better energy decisionsfor more efficient energy consumption in smart cities

copy The Author(s) 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 40International License (httpcreativecommonsorglicensesby40) which permits unrestricted use distribution andreproduction in any medium provided you give appropriate credit to the original author(s) and the source provide a link to theCreative Commons license and indicate if changes were made

Jawhar et al Journal of Internet Services and Applications (2018) 926 Page 2 of 16

Applications of these smart energy services are used tosupport smart grids and smart buildings as well as pro-vide better utilization of renewable energy Other smartservices involve structural health monitoring as well asreal-time monitoring of water networks bridges tun-nels train and subway rails and oil and gas pipelinesAdditional services include smart services for environ-mental monitoring and smart services for public safetyand securityThese smart city services do not only need the various

advanced technologies discussed here but also need reli-able and robust networking and communication infras-tructures to enable efficient exchange of messages amongthe different components of the systems that provide aparticular service Smart city services are designed atdifferent scales which require various networking andcommunication technologies for their implementationand operations Furthermore different network and com-munication models and approaches can be utilized forsmart city services This paper investigates the commu-nication and network issues of smart city systems It alsoinvestigates networking technologies architectures andcommunication requirements for such systems The suit-ability of existing network protocols for different smartcity services will be discussed Although there are sig-nificant research efforts to investigate different issues insmart cities and provide solutions for these issues verylittle research has been done to investigate the network-ing and communication parts of smart city systems whichconstitute the main objective of this paperThe rest of the paper is organized as follows Section 2

provides an overview of related work in this fieldSection 3 includes an overview of some smart city appli-cations Section 4 presents networking architectures andcommunication requirements for smart city applicationsSection 5 offers an illustration of selected smart citysystems Section 6 discusses open issues in the area ofnetworking and communication for smart city systemsFinally Section 7 concludes the paper and provides somefuture research directions

2 Related workThere are a few papers published addressing the networkand communication issues for smart cities In this sectionwe discuss some of the work presented in these papersZanella et al [2] provided an interesting study that ana-lyzed the current technologies available for deploying IoTfor a smart city In addition the authors presented anddiscussed a proof-of-concept implementation of IoT inthe city of Padova Italy There are generally two com-mon approaches to offer data access to things in IoT Thefirst is using multi-hop mesh networks with short-rangecommunication in the unlicensed spectrum among thenetwork nodes The second is using long-range cellular

technologies in the licensed frequency band A new com-munication technology is introduced to provide alter-native connectivity for IoT named LPWAN (Low-PowerWide Area Network) [12] From its name this com-munication technology can provide low-rate long-rangetransmission in the unlicensed frequency bands using startopology These features can be very useful for some smartcity applications Leccese et al [13] introduced a smartcity application of fully controlled street lighting using aZigBee Sensor Network and WiMAX while the applica-tion is controlled by Raspberry-Pi Card Sanchez et al [14]introduced a smart city testbed for IoT experimentationnamed SmartSantanderSome work more related to our contribution in this

paper is investigating network architectures for smartcities Wan et al [15] presented an event-based communi-cation architecture that helpsmanage and facilitates coop-eration among Machine to machine (M2M) componentsin smart cities The authors also conducted a case studyusing this architecture for vehicular applications Gauret al [16] proposed a multi-level smart city architecturebased on semantic web technologies This architectureis mainly for wireless sensor networks applications in asmart city It consists of 4 levels Data Collection DataProcessing Integration and Reasoning and Device Con-trol and Alerts Jin et al [17] studied distinct IoT networkarchitectures with focus on Quality of Service (QoS) fordifferent smart city applicationsThe last paper [17] is the closest to our contribu-

tions in the paper thus we will elaborate on it Theauthors presented five different IoT network architec-tures (1) Autonomous Network Architecture (2) Ubiqui-tous Network Architecture (3) Application-Layer OverlayNetwork Architecture (4) Service-Oriented Architectureand (5) Participatory Sensing The autonomous networkarchitecture is usually not directly linked to public net-works such as the Internet It can be connected thoughgateways if this link is needed This network architectureis suitable for some smart city applications such as auto-matic parking management in which most of the networkconnections are mainly to support the application TheQoS requirements in this network architecture are mainlydependent on the requirements of the applicationIn the ubiquitous network architecture smart objects

including the sensors and actuators are part of the Inter-net Information from these smart objects can be obtainedby authorized users and applications through the Internetdirectly or through intermediate servers which serve assinks that gather data from the connected smart objectsThe serversrsquo option can provide better scalability andresource efficiency than the direct access option Thesmart objects can be connected to the Internet throughmultitier and multiradio This network architecture issuitable for smart city applications such as structure


health monitoring in which multiple software applica-tions can collect and analyze different structuresrsquo healthin a city The QoS of such network architecture can bechallenging due to the level of heterogenous networkcomponents used The application layer overlay networkarchitecture is suitable for large scale networks with alarge number of distributed nodes These nodes can belogically structured in clusters with cluster heads thatcan run in-network data processing task to reduce net-work traffics In this network architecture ad hoc andormesh are used One application of this network archi-tecture is using a wireless sensor network (WSN) forenvironmental monitoring The QoS for applications suit-able for such architecture is generally tolerable to somelevel The service-oriented network architecture is basedon an innovative network architecture called Informa-tion Driven Architecture (IDRA) [18] In this architecturedifferent network functions such as addressing namingforwarding and routing are provided as network servicesThese network services can be utilized to provide differ-ent network configurations to suit different applicationsAlthough this approach can be very useful flexible andcan provide advanced network features it requires newnetwork components technology The participatory sens-ing network architecture is considered a special case anda new model of IoT In this model residents through theirconsumer devices collect analyze and share sensor dataThis can be called ldquohuman-as-a-sensorrdquo In this modewireless communications such as WiFi GPRS and 3G areused Some possible applications of this architecture areenvironmental monitoring intelligent transportation andhealthcare QoS in such network can be challenging ashumans are the main source of data and humans can belazy privacy-worried error-prone and misbehavingUnlike other related work our contributions in this

paper is mainly in investigating networking architec-tures focusing on the communication characteristics andrequirements of the main smart city applications includ-ing smart buildings smart grids gas and oil pipelinemonitoring and control smart water network intelli-gent transportation manufacturing control and monitor-ing and unmanned aerial vehicle applications for smartcities The works we studied usually focus on a sin-gle attribute or characteristic such as quality of servicein [17] We considered several communication charac-teristics and requirements including bandwidth delaytolerance power consumption reliability security het-erogeneous network support network type and mobilitysupport In addition we studied the suitability of differentnetworking protocols for different smart city applicationsThese protocols are IEEE 802154 (Zigbee) IEEE 802151(Bluetooth) IEEE 80211a IEEE 80211b IEEE 80211gIEEE 80211n IEEE 80216 (WiMAX) Cellular 3G Cel-lular 4GLTE and Satellite With this we are providing

a comprehensive study in networking architectures andprotocols for smart city systems

3 Smart city applicationsDevelopment and operation of smart city applicationscan face many challenges To identify and understandthese challenges we discuss some important smart cityapplications used or proposed for different domains Wehighlight their benefits as well as their development andoperational challenges This will help us identify the typeof support needed by the networking platforms designedfor smart city applicationsIn the energy domain smart city applications are used

to add values such as efficiency reliability and sustainabil-ity of the production and distribution of electric powerin smart grids [19] A smart grid is a renovated electri-cal grid system that uses information and communicationtechnology (ICT) to collect and act on available informa-tion about the behavior of suppliers and consumers in anautomated fashion A smart grid uses CPS to provide self-monitoring and advanced control mechanisms for powerproductions and consumer needs to increase grid effi-ciency and reliability In addition CPS systems are usedto control the processes of generating renewable energyfrom hydropower plants [20] and wind power plants [21]Furthermore some applications are used to monitor andcontrol energy consumptions in smart buildings [3] Thebuildingsrsquo equipment such as HVAC (Heating Ventilatingand Air-Conditioning) systems appliances and lightingsystems are controlled with CPS Smart building sys-tems are usually equipped with different types of sensornodes that monitor the current energy usage and envi-ronmental conditions These sensors report their obser-vations and measurements to a centralized monitoringand control system The control system implements intel-ligent algorithms to control the sub-systems used in thebuildings to optimize energy usage based on the sensedobservations and current operational and environmentalconditionsIn the transportation domain an important smart city

application area that recently received high attention isintelligent transportation Vehicular safety applicationsconstitute one of the most important classes of suchapplications There are many safety applications for vehi-cles including lane change warning messages emergencybreaking collision avoidance mechanisms and blind spotmonitoring These applications provide fully automaticor semi-automatic actions to enhance driving safety Themost important features of such applications are the real-time and reliability support in detection and responseAll aspects of vehicular safety applications includingthreat observations decision making communicationand actions must be reliable and able to run in real-timeThis imposes a serious restriction on how the software


is designed and how well it supports high levels ofintegration across all the devices involved to ensure real-time and reliable responses In addition self-driving carsare considered as important smart city applications [22]Since they practically integrate all the mentioned featuresin addition to vision and monitoring components to allowthe vehicle to navigate the roads based on sensed dataand intelligent software that interprets and responds tothis data in real-time Another intelligent transportationapplication include intelligent traffic light controls whichinclude monitoring devices across multiple locations toaccurately predict traffic patterns and adjust traffic lightsto optimize flow One example of such domain is dis-cussed in [23]In addition smart city systems can be used to protect

water networks and to make them smarter more efficientmore reliable and more sustainable CPS systems can beembedded within water networks to provide some moni-toring and control mechanisms and to add smart featuresto the operations of water distribution [24] One of thesefunctions is to provide early warning mechanisms to iden-tify problems in water networks For examples leaks andpipe bursts can be easily detected while fast and tempo-rary solutions can be applied to reduce water waste and tominimize further risks or damages to the networkOther smart city applications include greenhouse mon-

itoring that aims to provide efficient control for suitableclimate soil lighting and water level in greenhouses [25]In addition some applications involve autonomous oper-ation of unmanned vehicles using CPS systems Such sys-tems provide networks that connect the payloads on theunmanned vehicles like sensors actuators cameras stor-age communication devices and microcontrollers [26]Additional smart city systems are also used to auto-mate control monitor and enhance manufacturing pro-cesses [27] Finally monitoring and controlling oil andgas pipelines is another one of the applications for smartcities We discuss the corresponding architecture andfeatures of this and other important applications in thesection illustrating selected smart city systems later in thispaper

4 Smart city networking architectures andcommunication requirements

In this section we investigate the different networkingand communication requirements of the various smartcity applications as well as the protocols that can beused to connect the components used to support suchapplications

41 Networking characteristics requirements andchallenges of smart city systems

Table 1 describes the various smart city applications alongwith the appropriate networking protocols that can be

used the bandwidth requirements the delay tolerancepower consumption level reliability and security require-ments heterogeneity of the networking links whetherthey use wired communication wireless communicationor both and the mobility characteristics for each of theseapplications

411 Network protocolsAs shown in the table applications with short range com-munication such as smart buildings and smart waternetworks can use protocols from the personal area net-work (PAN) class such as IEEE 802154 (Zigbee) and801151 (Bluetooth) These protocols are generally char-acterized by lower bandwidth low energy consumptionand short range Applications requiring longer rangessuch as intelligent transportation and manufacturing andcontrol use protocols which are in the local area network(LAN) class such as IEEE 80211 (WiFi) Applicationsrequiring wide range communication such as UAVs andsmart grid can use protocols that are in the wide areanetwork (WAN) class such as IEEE 80216 (WiMAX) cel-lular and satellite All of these protocols have provisionsto support asynchronous and synchronous data connec-tions The former can be used with smart city applicationswith best effort traffic which can tolerate delay whilethe latter can be used with applications that generatetraffic requiring more stringent quality of service (QoS)requirements such as larger bandwidth and limited delaySuch applications involve real-time and multimedia com-munication In addition these protocols have reliabilityand security services However most of the security fea-tures require more processing and can cause added delayand energy consumption Consequently these considera-tions should be taken into account before enabling suchfeatures

412 BandwidthAlso the table shows that certain applications such asintelligent transportation have low bandwidth require-ments Others such as smart buildings gas and oilpipeline monitoring and UAVs require more bandwidthHowever even inside the same type of applications thebandwidth requirements can range from low to mediumor even high depending on the type of data that is gen-erated For example telemetric and control data such asUAV ground-to-air control commands only require smallbandwidth while UAVs taking images and videos andtransmitting them to ground base units require consider-ably larger bandwidth

413 Delay toleranceIn addition it is shown that some applications havelow tolerance for end-to-end delay Such applicationsinclude intelligent transportation This is the case since


Table 1 Networking characteristics and requirements of smart city applications

Smart cityapplication

AppropriateNet Prot

Band- width Delaytolerance

PowerConsump

Reliabil- ity Security HetNet

Wired wireless

Mobility

Smart buildings [3] IEEE 802154IEEE 802151

L M H L L M M H H H WDWL M

Smart grid [19] IEEE 80216 Cellular

L M L H M H H H M H WDWL L

Gas and oil pipelinemonitoring andcontrol [22]

IEEE 80216 Cellular

L M H L H L M H H M WL L

Smart waternetworks [24]

IEEE 802154 IEEE 80211IEEE 80216

L M L H L M H M WL L

Intelligenttransportation [54]

IEEE 80216IEEE 80211IEEE 802154Cellular

L M L H L M M H H H WDWL H

Manufacturingcontrol andmonitoring [27]

IEEE 802154 IEEE802151 IEEE 80211

L M L H L M M M M WDWL M

Unmanned aerialvehicle [5]

IEEE 80211 IEEE 80216Satellite

L M H L H L M H M H M WDWL H

IEEE 802151 Bluetooth IEEE 802154 Zigbee IEEE 80111 WiFi IEEE 80216 WiMAX H High M Medium L Low WD Wired WL Wireless

the data that is being transmitted needs to arrivewithin microseconds in order to allow the control sys-tems to react within an acceptable time frame to avoidcar imminent danger or life-threatening collisions Onthe other hand other smart city applications havehigher tolerance for delay These applications includeones that rely on the collection of information andmonitoring data for later analysis Examples of suchapplications include UAVs taking images for laterprocessing

414 Power consumptionPower consumption is also an important requirement forsmart city applications However as shown in the tablesome applications that have local high energy sourcessuch as smart grid systems can tolerate protocols withhigher power consumption levels Other applicationswhich have energy sources with limited capacities havemedium power requirements Such applications includeintelligent transportation Other applications have verylimited energy sources and require protocols with low orvery low energy consumption characteristics Such appli-cations include gas and oil pipeline monitoring smartwater networks and UAVs

415 ReliabilityReliability is another important parameter in smart cityapplications and the table shows that most applicationseither have medium reliability requirements such as smart

water networks while others have high reliability require-ments such as smart grid and intelligent transportation

416 SecurityWith respect to security most applications requiremedium to high security For example applications suchasmanufacturing control andmonitoring requiremediumsecurity while others such as smart grid have high secu-rity requirements due to the sensitivity of the data andcriticality of the functions that are performed

417 Heterogeneity of network protocolsMost smart city systems include networking protocolswhich connect the various components within the systemExamples of such systems include include smart buildingsand intelligent transportation In such cases these pro-tocols must be able to co-exist without interfering witheach other In addition appropriate mapping of the vari-ous control information inside the headers at the variouslayers of the networking stack of the different heteroge-neous protocols and networks must be done to ensureseamless and efficient operation

418 Wiredwireless connectivityThe table also shows that some smart city applica-tions such as gas and oil pipeline monitoring and UAVsmostly involve wireless communication Others such assmart buildings and intelligent transportation involveboth wired as well as wireless communication In such


cases communication within a particular physical systemcan use wired networking (eg inside a UAV) while wire-less communication can be used to connect the physicalsystem with other similar physical systems or backboneand infrastructure networks

419 MobilityFinally mobility is another important characteristic ofsmart city applications The table shows that some sys-tems have low or mediummobility such as smart grid gasand oil pipeline monitoring and smart water networksOther systems have highmobility such as intelligent trans-portation and UAVs Consequently the networking pro-tocols that are used to connect medium to high mobilitysmart city systems must be robust and adapt well tonode mobility without consuming too much bandwidthon control messages and related processing to readjust tochanges in the network topology

42 Additional issues and challengesIn addition to the requirements and characterization ofthe links between nodes in smart city systems we identifythe following additional issues and challenges whichmustbe considered

421 InteroperabilitySmart city systems rely on various heterogeneous net-working protocols at the physical and data link layerswhich use different medium access control (MAC) strate-gies Interoperability between these protocols is importantin order to provide seamless integration of the underly-ing technologies The IEEE 19051 protocol which wasdesigned to provide convergent interface between physi-caldata link layers and the network layer is intended toplay such a role for digital home networks [28] Develop-ment of similar protocols to expand the support mech-anism for smart city systems is a good area for futureresearch

422 AvailabilitySoftware and hardware availability are essential compo-nents of smart city systems due to the criticality and real-time nature of a lot of the related applications Softwareavailability can be achieved by ensuring that the variousservices are available to the corresponding applicationsOn the other hand hardware availability is obtained byensuring that the various devices that are needed to pro-vide networking contestability and efficient performanceare readily available anytime and anywhere One way toaccomplish these objectives is through redundancy ofboth software and hardware components and systemsThis was already considered and studied for IoT devices[29 30] Furthermore considerations to attain availabilityneed to be incorporated as a part of the design objectives

of networking and communication protocols for smartcity systems

423 PerformancePerformance is always an important consideration for anytype of architecture and this is also the case for smartcity systems In order to achieve this essential objectivemore evaluation needs to be done for the various net-working protocols at the various layers of the architectureespecially at the data link network and transport layersThese three layers are critical components in order to sup-port traffic of various QoS requirements In addition themiddleware layer can be used to provide proper interfaceand convergence services between these layers and theapplication layer

424 ManagementAnother important aspect of smart city system net-working is management of the thousands or even mil-lions of devices that are involved in many applicationsFor example to achieve energy management in smartbuildings thousands of sensor and actor devices can bedeployed in each building Efficient protocols are neededto provide effective management of fault configurationaccounting performance and security (FCAPS) aspectsof these devices The Light-weight machine-to-machine(LWM2M) [31] standard is being done by the OpenMobile Alliance to specify the interface between M2Mdevices and servers On the other hand the NETCONFLight protocol [32] is designed by IETF to provide mech-anisms to install manipulate and delete the configurationof network devices Similar efforts are encouraged forsmart city systems in order to offer standard mechanismsand services to efficientlymanage and control the commu-nication of devices at the various levels of the architecture

425 ScalabilityIt is important for smart city systems to be able to accom-modate new devices without appreciable loss in the qual-ity of the provided services and associated network trafficflows This can be accomplished through virtualizationand extensibility in the platforms and their operations Anextensible IoT architecture was proposed in [33] whichconsists of three layers Virtual object Composite virtualobject and Service layer The design must have objec-tives of automation intelligence and zero-configurationfor objects and related devices in order to achieve scal-ability and interoperability More research is desirable inorder to extend this strategy to smart city systems

426 Big data analyticsHuge amounts of data is collected by smart city systemsand the corresponding IoT devices that are spread outover a considerably larger geographic area Analyzing andextracting useful information from this data can provide


considerable advantages for businesses and governmentinstitutions In addition communication and collectionof a very large number of messages in a timely fashionaccording to their priority delay-tolerance and size is vitalto the efficient operation of smart city systems In order toreduce the amount of exchanged traffic local processingcompression and aggregation of the generated messagesneed to be done at the lower and intermediate levels of thenode hierarchy and geographic areas Consequently Moreresearch is needed to provide proper convergence andmapping of the networking parameters between the vari-ous layers of the networking stack at the data-generatingnodes (eg sensors IoT devices etc) the intermediaterouters processing servers (typically in the cloud) andactor nodes at the other end of the communication cycle

427 Cloud computingCloud computing is an important component of any smartcity as it can provide scalable processing power and datastorage for different smart city applications [34] Cloudcomputing has powerful processing capabilities large andscalable data storage and advanced software services thatcan be utilized to build different support services to provi-sion diverse smart city applications Cloud computing canbe used as the main control and management platformused to execute smart city applications Different sensorsand actuators of smart city applications can be connectedto the cityrsquos cloud computing services to collect pro-cess store the sensorsrsquo data and perform managementtasks for different smart city applications As the col-lected data from a smart city can also become big dataas huge amounts of data are collected throughout thecity Cloud computing can provide the necessary power-ful platforms for storing and processing this big data toenhance operations and planningThe communication between city sensors and actua-

tors and cloud computing can involve different commu-nication requirements to smoothly support smart cityapplications These requirements should be supportedby the network architectures deployed in the smart citySmart applications rely on the integration between sen-sors and actuators on one side and the cloud on theother and cannot performed well unless there is a goodnetwork that provides good communication services con-necting both sides Another issue that arises when usingcloud computing for a smart city is that the cloud ser-vices are either offered at a centralized location or acrossmultiple distributed platform in various locations Thedistributed cloud computing approach can provide betterquality and reliability support for different cloud appli-cations [35] However there is usually a need to providegood communication links among the distributed cloudcomputing facilities available in different places Anotherissue arising when using the cloud is the reliability and

performance of the networks connecting all componentson both sidesWith the Internet in themix there are prob-lems with delays lost packets and unstable connectionsCareful planning and management of network resourcesand communication models in addition to the design andarchitecture of the smart city application is necessary toaccount for these issues Yet there are some unavoidableaspects such as the transmission delays

428 Fog computingWhile cloud computing can provide many advanced andbeneficial services for smart city applications it can-not provide good provisions for distributed applicationsthat need real-time mobility low-latency data streamingsynchronization coordination and interaction supportservices This is mainly due to the transmission delaysimposed by the large distances to be covered between thesmart city censors and devices and the cloud platforms Inaddition it is difficult for cloud computing to manage anddeal with a large number of heterogenous sensors actua-tors and other devices distributed over a large-area Fogcomputing was lately introduced to offer more localizedlow latency and mobility services Fog computing allowsto move some functionalities from the cloud closer to thedevices [36] This approach aims to enable different IoTapplications through distributed fog nodes that providelocalized services to support these IoT applications In asmart city fog computing can complement cloud comput-ing to support smart city applications [37] While cloudcomputing can provide powerful and scalable services forsmart city applications fog computing can provide morelocalized fast-response mobility and data streaming ser-vices for smart city applications Furthermore integratingIoT fog computing and cloud computing as shown inFig 1 can provide a powerful platform to support differentsmart city applications Figure 2 shows a hierarchical rep-resentation where the IoT devices use amultihop topologyto reach the gateway connecting to the fog server Thisintegrated platform needs good networking and commu-nication support to efficiently handle the communicationbetween all these components This also includes a goodnetwork security support to avoid any threat vulnerabil-ity issues in the integration and in supporting smart cityapplications

43 Links between nodes in smart city systemsTable 2 describes the various networking protocols thatcan be used in smart city systems [38ndash40] The table showstheir main characteristics the physical and data link layerspecifications their data rates and transmission rangeWe can see that the applications requiring short range

such as smart buildings smart grid and smart water gen-erally can use the IEEE 802154 (Zigbee) protocol whichis a very short range protocol that is mainly designed


Fig 1 An Illustration of the integration of IoT fog computing and cloud computing to support smart city applications

for very small devices that have very limited energy It isintended to allow these devices to last up to several yearsusing the same battery In addition the IEEE 802151(Bluetooth) protocol can be used by such applicationsIt is a WPAN protocol which uses the 24 GHz bandIt employs a masterslave time division duplex (TDD)strategy with a 1 Mbps data rate and a range of 10to 100 mThe IEEE 80211abgn protocol can be used with

almost all smart city systems The IEEE 80211n proto-col which is the later version works in the 24 GHz and51 GHz ranges uses direct sequence spread spectrum(DSSS) and orthogonal frequency division multiplexing(OFDM) It employs a carrier sense multiple access withcollision avoidance (CDMACA) MAC strategy It allows

best effort operation using the distributed coordinationfunction (DCF) as well as reservation-based operationusing the point coordination function (PCF) The latterservice is useful for multimedia audio video and real-time data traffic which require QoS guarantees of certainparameters such as bandwidth delay and delay jitter Itsupports data rates from 15 to 150 Mbps and has acommunication range up to 25 mThe cellular 3G and 4G protocols can be used with

applications such as smart grid smart water UAVs andpipeline monitoring They use packet switching for datacommunication and optional packet or circuit switchingfor voice communication They use frequencies in the 800MHz to 1900 MHz 700 MHz and 2500 MHz rangesThey also use code division multiple access (CDMA) and

Fig 2 An hierarchical representation showing the integration of IoT fog computing and cloud computing to support smart city applications


Table 2 The various networking protocols that are useful for smart city applications

Protocol Maincharacteristics

Physical layer specs Data link layerspecs

Data rate Transmission range Smart cityapplication

IEEE 802154(Zigbee)

Energy saving veryshort range

24 GHz Band DSSS CSMACA 20 Kbps to 250Kbps

10 to 20 m Smart BuildingsSmart Grid SmartWater

IEEE 802151(Bluetooth)

Cable replacement 24 GHz BandFHSSFSK

MasterSlave TDD 1 Mbps 10 to 100 m Smart BuildingsSmart Grid SmartWater

IEEE 80211a Data networkinglocal area network

5 GHz Band OFDM CSMACADCFPCF

6 9 12 18 24 3648 54 Mbps

120 m outdoors All

IEEE 80211b Data networkinglocal area network

24 GHz Band DSSS CSMACADCFPCF

1 2 55 11 Mbps 140 m outdoors All

IEEE 80211g Data networkinglocal area network

24 GHz BandDSSS OFDM

CSMACA DFSPFS 6 9 12 18 24 3648 54 Mbps

140 m outdoors All

IEEE 80211n Data networkinglocal area network

24 GHz and 5 GHzBand DSSS OFDM


250 m outdoors All

IEEE 80216(WiMAX)

Metropolitan areanetwork

2 to 66 GHz BandOFDMA

TDD FDD 2 to 75 Mbps Up to 35 miles (56Km)

Smart Grid SmartWater UAVsPipelineMonitoring

Cellular 3G Wide area networkconnectivityDigital packetswitched for data

800 MHz to 1900MHz

CDMA HSDPA 144 Kbps (mobile)to 42 Mbps(stationary)

Depends on cellradius (1 Km toseveral Kmrsquos)


Cellular 4GLTE Same as 3G 700 MHz to 2500MHz

LTE and LTEAdvanced

300 Mbps to 1Gbps



Satellite Wide area network 153 GHz to 31 GHz FDMA and TDMA 10 Mbps (upload)and 1 Gbps(download)

Satellie can cover100rsquos of Kmrsquos toentire earth

UAVs PipelineMonitoringIntelligentTransportation

high-speed downlink packet access (HSDPA) as well aslong term evolution (LTE) advanced technology The datarates that are supported are 300 Mbps to 1 Gbps Thegeographic area that is covered is the entire city or coun-try without roaming and it has world-wide coverage ifroaming is usedSatellite communication can also be used with appli-

cations such as UAVs pipeline monitoring and intel-ligent transportation They typically use frequenciesin the range of 153 GHz to 31 GHz They alsoemploy frequency division multiple access (FDMA) andtime division multiple access (TDMA) at the datalink layer The data rate is between 10 Mbps (down-load) and 1 Gbps (upload) Geographically satellitecommunication covers the entire earth since handoffbetween satellites can be used to achieve such continuouscoverage

5 Illustration of selected smart city systemsIn this section five selected smart city systems are brieflypresented in order to illustrate some possible networkingand communication models that are used

51 Smart grid systemFigure 3 shows the general architecture of a smart grid sys-tem which is one of the essential applications in a smartcity As shown in the figure smart grid systems are dividedinto three categories (1) generation (2) transportationand (3) consumer In turn the consumer systems areseparated into three sub-categories (1) commercial (2)residential and (3) industrial Each of these sites usu-ally contains sensing and acting devices that are deployedto monitor and control the different mechanisms andmachines that are located on the premises These devicesform nodes in a mobile ad hoc network (MANET) or awireless sensor and actor network (WSAN) The nodescan communicate using multihop networking protocolsspecifically designed for MANETs and WSANs [41 42]Usually one (or more) of the nodes plays the role of a gate-way and it provides connectivity to the network at thatsite with the infrastructure LAN or the Internet Cloudcomputing platforms can also be used to provide storageanalysis processing and decision making services to thesmart grid network system [43] In addition the controlcenter and various users can collect information and issue

Jawhar et al Journal of Internet Services and Applications (2018) 926 0 of 16

Fig 3 The general architecture for a smart grid system used in a smart city

requests and commands to provide real-time control ofthe corresponding systems

52 Smart home energy managementIn a typical smart city the electric company will havedifferent rates for different time periods Typically theseperiods would be three Peak Mid-Peak and OFF-PeakAlso most homes would be equipped with environment-friendly local energy generation sources such as wind-mills solar panels or photovoltaic cells (PVC) In Fig 4 ageneral architecture of a smart home energy managementsystem is shown In this model when a specific service isrequested from a particular appliance (eg wash the laun-dry run the dishwasher use a robot to clean the pool etc)the energy management unit (EMU) is used to decidewhich energy source is used to supply the requested powerand the time to turn ON the corresponding electric appli-ance The user enters the requested task (or service) to becarried out by a particular appliance along with a dead-line (or an amount of acceptable delay) by which the taskneeds to be accomplished This allows the EMU to calcu-late the amount of maximum delay that can be toleratedfor the performance of the task Then it performs an algo-rithm which determines the source of the energy and thetime for executing the desired task by the indicated appli-ance The algorithm consists of the following logic If theamount of energy that is needed by the task is availablein the locally generatedstored energy then it runs theappliance immediately using the local energy storage as asource Otherwise it tries to shift the time of running the

appliance to the OFF-Peak period with the electric com-pany as a source using the maximum tolerable delay thatis calculated based on the user input If the delay does notallow such shifting it tries to shift the task executing totheMid-peak period Otherwise if the shifting is not pos-sible it executes the task during the current time periodThis type of energy management system provides consid-erable environmental benefits It also lowers the cost ofenergy for both the user as well as the electric company

53 Smart water systemsFigure 5 shows the general architecture of a smart watersystem which is another important smart city applicationThe system is used to monitor and control the irrigationof the soil with various types of crops using an optimizedprocess In general sensing devices are placed in selectedareas in the farm in order to monitor different parame-ters such as temperature and moisture of the soil Actordevices are used to control different activities such as thetime and amount of water that is provided by the sys-tem Both sensor and actor devices constitute nodes ina WSAN The nodes can have various topologies includ-ing mesh and star configurations In either case one ofthe nodes acts as a gateway to provide connectivity to thesink which in turn connects the WSANs to the backboneLAN or the Internet Cloud computing platforms canalso be used to provide storage analysis processing anddecision making services The figure also shows that dif-ferent databases can be used to provide plant and weatherrelated information to the system Such information is


Fig 4 The architecture of a home-based energy management system used in a smart city

typically combined with the collected sensing data tomake optimized decisions related to the time locationand amount of water that is released by the system Thesystem is also capable of issuing alert notifications to theusers whenever it is appropriate In addition the net-work is connected to the control center which oversees

the operations It also issues different configuration andcommand requests to supervise and manage the network

54 UAV and Commercial Aircraft Safety CommunicationDue to the several challenges of the new aeronauticalapplications including emerging UAV applications for

Fig 5 The general architecture for a smart water system used in a smart city


smart cities there are high needs to define new com-munication solutions which can effectively support thesenew applications The National Aeronautics and SpaceAdministration (NASA) in the United States and theEuropean Organization for the Safety of Air Navigation(EUROCONTROL) are leading the development of newcommunication systems A standard for UAS Controland Non-Payload Communication (CNPC) links is beingdeveloped in the United States to enable safe integrationof UAS operations within the National Airspace System(NAS) [44] NAS is the main aviation system in the UnitedStates that involves the US airspace airports and moni-toring and control equipment and services that implementthe enforced rules regulations policies and proceduresThis system covers the airspace of the United States andlarge portions of the oceans Some of its components arealso shared with the military air force For safe integrationof UAS operations in NAS sense and avoid techniquesaircraft-human interface air traffic management policiesand procedures certification requirements and CNPCare being studied [45] This integration allows UAVs tofunction within the airspace used for manned aircraftsused for carrying passengers and cargoCNPC links are defined to provide communication

connections to be used for aircraft safety applicationsand to enable remote pilots and other ground stationsto control and monitor the UAVs Figure 6 show anillustration of required communication links betweenthe UAVs commercial airlines and the control centerThis involves several issues including communicationarchitecture types as well as rate bandwidth frequencyspectrum allocation security and reliability require-ments Two communication architecture types wereproposed including line-of-sight (LOS) communication

which provides communication with unmanned air-crafts through ground-based communication stations andbeyond-line-of-sight (BLOS) communication which pro-vides communication with unmanned aircrafts throughsatellites The communication rates requirements forboth uplink (ground-to-air) and downlink (air-to-ground)were defined based on the size of unmanned aircraftsas shown in Table 3 The uplink rates are much lowerthan downlink rates as the uplink communication willbe mainly used to send small messages to control theunmanned aircrafts while the downlink communicationis used for different types of communication includ-ing video transmission The uplink rate was determinedto support transmission of 20 individual control mes-sages per second This rate is required to provide acomplete real-time ground control for a UAV using ajoystick [44]The supported density requirements of unmanned air-

crafts that use CNPC to the year 2030 are also specified[45] and shown in Table 4 The third column shows thenumbers of unmanned aircrafts (of different sizes) thatcan be supported if a terrestrial-based communicationlink of radius 100 Km is usedBased on the defined UAV density the required band-

width for CNPC links is 90 MHz divided into 34 MHzfor the terrestrial-based LOS CNPC links and 56 MHzfor the BLOS CNPC links [44] Two frequency spectrumranges were assigned by the 2012 International Telecom-munications Union World Radio Communications Con-ference (WRC-12) to be used by CNPC to provide reliableand real-time data transmissions These frequency spec-trums are from 960 MHz to 1164 MHz (L-Band) andfrom 5030 MHz to 5091 MHz However a portion of thefirst range will be shared with other legacy applications

Fig 6 UAV and commercial aircraft safety communication


Table 3 CNPC data communication rates

Aircraft size Uplink rate Downlink usages Downlink rate

Small(lt=55 kg)

2424 bps Basic services only 4008 bps

Mediumand Large(gt55 kg)



6925 bps Basic and weatherradar

34133 bps


6925 bps Basic weather radarand video

234134 bps

for surveillance and navigation purposes Another issueof CNPC links is the high security requirements Goodsecurity mechanisms should be used on CNPC links toavoid any possibility of spoofed control or navigation sig-nals that may allow unauthorized persons to control theUAVs [46] For meeting future communication capac-ity requirements in aeronautical communications a newair-ground communication system called L-Band Dig-ital Aeronautical Communication System (L-DACS) isbeing developed in Europe with funding from EURO-CONTROL L-DACS is the system in the Future Com-munication System (FCS) for L-band 960-1164 MHzL-DACS comprises of L-DACS1 [47] and L-DACS2[48] L-DACS1 is multi-carrier broadband Orthogo-nal Frequency-Division Multiplexing (OFDM)-based sys-tem while L-DACS2 is narrow band single-carrier withGaussian Minimum Shift Keying (GMSK) modulationsystem More information about L-DACS1 and L-DACS2including their benefits with the current aeronautical sys-tem and their physical and medium access layers can befound in [49]

55 Pipeline monitoring and controlIn this section we provide further discussion and illustra-tion of the smart city application of monitoring pipelinessystems as an example of infrastructure monitoringFigure 7 shows a CPS system for pipeline monitoring andcontrol In our previous work in [50 51] we present aframework for monitoring oil gas and water pipelinesusing linear sensor networks (LSNs) We defined an LSNto be a WSN where the sensors are aligned in linear formdue to the linearity of the structure or geographic area

Table 4 CNPC supported aircraft density

Aircraft size Density of UAVs No of UAVs withinin Space (km3) Radius of 100 km

Small 0000802212 1680

Medium 0000194327 407

Large 000004375 91

that is being monitored such as pipelines borders riverssea coasts railroads and more In the system illustratedin the figure the sensor nodes (SNs) are placed on thepipeline in order to monitor various important parame-ters such as temperature pressure fluid velocity chemicalsubstances leaks etc The collected data could either berouted to the sink which is placed at the end of thepipeline or pipeline segment using a multihop strategy[51] or it could be collected using a mobile node such asa low flying UAV [52] In the latter case the SN trans-mits its stored data to the UAV when it comes within itsrange The UAV which flies along the pipeline deliversthe collected data to the sink at the other end Once thedata arrives at the sink using either one of these strategiesit transmits the data to the infrastructure network usingany one of the communication networks that are availablein the area such as IEEE 80216 (WiMAX) cellular andsatelliteThe storage and processing servers that are used by the

CPS system can be a part of the cloud computing environ-ment The results are then sent in appropriate format tothe network control center (NCC) where they can be pre-sented to the NCC personnel using applications that allowvisualization of the pipeline structure The graphic repre-sentation that is displayed can show the pipelinersquos currentstatus as well as any anomalies that need more attentionand further inspection The NCC officers can then issuecertain commands back to the SNs in a particular hot spotAlternatively such commands might be automaticallyissued by the CPS system in accordance with algorithmsand configurations that are programmed by the sys-tem administrators The command messages intended toreach the target SNs are communicated via the networkthe sink and other SNs (if the multihop routing strategy isused) or the UAV (if UAV-based communication is used)An example of such commands could be (1) increase thequantity andor quality of the collected data (2) turn ONmore SNs in the hot spot and (3) turn ON additionalsensing or monitoring devices (that might be in sleep-mode to save energy) such as higher resolution camerasas well as audio or video monitoring equipment Alsothe NCC might dispatch specialized UAVs andor quickresponse teams for further inspection or to take remedialemergency actions (put out fire etc) In other cases thecollected data can be used to generate appropriate main-tenance schedules which could result in the service ofvarious parts of the pipeline according to a pre-set prioritystrategy

6 Open issuesIn this section we identify some of the most importantopen issues that need further research and investigation inthe area of networking and communication in smart citysystems


Fig 7 CPS for pipeline monitoring and control

61 Communication middleware for smart cityapplications

As a smart city network can consist of different devicescommunication technologies and protocols managingsuch a network can be very complex One of the pos-sible approaches to relax this complexity is to usea specialized communication middleware capable ofabstracting these communication details In addition thismiddleware can offer value-added features that are com-monly needed for different smart city applications andare not fully supported by existing network technolo-gies These features can include provisions and servicesfor security reliability scalability and quality of serviceprovisioning

62 Software defined network support for smart cityapplications

Software defined networking (SDN) is an approach toenable flexible and efficient network configuration toenhance a network SDN can provide many advantagesfor configuring city networks to support different appli-cations While there are some efforts in investigating thisapproach for supporting smart city applications [37] thereis room for developing more advanced management andnetworking mechanisms in SDN for efficient reliable andsecure network configurations in smart cities

63 New networks and network protocolsMost of the current communication infrastructure usesthe Internet infrastructure or the cellular networksAlthough these have proven to be efficient and usablethey often lack some necessary requirements for smartcity applications Issues in real-time responses mobilitysupport and the ability to handle huge volume of networktraffic When a smart city application is deployed city-wide and collects find grain data it will generate massivetraffic which could cause serious performance problemsfor the underlying network infrastructure Some work isunderway to add more capabilities such as the progres-sion from 3G to 4G and now 5G cellular networks [53]advances in MESH networks and investigation of moreefficient protocols on the existing networks Howeverthere is a lot to be done in this regard

64 Modeling and simulationThere are limited efforts in studying modeling and sim-ulation of the communication performance of differentsmart city applications on different network architecturesIt is important to develop different models and simula-tion tools to be able to design evaluate and plan forsuch networks In addition more detailed traffic pat-terns of different smart applications need to be compre-hensively studied Modeling and simulation of both the


traffic of smart city applications and the used networksrsquocapabilities may lead to better end results for the smartcity applications

7 Conclusions and future researchSignificant advancements in various technologies such asCPS IoT WSNs cloud computing and UAVs have takenplace lately The smart city paradigm combines theseimportant new technologies in order to enhance the qual-ity of life of city inhabitants provide efficient utilizationof resources and reduce operational costs In order forthis model to reach its goals it is essential to provideefficient networking and communication between the dif-ferent components that are involved to support varioussmart city applications In this paper we investigated thenetworking requirements for the different applicationsand identified the appropriate protocols that can be usedat the various system levels In addition we illustratednetworking architectures for five different smart city sys-tems This area of research is still in its initial stagesFuture studies can focus on important requirementsincluding routing energy efficiency security reliabilitymobility and heterogeneous network support Conse-quently more investigations and studies need to be donewhich should lead to the design and development ofefficient networking and communication protocols andarchitectures to meet the growing needs of the variousimportant and rapidly expanding smart city applicationsand services

AbbreviationsCPS Cyber-physical systems UAV Unmanned aerial vehicle WSN Wirelesssensor networks

AcknowledgmentsNot applicable

FundingThis research was not supported by any external funding source

Authorsrsquo contributionsIJ wrote the main sections of the paper NM contributed to the section onsmart city applications and illustration of smart city systems JA contributed tothe introduction and related work content All authors read and approved thefinal manuscript and contributed to various sections in the paper according totheir background

Ethics approval and consent to participateNot applicable

Competing interestsThe authors declare that they have no competing interests

Publisherrsquos NoteSpringer Nature remains neutral with regard to jurisdictional claims inpublished maps and institutional affiliations

Author details1Al Maaref University Old Airport Avenue PO Box 25-5078 Beirut Lebanon2Middleware Technologies Laboratory Pittsburgh Pennsylvania USA3Robert Morris University Moon Township Pennsylvania USA

Received 24 April 2018 Accepted 11 October 2018

References1 Watteyne T Pister KSJ (2011) Smarter cities through standards-based

wireless sensor networks IBM J Res Dev 55(12)1ndash72 Zanella A Bui N Castellani A Vangelista L Zorzi M (2014) Internet of

things for smart cities IEEE Internet Things J 1(1)22ndash323 Gurgen L Gunalp O Benazzouz Y Gallissot M (2013) Self-aware

cyber-physical systems and applications in smart buildings and citiesIn Proceedings of the Conference on Design Automation and Test inEurope pages 1149ndash1154 EDA Consortium

4 Ermacora G Rosa S Bona B (2015) Sliding autonomy in cloud roboticsservices for smart city applications In Proceedings of the Tenth AnnualACMIEEE International Conference on Human-Robot InteractionExtended Abstracts ACM pp 155ndash156

5 Mohammed F Idries A Mohamed N Al-Jaroodi J Jawhar I (2014) Uavs forsmart cities Opportunities and challenges In Unmanned AircraftSystems (ICUAS) 2014 International Conference on IEEE pp 267ndash273

6 Giordano A Spezzano G Vinci A (2016) Smart agents and fog computingfor smart city applications In International Conference on Smart CitiesSpringer pp 137ndash146

7 Clohessy T Acton T Morgan L (2014) Smart city as a service (scaas) afuture roadmap for e-government smart city cloud computing initiativesIn Proceedings of the 2014 IEEEACM 7th International Conference onUtility and Cloud Computing IEEE Computer Society pp 836ndash841

8 Al-Nuaimi E Al-Neyadi H Mohamed N Al-Jaroodi J (2015) Applications ofbig data to smart cities J Internet Serv Appl 6(1)25

9 Mohamed N Lazarova-Molnar S Al-Jaroodi J (2017) Cloud of thingsOptimizing smart city services In Proceedings of the InternationalConference on Modeling Simulation and Applied Optimization IEEEpp 1ndash5

10 Erol-Kantarci M Mouftah HT (2012) Suresense sustainable wirelessrechargeable sensor networks for the smart grid IEEEWirel Commun 19(3)

11 Gutieacuterrez J Villa-Medina JF Nieto-Garibay A Porta-Gaacutendara MAacute (2014)Automated irrigation system using a wireless sensor network and gprsmodule IEEE Trans Instrum Meas 63(1)166ndash76

12 Centenaro M Vangelista L Zanella A Zorzi M (2016) Long-rangecommunications in unlicensed bands The rising stars in the iot and smartcity scenarios IEEE Wirel Commun 23(5)60ndash7

13 Leccese F Cagnetti M Trinca D (2014) A smart city application A fullycontrolled street lighting isle based on raspberry-pi card a zigbee sensornetwork and wimax Sensors 14(12)24408ndash24

14 Sanchez L Muntildeoz L Galache JA Sotres P Santana JR Gutierrez VRamdhany R Gluhak A Krco S Theodoridis E et al (2014) SmartsantanderIot experimentation over a smart city testbed Comput Netw 61217ndash38

15 Wan J Di L Zou C Zhou K (2012) M2m communications for smart city Anevent-based architecture In Computer and Information Technology(CIT) 2012 IEEE 12th International Conference on IEEE pp 895ndash900

16 Gaur A Scotney B Parr G McClean S (2015) Smart city architecture and itsapplications based on iot Procedia Comput Sci 521089ndash94

17 Jin J Gubbi J Luo T Palaniswami M (2012) Network architecture and qosissues in the internet of things for a smart city In Communications andInformation Technologies (ISCIT) 2012 International Symposium on IEEEpp 956ndash961

18 De Poorter E Moerman I Demeester P (2011) Enabling direct connectivitybetween heterogeneous objects in the internet of things through anetwork-service-oriented architecture EURASIP J Wirel Commun Netw2011(1)61

19 Karnouskos S (2011) Cyber-physical systems in the smartgrid In IndustrialInformatics (INDIN) 2011 9th IEEE International Conference on IEEEpp 20ndash23

20 Miclea L Sanislav T (2011) About dependability in cyber-physical systemsIn Design amp Test Symposium (EWDTS) 2011 9th East-West IEEE pp 17ndash21

21 Sridhar S Hahn A Govindarasu M (2012) Cyberndashphysical system securityfor the electric power grid Proc IEEE 100(1)210ndash24

22 Berger C Rumpe B (2014) Autonomous driving-5 years after the urbanchallenge The anticipatory vehicle as a cyber-physical system arXivpreprint arXiv14090413


23 Cunningham R Garg A Cahill V et al (2008) A collaborativereinforcement learning approach to urban traffic control optimizationIn Web Intelligence and Intelligent Agent Technology 2008 WI-IATrsquo08IEEEWICACM International Conference on IEEE Vol 2 pp 560ndash566

24 Kartakis S Abraham E McCann JA (2015) Waterbox A testbed formonitoring and controlling smart water networks In Proceedings of the1st ACM International Workshop on Cyber-Physical Systems for SmartWater Networks ACM p 8

25 Gonda L Cugnasca CE (2006) A proposal of greenhouse control usingwireless sensor networks In Proceedings of 4thWorld CongressConference on Computers in Agriculture and Natural Resources OrlandoFlorida USA p 229

26 Mohamed N Al-Jaroodi J Jawhar I Lazarova-Molnar S (2014) Aservice-oriented middleware for building collaborative uavs J Intell RobotSyst 74(1-2)309ndash21

27 Lee J Bagheri B Kao H-A (2015) A cyber-physical systems architecture forindustry 40-based manufacturing systems Manuf Lett 318ndash23

28 Loacutepez P Fernaacutendez D Jara AJ Skarmeta AF (2013) Survey of internet ofthings technologies for clinical environments In Advanced InformationNetworking and Applications Workshops (WAINA) 2013 27thInternational Conference on IEEE pp 1349ndash1354

29 Macedo D Guedes LA Silva I (2014) A dependability evaluation forinternet of things incorporating redundancy aspects In NetworkingSensing and Control (ICNSC) 2014 IEEE 11th International Conference onIEEE pp 417ndash422

30 Silva I Leandro R Macedo D Guedes LA (2013) A dependability evaluationtool for the internet of things Comput Electr Eng 39(7)2005ndash18

31 (2014) Oma lightweight m2m Available httptechnicalopenmobileallianceorgtechnicaltechnical-informationrelease-programcurrent-releasesoma-lightweightm2m-v1-0-2

32 Perelman V Ersue M Schoumlnwaumllder J Watsen K (2012) NetworkConfiguration Protocol Light (NETCONF Light) Network

33 (2013) Commercial building automation systems Navigant ConsultingRes Boulder

34 Suciu G Vulpe A Halunga S Fratu O Todoran G Suciu V (2013) Smartcities built on resilient cloud computing and secure internet of thingsIn Control Systems and Computer Science (CSCS) 2013 19thInternational Conference on IEEE pp 513ndash518

35 Bolodurina I Parfenov D (2017) Development and research of models oforganization distributed cloud computing based on the software-definedinfrastructure Procedia Comput Sci 103569ndash76

36 Bonomi F Milito R Zhu J Addepalli S (2012) Fog computing and its role inthe internet of things In Proceedings of the first edition of the MCCworkshop on Mobile cloud computing ACM pp 13ndash16

37 Liu J Li Y Chen M Dong W Jin D (2015) Software-defined internet ofthings for smart urban sensing IEEE Commun Mag 53(9)55ndash63

38 Olenewa JL (2014) Guide to Wireless Communications Cengage Learn39 Stallings W (2005) Wireless Communications and Networks Prentice Hall

Pearson Education Inc Upper Saddle River40 IEEE 80211 IEEE 80216 httpenwikipediaorgwiki viewed December

10 201441 Goyal D Tripathy MR (2012) Routing protocols in wireless sensor networks

A survey In Advanced Computing amp Communication Technologies(ACCT) 2012 Second International Conference on IEEE pp 474ndash480

42 Jawhar I Mohamed N Agrawal DP (2011) Linear wireless sensor networksClassification and applications J Netw Comput Appl 34(5)1671ndash82

43 Nunes BAA Mendonca M Nguyen X-N Obraczka K Turletti T (2014)A survey of software-defined networking Past present and future ofprogrammable networks IEEE Commun Surv Tutor 16(3)1617ndash34

44 Kerczewski RJ Griner JH (2012) Control and non-payloadcommunications links for integrated unmanned aircraft operationsIn Report NASA Glenn Research Center Cleveland Ohio USA

45 (2012) Unmanned aircraft systems (uas) integrated in the nationalairspace system (nas) technology development project plan In NationalAeronautics and Space Administration

46 Zeng Y Zhang R Lim TJ (2016) Wireless communications with unmannedaerial vehicles opportunities and challenges IEEE Commun Mag arXivpreprint arXiv160203602 54(5)

47 Sajatovic M Haindl B Ehammer M Graupl T Schnell M Epple U Brandes S(2009) L-dacs1 system definition proposal Delivrable d2 EUROCONTROLTech Rep Version 10

48 Fistas N (2009) L-dacs2 system definition proposal Delivrable d2EUROCONTROL Tech Rep Version 10

49 Neji N De Lacerda R Azoulay A Letertre T Outtier O (2013) Survey on thefuture aeronautical communication system and its development forcontinental communications IEEE Trans Veh Technol 62(1)182ndash91

50 Jawhar I Mohamed N Agrawal DP (2011) Linear wireless sensornetworks Classification and applications Elsevier J Netw Comput Appl(JNCA) 341671ndash82

51 Jawhar I Mohamed N Shuaib K (2007) A framework for pipelineinfrastructure monitoring using wireless sensor networks In The SixthAnnual Wireless Telecommunications Symposium (WTS 2007) IEEECommunication SocietyACM Sigmobile Pomona California USApp 1ndash7

52 Jawhar I Mohamed N Al-Jaroodi J Zhang S (2014) A framework for usingunmanned aerial vehicles for data collection in linear wireless sensornetworks J Intell Robot Syst 74(1-2)437ndash453

53 Andrews JG Buzzi S Choi W Hanly SV Lozano A Soong ACK Zhang JC(2014) What will 5g be IEEE J Sel Areas Commun 32(6)1065ndash82

54 Fallah YP Huang C Sengupta R Krishnan H (2010) Design of cooperativevehicle safety systems based on tight coupling of communicationcomputing and physical vehicle dynamics In Proceedings of the 1stACMIEEE International Conference on Cyber-Physical Systems ACMpp 159ndash167

Abstract

Keywords

Introduction

Related work

Smart city applications

Smart city networking architectures and communication requirements

Networking characteristics requirements and challenges of smart city systems

Network protocols

Bandwidth

Delay tolerance

Power consumption

Reliability

Security

Heterogeneity of network protocols

Wiredwireless connectivity

Mobility

Additional issues and challenges

Interoperability

Availability

Performance

Management

Scalability

Big data analytics

Cloud computing

Fog computing

Links between nodes in smart city systems

Illustration of selected smart city systems

Smart grid system

Smart home energy management

Smart water systems

UAV and Commercial Aircraft Safety Communication

Pipeline monitoring and control

Open issues

Communication middleware for smart city applications

Software defined network support for smart city applications

New networks and network protocols

Modeling and simulation

Conclusions and future research

Abbreviations

Acknowledgments

Funding

Authors contributions

Ethics approval and consent to participate

Competing interests

Publishers Note

Author details

References

Jawhar et al Journal of Internet Services and Applications (2018) 926 of 16

Applications of these smart energy services are used tosupport smart grids and smart buildings as well as pro-vide better utilization of renewable energy Other smartservices involve structural health monitoring as well asreal-time monitoring of water networks bridges tun-nels train and subway rails and oil and gas pipelinesAdditional services include smart services for environ-mental monitoring and smart services for public safetyand securityThese smart city services do not only need the various

advanced technologies discussed here but also need reli-able and robust networking and communication infras-tructures to enable efficient exchange of messages amongthe different components of the systems that provide aparticular service Smart city services are designed atdifferent scales which require various networking andcommunication technologies for their implementationand operations Furthermore different network and com-munication models and approaches can be utilized forsmart city services This paper investigates the commu-nication and network issues of smart city systems It alsoinvestigates networking technologies architectures andcommunication requirements for such systems The suit-ability of existing network protocols for different smartcity services will be discussed Although there are sig-nificant research efforts to investigate different issues insmart cities and provide solutions for these issues verylittle research has been done to investigate the network-ing and communication parts of smart city systems whichconstitute the main objective of this paperThe rest of the paper is organized as follows Section 2

provides an overview of related work in this fieldSection 3 includes an overview of some smart city appli-cations Section 4 presents networking architectures andcommunication requirements for smart city applicationsSection 5 offers an illustration of selected smart citysystems Section 6 discusses open issues in the area ofnetworking and communication for smart city systemsFinally Section 7 concludes the paper and provides somefuture research directions

2 Related workThere are a few papers published addressing the networkand communication issues for smart cities In this sectionwe discuss some of the work presented in these papersZanella et al [2] provided an interesting study that ana-lyzed the current technologies available for deploying IoTfor a smart city In addition the authors presented anddiscussed a proof-of-concept implementation of IoT inthe city of Padova Italy There are generally two com-mon approaches to offer data access to things in IoT Thefirst is using multi-hop mesh networks with short-rangecommunication in the unlicensed spectrum among thenetwork nodes The second is using long-range cellular

technologies in the licensed frequency band A new com-munication technology is introduced to provide alter-native connectivity for IoT named LPWAN (Low-PowerWide Area Network) [12] From its name this com-munication technology can provide low-rate long-rangetransmission in the unlicensed frequency bands using startopology These features can be very useful for some smartcity applications Leccese et al [13] introduced a smartcity application of fully controlled street lighting using aZigBee Sensor Network and WiMAX while the applica-tion is controlled by Raspberry-Pi Card Sanchez et al [14]introduced a smart city testbed for IoT experimentationnamed SmartSantanderSome work more related to our contribution in this

paper is investigating network architectures for smartcities Wan et al [15] presented an event-based communi-cation architecture that helpsmanage and facilitates coop-eration among Machine to machine (M2M) componentsin smart cities The authors also conducted a case studyusing this architecture for vehicular applications Gauret al [16] proposed a multi-level smart city architecturebased on semantic web technologies This architectureis mainly for wireless sensor networks applications in asmart city It consists of 4 levels Data Collection DataProcessing Integration and Reasoning and Device Con-trol and Alerts Jin et al [17] studied distinct IoT networkarchitectures with focus on Quality of Service (QoS) fordifferent smart city applicationsThe last paper [17] is the closest to our contribu-

tions in the paper thus we will elaborate on it Theauthors presented five different IoT network architec-tures (1) Autonomous Network Architecture (2) Ubiqui-tous Network Architecture (3) Application-Layer OverlayNetwork Architecture (4) Service-Oriented Architectureand (5) Participatory Sensing The autonomous networkarchitecture is usually not directly linked to public net-works such as the Internet It can be connected thoughgateways if this link is needed This network architectureis suitable for some smart city applications such as auto-matic parking management in which most of the networkconnections are mainly to support the application TheQoS requirements in this network architecture are mainlydependent on the requirements of the applicationIn the ubiquitous network architecture smart objects

including the sensors and actuators are part of the Inter-net Information from these smart objects can be obtainedby authorized users and applications through the Internetdirectly or through intermediate servers which serve assinks that gather data from the connected smart objectsThe serversrsquo option can provide better scalability andresource efficiency than the direct access option Thesmart objects can be connected to the Internet throughmultitier and multiradio This network architecture issuitable for smart city applications such as structure























AppropriateNet Prot


PowerConsump


Wired wireless

Mobility






IEEE 80216 Cellular



IEEE 802154 IEEE 80211IEEE 80216






IEEE 802154 IEEE802151 IEEE 80211












































IEEE 802154(Zigbee)









6 9 12 18 24 3648 54 Mbps

120 m outdoors All







140 m outdoors All




250 m outdoors All

IEEE 80216(WiMAX)






800 MHz to 1900MHz





LTE and LTEAdvanced

300 Mbps to 1Gbps
































Small(lt=55 kg)






34133 bps



234134 bps





Small 0000802212 1680

Medium 0000194327 407

Large 000004375 91
















































































Abstract

Keywords

Introduction

Related work




Network protocols

Bandwidth

Delay tolerance

Power consumption

Reliability

Security



Mobility


Interoperability

Availability

Performance

Management

Scalability

Big data analytics

Cloud computing

Fog computing



Smart grid system


Smart water systems



Open issues






Abbreviations

Acknowledgments

Funding



Competing interests

Publishers Note

Author details

References























AppropriateNet Prot


PowerConsump


Wired wireless

Mobility






IEEE 80216 Cellular



IEEE 802154 IEEE 80211IEEE 80216






IEEE 802154 IEEE802151 IEEE 80211












































IEEE 802154(Zigbee)









6 9 12 18 24 3648 54 Mbps

120 m outdoors All







140 m outdoors All




250 m outdoors All

IEEE 80216(WiMAX)






800 MHz to 1900MHz





LTE and LTEAdvanced

300 Mbps to 1Gbps
































Small(lt=55 kg)






34133 bps



234134 bps





Small 0000802212 1680

Medium 0000194327 407

Large 000004375 91
















































































Abstract

Keywords

Introduction

Related work




Network protocols

Bandwidth

Delay tolerance

Power consumption

Reliability

Security



Mobility


Interoperability

Availability

Performance

Management

Scalability

Big data analytics

Cloud computing

Fog computing



Smart grid system


Smart water systems



Open issues






Abbreviations

Acknowledgments

Funding



Competing interests

Publishers Note

Author details

References
































IEEE 802154(Zigbee)









6 9 12 18 24 3648 54 Mbps

120 m outdoors All







140 m outdoors All




250 m outdoors All

IEEE 80216(WiMAX)






800 MHz to 1900MHz





LTE and LTEAdvanced

300 Mbps to 1Gbps
































Small(lt=55 kg)






34133 bps



234134 bps





Small 0000802212 1680

Medium 0000194327 407

Large 000004375 91
















































































Abstract

Keywords

Introduction

Related work




Network protocols

Bandwidth

Delay tolerance

Power consumption

Reliability

Security



Mobility


Interoperability

Availability

Performance

Management

Scalability

Big data analytics

Cloud computing

Fog computing



Smart grid system


Smart water systems



Open issues






Abbreviations

Acknowledgments

Funding



Competing interests

Publishers Note

Author details

References























Small(lt=55 kg)






34133 bps



234134 bps





Small 0000802212 1680

Medium 0000194327 407

Large 000004375 91
















































































Abstract

Keywords

Introduction

Related work




Network protocols

Bandwidth

Delay tolerance

Power consumption

Reliability

Security



Mobility


Interoperability

Availability

Performance

Management

Scalability

Big data analytics

Cloud computing

Fog computing



Smart grid system


Smart water systems



Open issues






Abbreviations

Acknowledgments

Funding



Competing interests

Publishers Note

Author details

References













































































Abstract

Keywords

Introduction

Related work




Network protocols

Bandwidth

Delay tolerance

Power consumption

Reliability

Security



Mobility


Interoperability

Availability

Performance

Management

Scalability

Big data analytics

Cloud computing

Fog computing



Smart grid system


Smart water systems



Open issues






Abbreviations

Acknowledgments

Funding



Competing interests

Publishers Note

Author details

References

RESEARCH OpenAccess Networkingarchitecturesandprotocols ...

Documents

Transcript of RESEARCH OpenAccess Networkingarchitecturesandprotocols ...