1

Mobile Phone Computing and theInternet of Things: A Survey

Andreas Kamilaris∗ and Andreas Pitsillides†∗Insight Centre for Data Analytics, National University of Ireland, Galway, Ireland

Email: andreas.kamilaris@insight-centre.org†Department of Computer Science, University of Cyprus, Nicosia, Cyprus

Email: andreas.pitsillides@cs.ucy.ac.cy

Abstract—As the Internet of Things and the Web of Thingsare becoming a reality, their interconnection with mobile phonecomputing is increasing. Mobile phone integrated sensors offeradvanced services, which when combined with web-enabled real-world devices located near the mobile user (e.g. body areanetworks, RFID tags, energy monitors, environmental sensorsetc.), have the potential of enhancing the overall user knowledge,perception and experience, encouraging more informed choicesand better decisions. This paper serves as a survey of themost significant work performed in the area of mobile phonecomputing combined with the Internet/Web of Things. A selectionof over 100 papers is presented, which constitute the mostsignificant work in the field up to date, categorizing these papersinto ten different domains, according to the area of application(i.e. health, sports, gaming, transportation, agriculture), thenature of interaction (i.e. participatory sensing, eco-feedback,actuation and control) or the communicating actors involved(i.e. things, people). Open issues and research challenges areidentified, analyzed and discussed.

Index Terms—Mobile Computing, Mobile Phone Applications,Internet of Things, Web of Things, Sensors, Survey.

I. INTRODUCTION

Technologies such as wireless sensor networks, short-rangewireless communications and radio-frequency identification(RFID) have allowed the Internet to penetrate in embeddedcomputing [1], [2]. The Internet of Things (IoT) [3] is be-coming reality, as everyday physical objects are becomingequipped with sensors and actuators, being (uniquely) ad-dressable and interconnected, allowing the interaction withthem through the Internet. Porting the IP stack on embeddeddevices was a successful effort [4], [5] and, together withthe introduction of IPv6 (which provides extremely largeaddressing capabilities), facilitate the merging of the physicaland the digital world, enabling the IoT to grow faster.

Building upon the notion of the IoT, the Web of Things(WoT) [6], [7], [8] reuses well-defined Web techniques tointerconnect this new generation of Internet-enabled physicaldevices. While the IoT focuses on interconnecting heteroge-neous devices at the network layer, the WoT can be seen as apromising practice to achieve interoperability at the applicationlayer. It is about taking the Web as we know it and extendingit so that anyone can plug devices into it.

The rise of multi-sensory mobile phones with Internetconnectivity has helped to reduce the barriers for associatingmobile computing with the IoT/WoT. Mobile phone integrated

sensors offer advanced capabilities such as measuring prox-imity, acceleration and location or record audio/noise, senseelectromagnetism or capture images and videos [9]. Thesesensing services, when combined with web-enabled physicalentities located near the user, have the potential of enhancingthe overall user knowledge and experience, helping the userto take more informed choices [10], [11].

Although there exist several survey papers focusing onadvances related to the IoT/WoT [12], [13], [14], [15], notany papers -as far as we can ascertain- have yet surveyed thegrowing interconnection between mobile computing and theIoT/WoT. This paper aims to address this gap, presenting themost significant work in this area. Section II describes themethodology followed to develop this survey, while SectionIII comments on the important findings of our study. Then,Section IV discusses the overall observations during this studyand identifies open issues in this research domain, and finallySection V concludes the paper.

II. METHODOLOGY

Before describing the important work in the field, themethodology in collecting this information is explained, whichinvolved three steps: a) collection of the state of the art work,b) clustering of related work, and c) analysis of related work.

In the first step, a keyword-based search for conference pa-pers and articles in well-known scientific databases (e.g. IEEEXplore, ACM, DBLP, ScienceDirect, CiteSeerX) and searchengines was performed. Various keywords were used suchas ”mobile computing”, ”internet”, ”web”, ”web of things”and combinations of them. Existing surveys on the IoT/WoTand mobile sensing [12], [13], [14], [16], [15], [9] were alsostudied for relevant efforts. The focus was to pick only paperswhich followed the main concepts and design principles ofthe IoT/WoT [6], [7], excluding papers that used proprietaryprotocols and in which vendor lock-in was evident. This meansthat some popular papers in mobile pervasive computing mighthave been excluded from our survey. Also, some cutting-edgeresearch papers related to the exciting domain of feature- anddepth-based identification, tracking of and interaction with thephysical world have also been excluded, as these works havenot yet become (widely) applied on mobile devices, due totheir large processing requirements. Thus, 68 papers in totalwere firstly identified, performing then a depth search on theirmost relevant references, increasing the list of papers to 102.

2

In the second step, related work was categorized in clusters1

according to their topic/area of application. Ten clusters werecreated, in which the papers were placed with respect to theirrelevance to each cluster. These clusters are explained in detailin the next section. It is noted that these clusters were aimedto cover more or less the whole spectrum of combining mobilecomputing with the IoT/WoT.

Some clusters had more studies performed than otherswhile some other clusters involved also business cases (notonly research efforts). In the final step, each cluster wasexamined carefully, studying and analyzing each paper sep-arately, recording its summary, contribution and impact to thecommunity, and its overall novelty and importance. After thisprocedure, at most 8-12 more significant works per clusterwere selected, which are described in the next section. Hence,the rest of this survey is based on the analysis performed onthe selected papers, as organized in the ten clusters.

III. MOBILE COMPUTING AND INTERNET/WEB OF THINGS

We define the domain of mobile computing and IoT/WoT asthe research area that involves case studies, prototypes, demon-strations, applications and business cases of the IoT/WoT,through mobile phones, where the user of the mobile phoneinteracts with cyber-physical things that are enabled to theinternet/web, through his/her phone device, exploiting at thesame time the sensing capabilities of the phone.

As mentioned in Section II, related work is divided indifferent clusters or categories, which represent either thearea of application (e.g. health, sports, gaming, agriculture,transportation), the nature of interaction/communication (e.g.participatory sensing, eco-feedback) or more general the typeof interaction (e.g. with things, social interactions with otherusers). The ten different categories identified are graphicallypresented in Figure 1, together with the most common tech-nologies used at each category (see also Table I), and are listedbelow:

A. Participatory Sensing. Involves mobile systems encour-aging users to record and share information, towards theco-creation of advanced knowledge.

B. Eco-Feedback. Involves mobile applications that providefeedback to the users about various environmental phe-nomena or events, or about their personal consumption(water, electricity, energy etc.).

C. Actuation and Control. Related to mobile applicationsthat control some physical devices or actuate someevents related to these devices. A typical example ishome automation applications.

D. Health. The mobile phone acts as an intermediary be-tween body sensors and the web. It receives informationabout the health status of the user, measured by varioussensor devices installed on the body of the user andthen uploads/shares this information to the web for betteranalysis, comparisons and feedback.

1The representation in clusters is suggested by the authors in order toprovide a means to ease the readability and comprehension of the large amountof papers, and to group them in what seems to us a sensible classification.Some papers could have been included in more than one cluster. We selectedthe most relevant one at each case.

Fig. 1. Mobile computing and IoT/WoT: Application categories.

E. Sports. Includes a combination of physical sensors andmobile applications which are used during sport activ-ities to record various metrics and help to improve theperformance of the athlete user.

F. Agriculture. Related to smart farming practices to im-prove productivity, management of livestock and in-crease consumer satisfaction and transparency.

G. Gaming. This category is about virtual games whichconsider the physical presence or status of the mobileuser to enhance the gaming experience.

H. Transportation. A growing domain in which the sensingfeatures of the mobile phone are harnessed for betterdriving experience and convenience of parking.

I. Interaction with Things. This is the broader categoryas it relates to efforts which focus on interacting withweb-enabled physical entities available in the nearbyenvironment, as for example tagging technologies.

J. Social Interactions with People. If we broaden the WoTto involve humans, we can identify various mobile appsthat combine information from online social networkingsites with information from the mobile phones of nearbyusers, to offer an augmented social experience.

In the following subsections, each category is presented, list-ing the most significant work performed. General observationsare discussed in Section IV.

A. Participatory Sensing

Participatory sensing involves the tasking of mobile de-vices to form interactive, participatory systems that enableindividuals in the general public to gather, analyze and sharelocal information, towards the co-creation of knowledge oraddressing environmental challenges [17].

The MetroSense project [18] aims at transforming themobile device into a social sensing platform. It is based on

3

Fig. 2. Sound exposure through the NoiseTube project (source: [22]).

a three stage framework which involves sensing, learning andsharing. In the sense stage, MetroSense leverages mobility-enabled interactions between human-carried mobile sensors,static sensors embedded in the civic infrastructure and wirelessaccess nodes providing a gateway to the Internet, to supportthe delivery of application requests to the mobile devices andthe delivery of sampled data back to the application.

Goldman et al. [19] show the significance of participatorysensing for our daily lives, namely its impact on climaticchange. GPS-equipped mobile phones are used to photographdiesel trucks, in order to understand community exposure toair pollution. Odourmap [20] is a communication platformon which citizens and authorities can be involved into theodour management process. Portolan [21] builds signal cov-erage maps performing network monitoring based on crowdsourcing. In NoiseTube [22], mobile phones are used asnoise sensors, to measure the personal exposure to noise ofcitizens, in their everyday environment. Figure 2 presents thesound exposure of a user during a walk in Paris through theNoiseTube project.

The Common Sense project [23] derives design principlesfor describing data collection and knowledge generation fromremote air quality data. In [24], a case study is performed,involving sensors deployed in public areas, shared by differentcommunities. Users get informed of environmental conditionsdirectly from these sensors. The findings indicate sensitivityto environmental factors from the people involved.

Furthermore, participatory sensing can be extended to thehealth domain. A very nice example is about smartphoneapplications for melanoma detection [25]. Users use theirphones’ camera to record unusual spots on their body, andthen share with the online community, patients and generalistclinician users who provide feedback and informal diagnosis.

Related work indicates that mobile sensing, both participa-tory and remote, is beneficial to the users and engages them insustainable actions that improve their community [19], [24].Mobile users become active members of their communitiesand raise their awareness towards their surroundings, whilebarriers for delivering large-scale environmental campaignsare reduced [17]. It constitutes an excellent approach forco-creating advanced knowledge based on a participatoryscheme, and it could be used even for medical advice [25].We need to comment however that these positive effects areoften transient and only persist as long as users engageactively with the application. The evolution of participatory

Fig. 3. The eMeter system (source: [30]).

sensing is mobile crowd sensing for large-scale sensing [26],[27], which enables a broad range of applications includingurban dynamics mining, public safety, traffic planning andenvironment monitoring.

B. Eco-Feedback

Eco-feedback mobile applications provide feedback to theusers about their impact on the physical environment, suchas their personal energy footprint (water, electricity, drivinghabits etc.) or useful information and knowledge about theexisting local environmental conditions.

Regarding eco-feedback on personal consumption, Ener-gyLife [28] developed a mobile application to provide electric-ity consumption feedback about domestic electrical appliancesand conservation tips. In UbiLense [10], users can utilize theirmobile phones as magic lenses to view the energy consumptionof their home devices just by pointing on them with thephone’s camera. Ambient meter [29] is a mobile device thatdisplays the level of energy consumption of the place it iscurrently located in, by changing its color from green (i.e. theamount of energy consumed in the room is low) to red (i.e. alot of energy is consumed).

Moreover, the eMeter system [30] allows users to interac-tively monitor, measure, and compare their energy consump-tion at a household and device level, by making use of asmart electricity meter and getting consumers ”into the loop”of energy monitoring, as displayed in Figure 3. Disaggregationof the electrical consumption of individual appliances withina household by means of a mobile application is discussed in[31], with accuracy rates of 87%.

Social Electricity [32] is a mobile application that allowspeople to compare their energy consumption with their onlinefriends, neighbors, similar peers etc., in order to perceivewhether their footprint is high. In this case, the applicationassists the user to reduce his/her consumption through person-alized tips and educational material. OPOWER is a US com-pany that offers similar services through mobile phones [33].PowerPedia [34] enables users to identify and compare theconsumption of their residential appliances to those of othersthrough a mobile application that uploads the measured datato a community platform. It thus helps users to better assesstheir electricity consumption and draw effective measures tosave electricity. Furthermore, products such as CubeSensors[35] combine sensors and a mobile application to help peopleunderstand how their home or office is affecting their health,comfort and productivity. LoseIt [36] is an example of a nutri-tion tracking tool, which offers a barcode scanner for providing

4

Fig. 4. An example of an urban mashup (source: [11], [38]).

information on various processed food options available at thesupermarkets.

Eco-feedback through mobile phones can also help tomitigate the negative effects of the increasing urbanism, to thepeople and the city as a whole. Urbanism implies that trafficis increased, levels of pollution are rising, some areas becomedirty and polluted while health and security are compromised.

To address some of the challenges implied by increasingurbanism, UrbanRadar [11] is a location-based application thatdiscovers and interacts with environmental services offered byWeb-enabled urban sensors. UrbanRadar allows the user tocreate urban mashups, defined as opportunistic web mashupsthat integrate real-world services, validated only when thelocal environmental conditions support the sensor-based Webservices defined by these mashups. An example urban mashupis displayed in Figure 4. In this example, a sensor measuringlevels of noise, a pollution sensor and a street camera combinetheir services to infer the traffic conditions existing at a citycenter. Air quality and pollution is studied in [37], usingspecialized sensors embedded in prototype mobile phones.

By means of UrbanRadar, the work in [38] investigates anddiscusses the acceptance, influence, usefulness and potentialof these services to mobile users, towards the vision of areal-time digital city. This case study suggested eleven designprinciples, which the authors consider important for futuremobile applications that deal with remote sensing of thephysical and urban environment.

Timely eco-feedback can influence residents to reduce theirconsumption by a fraction of 5-15%, through more informedchoices and better energy management [39]. It is importanthowever to consider the fact that the positive effect mightnot be persistent for long time [40]. Mobile applicationsmake the eco-feedback experience more personal, effective andconvenient. Mobile applications on urban remote sensing cancontribute toward the active involvement of citizens with theirurban landscape.

C. Actuation and Control

In this category, papers are identified that use mobile devicesas the tools to control physical devices such as electricalappliances. The most popular application domain is home au-tomation, but other areas such as offices, factories and commonspaces could be controlled through mobile applications.

Regarding home automation and control, it is speculatedthat Internet technology, in line with IoT, will become thefuture standard [41], [42], offering advanced interoperabilitybetween heterogeneous home devices and appliances.

The Aware Living Interface System (ALIS) [43] provides amobile application for feedback and control of a smart home,such as allowing the resident to adjust the lights or shadesfor a house facade, with a single control. A mobile phoneinterface that allows users to monitor, control, and measure theconsumption of single appliances at their home is presentedin [44], where the mobile phones are connected to smartelectricity meters through web protocols.

Companies offering home automation services such asEnOcean [45] are producing solutions for home and buildingautomation, based on Internet and web principles, enablingthe control of the house/building environment through mobilephone applications. An interesting start-up company is n.thing[46], which has developed Planty, a system that assists plantsgrowing through embedded sensors (soil humidity, tempera-ture and light) and mobile applications. Planty’s status can bemonitored in real-time remotely through the mobile phone andcontrol its watering through a water pump whenever needed.

Finally and more generally, a mobile phone applicationfor creating physical mashups is introduced in [47]. Physicalmashups [29] are defined as web mashups that combinesensory services using the classic techniques of Web mashups.

This category offers mobile applications that control the realworld by actuating physical devices such as home appliances.A strong focus of related work is on home automation as wellas automation in offices/buildings towards more convenienceand energy savings.

D. Health

In health-related mobile applications, the mobile phone actseither as an intermediary between body area sensors andthe web or the sensors of the phone are used to identifyabnormal behavior or indication of some disorder. In thefirst case, the mobile device receives information about thehealth status of the user, monitored by body sensors, and thenshares this information to the web for further analysis andfeedback by experts. In general, a wireless body area networkis a wireless networking technology that interconnects tinynodes with sensor or actuator capabilities in, on, or around ahuman body [16]. A system architecture of a wireless bodyarea sensor network for ubiquitous health monitoring throughmobile phones is presented in [48].

Zhong et al. [49] describe a Bluetooth-based body sensornetwork platform for physiological diary applications, usinga wrist-worn device as the user interface, which displaysinformation received from a mobile phone. The system de-scribed in [50] monitors the physiological signs of patients(electrocardiogram diagnosis) and records this informationon a mobile application, which then analyzes the data andforwards it if needed to the medical center through the web.myHealthAssistant [51] is a phone-based body sensor networkthat captures the wearer’s exercises throughout the day, bothdaily activities and specific gym exercises, and then allows

5

comparisons and competitions with other people through theweb. Rodrigues et al. [52] present an application that acquires,processes, stores, and displays medical data, thus enablinga full body sensor networking interface. A smart device foralerting a person’s critical health condition is presented in [53].

In the case when the sensors of the phone device are usedto identify health issues, Madan et al. [54] use mobile phone-based co-location and communication sensing to measurecharacteristic behavior changes in symptomatic individuals,reflected in their total communication, interactions with respectto time of day, diversity and entropy of face-to-face interac-tions and movement. Using these extracted mobile features, itis possible to predict the health status of an individual, withouthaving actual health measurements from the subject. Moreexamples include identifying depressive symptoms based onuser’s daily life behavior [55], therapy of insomnia [56] andstress recognition [57]. HealthyOffice exploits mood recogni-tion to improve employees’ health and productivity [58].

Concerning nutrition, FoodCam [59] is a real-time mobilefood recognition system recording a user’s eating habits.BodyBeat [60] is a mobile sensing system for capturing adiverse range of non-speech body sounds in real-life scenarios,such as sounds of food intake and breath, which contain invalu-able information about our dietary behavior and respiratoryphysiology.

Finally, the CrossCheck study [61] includes 150schizophrenic patients who use a smartphone applicationthat monitors their typical locations, when they are walking,running or sedentary, and detects and records the durationand frequency of conversations as well as sleep patterns. Itthen sends this data to health care providers who determinethe improvement on the patients’ health.

Mobile applications provide an unobtrusive way of man-aging the monitoring of the user’s health status through bodysensors and have the potential of taking locally some decisionsabout critical situations. In such cases, their communicationcapabilities are used to immediately alert caregivers andmedical centers remotely. Applications such as [54], [55],[57], [61] offer unique monitoring opportunities that allowthe patients to have a normal life, being at the same timemonitored for abnormal behavior that could indicate risks tothemselves or to other people. It is a much promising researchdomain.

E. Sports

Sports and recreational activities constitute one of the mostrapidly growing areas of personal and consumer-oriented IoTand WoT technologies [62]. WoT-enabled mobile applicationsfocusing on sports include various physical sensors, installedusually on the body or the clothes of the user, and whichare used during sport activities to record various metrics andhelp to improve the user’s performance. The mobile phonein these cases records the measurements of the sensors andshares this information to the web. A typical example is theBikeNet application [63], which aims to give a holistic pictureof the cyclist experience, not only by measuring variousmetrics (speed, distance traveled, calories burned, heart rate

Fig. 5. Snapshot of the BikeNet application (source: [63]).

etc.) but also by sharing this information within the onlinecycling community. A snapshot of the web portal of BikeNetis illustrated in Figure 5.

Also, measuring physical activity and promoting an activeway of life is a subject of research. Mobile Teen [64] uses themobile phone’s built-in motion sensor to automatically detectlikely sedentary behavior. Fit Buddy [65] helps users tracktheir personal fitness statistics focusing on step counting fromboth walking and running using a smartphone.

Commercial solutions also exist, such as the Nike+ SportsSensor [66], which puts a super smart sensor in shoes anduses pressure data in combination with an accelerometer tocalculate movement. This information can then be recordedby smart watches and mobile phones and then shared throughthe web, to challenge online friends and other online users, setpersonal goals, train smarter, improve performance etc. AppleWatch [67] has an excellent design and co-exists with theuser’s mobile device to track daily activity, encouraging theuser to keep moving for health and fitness. Similar productsand services are offered by other large companies such asGarmin [68] and Samsung [69] (smartwatches), as well asSuunto [70] (sports watches, dive computers and precisioninstruments).

The aforementioned mobile applications make the sportsexperience more entertaining and fun for the user, helpinghim/her to improve performance through competitions andcomparisons with friends, similar peers and the online commu-nity, encouraging a more active way of living. This communitycan create new useful knowledge, as for example the sharingof nice cycling pathways [63] recently discovered.

F. Agriculture

Smart farming [71] is about real-time data gathering, pro-cessing and analysis, as well as automation technologies on thefarming procedures, allowing improvement of the overall farm-ing operations and management and more informed decision-making by the farmers. Mobile applications interacting withthe agricultural environment can facilitate management ofcultivation and/or livestock, as well as consider the wholevalue chain from farm to fork, by considering consumertransparency over the product he/she buys.

Examples include Herdwatch [72], a herd managementsystem which allows cattle farmers manage their beef or dairyherds via a smartphone or tablet. MooMonitor+ [73] monitorseach individual cow’s health and fertility by means of wearable

6

collars installed on the animals, allowing farmers to monitortheir entire herd, from their phone.

Map your meal [74] is an agriculture-related smartphoneapplication which aims at enhancing the public awarenessunderstanding of global interdependence via exploring theglobal food system. With this smartphone application, farmersand consumers can scan the products to see their fairness andhow much green they are. Finally, FoodLoop [75] ties grocerinventory system and smart tagging to consumer-facing mobileapplications to provide real-time deals on ”nearly expired”products, supporting food waste reduction.

As agriculture is a highly unpredictable and risk-pronedomain, smart sensing and real-time monitoring is importantto prevent pests and animals diseases, react fast to changingenvironmental conditions, optimizing operations and produc-tivity. Hence, combining IoT sensing technology with mobileapplications, involving online information and web services[74], [75], permits faster reaction to unpredictable events andbetter decision making.

G. GamingThis category is about virtual games which take into account

the physical presence and characteristics of the mobile user toenhance his/her gaming experience.

The use of personal sensor streams in virtual worlds isdemonstrated in Second Life [76], which is a virtual worldsimulator where people lead virtual lives by using personalavatars. Accelerometer data is collected from a user mobilephone and classified into various activity states (sitting, stand-ing and running). These activities are then injected into SecondLife and interpreted by the user’s avatar [77]. Greenet [78] isan augmented reality mobile game dedicated to recycling. Thecamera of the mobile phone is used to detect whether the useris recycling various items. Object markers and bin markersare used to augment the physical world with this digital,mobile gaming experience. Users earn points by recycling, andcompete with each other for more points through the web.

Fish’n’Steps [79] is a social game which links a player’sdaily foot step count, measured by a pedometer, to the growthand activity of an animated virtual character, a fish in a fishtank. As further encouragement, some of the players fish tanksincluded other players fish, thereby creating an environmentof both cooperation and competition.

Augmented reality is expected to be the future in gaming,and early examples include Monopoly Zapped and Game ofLife2 [80]. The board games play out like the original versionsbut in the center of the board, an iPhone running the respectiveonline application is placed to spice up the game. In MonopolyZapped, the iPhone application turns the iOS device into abanking unit, which makes counting a lot less of a hassle.

In general, online gaming integrations with the real worldaugment the user’s gaming experience and offer educationalvalue, targeting in some cases good causes such as envi-ronmental protection or maintaining good health and well-being. Augmented reality would offer in the near future novel,

2These augmented reality games do not constitute WoT applications. Theirexamples are mentioned by the authors only to show the potential of cyber-physical games.

unique gaming experiences blending digital and physicalworld. Examples include mimicking with high accuracy peo-ple’s postures and movements in real life, transforming theminto digital actions in the game [76], or harnessing the nearbyphysical environment as part of the digital game [78], [80].A more futuristic example could be distributed gaming, wherethe players’ avatars are teleported at another player’s placeor in a common place.

H. Transportation

Smart WoT-based mobile applications targeting transporta-tion relate to more informed driving, easier parking and moreconvenient city transit by means of public transport.

VTrack [81] uses position samples from drivers’ phonesto monitor traffic delays. Similarly, Mobile Millennium [82]is a research project that includes a pilot traffic-monitoringsystem that uses the GPS in mobile phones to gather trafficinformation, process it, and distribute it back to the phones inreal time. A road condition monitoring and alert application ispresented in [83], harnessing the compass and connectivityon user’s mobile device to provide road condition basedalerts to the user. Park Here! [84] is a smart parking systembased on smartphones’ embedded sensors and short rangecommunication technologies, towards the automatic detectionof parking actions. Find My Car [85] makes use of GPSservices and internet connectivity to help drivers find theirway back to their parked car.

Waze [86] is perhaps the most popular mobile app forcar drivers, with a community of 50 million users. It pro-vides up-to-date traffic conditions, live-routing and maps,comprehensive voice-assisted navigation, alerts about roadhazards, and even notifications when a Facebook friend isheading towards the same destination. INRIX Traffic [87]offers similar services, automatically learning a user’s drivinghabits, personalizes routes to avoid traffic, recommends tripsand departure times, providing also parking services, based onlocation, vehicle type and parking duration.

Finally, examples of mobile applications that combine user’slocation with public transportation information for convenientcity transit involve GoMetro [88] (Los Angeles), Muni+ [89](San Francisco) and Gothere [90] (Singapore).

Transportation is an area where the authors expect moreresearch efforts to rise, exploiting the sensors of the mobiledevice for providing better and safer driving experience.Commercial mobile apps focusing on driving assistance, suchas Waze and INRIX Traffic are gaining popularity, especiallyin the US.

I. Interaction with Things

This category is the broader one, and relates to efforts tar-geting interaction in general with web-enabled things, locatedin the nearby environment.

At first, a framework for supporting mobile interactions withreal-world objects is presented in [91]. In [92], Frank et al.present the architecture, design and evaluation of a searchsystem that relies on sensor-enabled mobile phones to discoverlost objects. MobileIoT Toolkit [93] connects the Electronic

7

Product Code (EPC) network to mobile phones for reading andinterpreting various tags (RFID, NFC, barcode). Furthermore,a toolkit for barcode recognition and resolving on mobilephones equipped with cameras is presented in [94]. Similarly,BIT [95] is a unified system architecture for providing mobilephone-based digital services related to tagging technologies.Tales of Things [96] is a tagging service that uses QR codesand RFID tags to enable people to attach stories and memoriesto any object. The scanning of readable and writable tagsthrough mobile phones allows stories to be replayed andadded. Scandit [97] is a company that claims to providethe highest quality in mobile barcode scanning solutions forsmartphones, tablets and wearable devices.

Other approaches target the sales market, to improve con-sumer awareness and optimize the sales process. The MobileSales Assistant [98] aims at improving sales in retail stores,by enabling shop assistants to check the availability andstock information of products directly with an NFC-enabledmobile phone. My2cents [99] allows consumers to have accessto comments about a product via their mobile phone andshare their own product experience with other consumers andwithin their social networks. APriori [100] makes productrecommendation available for mobile users, who utilize theirphones to identify tagged consumer products. Also, companiessuch as Square [101] and PayPal [102] offer credit cardpayment systems connected directly to mobile phones.

Interacting with physical things through mobile applicationsincludes a variety of technologies such as RFID, NFC, bar-codes, QR codes, which can be either identified and sensedby the phone’s camera (barcodes, QR codes), or sensedby special equipment (RFID readers, credit card readers).Local information is combined with information from theweb to create advanced knowledge or to implement paymenttransactions.

J. Social Interactions with People

If we broaden the notion of ”things” to include also (nearby)people, then we can refer to BlueAware [103], a mobilephone-based system that uses Bluetooth hardware addressesand a database of user profiles to cue informal, face-to-faceinteractions between nearby users who do not know each other,but probably should, as they have similar profiles, interestsand characteristics. This concept of Familiar Strangers is alsodemonstrated in [104], where Jabberwockies, designed forBluetooth enabled mobile phones, investigate the constraintsof our feelings and affinities with strangers in pubic places.

EmotionSense [105] is a passive monitoring smartphoneapplication that can autonomously capture emotive, behav-ioral, and social signals from smartphone owners. Similarly,SociableSense [105] is a smart phone-based platform forproviding real-time feedback to users to foster and improvesocial interactions. Also, CenceMe [106] is a mobile appli-cation serving a similar goal, by combining the inference ofthe presence of individuals with sharing of this informationthrough social networking applications such as Facebook andMySpace. In a user study where 22 people used CenceMecontinuously over a three week period in a campus town, it

was observed that participants found it as a new way to connectwith people, with the presence of their friends always nearby.

Finally, Kostakos et al. [107] use a Bluetooth-based infras-tructure to identify characteristics of urban environments suchas mobility patterns, social and spatial structures etc. Similarly,the Wireless Rope [108] is a framework to study the notionof social context and the detection of social situations usingBluetooth-based mobile devices for proximity detection, andits effects on group dynamics.

Low-power short-distance wireless technology (e.g. Blue-tooth, RFID) can be used to improve interactions with nearbyindividuals, for improving physical social experiences.

IV. DISCUSSION AND OPEN ISSUES

As the previous section shows, an increasing number ofworks are combining mobile computing and the IoT/WoT toprovide more personalized, enriched services to the users.In this section, the general findings are discussed and openchallenges in this domain are identified.

A. Analysis of Related Work

Through this review of related efforts, the following eightinteraction possibilities have been identified:

1) The built-in sensors of the mobile phone are used ina participatory manner to co-create knowledge on theinternet/web (e.g. [19], [22], [23], [24], [25], [81], [82]).

2) The mobile phone is used as a tool for getting feedbackabout the nearby environmental conditions and/or thepersonal energy footprint of the user using internet/webprotocols (e.g. [28], [10], [29], [30], [31], [32], [34]).

3) The mobile phone uses internet/web principles to controlphysical devices such as electrical appliances, which arelocated either nearby or remotely. (e.g. [43], [44], [45],[46], [47], [29]).

4) The mobile phone acts as an intermediary betweenbody sensors and the internet/web, for health/sportsapplications (e.g. [49], [50], [51], [54], [61], [63], [66]).

5) The built-in sensors of the mobile phone are used toenhance the gaming experience of the user in web-based(e.g. [77]), mobile (e.g. [78], [79]) or cyber-physicalgames (e.g. [80]).

6) The mobile phone interacts with real-world objects,using knowledge available on the web to better informthe user about these physical things (e.g. [93], [100],[96], [98], [99]).

7) The mobile phone becomes a credit card payment sys-tem (e.g. [101], [102]).

8) Low-power short-distance wireless communication (e.g.Bluetooth) and information from online social network-ing sites is used to promote face-to-face interactions withnearby mobile users who share similar interests (e.g.[103], [104], [106]).

The most common sensing technologies employed in eachof the ten categories, together with the type of data analysisinvolved, are listed in Table I (see also Figure 1). Thetechnologies employed are either embedded natively ontothe mobile device or involve third-party sensors interacting

8

directly with the mobile phone. Data analysis is performedeither on the mobile phone or on the crowd/web. The latteris needed for more computationally expensive operations anddata integration from various mobile phones/sensors.

The last column of Table I indicates the estimated marketvalue per category in the next 5-10 years combining a numberof sources, including [109], [110], [111], [112]. Accordingto them, several trillion US dollars will be spent on IoTsolutions (indicatively, estimates by [109] $6 trillion, [110]$4-11 trillion, [111] $14.4 trillion). Tech leaders predict thegreatest potential revenue growth for IoT in the next threeyears is in consumer and retail markets (22%), with IoT/WoTexpected to see significant revenue growth in technologyindustries (13%), aerospace and defense (10%), and education(9%). The share of the market for mobile phone computing/IoThas not been analysed to date by industry, at least publicly, butwe expect it will be a sizable share of the IoT market. Ourown assessment is that actuation and control, gaming, healthand transportation are the domains with the highest marketvalue and potential.

In regard to the impact of these categories to people’slives, mobile applications in health, sports, transportation andgaming seem to be highly impactful, especially those that con-stitute commercial products (i.e. Nike+, Apple Watch, SecondLife, Waze, GoMetro). Participatory sensing-based applica-tions have been successfully used only in local projects, ad-dressing particular community problems for specific time. Theengagement of the users in the long run remains questionable.Mobile applications in eco-feedback and actuation/control aregaining momentum, in domains such as home and building au-tomation, and industrial control. Finally, the impact of mobileapps in agriculture and real-life interaction with things/peopleremains to be seen, being low at the moment.

B. Open Challenges

Some general IoT/WoT challenges are discussed in relatedsurveys [12], [13], [14], [15]. Hence, the focus is on particularissues that appear in mobile computing when integrated tothe IoT/WoT, discovered through this study, which are listedbelow.

Heterogeneity, either because of IoT/WoT devices/servicesthat provide proprietary communication protocols or becauseof heterogeneous consumers of data with different require-ments and needs. In the former case, although the communi-cation at the application layer becomes standardized throughthe WoT principles, the lower-level interaction remains achallenge. In the latter scenario, some users might ask for real-time information while some others might need historical data.Their needs regarding data quality, spatial resolution, and sam-pling rates vary, hence IoT/WoT and mobile computing needto be open to support diverse applications and requirements.

Continuous sensing. Some IoT/WoT-enabled mobile appli-cations require continuous sensing, signal processing, analysisand inference which asks for more computation, memory, stor-age, sensing, and communication bandwidth. Recommendedapproaches include sending only compressed summaries overthe air (rather than raw data) or involve delay-tolerant models

adaptable to the application needs. Opportunistic sensing couldbe another option [18], [9], in which sensing is a secondary,low-priority operation on the mobile device.

Crowd sensing. Participatory sensing as well as mobilecrowd sensing have various open issues, which are identifiedin related work [26], [27]. Some of these issues in particularinclude:

• How can the sensing opportunities and sensing quality bemeasured?

• How to deal with incomplete, noisy and unreliable data?• How many mobile users can provide enough sensing

opportunities to achieve the required sensing quality?• How to tackle the fact that people always move around

a set of popular locations, instead of purely randommovements?

• How to tackle the fact that each individual shows prefer-ence for some particular locations?

• How to avoid using up significant battery that couldprevent users from accessing their usual services?

The context problem occurs because phones may beexposed to events for too short a period of time (e.g. if the useris traveling quickly, if the event is local and spontaneous (e.g.a sound) or the sensor requires more time to gather a sample(e.g. air quality sensor). Phone calls could also interfere incollecting relevant context. An idea would be to leverage co-located mobile phones for localized participatory sensing.

Security is a top priority in mobile computing, and be-comes more important when interacting with the IoT/WoT.One practical consideration is ensuring security of sharedresources (e.g. WoT-enabled devices/services) against misuseby unauthorized mobile users. Relay infrastructures for secureaccess to embedded systems such as Yaler [113] could formpart of the solution here.

Privacy in terms of sharing personal information throughthe web is one of the most important research challenges. Howmuch privacy do applications such as CrossCheck [61] and thework in [54] provide? Additionally, various privacy threatshave been reported in mobile RFID applications [114]. Someapproaches to address privacy include AnonySense [115],which suggests a notion of anonymity through k-anonymoustasking and Social Access Controller (SAC) [116], which is anauthentication proxy between (mobile) users and smart things.Perhaps more interesting is the work in [117], preserving theanonymity of sensor reports without reducing the precision oflocation data.

Reliability becomes important especially in health moni-toring applications that involve body area networks. In thesecases, measurements need to be reliable and precise and thecorresponding mobile applications needs to be able to identifyanomalies and faults at the data, which could create falsealarms. These concerns are discussed in [52] too. Reliabilityrelates also to how well mobile phones can interpret humanbehavior (e.g. sitting, sleeping, talking) from low-level mul-timodal sensor data, or similarly how accurately can theyinfer the surrounding context (e.g. pollution, weather, noiseenvironment) from low-level measurements. UrbanRadar [11]proposed urban mashups as a way to infer various environ-mental conditions from more basic WoT-based services.

9

Category Popular technologies used Type of data analysis Market valueParticipatorySensing

GPS, camera, microphone. Location-based search, map visualizations, in-formation sharing.

N/A

Eco-Feedback

Camera, energy monitors, smartmeters, barcodes, environmentalsensors e.g. air quality.

Historical comparisons, information retrieval,stream data processing.

$200-750Billions

Actuation andControl

Smart electricity meters, smart ap-pliances, light switches, smart fac-tory sensors/actuators.

Rule-based inference, adaptive reasoning, opti-mizations.

$1.3-4.6Trillions

Health Body area sensors, GPS, micro-phone, accelerometer, communica-tion and conservation sensing (e.g.phone calls, SMS).

Personalization and profiling, anomaly detec-tion, emotions detection, stream data process-ing, machine learning.

$110-900Billions

Sports Body area sensors, motion sensors,GPS, pressure (shoe) sensors, pe-dometers.

Personalization and profiling, historical com-parisons, performance visualizations, statistics,information sharing.

$200-450Billions

Agriculture Wearable collars, GPS, barcodes,RFID tags, plant (soil) sensors.

Historical comparisons, information retrieval,stream data processing, optimizations.

$140-200Billions

Gaming Accelerometer, gyrometer, camera,pedometer, barcodes, RFID tags.

Activity recognition, machine learning, infor-mation sharing, image and video processing.

$450-635Billions

Transportation GPS, compass, carrier connectivity,RFID tags.

Location-based search, big data analysis, streamdata processing, anomaly detection, machinelearning, information sharing, image and videoprocessing.

$500-740Billions

Interactionwith Things

NFC technologies, RFID tags, bar-codes, QR codes, credit card de-vices.

Location-based search, information retrieval,information sharing, personalization and profil-ing, recommender systems, transactions.

$70-150Billions

SocialInteractionswith People

Bluetooth, GPS, camera, micro-phone.

Location-based search, information retrievaland sharing, personalization and profiling, emo-tions detection, recommender systems.

$170-450Billions

TABLE ISENSING TECHNOLOGIES AT EACH CATEGORY OF IOT/WOT-BASED MOBILE APPLICATIONS.

Search and Discovery is crucial for locating nearby, localand/or relevant real-world devices (and/or people), exploitingtheir services for the creation of more advanced knowledge.Various works and applications such as [35], [11], [29], [103],[104], [106] require discovery of relevant physical devices.The WoT has not yet standardized any technique for real-time discovery of physical entities and their services, but earlyefforts in this direction have been proposed [118], [119].

Personalization is expected to give significance to theinteraction with the IoT/WoT through mobile computing.Planty [46] could understand the user’s sense of aestheticsand help him create his personal garden. Sports apps [66],[67], [68], [70] could record the personal characteristics andachievements of the users to suggest realistic short- and long-term goals. My2cents [99] could record the buying habits ofpeople and propose them relevant nearby offers. Imagine alsowalking into a pharmacy and your phone suggesting vitaminsand supplements, depending on your health monitoring met-rics, as recorded by body area sensors [50], [51], [52]. Mobilecomputing, personalized to a user’s profile, empowers him/herto make more informed decisions across a spectrum of WoT-enabled services.

Emotions analysis relates to personalization as it allows tooffer better services to users according to their reactions inrelation to the features provided by the mobile applications.The most important efforts in understanding emotions andmeasuring feelings by monitoring the affective states of themobile user are discussed in [120], concluding that affectivesensing offers exciting research opportunities. Research in thisarea involves sensing unconscious emotions too [121], aimingto quantify and log aspects of our behavior we are not aware

of. Mood recognition is applicable in working environmentstoo to benefit employees’ health and productivity [58].

Persuasion in mobile computing is an open area, cre-ating design elements and exploiting various psychologi-cal/sociological factors which affect users to change theirbehavior towards better quality of living, more sustainable life,environmental awareness etc. [122], [123]. The connectionbetween mobile computing and IoT/WoT offers great newopportunities for examining which design characteristics ofmobile phone applications and which motivational factors mayinfluence users to question and change their behavior and wayof living. As an early effort, the work in [38] identified 11design principles for remote sensing mobile applications. Abig challenge related to persuasion is how to address the factthat people tend to lose their interest after being exposed tothe feedback for some weeks or months [40].

Big data analysis is a necessary step in most participatorysensing applications, in order to extract useful informationfrom vast amounts of real-world data streams. Related tothis, an open, standard methodology for IoT/WoT analyticsproblems is still non-existent. Important to researchers in thisdomain is the availability of relevant datasets. CrowdSignals.io[124] aims to create the largest set of rich, longitudinal mobileand sensor data recorded from smartphones and smartwatches,including geo-location, sensor, system and network logs, userinteractions, social connections, and communications as wellas user-provided ground truth labels and survey feedback.

Business cases are required to prove the market potentialof the IoT/WoT and mobile computing, showcasing that themerging of digital and physical world through web tech-nologies and mobile computing is profitable to enterprises

10

worldwide. Evrythng [125] is a start-up company aiming todemonstrate, through various case studies [126], the economicviability of IoT/WoT-based investments. Solutions offeredinclude brand protection, inventory management, consumerloyalty and engagement, data monitoring and analysis etc.

Each of the issues listed above affect in a varying degree theten different categories in which related work was divided inthis survey. According to the authors’ point of view, based onthe overall research performed in this paper, the relevance andimportance of each of the aforementioned issues in relationto each category is assessed in Table II, according to thefollowing scale: Essential (E), Very Important (VI), Important(I), Less Important (LI) and Not Applicable (NA).

Observing Table II, aspects of security, privacy and relia-bility are essential in ”sensitive” domains such as health andtransportation. Search and discovery is relevant only wheninteracting with things or people, while big data analysismakes sense in open systems, when large streams of data areinvolved. Heterogeneity and context are important in almostevery category, while business cases are needed across thewhole mobile IoT/WoT realm, not only to prove the value ofmobile IoT/WoT but also to analyze the barriers for marketuptake of various solutions especially in health, gaming andsports. Moreover, personalization, persuasion and emotionsanalysis are relevant in applications involving feedback, suchas in eco-feedback, health, sports and transportation.

Finally, other mobile phone application domains which willpotentially be important in the future include smart cities[127] (i.e. more interactive and informed mobile phone assis-tants), tourism [128] (i.e. personalized location-based touristguides), micro-grids of electricity [129] (i.e. users could tradeelectricity produced by local renewable energy sources) andsmart metering [130], [131], disaster management [132] (i.e.mobile phones could form ad hoc wireless networks contribut-ing to rescue operations), collaborative economy [133] (i.e.users share things and services and perform legal transactionsthrough their mobile phones) and nano-networking [134] (i.e.nano-sensors inside the human body would collaborativelycollect health-related data, transmitting it to the user’s mobiledevice for anomaly detection).

V. CONCLUSION

This paper performed a survey on the most significantefforts in the area of mobile computing combined with theIoT/WoT, an exciting research domain in which mobile phoneapplications exploit the sensing of the real world throughinternet/web technologies for providing better information andmore advanced knowledge to the user, helping him/her to takemore informed decisions during everyday life.

More than 100 papers have been identified, analyzed anddivided in ten different categories, which represent either thearea of application (i.e. health, sports, gaming, transporta-tion, agriculture), the nature of interaction (i.e. participatorysensing, eco-feedback, actuation and control) or the com-municating actors involved (i.e. things, people). Open issuesand research challenges at this research domain have beenidentified and discussed.

Summing up, the practice of combining mobile computingand the IoT/WoT seems to offer tremendous new opportunitiesin many real-life domains, enabling the seamless integrationof devices, services and information that can create advancedknowledge, more informed reasoning and decision-making,as well as encouraging big data analysis, more use of per-suasive and engagement techniques, emotions analysis andpersonalization. However, this openness comes together withvarious risks in privacy, security and reliability of information,and also various challenges such as search and discovery ofdevices, services and information on the fly, which is stillan open issue in the WoT world [118]. Finally, the mostcritical element seems to be a need for large business caseswhich could prove the market value of mobile computing andthe IoT/WoT, and which could eventually lead to wide-scaleadoption of this practice, inspiring even more research anddevelopment in satisfying the risks and challenges at this field.

VI. ACKNOWLEDGEMENTS

This research has been partially supported by Science Foun-dation Ireland (SFI) under grant No. SFI/12/RC/2289 and EUFP7 CityPulse Project under grant No.603095. http://www.ict-citypulse.eu

REFERENCES

[1] Lars Schor, Philipp Sommer, and Roger Wattenhofer. Towards aZero-Configuration Wireless Sensor Network Architecture for SmartBuildings. In Proc. BuildSys, Berkeley, California, USA, 2009.

[2] Dogan Yazar and Adam Dunkels. Efficient Application Integration inIP-based Sensor Networks. In First Workshop On Embedded Sens-ing Systems For Energy-Efficiency In Buildings (BuildSys), Berkeley,California, November 2009. ACM.

[3] Neil Gershenfeld, Raffi Krikorian, and Danny Cohen. The Internet ofThings. Scientific American, 291(4):76–81, 2004.

[4] Adam Dunkels, Thiemo Voigt, and Juan Alonso. Making TCP/IPViable for Wireless Sensor Networks. In Proceedings of InternationalConference on Embedded Wireless Systems and Networks (EWSN),Berlin, Germany, 2004.

[5] Jonathan W. Hui and David E. Culler. IP is dead, long live IP forwireless sensor networks. In Proceedings of the Sixth Conference onNetworked Embedded Sensor Systems (SenSys), pages 15–28, Raleigh,NC, USA, November 2008. ACM.

[6] Erik Wilde. Putting things to REST. Technical Report UCB iSchoolReport 2007-015, School of Information, UC Berkeley, November2007.

[7] Dominique Guinard, Vlad Trifa, and Erik Wilde. Architecting aMashable Open World Wide Web of Things. Technical Report No.663, Dept. of Computer Science, ETH Zurich, February 2010.

[8] Andreas Kamilaris, Andreas Pitsillides, and Michalis Yiallouros. Build-ing Energy-aware Smart Homes using Web Technologies. Journal ofAmbient Intelligence and Smart Environments (JAISE), 5(2):161–186,2013.

[9] Nicholas D Lane, Emiliano Miluzzo, Hong Lu, Daniel Peebles,Tanzeem Choudhury, and Andrew T Campbell. A survey of mobilephone sensing. IEEE Communications Magazine, 48(9):140–150, 2010.

[10] Marco Jahn, Marc Jentsch, Christian R. Prause, Ferry Pramudianto,and Amro Al-Akkad andRen Reiners. The Energy Aware SmartHome. Proceedings of the Fifth International Conference on FutureInformation Technology (FutureTech), pages 1–8, May 2010.

[11] Andreas Kamilaris, Nicolas Iannarilli, Vlad Trifa, and Andreas Pit-sillides. Bridging the Mobile Web and the Web of Things in UrbanEnvironments. In First Workshop on the Urban Internet of Things, inProc. of IoT 2010, Tokyo, Japan, 2011.

[12] Deze Zeng, Song Guo, and Zixue Cheng. The web of things: A survey.Journal of Communications, 6(6):424–438, 2011.

[13] Tein-Yaw Chung, Ibrahim Mashal, Osama Alsaryrah, Van Huy, Wen-Hsing Kuo, and Dharma P Agrawal. Social web of things: a survey. InProceedings of International Conference on Parallel and DistributedSystems (ICPADS), pages 570–575. IEEE, 2013.

11

Issue /Category

ParticipatorySensing

Eco-Feedback

Actuation andControl

Health Sports Agriculture

Gaming

Transportation

Interactionwith

Things

SocialInteractions

withPeople

Heterogeneity

I I VI E LI E LI E VI VI

Continuoussensing

E VI LI E VI VI VI E VI VI

Contextproblem

E I LI E I I LI E I I

Security LI E E E I VI LI E VI VIPrivacy VI E E E VI VI I E VI E

Reliability VI VI E E VI VI I E I ISearch andDiscovery

I I I LI LI LI LI VI E E

Personalization

LI I LI E E I VI I VI VI

Emotionsanalysis

NA I NA VI VI NA I VI VI VI

Persuasion I E LI E I I LI I I IBig dataanalysis

E VI VI E I E I E I LI

Businesscases

VI E E E VI E VI E E E

TABLE IIOPEN CHALLENGES AND ISSUES AFFECTING EACH CATEGORY OF IOT/WOT-BASED MOBILE APPLICATIONS.

[14] Luigi Atzori, Antonio Iera, and Giacomo Morabito. The internet ofthings: A survey. Computer networks, 54(15):2787–2805, 2010.

[15] Gerben G Meyer, Kary Framling, and Jan Holmstrom. Intelligentproducts: A survey. Computers in industry, 60(3):137–148, 2009.

[16] Huasong Cao, Victor Leung, Cupid Chow, and Henry Chan. Enablingtechnologies for wireless body area networks: A survey and outlook.IEEE Communications Magazine, 47(12):84–93, 2009.

[17] Alan Chamberlain, Mark Paxton, Kevin Glover, Martin Flintham,Dominic Price, Chris Greenhalgh, Steve Benford, Peter Tolmie, EimanKanjo, Amanda Gower, et al. Understanding mass participatorypervasive computing systems for environmental campaigns. Personaland Ubiquitous Computing, 18(7):1775–1792, 2014.

[18] Andrew T. Campbell, Shane B. Eisenman, Nicholas D. Lane, EmilianoMiluzzo, Ronald A. Peterson, Hong Lu, Xiao Zheng, Mirco Musolesi,Kristof Fodor, and Gahng-Seop Ahn. The rise of people-centricsensing. IEEE Internet Computing, 12(4):12–21, 2008.

[19] J. Goldman and et al. Shilton, K. Participatory sensing: A citizen-powered approach to illuminating the patterns that shape our world.Woodrow Wilson International Center for Scholars.

[20] Olfasense GmbH. Odourmap Citizen Monitoring App, 2016.https://play.google.com/store/apps/details?id=de.ecolony.odourmap.

[21] Adriano Faggiani, Enrico Gregori, Luciano Lenzini, Valerio Luconi,and Alessio Vecchio. Smartphone-based crowdsourcing for networkmonitoring: Opportunities, challenges, and a case study. IEEE Com-munications Magazine, 52(1):106–113, 2014.

[22] Nicolas Maisonneuve, Matthias Stevens, Maria E. Niessen, PeterHanappe, and Luc Steels. Citizen noise pollution monitoring. InProceedings of the 10th Annual International Conference on DigitalGovernment Research, pages 96–103, 2009.

[23] Wesley Willett, Paul Aoki, Neil Kumar, Sushmita Subramanian, andAllison Woodruff. Common sense community: Scaffolding mobilesensing and analysis for novice users. In Pervasive Computing, pages301–318. Springer Berlin Heidelberg, 2010.

[24] Stacey Kuznetsov and Eric Paulos. Participatory Sensing in PublicSpaces: Activating Urban Surfaces with Sensor Probes. In ACMDesigning Interactive Systems (DIS), Aarhus, Denmark, 2010.

[25] AP Kassianos, JD Emery, P Murchie, and FM Walter. Smartphone ap-plications for melanoma detection by community, patient and generalistclinician users: a review. British Journal of Dermatology, 2015.

[26] Salil S Kanhere. Participatory sensing: Crowdsourcing data frommobile smartphones in urban spaces. In Proceedings of the 12th Inter-national Conference on Mobile Data Management (MDM), volume 2,pages 3–6. IEEE, 2011.

[27] Huadong Ma, Dong Zhao, and Peiyan Yuan. Opportunities in mobilecrowd sensing. IEEE Communications Magazine, 52(8):29–35, 2014.

[28] Giulio Jacucci, Anna Spagnolli, Luciano Gamberini, Alessandro Cha-lambalakis, and Christoffer Bjorkskog. Designing Effective feedbackof Electricity Consumption for Mobile User Interfaces. PsychNologyJournal, 7(3):265–289, 2009.

[29] Dominique Guinard and Vlad Trifa. Towards the web of things: Webmashups for embedded devices. In Workshop on Mashups, EnterpriseMashups and Lightweight Composition on the Web (MEM), Proc. ofWWW, page 15, Madrid, Spain, 2009.

[30] Friedemann Mattern, Thorsten Staake, and Markus Weiss. ICT forGreen: How computers can help us to conserve energy. In Proceedingsof the 1st International Conference on Energy-Efficient Computing andNetworking, pages 1–10. ACM, 2010.

[31] Markus Weiss, Adrian Helfenstein, Friedemann Mattern, and ThorstenStaake. Leveraging smart meter data to recognize home appliances. InProceedings of the Conference on Pervasive Computing and Commu-nications (PerCom), pages 190–197. IEEE, 2012.

[32] Andreas Kamilaris, , Andreas Pitsillides, and Christos Fidas. SocialElectricity: A Case Study on Users Perceptions in Using GreenICT Social Applications. International Journal of Environment andSustainable Development (IJESD), 15(1):67–88, January 2016.

[33] OPOWER. Social Application, 2016. https://social.opower.com/.[34] Markus Weiss, Adrian Merkle, Thorsten Staake, and Elgar Fleisch.

Towards a powerpedia-a collaborative energy encyclopedia. In FirstWorkshop on Ubiquitous Computing for Sustainable Energy, Proc. ofUbicomp 2010, 2010.

[35] CubeSensors. Make your home healthier, your office more productive,2016. https://cubesensors.com.

[36] FitNow, Inc. LoseIt, 2016. http://loseit.com/.[37] Richard Honicky, Eric A Brewer, Eric Paulos, and Richard White. N-

smarts: networked suite of mobile atmospheric real-time sensors. InProceedings of the Second ACM SIGCOMM workshop on Networkedsystems for developing regions, pages 25–30. ACM, 2008.

[38] Andreas Kamilaris and Andreas Pitsillides. The Impact of RemoteSensing on the Everyday Lives of Mobile Users in Urban Areas. InProceedings of the 7th International Conference on Mobile Computingand Ubiquitous Networking (ICMU), Singapore, jan 2014.

[39] Sarah Darby. The effectiveness of feedback on energy consumptionA review for defra of the literature on metering, billing and directdisplays. Environmental Change Institute, University of Oxford, 2006.

[40] Sarah Darby. Making it Obvious: Designing Feedback into EnergyConsumption, pages 685–696. Springer Berlin Heidelberg, 2001.

[41] Matthias Kovatsch, Markus Weiss, and Dominique Guinard. Embed-ding Internet Technology for Home Automation. In Proceedings of the15th Conference on Emerging Technologies and Factory Automation(ETFA), Bilbao, Spain, September 2010. IEEE.

12

[42] Carles Gomez and Josep Paradells. Wireless home automation net-works: A survey of architectures and technologies. IEEE Communica-tions Magazine, 48(6):92–101, 2010.

[43] Lynn Bartram, Johnny Rodgers, and Rob Woodbury. Supportingsustainable living: aware homes and smart occupants. In Proceedingsof the International Joint Conference on Pervasive and UbiquitousComputing (UbiComp), 2010.

[44] Markus Weiss, Friedemann Mattern, Tobias Graml, Thorsten Staake,and Elgar Fleisch. Handy feedback: Connecting smart meters withmobile phones. In Proceedings of the 8th international conference onmobile and ubiquitous multimedia, page 15. ACM, 2009.

[45] EnOcean. Sensors for smart homes and the Internet of Things, 2016.https://www.enocean.com/en/.

[46] N.Thing. Planty, 2016. https://nthing.net/planty.[47] Dominique Guinard. Mashing up your web-enabled home. In

Proceedings of the 10th International Conference on Current Trendsin Web Engineering (ICWE), pages 442–446, Vienna, Austria, 2010.Springer-Verlag.

[48] Chris Otto, Aleksandar Milenkovic, Corey Sanders, and Emil Jovanov.System architecture of a wireless body area sensor network for ubiqui-tous health monitoring. Journal of Mobile Multimedia, 1(4):307–326,2006.

[49] Lin Zhong, Mike Sinclair, and Ray Bittner. A phone-centered bodysensor network platform cost, energy efficiency & user interface. InInternational Workshop on Wearable and Implantable Body SensorNetworks (BSN). IEEE, 2006.

[50] Wan-Young Chung, Chiew-Lian Yau, Kwang-Sig Shin, and Risto Myl-lyla. A cell phone based health monitoring system with self analysisprocessor using wireless sensor network technology. In Proceedings ofthe 29th Annual International Conference of the IEEE Engineering inMedicine and Biology Society (EMBS), pages 3705–3708. IEEE, 2007.

[51] Christian Seeger, Alejandro Buchmann, and Kristof Van Laerhoven.myhealthassistant: A phone-based body sensor network that capturesthe wearer’s exercises throughout the day. In Proceedings of the 6thInternational Conference on Body Area Networks (BodyNets), Brussels,Belgium, 2011.

[52] Joel Rodrigues, Orlando Pereira, and Paulo Neves. Biofeedback datavisualization for body sensor networks. Journal of Network andComputer Applications, 34(1):151–158, 2011.

[53] Mohan Prakash, U Gowshika, and T Ravichandran. A Smart DeviceIntegrated with an Android for Alerting a Persons Health Condition:Internet of Things. Indian Journal of Science and Technology, 9(6),2016.

[54] Anmol Madan, Manuel Cebrian, David Lazer, and Alex Pentland.Social sensing for epidemiological behavior change. In Proceedingsof the 12th ACM International Conference on Ubiquitous Computing,pages 291–300. ACM, 2010.

[55] Sohrab Saeb, Mi Zhang, Christopher J Karr, Stephen M Schueller,Marya E Corden, Konrad P Kording, and David C Mohr. Mobilephone sensor correlates of depressive symptom severity in daily-lifebehavior: an exploratory study. Journal of medical Internet research,17(7), 2015.

[56] Robbert Jan Beun. Persuasive strategies in mobile insomnia therapy:alignment, adaptation, and motivational support. Personal and Ubiq-uitous Computing, 17(6):1187–1195, 2013.

[57] Mikhail Sysoev, Andrej Kos, and Matevz Pogacnik. Noninvasive stressrecognition considering the current activity. Personal and UbiquitousComputing, 19(7):1045–1052, 2015.

[58] Alexandros Zenonos, Aftab Khan, Georgios Kalogridis, Stefanos Vat-sikas, Tim Lewis, and Mahesh Sooriyabandara. HealthyOffice: Moodrecognition at work using smartphones and wearable sensors. InProceedings of the International Conference on Pervasive Computingand Communication Workshops, pages 1–6. IEEE, 2016.

[59] Yoshiyuki Kawano and Keiji Yanai. Foodcam: A real-time mobile foodrecognition system employing fisher vector. In MultiMedia Modeling,pages 369–373. Springer, 2014.

[60] Tauhidur Rahman, Alexander T Adams, Mi Zhang, Erin Cherry, BobbyZhou, Huaishu Peng, and Tanzeem Choudhury. Bodybeat: A mobilesystem for sensing non-speech body sounds. In Proceedings of the 12thAnnual International Conference on Mobile systems, Applications, andServices, pages 2–13. ACM, 2014.

[61] Northwestern Universitys Center for Behavioral Inter-vention Technologies. The CrossCheck study, 2016.http://spectrum.ieee.org/biomedical/diagnostics/smartphone-app-keeps-watch-over-schizophrenic-patients.

[62] Partha Pratim Ray. Internet of Things for Sports (IoTSport): An archi-tectural framework for sports and recreational activity. In Proceedings

of the Conference on Electrical, Electronics, Signals, Communicationand Optimization (EESCO), pages 1–4, January 2015.

[63] Shane B Eisenman, Emiliano Miluzzo, Nicholas D Lane, Ronald APeterson, Gahng-Seop Ahn, and Andrew T Campbell. Bikenet: A mo-bile sensing system for cyclist experience mapping. ACM Transactionson Sensor Networks (TOSN), 6(1):6, 2009.

[64] Genevieve Fridlund Dunton, Eldin Dzubur, Keito Kawabata, BrendaYanez, Bin Bo, and Stephen Intille. Development of a smartphoneapplication to measure physical activity using sensor-assisted self-report. Emerging Technologies to Promote and Evaluate PhysicalActivity, pages 88–100, 2014.

[65] Kasidit Wijitsopon, Chavalit Panichayanubal, and Pusadee Seresang-takul. Fit Buddy: A Mobile Application for Fitness Tracking Usingthe Always-On Low-Power Sensor. In Mobile and Wireless Technology2015, pages 155–162. Springer, 2015.

[66] Nike. Nike+ Sport Sensors, 2016. http://www.nikeplus.com.br/.[67] Apple. Apple Watch, 2016. http://www.apple.com/watch/.[68] Garmin. Watches and Wearable Technology, 2016.

https://buy.garmin.com/en-US/US/wearables/c10002-p1.html.[69] Samsung. Gear S2, 2016. http://www.samsung.com/us/explore/gear-

s2/.[70] Suunto. Sports watches, dive computers, compasses and precision

instruments, 2016. http://www.suunto.com/.[71] Rodolfo Bongiovanni and Jess Lowenberg-DeBoer. Precision agricul-

ture and sustainability. Precision agriculture, 5(4):359–387, 2004.[72] Herdwatch, 2016. https://www.herdwatch.ie/.[73] DairyMaster. MooMonitor+, 2016. http://www.dairymaster.com/heat-

detection/.[74] Map your meal Europe Aid funded project, 2016.

http://www.mapyourmeal.org/.[75] FoodLoop GmbH. FoodLoop, 2016. https://www.foodloop.net/en/.[76] Second Life. The largest-ever 3D virtual world created entirely by its

users, 2016. http://secondlife.com/.[77] Mirco Musolesi, Emiliano Miluzzo, Nicholas D Lane, Shane B Eisen-

man, Tanzeem Choudhury, and Andrew T Campbell. The second lifeof a sensor: Integrating real-world experience in virtual worlds usingmobile phones. Technology, 100:3, 2008.

[78] Leila Alem and Weidong Huang. Recent trends of mobile collaborativeaugmented reality systems. Springer Science & Business Media, 2011.

[79] James J Lin, Lena Mamykina, Silvia Lindtner, Gregory Delajoux, andHenry B Strub. Fishnsteps: Encouraging physical activity with aninteractive computer game. In Proceedings of the International JointConference on Pervasive and Ubiquitous Computing (UbiComp), pages261–278. Springer, 2006.

[80] Hasbro. Game of Life and Monopoly get zAPPed, 2016.http://www.ubergizmo.com/2012/02/game-of-life-and-monopoly-get-zapped/.

[81] Arvind Thiagarajan, Lenin Ravindranath, Katrina LaCurts, SamuelMadden, Hari Balakrishnan, Sivan Toledo, and Jakob Eriksson. Vtrack:accurate, energy-aware road traffic delay estimation using mobilephones. In Proceedings of the 7th ACM Conference on EmbeddedNetworked Sensor Systems, pages 85–98. ACM, 2009.

[82] UC Berkeley, Nokia and NAVTEQ. Mobile Millennium, 2016.http://traffic.berkeley.edu/.

[83] Avik Ghose, Provat Biswas, Chirabrata Bhaumik, Monika Sharma,Arpan Pal, and Abhinav Jha. Road condition monitoring and alertapplication: Using in-vehicle smartphone as internet-connected sensor.In Pervasive Computing and Communications Workshops, in Proc. ofPERCOM, pages 489–491. IEEE, 2012.

[84] Rosario Salpietro, Luca Bedogni, Marco Di Felice, and LucianoBononi. Park here! a smart parking system based on smartphones’embedded sensors and short range communication technologies. InProceedings of the IEEE World Forum on Internet of Things (WF-IoT),pages 18–23, 2015.

[85] eLibera. Find My Car, 2016. http://www.elibera.com/eLibera/libapps-en.html.

[86] Waze Mobile. Waze, 2016. https://www.waze.com/.[87] INRIX. INRIX Traffic, 2016. http://inrix.com/mobile-apps/.[88] Metro. GoMetro, 2016. https://www.metro.net/mobile/metro-mobile-

app/app-info/.[89] Sky Highways. Muni+, 2016. http://muniapp.us/.[90] Gothere, 2016. http://gothere.sg/iphone.[91] Enrico Rukzio, Sergej Wetzstein, and Albrecht Schmidt. A framework

for mobile interactions with the physical world. Proceedings of WirelessPersonal Multimedia Communication (WPMC), 2005.

13

[92] Christian Frank, Philipp Bolliger, Christof Roduner, and WolfgangKellerer. Objects calling home: Locating objects using mobile phones.In Pervasive Computing, pages 351–368. Springer, 2007.

[93] Dominique Guinard, Felix Von Reischach, Florian Michahelles, andElgar Fleisch. MobileIoT Toolkit: Connecting the EPC Network toMobilePhones. In Proceedings of the Conference on Human-ComputerInteraction with Mobile Devices and Services (MobileHCI), pages 115–126, 2008.

[94] Robert Adelmann, Marc Langheinrich, and Christian Florkemeier.Toolkit for bar code recognition and resolving on camera phones -jump starting the internet of things. GI Jahrestagung (2), 94:366–373,2006.

[95] Christof Roduner and Marc Langheinrich. BIT A framework andarchitecture for providing digital services for physical products. InProceedings of the Internet of Things Conference (IOT), pages 1–8,Tokyo, Japan, 2010. IEEE.

[96] Ralph Barthel, Andrew Hudson-smith, Martin De Jode, and BenjaminBlundell. Tales of things the internet of oldthings: Collecting storiesof objects, places and spaces. In Proceedings of the Internet of ThingsConference (IOT), Tokyo, Japan, 2010. IEEE.

[97] Scandit. Barcode scanning evolved, 2016. http://www.scandit.com/.[98] S. Karpischek, F. Michahelles, F. Resatsch, and E. Fleisch. Mobile

sales assistant - an nfc-based product information system for retailers.In First International Workshop on Near Field Communication (NFC),pages 20–23, February 2009.

[99] Stephan Karpischek and Florian Michahelles. my2centsdigitizingconsumer opinions and comments about retail products. In Proceedingsof the Internet of Things Conference (IOT), pages 1–7, Tokyo, Japan,2010. IEEE.

[100] Felix Von Reischach and Florian Michahelles. Apriori: A ubiquitousproduct rating system. In 4th International Workshop on PervasiveMobile Interaction Devices (PERMID), Proc. of Pervasive, 2008.

[101] Square. Square Register, 2016. https://squareup.com/pos.[102] PayPal Here, 2016. https://www.paypal.com/webapps/mpp/credit-card-

reader.[103] Nathan Eagle and Alex Pentland. Social serendipity: Mobilizing social

software. IEEE Pervasive Computing, 4(2):28–34, 2005.[104] Eric Paulos and Elizabeth Goodman. The familiar stranger: anxiety,

comfort, and play in public places. In Proceedings of the SIGCHIConference on Human Factors in Computing Systems, pages 223–230.ACM, 2004.

[105] Neal Lathia, Veljko Pejovic, Kiran K Rachuri, Cecilia Mascolo, MircoMusolesi, and Peter J Rentfrow. Smartphones for large-scale behaviorchange interventions. IEEE Pervasive Computing, (3):66–73, 2013.

[106] Emiliano Miluzzo, Nicholas D Lane, Kristof Fodor, Ronald Peterson,Hong Lu, Mirco Musolesi, Shane B Eisenman, Xiao Zheng, andAndrew T Campbell. Sensing meets mobile social networks: thedesign, implementation and evaluation of the cenceme application.In Proceedings of the 6th ACM Conference on Embedded NetworkedSensor Systems, pages 337–350. ACM, 2008.

[107] Vassilis Kostakos, Tom Nicolai, Eiko Yoneki, Eamonn O’Neill, HolgerKenn, and Jon Crowcroft. Understanding and measuring the urbanpervasive infrastructure. Personal Ubiquitous Computing, 13(5):355–364, 2009.

[108] Tom Nicolai, Eiko Yoneki, Nils Behrens, and Holger Kenn. Exploringsocial context with the wireless rope. In Workshop on the Move toMeaningful Internet Systems (OTM), pages 874–883. Springer, 2006.

[109] John Greenough. The Internet of Things Will Be TheWorld’s Most Massive Device Market. Business Insider UK,2014. http://uk.businessinsider.com/how-the-internet-of-things-market-will-grow-2014-10.

[110] James Manyika, Michael Chui, Peter Bisson, Jonathan Woetzel,Richard Dobbs, Jacques Bughin, and Dan Aharon. Unlockingthe potential of the Internet of Things. McKinsey Insights, 2015.http://www.mckinsey.com/business-functions/business-technology/our-insights/the-internet-of-things-the-value-of-digitizing-the-physical-world.

[111] Joseph Bradley, Joel Barbier, and Doug Handler. Embracing theInternet of Everything to Capture Your Share of $14.4 Trillion. CiscoWhite Paper, 2013. http://www.cisco.com/c/dam/en us/about/ac79/docs/innov/IoE Economy.pdf.

[112] Harman International. Projections for 2025, 2016.http://www.harman.com/.

[113] Yaler. A relay infrastructure for secure access to embedded systems,2016. https://www.yaler.net/.

[114] Hyangjin Lee and Jeeyeon Kim. Privacy threats and issues in mobilerfid. In Proceedings of the First International Conference on Avail-ability, Reliability and Security (ARES). IEEE, 2006.

[115] Cory Cornelius, Apu Kapadia, David Kotz, Dan Peebles, Minho Shin,and Nikos Triandopoulos. Anonysense: privacy-aware people-centricsensing. In Proceedings of the 6th International Conference on MobileSystems, Applications, and Services, pages 211–224. ACM, 2008.

[116] Dominique Guinard, Mathias Fischer, and Vlad Trifa. Sharing usingsocial networks in a composable web of things. In Proceedingsof the 8th Conference on Pervasive Computing and Communications(PERCOM), pages 702–707. IEEE, 2010.

[117] Karen P Tang, Pedram Keyani, James Fogarty, and Jason I Hong.Putting people in their place: an anonymous and privacy-sensitiveapproach to collecting sensed data in location-based applications. InProceedings of the SIGCHI conference on Human Factors in Comput-ing Systems, pages 93–102. ACM, 2006.

[118] Andreas Kamilaris, Koula Papakonstantinou, and Andreas Pitsillides.Exploring the Use of DNS as a Search Engine for the Web of Things.In Proceedings of the IEEE World Forum on Internet of Things (WF-IoT), Seoul, South Korea, March 2014.

[119] Benedikt Ostermaier, Kay Romer, Friedemann Mattern, MichaelFahrmair, and Wolfgang Kellerer. A Real-Time Search Engine for theWeb of Things. In Proceedings of the Internet of Things Conference,Tokyo, Japan, December 2010.

[120] Eiman Kanjo, Luluah Al-Husain, and Alan Chamberlain. Emotions incontext: examining pervasive affective sensing systems, applications,and analyses. Personal and Ubiquitous Computing, 19(7):1197–1212,2015.

[121] Leonid Ivonin, Huang-Ming Chang, Wei Chen, and Matthias Rauter-berg. Unconscious emotions: quantifying and logging something weare not aware of. Personal and Ubiquitous Computing, 17(4):663–673,2013.

[122] Jon Froehlich, Leah Findlater, and James Landay. The Design of Eco-feedback Technology. In Proceedings of the SIGCHI Conference onHuman Factors in Computing Systems, CHI ’10, pages 1999–2008,New York, NY, USA, 2010. ACM.

[123] Brian J. Fogg. Persuasive technology: Using computers to change whatwe think and do. Ubiquity, 2002(5), December 2002.

[124] CrowdSignals.io. Community’s Largest Labeled Mobile and SensorDataset, 2016. http://crowdsignals.io/.

[125] Evrythng. IoT Smart Products Platform, 2016. https://evrythng.com/.[126] Evrythng News. Evrythngs partnership with Avery Dennison,

2016. https://evrythng.com/big-switch-10-billion-apparel-industry-products-born-digital-digital-identities/.

[127] CityPulse EU FP7 Project. Demonstrations, 2016. http://www.ict-citypulse.eu/page/content/tools-and-data-sets.

[128] Andreas Kamilaris and Andreas Pitsillides. A web-based tourist guidemobile application. In Proceedings of the International Conferenceon Sustainability, Technology and Education (STE), Kuala Lumpur,Malaysia, November 2013.

[129] Andreas Kamilaris, Yiannis Tofis, Chakib Bekara, Andreas Pitsillides,and Elias Kyriakides. Integrating Web-Enabled Energy-Aware SmartHomes to the Smart Grid. International Journal On Advances inIntelligent Systems, 5(1), July 2012.

[130] Andreas Kamilaris, Vlad Trifa, and Dominique Guinard. Buildingweb-based infrastructures for smart meters. In Energy Awarenessand Conservation through Pervasive Applications Workshop (Pervasive2010), Helsinki, Finland, May 2010.

[131] Andreas Kamilaris, Giannis Kitromilides, and Andreas Pitsillides.Energy Conservation through Social Competitions in Blocks of Flats. Inthe 1st International Conference on Smart Grids and Green IT Systems(SMARTGREENS), Porto, Portugal, April 2012.

[132] Panayiotis Kolios, Andreas Pitsillides, Osnat Mokryn, and KaterinaPapadaki. Explore and exploit in wireless ad hoc emergency responsenetworks. In Proceedings of the International Conference on Commu-nications (ICC), pages 452–458. IEEE, 2014.

[133] Jeremiah Owyang and VBProfiles. Meet the In-teractive Collaborative Economy Landscape, 2015.http://pages.vbprofiles.com/Collaborativeeconomy.

[134] IF Akyildiz, M Pierobon, S Balasubramaniam, and Y Koucheryavy.The internet of bio-nano things. IEEE Communications Magazine,

53(3):32–40, 2015.