Springer Professional. Back to the search result list. Table of Contents. Hint Swipe to navigate through the chapters of this book Close hint. Abstract RFID projects compete with other IT projects and therefore need to show a better performance in order to justify the corresponding investments Lee and Lee Current approaches to measure the performance of RFID and Internet of Things-related projects, including costs and benefits, will be analysed in this chapter and their uneven allocation among stakeholders will be shown.
Additionally, there will be a look at how costs and benefits can be harmonised between participants of a supply chain using CBS. The findings will show that the existing methodologies have several shortcomings. An alternative approach to performance measurement and CBS, which relies on pricing and selling information, will be introduced. Please log in to get access to this content Log in Register for free.
To get access to this content you need the following product:. Springer Professional "Technik" Online-Abonnement. Springer Professional "Wirtschaft" Online-Abonnement. Agarwal, V. Assessing the benefits of Auto-ID technology in the consumer goods industry. Al-Kassab, J. Ranasinghe, Q. Heidelberg, Germany: Springer. Baars, H. Toronto, Canada. Bensel, P. Straube Ed. Paris, France. Bovenschulte, M. Chan, J. Rail transit OD matrix estimation and journey time reliability metrics using automated fare data. Cheremushkin, S. Collins, J. Department of the Navy Chief Information Officer.
Tutorial: The Value of Information. Dittmann, L. Dutta, A. Production and Operations Management, 16 5 , pp. Erdman, L. Feinbier, L. The benefits of RFID for slab- and coil-logistics. Fleisch, E. Die betriebswirtschaftliche Vision der Internets der Dinge. Mattern Eds. Berlin, Germany: Springer. Gille, D. Golden, T. Acton, H. Tuunainen Eds. Grote, W. Knoblauch Eds. Nachlieferung pp. Hardgrave, B. Haasis, H. Scholz-Reiter Eds. Harley, S. Effektive Doppelspielfahrten.
Verwandte Produkte. Mehr erfahren. Elektro-Handstapler Jungheinrich HC Dieselstapler Jungheinrich DFG Newsletter Jetzt anmelden. Social Media. Haben Sie Fragen? Kontaktieren Sie uns. Could Ubiquit- ous Computing undermine the autonomy and choices that are said to characterise mankind? A fundamental step towards answering this key question is to define the power relationship between human beings and computer systems.
When we compare ourselves to computer systems, we tend to question our human skills and capaci- ties: Who makes faster and better decisions? Who do we trust to tell the truth? Who will evolve more rapidly? All too often, we display a latent disposition to trust the power of the machine more than the human subject. But what do such views imply?
Do we risk slipping into a perspective that views man as inferior to computer systems and that questions human power and decision-making? Do we give up the autonomy and free- dom of choice that we are so proud of? In contrast to what many science fiction novels tell us, automation scholars regularly show that the overall superiority of machines is not a given Sheridan In his acclaimed essay, Fitts tries to objectify the competence relationship between men and machines for the engineering sciences Fitts He states that machines outperform humans in terms of fast reaction time, the use of strong power in a soft and precise way, the complete de- letion of information, or deductive argumentation.
Despite considerable advancements in computing since the s, this fundamental view of the man-machine relationship still has virtue. Machines may be getting better at making complex decisions, but complexity also adds cost and risk to machine operations. The tradeoff between these risks and costs and the efficiency of control delega- tion, this fundamental decision, regarding the distribution of work between men and machines, remains a grey area Sheridan Are fully auto- mated airplane cockpits safer than human pilots?
Unser Industrie-PC – Ihr mobiler Arbeitsplatz
Are electronic voting machines better at counting ballots than electoral staff? Do video control systems prevent crime more efficiently than human guards? It is within this alchemistic grey area of control-allocation decisions that the Idea of Man comes into play. Do we opt for men or for computer systems? Whe- rever an objectively detectable superiority of people over machines is not a given, the Idea of Man can help people decide whether human beings will be allowed to maintain control. Hence, the Idea of Man manifests itself in both, behavioral rules of social interac- tion and values that underlie positive cooperation between humans.
Values and behavioral rules define how Man should be. Today, this idea of how Men should be is undeniably affected by pluralism: in a rapidly changing global society, no value monopoly exists. However, we still share ethical norms that are widely accepted; such norms are reflected, for exam- ple, in international agreements like the Universal Declaration of Human Rights of the United Nations.
These values and norms are also meaningful for technical design. When com- puter systems become social actors, interact with human beings in their everyday lives, and handle tasks for them, people expect them to act like people. Socially developed norms of interaction and behavior are conferred upon machines Reeves and Nass But which norms will help us uphold our Idea of Man in an evolving Internet of Things landscape?
Machine developers there- fore have a tremendous influence on the Idea of Man that is embedded in ma- chines. On a macro level, system developers make fundamental decisions about the role a ma- chine is allowed to play with regard to people. For example, privacy policies may ostensibly be in place, while, at the back end, their technical implementation is neither supervised nor permanently adhered to. Developers choose the values a system lives up to see above. On a micro lev- el, these decisions are translated into concrete machine actions.
On a macro level, the values that Friedmann summarises could inform developers about key points of system development Friedman and Kahn However, these macro level principles must be translated into concrete micro level system design guidelines. Second, they design the way machines treat humans contact. These two areas are frontend design decisions. Third, engineers determine the way machines act at the back end and to what extent such actions are transparent and subject to influence by users. Consequently, at the micro level, the Idea of Man manifests itself in how programmers design frontend interaction and backend behavior.
Figure 2. But can they? Law regulates that every vehicle must be equipped with a seatbelt warning system. However, it is the man- ufacturer or vehicle developer who, on the micro level, determines its concrete de- sign, making decisions, such as: Can drivers manipulate the system by turning off the warning signal? Manipulation How does the vehicle at the frontend warn its drivers: by means of a drown-out, shrill acoustic signal that forces them to per- form the desired action? Or does it discreetly remind them that it would be wiser to fasten their seatbelts the moment they start the engine?
Contact And finally, how does the vehicle behave at the back end? Will it make that information available to insurance compa- nies in case of an accident? And do drivers have the right to access and delete this information? Backend Behavior This example shows the tremendous impact, a single component of the Idea of Man, namely control, can have on concrete micro decisions in technological de- sign. It also illustrates the broad margin that designers enjoy on the micro level, one that allows them to develop a system in a variety of ways. When asked how data protection is taken into ac- count during prototype development, nearly all interviewees responded with one or more of the following arguments Lahlou et al.
The reason for this lag of con- sciousness may be that the engineering sciences have been primarily interested in enhancing technical functionality. Mass market technologies, such as home IT, mobile communications, video games, or navigational systems are key drivers of technol- ogical progress today. Yet, less emphasis has been put on how to respect ethical system behavior systematically when designing backends. While a few scientists have tried to raise awareness of ethical issues in computer science for decades e. Consequently, we lack knowledge about what constitutes socially acceptable technology.
Such processes should be developed. Some are already taking first steps to develop micro level mechanisms for ethical engineering i. But few of their concepts and ap- proaches are integrated into the teaching of computer and engineering sciences. Commitment from practitioners of the computer and engineering sciences, however, will ultimately not be enough to craft technology that is more socially compliant. Therefore, management must emphasise socially acceptable technical design. Yet, because companies are driven by profitability, they limit themselves to meeting basic legal regulations in an effort to make the development of systems as cost-effective as possible and maximise the potential use of their technical systems such as data collection.
As a result, developers often try to avoid the topic until they are forced to confront it by the market or regulators. The reaction time of lawmakers, however, is often too slow to affect rapidly developing technical markets. Especially in Europe, there is a latent fear of over- regulation; politicians want to reduce the risk of stifling the innovative spirit of technology markets by limiting ethical regulations. Some experts argue that mar- ket mechanisms should be responsible for sanctioning socially incompatible tech- nological designs and rewarding socially compatible ones.
Would economic incentives justify private investment in socially compatible technologies? As a result of strong pressure and negative reactions from clients, social network operators now allow clients to adjust privacy settings for their data. Another possible scenario is that clients, who begin to value sustainable, ethical technological designs, might be willing to pay more for them than for traditional designs. In some cases, such as organic food, consumer markets have developed in this direction.
However, it remains unclear, whether markets that are less transpa- rent and more technically complex, like IT services, can become clear enough for clients to understand the added value of socially compliant services. Spiekermann backend practices Bizer et al. If transparency increases and more informa- tion about common backend operations become available, markets might be forced to change completely. Even if clients recognise one technology as more value-sensitive than another, they might not pay more to avoid the risk of long-term damages.
For example, In- ternet users do not pay much attention to their privacy on the World Wide Web. They seem to value the short-term advantages of Internet services more than they fear the long-term potential loss of their privacy. People exhibit such lax behaviors because they have difficulty evaluating risks. They often underestimate discount long-term risks and overvalue short-term benefits Acquisti and Grossklags Time and again, govern- ments have launched research programs to analyse the ethical aspects of compute- risation, while impact assessment studies have addressed the social implications of technology.
This essay is only a small contribution to the endeavor of making our technical environment more humane. It deals with the specific notion of the Idea of Man and its potential value for the technical design of systems and networked environ- ments, such as an Internet of Things. It shows that the Idea of Man can act on three levels: first, it enables us to reflect upon the power relationship between hu- man beings and machines on a higher level; second, a decomposition of the Idea of Man helps us to identify concrete values that should impact technical design at a macro level; and third, a conscious sharing of an Idea of Man supports the re- spect of values at a micro level, where developers make daily decisions about how to structure interactions between men and machines.
Auf dem Weg in eine andere Moderne. Based on that, we elaborate three concepts forming a basis for new creation paradigms in such smart spaces, potentially leading to new DiY-enabling functions in Internet of Things service creation environments: the Call-Out Internet of Things, the Smart Composables Internet of Things, and the Phenomena Internet of Things.
Next to a discussion of applicable state-of-the-art for implementing parts of these concepts, we show first experimental grounding for them, as part of the ongoing exploration process. In this section, we first look broadly at DiY as a cultural practice and discuss some core characteristics. We then make the transposition from the cultural practice to what this may imply for application creation and for context- aware environments, which are the necessary building blocks for reaching the goal of enabling the masses to become creative in smart spaces.
Roelands et al. Until the development of dedicated DiY stores in the s, people who wanted to decorate, repair or modify their own home had to venture into the spe- cialised world of the traditional builders merchant Roush Companies mak- ing and selling tools and materials to amateur rather than professional customers undoubtedly were the promoters of the idea of DiY.
In the s lots of DiY shops and magazines were initiated. And this happened with great success. At the basis of the rise of DiY as a cultural practice were different drivers. Firstly, the economic changes brought that more people than only the rich part of the popula- tion had money to invest in home interior and decoration. Secondly, there was the fact that work hours became ever more expensive and the related rise of the DiY stores. But there are more than only these economical reasons that made DiY at- tractive.
DiY offers people pleasure by creating personalised artefacts or tune existing ap- plications to their ultimate wishes. They claim that par- ticipation in gardening, sports and home improvement constitutes a form of every- day resistance to the alienating effects of contemporary society. The contemporary society indeed is characterised by excessive consumerism, globalisation and eco- nomic inequalities between persons and groups, alienating us from our environ- ment and ourselves.
The first is someone who works alone, typically in a per- sonal closed environment like a garage or attic, in a very dedicated way. The sec- ond can more often be found in a community of likely interested people. They col- laborate in producing their invention of creation, so they are willing to make it public before it is finished and discuss about it with their companions. We speak of DiY, but out of research we learned that less institutionalised channels, personal networks of family, friends and neighbours are crucial for indi- vidual experiences of DiY Shove et al. A pro-am is an amateur that pursues activities out of the love for it, but at the same time setting a professional stan- dard Leadbeater and Miller A lead-user is at the leading edge of an important market trend, and so is currently experiencing needs that will later be experienced by many users in that market.
With the attempt to describe the different roles and activities of a person doing DiY activities it becomes clear that a complex net of practices and social relations are at the basis of a DiY culture. These activities are also related to the kind of DiY activity people are executing; knitting pullovers, making a bench for a dog, designing an operating system, making a YouTube movie, and so on.
But, overall, a DiY activity has some common characteristics. It is about connecting, about tak- ing control and about diversification. Gauntlett gives a good insight in the social aspects of creating. He distinguishes three ways on how making is connecting, and, therefore, in essence indicated that DiY is about com- munication. Making is connecting because you connect things together materials, ideas or both to make something new. Making is connecting because arts of creativity usually involve, at some point, a social dimension and connect us with other people.
Making is connecting because through making things and sharing them in the world, we increase our engagement and connection with our social and physi- cal environments. The software culture is very much based on the reuse of code. The recombination of components and mash-up systems are other examples. The existence of online communities also is highly important for the DiY communities in the physical world. Not only can DiY-ers rely better on their local network for help, or people with the same interests, via these online communities, their community effectively gets world scale and world level, with reputation becoming a stronger factor.
Therefore, one needs capabilities as well as tools with particular attributes, openness being an important attribute of that. The aim of handing over the control to the creator or end user can be put in the discussions on innovation and technology. Paul Dourish , in his design view, focuses on the fact that users are not to be perceived as passive recipients of predefined technologies, but as actors determined by the circumstances, contexts and consequences of technology use Dourish It can be seen as a reaction against ex- cessive consumerism, globalisation and economic inequalities between persons and groups, which alienates us from our environment and from ourselves.
Thus, DiY also is about diversification. Own application creation can currently only be seen as an activity for the happy few. Applica- tions like ZohoCreator and LongJump are not aiming at end-users to create appli- cations, but at an audience not much less but professionals. The role of open Ap- plication Programming Interfaces APIs for application creation cannot be underestimated in this respect, like it is a current topic in the context of iPhone, Blackberry, Facebook, Twitter, and other specific ICT environments.
First, the vision on context awareness is very much technological driven and often does not take into account the meaning of context for the person that is act- ing in the particular environment. This problem lies at the origin of typical context-aware applications today being far from appealing. Because of these intrinsic characteristics, context cannot be defined as a fixed computational structure, and rather is an interesting but hard-to-capture concept. Second, context-awareness seems to imply loss of control for the person it ap- plies to. While issues as privacy, autonomy and control were in the picture from the start, these issues seem very hard to ad- dress.
Both these issues are essential arguments for the importance of DiY in smart spaces. As our multidisciplinary research meth- odology moreover involves users in the validation of mock-ups and proof-of- concepts as well as in the creation process itself, as active participants via e. The final value assessment will follow from user feedback experimentation in the ongoing work, but already now these concepts can help confining the problem area.
In turn, this will make the actual analysis of their potential merits and tech- nological feasibility more practicable. Concrete experimentation around the con- cepts therefore does not need to resort to one very narrowed-down application domain a priori, but can rather try to apply the concept in multiple concrete do- mains to enrich it generically, without ending up in an explosively broad DiY scope. However, first, as a basis, we introduce a typology of what kinds of DiY crea- tion are imaginable in smart spaces. While acts of DiY show to have an important potential in the Internet of Things, as discussed previously, we should indeed first identify what these DiY creation acts could be in this context.
As illustrated by Figure 3.
ETX 513/ 515
Today, several examples of that exist in the web, as we will discuss further on. As such, people can be creative in shaping the tangible interaction front-end to the Internet of Things. Call-outs, as meant here, may entail the traditional variety of information prop- erties of locations, objects or other aspects of the surroundings, but especially can also be behavioural descriptions, describing a local interaction pattern, implying requests for interaction and an opportunity for adding new elements or actors in an open-ended machine or process.
A Call-Out Internet of Things would, moreover, support the exchange and reuse of these properties across contexts of place, time or embodiment. While examples of implementations exist that fall under this definition of the call-out concept, we use it as an instrument to get a deeper understanding of it as a new medium in ambient experiences. Thus, in this section we discuss how this concept is currently applied and what we see as future challenges in mass creativ- ity in the Internet of Things from this perspective.
In fact, call-outs are commonly known in our culture already and are used in a variety of communication applications. People can experience call-outs as a com- munication medium in their own intimate social space as well as the broader pub- lic surroundings. Call-outs have the potential to be used for exclaiming aloud and with surprise, e. For example, commercial electronic billboards in city shopping streets are competing to get their messages across. The call-out balloons in comics, depicting dialogues and supporting the structure of the narrative, are another effective example of attention-grasping communication.
- Integrated Billing Solutions in the Internet of Things | SpringerLink!
- Mit SIMATIC ITP1000 always on track;
- Back to Basics - Les Ressources Familiales (French Edition)!
- Exhibitor press releases.
- Les Cinq cents millions de la Bégum (Illustré) (French Edition);
As early as in the Middle Ages, Leonardo da Vinci masterly practiced the technique of adding text captions to complex drawings and sketches to explicitly communicate on innova- tive compositions. In this case, we can see call-outs as a way to expose otherwise hidden meaning and insights into structure. An example is the Leo- nardo da Vinci styled exploded phone drawing by which artist Kevin Tong cap- tures the imagination of H. Wells and the brilliance of Jonathan Ive In the networked society of today, at least three types of call-outs are practiced.
These personally, culturally, and socially driven reflections and annotations aug- ment locative meaning and stimulate interaction in this way. The Augmented Reality AR browser Layar 21 is a good example, where people can browse knowledge layers in overlay to the camera view, position- based. In this way, knowledge attached by people — or commercial organisations — can be experienced in its real geo-spatial context by others, giving the environ- ment new collective meaning, e.
An interesting Layar layer is the application Tweeps Around, which queries Twitter for posts labelled with an exact location With this example, the link to social networks is indeed made, hinting at a trend towards much richer geo-aware variants of the popular communication means. Another currently popular example of that trend is in fact Foursquare 23, where people earn community recognition and sometimes rebate vouchers by checking- in often, in particular venues such as public places, restaurants and other points of — often commercial — interest, having the community comments at the place as call-outs.
Wikitude World Browser is yet another example of an AR browser, leveraging a location-based style of Wikipedia A lot of creative development activities are organised and supported, fostering open and collaborative development by the masses extending the Wikipedia-style spirit to a location-based experience. Wiki- tude Drive is the first mobile AR satellite navigation system currently being tri- alled. Historically, this was one of the first expressions of the emergence of an Internet of Things.
Here, the key is that the physical objects are approached at short distance — touch — and that access to augmented media is achieved by reading an attributed object identifier. Today, user support for managing online identifiers and the associated media is offered via various online portal services, such as Thinglink 26, Tales of Things 27, ThingD 28 and the Touchatag 29 platform, which pioneered the RFID tagging scene and now offers both business-to-consumer B2C as well as business-to-business B2B interfaces, e.
Moreover, with the mixing with social network effects, as seen in several of the mentioned examples, we can expect that locative space will be shaped by DiY creation acts by the masses. See in this respect, e. Finally, can call-outs become a way to pinpoint instructions for acting, or even computing? This gives rise to the concept of the Smart Composables Internet of Things, as discussed in the next section. Querying can be done by context, possibly in relation to Phenomena see later section on the Phenomena Internet of Things concept and the context of sur- rounding objects, e.
A related classification of smart objects based on their awareness, their rep- resentation and their interaction can be found in Kortuem et al. Sometimes the creator is an individual creating an object for personal use, while in other cases the creator is an industrial actor who creates objects for mass consumption. In the figure below, we denote this as self-made and ready-made smart objects, respectively.
The purpose of a smart object may be to play a role in any application — or at least in a broad range of applications — or it may serve as a component in one spe- cific application. We call this open-ended versus specific smart objects. Littlebits 35, in the upper left quadrant, is an example of a smart object that can be created by an indi- vidual through combination of different electronic components with the purpose of creating any application that the person can think of.
(PDF) Architecting the Internet of Things | Luis Cancino Muñoz - magoxuluti.tk
In the opposite quadrant, lower right, we see examples of smart objects that are created by industry for a specific domain. Chumby 36 is an early example of such a smart object connected to the internet, with applications such as morning wake up calls as well as serving as a window to your favourite social networks. However, the two most important quadrants for the discussion in the Smart Composables Internet of Things are the upper right and the lower left categories The BUG 38 is an example of a ready-made, open-ended smart object, consisting of a modular hardware kit, out of which individuals can create standalone smart ob- jects by combining kit parts.
The BUG can, however, also be used to augment an everyday object for a domain-specific application, thus moving to the lower left quadrant of the diagram. An example of such an augmented object that is self- made for a specific goal, is a chair equipped with the BUG components, for exam- ple to detect whether a person is sitting on it or not. The BUG components in such case could, at the same time, be used to, for example, provide the person with an auditory feedback when someone rings the door. Composables are the smart objects in the upper right quadrant of our diagram.
Their strength lies in the creativity that they give to individuals to compose or de- compose, and to connect and disconnect with materials, people and society. Al- though composables are open-ended, they can be components of a domain-specific kit that supports the user to create domain-specific smart objects. If networks of composables exchange data by means of sensors and actuators, according to the Smart Composables Internet of Things concept, their context of use needs to be known in order for anyone to understand what this object interac- tion means.
Beneath this user-level meaningful exchange of data between located, identi- fied objects and object parts, also the technological means are needed to do the ac- tual data exchange. This is judged according to their openness for integration in a physical smart object design, as discussed here. All four use cases were based on the use of the same sen- sor a greyscale vision sensor and one actuator a dispensing actuator. The chosen example mock-up cases were inspired by small real-world problems, which we formulated as design goal questions from the user perspective: 1.
The duster case: how would people create a duster that detects spots on the floor, and automatically cleans them up? The door case: how would people augment a door to sprinkle a nice fragrance in the room, whenever the door is opened? The plant case: how would people create a flowerpot that automatically pro- vides the contained plant with the right amount of water — in time, and without any manual human intervention? The garage case: what can people build to avoid that fresh oil spots in their car garage make their shoes and carpet dirty?
Common to all cases was that we assumed a base object with an initial function a door is used to close a room, a duster is used to get rid of dust, etc. The following figures show the artefacts resulting from the mock-up exercise. The right picture in Figure 3. The left picture in Figure 3. Finally, the smart oil cleaner mock-up is shown in the right picture of Figure 3. Here, the experimenter conceptualised an autonomously driving platform for pluggable composables such as the oil stain detector and a sawdust dispenser, the platform having the same sensor data connection and aggregation function as with the smart plant pot.
The stickers that some experimenters used for labelling and giving meaning to particular composables or the composed whole, indicate various ways how software or call-out technologies could be used in real smart objects as would be composed by non-technical, kit- supported creators. The notion of a physical platform that makes composable eas- ily pluggable, as introduced in two of the mock-ups, may offer a new approach for the practical implementation of a particular type of composition call-out.
Of crucial importance in this respect is which patterns are of real value to users, implying that close user in- volvement in the iterative identification of these phenomena is essential for maxi- mising the potential of adoption in user-generated or other applications. So, in the Phenomena Internet of Things, higher abstractions of user context-awareness, considering long-lived patterns in personal life and society, are aimed to be de- rived from crowdsourcing across user groups, geography or application domains.
Of course, the crucial question in this perspective is: Which patterns are of real value to users? From that, Phenomena can be identified by the crowd, in line with the continuing crowdsourcing trend Howe , searching such relevant patterns by leveraging massive dimensions of scale, over long time spans, within or across geographical locations and in different application contexts.
This Phenomena effect is in par- ticular strengthened by the growing amount of personal data becoming available in the Internet of Things. In the next subsections we shortly discuss each of the four ingredients. In an example such as Noisetube, massive data is collected due to large num- bers of people contributing their personal mobile noise measurements. Clearly missing in this emerging Web-of-Things, is a collective identification of valuable, abstractable patterns, Phenomena, which could trigger much richer application possibilities, in the least already because many such potential patterns are simply not recognised yet as valuable elements for influencing the behaviour of newly created applications, at various expertise levels of the creation process.
Another example is the data collected through detection of user activities in a smart house, with for example hourly, daily or weekly repeated patterns becoming apparent over long time spans. Therefore, a sound starting point may be a system defaulting to an assumption of all data being strictly personal and only for personal use, further only requesting user intervention upon specific pattern proposals instead of demanding the con- figuration of controlling rules for the entire space.
The identification of the most relevant patterns as candidate Phenomena even allows for the crystal- lisation and optimisation of them into new enabling data brokerage and exposure functions, that can become new services or products for actors that want to engage in a business based on the Phenomenon. For example, from the previously listed means for user appreciation feedback, if many of the most popular visualisations as chosen by users are invariant to par- ticular parts of the collected data streams, this may mean that indeed that part of the data is less relevant in general.
Or, patterns that support application behaviour that is systematically approved, respectively disapproved by users, may grow, re- spectively diminish, in importance as candidate Phenomena.
In contrast to that, a second important incentive for users in leveraging new Phenomena — whether still emerging in a Phenomena network, or already institutionalised in a brokerage service — lies in applying a Phenomenon in an application. Either the awareness for the Phenomenon through its use in the application, or the entire application by itself, is then discovered or user-generated.
In this way, an additional wave of user value emerges from the identification of the Phenomena. Many of the existing context-aware applications can be seen as canonical ex- amples of Phenomena-triggered application behaviour. In the domain of assisted living, several research activities tried to track user activity over time and monitor health, often assuming a given scenario a classification task , or analysing spe- cific sequences a time series analysis task. In the vast amount of literature on computer vision for activity recognition Moeslund and Granum , motion patterns include variations of neural networks and hidden Markov models.
An in- tensive area of research is the area of smart homes 42, 43, 44, 45, where contextual in- formation is gathered from many different kinds of sensors around the house or office, often also using layered hidden Markov models, naive Bayesian net- works, or decision trees to identify particular context situations Desai et al. In the domain of wearable and mobile computing, principal component analysis, Kohonen self-organising maps, k-means clustering, or again first order Markov models are used to detect user status from wearable sensors Oliver et al.
Finally, some original approaches use an ontology, e. While closing the loop with users iteratively providing appreciation feedback on identified patterns has already been considered for many applications in the pervasive or ubiquitous computing domain, the more generic approach of leverag- ing this Ingredient 2 to identify Phenomena, is only starting to be analysed for its potential. These new services, however, do not offer a higher pattern abstraction level, as a commonality for many, and so leave it entirely to the user how to interpret the data in various visu- alisations, often even exclusively oriented to visualising long-term data trends, without any real-time applications.
In a classical approach, such an applica- tion would require the user to configure, or even design a complex set of context rules, triggered by carefully chosen conditions, and resulting in a well-structured sequence of actions. Crucial for the Phenomena Internet of Things approach thus is that this configuration complexity is hidden from the user dramatically better, by deriving an internal set of rules, conditions and actions, indirectly rather than explicitly based on appreciation feedback of the user. From the monitored activity, which may already be tailored by the user to con- tain especially relevant clues, new patterns are mined, correlating them with ini- tially manual lighting and music selections.
The user not only becomes aware of the underlying pat- terns the system discovers, but also implicitly appreciates their relevance from the personal user perspective. Beyond that stage, users have at their disposal a well-trained system that de- tects and possibly even predicts personally defined atmospheres, which they now can start leveraging in other smart home applications, like presence-based com- munication applications or home energy management services. From use of the data in these additional applications, new types of user feedback, yet refining the personal atmosphere model, can follow, making the system still more accurate.
Here, the process starts with the classical analysis of crowd traffic patterns and visualisation of them in an attractive way on geographical maps for citizens. Phe- nomena, in this context, could be public phenomena concerning e. Based on these, a routing application could derive route change advice to users, and get feedback by people actually following the advice or not, or giving other appreciation when an advised route is eventually followed. In this way, a model is grown about the collective awareness and behaviour upon changing city traffic conditions. From the user feedback, the system learns what traffic phenomena are relevant to people, making them change route plans, or it learns about obstructions it did not explicitly detect in the first place very lo- cal peak hour traffic congestion effects, unregistered road works or damage, etc.
Here again, other applications can start using the identified Phenomena as ad- vanced context triggers for smart adaptation. In this way, the citizens also influence decisions and actions of local government and other stakeholders. Figure 3. The Prototype System From the target goals stipulated above, we built a first working prototype in an ini- tial technical step towards the evaluation of the Phenomena Internet of Things concept in a user community.
The stream of tuples is saved at a remotely located MySql database, from which real-time visualisation or any — possibly third party — application is made possible using a Google visualisation API, taking the MySql server as an external data source. For instance, as shown in Figure 3. The visualisation on the right in the figure shows a more sophisticated instrumentation, allowing users to selects parameters according to their preferences. For this purpose, we plan to build a data broker agent, also cooperating with a Bell Labs research department specialising in the data mining aspects and related relevant techniques.