A survey on user studies and technical aspects of mobile multimedia applications

Entertainment Computing 2 (2011) 175–190

Contents lists available at ScienceDirect

Entertainment Computing journal homepage: ees.elsevier.com/entcom

A survey on user studies and technical aspects of mobile multimedia applications Shelley Buchinger ⇑, Simone Kriglstein, Sabine Brandt, Helmut Hlavacs University of Vienna, Faculty of Computer Science, Entertainment Computing Research Group, Lenaugasse 2/8, 1080 Vienna, Austria Workflow Systems and Technology Research Group, Rathausstraße 19/9, 1010 Wien, Austria University of Veterinary Medicine, Equine Biotechnology Unit, Veterinärplatz 1, 1210 Vienna, Austria

a r t i c l e

i n f o

Article history: Available online 26 February 2011 Keywords: Mobile TV QoE Human Centered Design Survey

a b s t r a c t Some years ago, Mobile TV has been introduced in several countries all over the world. It was expected that it would play a major role among traditional TV services. Unfortunately, the success has been limited at the beginning. Since Mobile TV brings new aspects into television, like small screens, consumption in noisy surroundings, etc., it also represents a new challenge on how to create, transfer and present content that maximizes the consumer experience. Today, some of these issues have been solved. Due to the introduction of smart phones and the large amount of available applications customers are starting to use their mobile phones for several purposes including mobile multimedia services. As a consequence, the concept for Mobile TV has changed significantly. In the past, research has often been focusing on one particular aspect of this new TV scheme, as well as surveys on this research often neglected aspects that still might be of interest when trying to understand the dependencies of Mobile TV content and presentation to perceived quality. In this survey paper we want to discuss challenges and requirements in a comprehensive way, trying to shed light on all relevant aspects of Mobile TV. The aim of this paper is to give a good overview about the state of the art with the focus of users’ need and experiences. A large collection of technical aspects and research results represents a special interest of this study. Finally, we want to discuss a framework for mobile multimedia applications which is relevant for further research work. Ó 2011 International Federation for Information Processing Published by Elsevier B.V. All rights reserved.

1. Introduction Mobile TV and digital video content services represent an important business of the future. According to forecasts user demand will grow significantly by 2011 [72]. The realization of Mobile TV leads to an important change of television that we know today. Wippersberg compares the current situation to the early fifties of the twentieth century, when television started to enter our homes [101]. At this time, similar to now, the screen size was dramatically reduced leading to a completely new scenario with new problems and challenges. Due to different consumption scenarios and time schedules, a new concept for the content production and consumption had to be made for television. Wippersberg states that this needs to happen again for Mobile TV. In the first period, mostly early adopters used Mobile TV. They were limited to the available services offered at the market, e.g., broadcasting and video on demand. Due to technical as well as conceptual reasons that will be discussed later in this paper the initial success remained below most expectations. However, since then the concept of Mobile TV has changed. Currently, the number ⇑ Corresponding author. Tel.: +43 1427739653. E-mail address: [email protected] (S. Buchinger).

of available applications on the market has considerably increased. This has mainly occurred for two reasons: (1) more sophisticated devices such as smart phones are available on the market (2) users have the possibility to contribute their applications to the community for example via App Store or Market. The consumers get more in more accustomed to use their mobile phones for a larger variety of purposes including mobile multimedia services. Furthermore, the use of multimedia data in general becomes more and more popular, e.g., in social platforms or video responses in discussion forums instead of text messages. In [15] the ongoing evolution of Mobile TV is described in more detail from the social point of view. As possible perspective for the future Chorianopoulos states that ‘‘the introduction of content-enriched communication services is a worthwhile direction’’. The open development platforms play an important role for the progress experienced in this area. In order to understand the needs and success parameters of mobile multimedia services, a large amount of user studies and experiments have been performed. Many user studies [72,37,14,6, 74,33,73] show that one of the main critical aspects for the success of Mobile TV is content. Special made-for-mobile content results in much higher user satisfaction when compared to adopted standard TV content.

1875-9521/$ - see front matter Ó 2011 International Federation for Information Processing Published by Elsevier B.V. All rights reserved. doi:10.1016/j.entcom.2011.02.001

176

S. Buchinger et al. / Entertainment Computing 2 (2011) 175–190

However, because of the involved costs, made-for-mobile programmes are still the exception, and generally available content is recoded for the smaller screens of mobile devices. This represents one of the reasons why Mobile TV is still in its infancy and the number of users is very limited. Evaluations showing a considerable number of consumers, like one million in Italy [17], have to be considered with care. This is due to the fact that usually persons having flat fee agreements which include Mobile TV are counted as Mobile TV consumers, even if they are not actually using Mobile TV at all. A number of surveys has been published that offer overviews for research and development of Mobile TV. The surveys present research results from different studies. Yet they show that research on similar questions has yielded completely different results in the past. For example user studies carried out in Germany [6,74], Austria [73], or in South Japan [64] show that Mobile TV will be mainly viewed during short waiting times, while in similar studies carried out in the United Kingdom and South Korea [14] or Belgium [95], the main application area is at home. The reason for this divergence can be understood when reading user comments recorded during the experiments. Culture, mentality and age have a high impact on this research issue and lead to different outcomes, e.g., in the United Kingdom and in Belgium people do not want to disturb other people in public transports or on the street, and do not want to compromise their privacy. In this paper a comprehensive overview of this earlier work is given, by putting a major focus on the user’s point of view, thus combining the Quality of Experience (QoE) approach with the traditional Human Centered Design approach. Furthermore, parameters depending on cultural aspects and on different user profiles are highlighted. Hereby, previous surveys are updated, e.g., a study on Mobile TV for 3D movies [46], but also technical aspects are included. As a consequence of this state of the art analysis open issues for Mobile TV are deduced and verbalised. Furthermore, the most recent developments of mobile multimedia are analyzed and an existing QoE framework adapted to the current mobile multimedia process is presented. 2. Methodology, framework and structure In this section, the applied methodology is described and discussed. Since we aim at putting the user in the center of our investigation, we provide some related work on Human Centered Design and Quality of experience. Furthermore, we intend to visualize the results of this investigation by presenting a framework for QoE of Mobile Multimedia. Therefore, existing work on QoE frameworks will be presented. Finally, before starting to describe the results of previous mobile multimedia studies, we motivate, describe and justify the structure selected for this survey. 2.1. Methodology 2.1.1. Human Centered Design Several research papers focus on results elucidating the users’ needs and motivations in regard to interactive television as well as Mobile TV, since commercial success of a TV product depends on the users’ requirements and its viability [77,59]. Eronen [21] points out that a good usability is an essential aspect of television. This is due to the fact that Mobile TV is mainly consumed in leisure time, and users do not want to spend much time to find out how to use it. Furthermore, Eronen also states that for new devices and applications, the focus of Human Computer Interaction (HCI) research, which usually concentrates on usability, has to shift more to user experiences, which are relevant for the use of the product according to user needs [22]. Human Centered

Design (HCD), also known as User Centered Design (UCD) involves users in the design and development process for a better understanding of their needs, tasks and expectations [91,35,22]. The aspects of Human Centered Design correlate with the understanding of User Experience (UX), which Hassenzahl and Tractinsky [39] describe as a ‘‘consequence of a user’s internal state (predispositions, expectations, needs, motivation, mood, etc.), the characteristics of the designed system (e.g., complexity, purpose, usability, functionality, etc.)’’. Especially for mobile phones the user interface and its usability often depends on the usage model and its layout [46] ‘‘and the context (or the environment) within the interaction occurs’’. Based on these factors, design solutions by prototyping can be developed, which are evaluated by their usability and by how they suit user needs in the next step of the Human Centered Design process. Currently, users generally watch television in private environments, while mobile phones are often used in public areas. There are many interesting studies about interactive television performed in a home environment, yielding results that may have a high influence on mobile television as well. Therefore publications on interactive television that might have an impact on Mobile TV are equally considered in this work. These additional insights can be particularly helpful for further design solutions. Finally approaches and applications that are estimated to be promising. 2.1.2. Quality of experience (QoE) An overall picture from the user’s point of view is given by Quality of Experience (QoE). In this context, quality considers the entire user experience and is measured by collecting and analyzing subjective judgements. Standardization attempts to define this term are still at an early stage. The ITU-T Study group 12 formulated the Question 1/12 of the study period Study Period 2009– 2012 ‘‘QoS/QoE coordination and bridging the standardization gap’’ and approved in December 2008 the pre-published ITU-T Recommendation G.1080 [92] defining user requirements for Quality of Experience for IPTV services. The recommendation contains the following definition: ‘‘Quality of experience (QoE) is as the overall acceptability of an application or service, as perceived subjectively by the end user. NOTE 1: Quality of Experience includes the complete endto-end system effects (client, terminal, network, services infrastructure, etc.). NOTE 2: Overall acceptability may be influenced by user expectations and context.’’ Previous definitions have been summarized in [86] leading to a more technical approach: ‘‘The users’ perceived experience of what is being presented by the application layer, where the application layer acts as a user interface front-end that presents the overall result of the individual Quality of Services.’’ This definition has been expressed in 2003 and it does not match the current point of view implying more factors, e.g., environmental influences or device properties. Based on the Human Centered Design approach, it is necessary to find answers to the following questions: Where and when will Mobile TV be consumed? What kind of service and content should be made? How should it look like and how will it be consumed? Initial attempts to answer these questions are available [72,37,54], but results still need to be refined. 2.2. QoE framework A certain amount of research in the area of quality of experience has already been done. Most members of the research community agree that an interdisciplinary approach needs to be followed. However, except for the definitions of QoE presented above no pro-

S. Buchinger et al. / Entertainment Computing 2 (2011) 175–190

cedure for assessing QoE has been standardized yet. In this section, a selection of proposed QoE frameworks will be presented. The aim is to elucidate several perspectives ranging from frameworks providing a high level description, followed by some more detailed approaches down to QoS related QoE concepts. Moeller et al. [65] propose a taxonomy of QoS and QoE aspects that consists of three layers: (1) QoS-influencing factors (2) QoS interaction performance and (3) QoE aspects. The QoE aspects are specified to be of hedonic or pragmatic nature according to the definitions of Hassenzahl [38]. In this way, the entire picture comprising user as well as technical aspects are taken into account. An extension of this work has been published by Geerts et al. in [31]. Their aim is to build a framework that is linked with appropriate QoE measurement methods. In this scheme, according to the QoE definition, different layers of context and user expectations are taken into account. Furthermore, changes in use over time are integrated in this model. This aspect seems to be very useful for describing the current developments in mobile multimedia. A more detailed view on the QoE in combination with a concrete procedure for measurements is provided in [18] and [49]. In both cases agents are used for measuring single quality aspects and overall results are additionally based on users’ feedback. Fiedler et al. provide a theoretical view on the subject in [27] where mathematical mapping functions between QoE and QoS are proposed. A specific framework focusing on the relation between QoE and QoS parameters in distributed interactive multimedia environments has been published in [102]. Even if the context and the evolution over time is not expressed as explicit as in [31] the approach presented in [102] seems to fit best for our purposes because of the following reasons: As it can be observed in Fig. 1 the influences on (1) performance gains, (2) exploratory behaviors and (3) technology adoption are pointed out. Since mobile multimedia is a new area especially (2) and (3) seem to be essential for our case. The mentioned QoE terms as they have been defined in [102] are explained below. Flow (psychology): defines the mental state in which a person has the feeling of full involvement and success in the process of the activity under the assumption that the request is high enough to require the full concentration of the person. Telepresence: allows users to have the feeling as if they are present (e.g., conversation with mobile phone) or even visually present (e.g., video call or video conferencing). Acceptance: defined as ’’the degree to which a person believes that using a particular system would be free of effort’’ [102]. It depends on the factors perceived ease of use and perceived usefulness (e.g., if the

177

navigation for Mobile TV is easy to use and includes useful features). These three main dimensions of cognitive perceptions influence the following three behavioral consequences [102]: Exploratory behaviors: describes if users explore spontaneously the mobile multimedia applications with no plans or goals (e.g., try a new feature, which was developed for recording mobile videos). Performance gains: the goal of this dimension is to increase user’s performance for solving tasks. Performance gains depend on the application environments and on the task requirements. Technology adoption: describes the adoption or acceptance of a new mobile multimedia application or feature (e.g., to navigate the Mobile TV channels with the use of speech and the acceptance of this feature by users). In [102] the relation between QoS and QoE characteristics can be studied. The framework of this survey describing the past and current evolution of mobile multimedia will be built on top of the DIME framework where the focus is put on the QoE elements only. 2.3. Structure Since we decided to undertake our analysis according to QoE criteria, we had to identify QoE key factors for mobile multimedia. Then, we sorted the collected information in respect to the defined criteria yielding to conclusions for each single aspect. QoE key factors for Mobile TV have been identified by [54,14,60]. Although all key factors are equally important, users and their motivations to use Mobile TV play an essential role and influence other key factors. Therefore, these factors are related to the Human Centered Design approach to consider users’ needs and the resulting key factors represent personal views on this subject. An overview is provided in Fig. 2 where the user is positioned in the center of the key factors described below: User: the profile of typical users, including user motivation, future trends and needs. Mobile device: limits and novel features of Mobile TV offered by mobile devices, features that are still missing. Context: the aim is to find where, when and how long Mobile TV is consumed. Mobile services and features: the focus is to analyze which services at what level of interactivity promise to be successful in the future. Content: content types as requested by users, user generated content.

Fig. 1. Quality framework in DIME [102].

178

S. Buchinger et al. / Entertainment Computing 2 (2011) 175–190

Fig. 2. QoE key factors for Mobile TV.

Costs and commercials: how to finance Mobile TV, fee types and amounts users are willing to pay, the role of commercials and how they need to be introduced. Technical performance: which technical parameters concerning audio, video and transmission need to be considered to improve the quality? Finally, results will be taken together in a framework providing a general view of the topic as depicted in Fig. 3. The key factors represent the relevant elements leading to the mentioned behavioral consequences that are required for the success of mobile multimedia. The offered content, the context of the mobile multimedia applications and the cost model have an important impact on the behavioral consequences. For example, if a user wants to watch

the news with his/her mobile phone, it is necessary that the desired content is available and that the context in which s/he wants to watch the news is considered (e.g., environmental noise, in the morning, daylight). Furthermore, her/his acceptance to watch the news can depend on the cost model (e.g., a monthly flat rate or ’’pay per view’’). However, the behavioral consequences are also influenced by (a) the design of the mobile devices (e.g., touch screen versus buttons), (b) mobile services and features which are developed for mobile multimedia applications and (c) technical performances. For example, if users do not watch Mobile TV, because the screen size of the mobile phone is too small, a possible solution would be to design new mobile phones which maximize their display (e.g., iPhone or HTC). In addition, the content, the context and the cost model influence the development of mobile devices and services as well as the technical performances and vice versa. For example, a new interface design, which integrates the possibility to navigate a mobile multimedia application with gestures, can open new ways in which people will use it. The following sections of this paper describe each key factor in detail and aim to determine the parameters which depend on cultural aspects, and those which yield constant result in different types of context. Furthermore, minimal technical pre-requisites for a successful mobile multimedia service are highlighted and presented in more detail than in recent work, in order to draw attention to yet unresolved issues in this field. 3. User The most critical factor for the success of a mobile multimedia service is the user. He decides if and how Mobile TV will exist in the future. Therefore, first efforts are dedicated to the analysis of user needs and expectations. For that purpose typical user profiles in the current situation are discussed. In this section, we present several studies describing user profiles and motivation. Some studies address interactive television viewers, but also yields valuable background information on potential Mobile TV users that should be taken into consideration in Mobile TV design. Similar to interactive TV [21], viewers are changing their passive television viewing towards an active consumer behavior.

Fig. 3. First approach of our framework for mobile multimedia applications.

S. Buchinger et al. / Entertainment Computing 2 (2011) 175–190

3.1. User profiles User profiles describe typical users and yield information about the users’ age, culture, leisure time activities, work and their relationship to modern technology. Eronen [21] defines three main user profiles for interactive television, based on their home, work and technology orientation: Pioneers, High-Fliers and ComfortLovers. Eronen combined the results of a questionnaire study, which aimed at finding out more about the scope of the new media in Finland, with those of the following focus group study: The user profiles represented mainly younger users. The median age of Pioneers was 21 and they wanted to have the newest and fanciest electronic entertainment possible. High-Fliers had a median age of 29 years and they were strongly technology-oriented. They wanted to read personal content (e.g., emails) on their own screen. The median age of the Comfort-Lovers is 36. They were not so flexible with media usage and they preferred to use services which were easy to handle. Furthermore, persons were generally open for new technologies. Yet, they preferred to wait until they would know more about their personal benefit of such new interactive services. Jumisko-Pyykkö et al. [46] investigated the acceptance of 3D movies for mobile devices and pointed out that the typical Mobile TV user is a well-educated male, aged 23–35 years. His major motivation to use this service is to shorten waiting times (e.g., for a public transport), to be up to date with the daily news while being on the move, or to get entertained. In [105] the association Zukunft Digital stated that Mobile TV attracts mainly young persons, aged 19–29 years, who are technologically interested ‘‘heavy-users’’. Furthermore they found out that women would like to consume other content than men, for commercials, no gender specific difference has been noticed. In [61] Mobile TV is considered to be attractive to one sixth of the users and is especially attractive to users between 18 and 44 years old, in Latin America, Asia and Eastern Europe. Rice and Alm focused their work [80] on the requirements of older adults (average age 71 years) to design an interactive television interface. Older users often have difficulties in understanding the technical language. They prefer larger buttons and voice-activated devices. The interface design should be simple and have only limited functions. Svon [93] analyzed how young new media-responding people are and how they used this media in their everyday life. She noted that mobile phones are widespread among young people (ages 8–15) and that nearly all possess their own phone. The Vanguard study [93] has shown that girls use mobile phones as their preferred medium in addition to computers. Television was a constant companion for the younger people and the study has shown that the numbers of channels, the quality of the device (e.g., sound and size) as well as the comfort (e.g., sofa) are essential factors. The survey also contained explicit questions about Mobile TV, but only few responded. Reasons were that the costs of mobile phone were too high and program offers too limited. The small number of persons using mobile television noted that the lines were too slow, thus humpering Mobile TV consumption. Generally, mobile phones are additionally used to store images, small videos or movies, music and ring tones. Furthermore, the study showed that young people used a wide range of mobile services. Also most recent studies such as the [28] reveal that early adopters are attracted by Mobile TV. The difference to previous results concerns the age of potential users. Not only young consumers aged between 18 and 29 are interested in mobile multimedia also 50% of users aged between 30 and 39 are attracted. Furthermore, it seems that users’ age does not represent anymore the most important criterion for the adoption of new mobile multi-

179

media services. Consumers can be divided by type of hardware owners [28], e.g., Netbook, smartphone, mp3, laptop or cell phone owners or by consumer type as it has been recently published in [87]. There, consumers have been segmented into five groups: mobirati who would use their phone for every day lives’ tasks, the social connectors using the phone to stay connected with their friends and social world, the mobile professionals are attracted by the extra features and would retrieve information in several ways, the pragmatic adopters use the cell phone mainly for calling and the basic planners use it only in case of emergencies. Taken together, until 2008 all studies agree that typical Mobile TV users are young persons. Through more recent studies it can be noticed that the picture is changing. Also up to 40 year old users seem to be interested. Consumers’ affinity to hardware and their motivation of use seem to become the dominant aspects. Gender-related predictions greatly depend on service features and proposed contents so that no significant correlation between Mobile TV consumption and gender could be determined.

3.2. User motivation There are many occasions inviting users to consume Mobile TV. In [72] it is mentioned that the ability to watch television at any time anywhere gives users a feeling of control and security. As a global picture people want to use Mobile TV to  kill time while waiting, i.e., for a public transport,  kill time during extended waiting periods, i.e., waiting for friends in bars, or in a car,  being always up to date, i.e., with popular events or news  create a private sphere,  feel less lonely, i.e., lunch break at work,  relax, i.e., at home in the bedroom or in the bath,  be entertained,  own, share and exchange content,  novelty, the desire to be the first, or  create and consume personalized content. Based on the traditional television, Svon pointed out [93] that the main motivations for younger people to use television in general is to relax, watch their favorite shows/movies or use it as background medium. The study [5] also showed that viewing television is often a background activity at home. A particular case of Mobile TV – Mobile 3D TV – has been recently investigated [46]. It has been noted that Mobile 3D TV is not likely to constitute a background medium (only 17.3% used it as a secondary task). On the contrary, users expect a deeper immersion from this service and intend to use it as main activity. In [61] it is stated that persons who own a multimedia or Mobile TV device will mainly use it as such. This means that just the fact of having the opportunity to use Mobile TV represents a motivation for its use. Several studies [14,69,60,10,77,70] show that users mostly want to kill time during waiting periods. Users are more motivated to use their mobile phone to watch videos during a short time period (e.g., waiting for a bus) instead to watch it on a laptop, because it takes more time to start it [57]. Furthermore the possession of content is of huge interest [69]. It has been noticed that people will use Mobile TV for relaxing [14]. These observations lead immediately to the question as to where people want to use Mobile TV, because relaxing and waiting occurs in different places. This issue will be analyzed in detail in Section 5.1. The trend of media entertainment tends to personalized content, which depends on user needs, tastes and preferences [104]. Mobile phones are considered to be very private devices. This is shown by a study [5] where participants are reluctant to share

180

S. Buchinger et al. / Entertainment Computing 2 (2011) 175–190

them with anybody else. This result is confirmed for Mobile TV [69,14], where individual viewing is the predominant case. In [69,14] subjects often used Mobile TV to create privacy. However, in particular young users sometimes shared their experience devices are lent. The possibility to personalize content represents one reason for the growing popularity of Mobile TV. Hussain et al. [42] pointed out that the number of traditional broadcast viewers decreases because of the increasing use of Internet or Mobile TV. Another study [88] showed that Mobile TV is attractive because people like to use their own personal TV at home. It can be used as a tool to get a closer interaction with television content.

4. Mobile device The big change from traditional television to Mobile TV is the device itself, its small size, operability and reliability. Results from user studies in relation to mobile devices found in different countries like UK, USA, Japan, Finland, South Korea, Switzerland and Austria are very coherent and in some cases even surprisingly similar. Some issues that have been identified in these studies have already been addressed nowadays. For example the screens have become slightly larger or touch screens provide the opportunity to display simple user interfaces for multimedia players. This might be one of the reasons why smartphones have become more and more popular. In [100] it is stated that mobile phones could easily outnumber PCs by 2013.

4.1. Screen and phone size In many cases the first reaction of new Mobile TV users is ‘‘but the screen is so small!’’ (see an example in Fig. 4). Whereas users want the screen to be as large as possible, they do not want their phones to be too big [54]. Larger image sizes and landscapeoriented use of the display might be preferred [54,73]. On the other hand, in particular female users do not want the weight and size of the handset to be increased [50]. During the FIFA 2010 World Cup MobiTV Inc. has observed [48] that users using mobile phones with larger screens revealed to have a longer viewing time.

4.2. Battery and memory Beside the screen size, the limited battery life has been identified as a main barrier [69,14,54,73]. High battery consumption of Mobile TV compromises mobile phone functionality. As a consequence, time-slicing for DVB-H transmissions has been developed, resulting in power savings of up to 90% [67]. This is due to the fact that the receiver part of the device can be put into a low-power

Fig. 4. A Mobile TV device.

sleeping mode when uninteresting transmissions transmission slices occur. The increase of battery lifetime and the decrease of power consumption are subjects of current research. Furthermore, a lot of memory is required for Mobile TV application and if users have not enough space (e.g., they stored too many photos) it would not work, which can have a negative influence on the usage behavior [47].

4.3. TV buttons Several studies [54,90] point out that Mobile TV must be easy to use. People are not willing to navigate through menus and therefore the need for special TV buttons on the phone has been expressed. In particular the sound management must be quick and simple.

4.4. Sound control and headphones As already mentioned above the sound control is very important for the Mobile TV user, i.e., people do not want to miss an announcement from the public transport and hence need to turn down or mute the sound rapidly. Two studies provide more indepth results on this issue, one has been performed in Japan [64] and the other one in Belgium [95]. Both studies, though having been carried out in different cultures, return the same result: the involved subjects did not want to use headphones when consuming mobile TV. Reasons for this behavior can vary. One explanation is that headphones are seldom used when watching traditional TV. Therefore, people simply do not associate television with headphones and do not think about using them. Another reason is that some people do not want to carry more than the mobile device with them; others may prefer to remain in contact with their surrounding. Not all the users reject headphones and there are situations where users definitely prefer to wear them. An example is given in [69] where a boy describes a situation in which he uses the headphones to become unavailable for his younger sister and brother. According to Oksman et al. [70], in Finland Mobile TV is not often used without headphones in public places and transport. It has to be considered that some mobile customers are already used to wear headphones when listening to music on the way. The specific circumstances defining the willingness to wear headphones actually represent an open issue that is worth to be further explored. One assumption might be that the acceptance of headphones depends on the user profile. The implications of these studies are quite important as they may concern a variety of issues. It seems that content, at least for a significant user group, has to be produced in a completely different manner. When remembering old silent movies, Mobile TV actors must show a new profile. In China a made-for-mobile soap opera has been produced [34]. It contains very little dialogues and limited gestures. An actor told the press: ‘‘Your emotions should come only from facial expressions.’’ News transmissions today are usually presented by a talking head, a scenography that does not make sense when there is no sound. If research results of [64,95] are confirmed by further studies, then the traditional talking head for news content should be substituted by a sequence of images, probably supported by some textual information. But even if costumers are using their headphones the sound perception still represents a problem. While commuting the loudness and shape of environmental noise continuously changing and is likely to occasionally mask the sound of presented content. First, it is worth to understand the Mobile TV users’ context in more detail.

S. Buchinger et al. / Entertainment Computing 2 (2011) 175–190

5. Context Most user studies comprise questions related to the mobile service. Researchers and companies need to know which service should be used at which time, how and where. The ideas and concepts for Mobile TV are the same in all studies, but user priorities vary. The culture of the country in which the study has been performed and the age of the test subjects are the predominant factors leading to these discrepancies. 5.1. Physical context Location of mobile TV use is a highly controversial issue. Studies performed in the UK [14], South Korea [14] and Belgium [95] reveal that people clearly want to watch Mobile TV at home. In other studies performed in Germany [6,?], Austria [73], or Japan [64] point out that Mobile TV is mainly consumed during waiting times or on the go. At home, people want to relax, create their private sphere for a short time, e.g., to calm down for some minutes when arriving at home from work before starting to play and chat with their children. Home is the most likely location in studies performed in the UK, USA and Belgium. In the UK and in Belgium people are afraid to disturb the others, and also want to keep their privacy. They do not want strangers to observe what they currently view on their screens. In the USA people mainly use the car to go to work or to school. Hence they do not have the opportunity to watch TV because they drive. The passengers in the back of the car can use it, but in this case the transport happens in a private setting similar to home. In studies performed in countries where public transport is used a lot, and the education to behave in a polite way is not as strong as in the UK or Belgium, Mobile TV consumption occurs during waiting times and in public transportation. In South Korea [14], Mobile TV is mainly consumed at home and while commuting. In this study, emphasis is put on the fact that this observation is highly culture depending and hence does not reflect general trend. Jumisko-Pyykkö et al. [46] found that the majority of respondents use mobile TV with 3D content in lounges (e.g., airport) or public transports. This means that future research and service providers must investigate very carefully the cultural background of the user they want to reach with their product. It is not enough to rely on one well performed user study to know why and where the service will be used in future, and even more so as business success will depend on this choice. 5.2. Temporal context In order to set up adequate programmes and to be prepared for high bandwidth peaks, it is interesting to know when Mobile TV is likely to be viewed and for how much time. Many studies have been published on that purpose and it is very interesting to notice that nearly all recent studies agree about usage durations and times. It has always been stated that Mobile TV content must be short, initially starting by approximately 10 minutes [73], and it has been reduced from year to year, reaching now a length of 1 to 3 min for one content type [74,33], because users use Mobile TV usually as a time-killer for short time periods [57]. Lievens et al. [57] point out that users want to browse quickly through the different channels and often this is the only thing that they do during that period. The same effect happened for the total amount of consumption per day. It is estimated to range between 5 and 40 minutes per day [72]. According to [10], results show that users spend approximately 20 min per day, and the watching time differs from traditional television peak hours. Reasons are that the participants watched Mobile TV while traveling or also at work to

181

get informed about the latest news, which is also confirmed by the study of Kaasinen et al. [47]. However, other studies do even agree on the prime times:  Early in the morning,  at lunch, and  early in the evening, before dinner. The early evening prime time is the most popular [10]. Japanese test persons stated that this is due to the higher number of spare places to sit in public transports [64]. In contrast, watching Mobile TV in the morning while standing in the bus is highly uncomfortable. The lunch break at work or in the courtyard of the school reached the second place. Nearly all studies [6,73,29] agree about these prime times. But they also agree on the fact that with respect to traditional television consumption, these times will undergo more variations. The study of Kaasinen et al. [47] shows that the usage of Mobile TV does not influence the usage of other media and that Mobile TV is often used as extension to the traditional television. Jumisko-Pyykkö et al. [46] found that Mobile 3D TV is suitable for time filling situations and stated that the most interesting option to use it was during a long journey. Other options were the use during waiting situations, commuting or short coffee or lunch breaks. Furthermore, they pointed out that the optimal length of viewing ranged from a couple of minutes to 15 min. This is in accordance to the results obtained by Oksman et al. [70]. However, in this work no prime time could be identified for Mobile TV. The physical context has been identified to be the predominant factor. Another possibility consists in taking the social context into account. Since the social connectors have been identified in [87] to be one important user group it is indispensable to shed light on this aspect. 5.3. Social context The traditional TV viewing in the living room is often not an isolated viewing activity, e.g., it allows viewers to guess with quiz show participants or they record and share TV content with friends which can introduce discussions about the content [15,84]. Therefore, it is often not enough to only pay attention to usability issues [15]. Chorianopoulos [15] declares that mobile devices will play an important role for the next generation of social TV services. In contrast to the traditional television behavior Mobile TV is often used in combination with other activities that users perform simultaneously [57]. Social activities (e.g., chatting while remote viewers watch the news or sending cartoons to other viewers) are possible because communication features can be integrated with television viewing [36,84]. Sørensen et al. [89] confirm that users would like to use their mobile phones as a shared device with others to produce and share mobile videos with communities or groups. Mitchell et al. [62] show that the usage of mobile videos can be divided in individual and collaborative uses, which are influenced by various social factors. For example, although individually viewing is used only to pass time (e.g., waiting spaces), they observed that people use mobile videos as a way of managing relationships with others [62]. Another example is that people can use mobile video to appear busy during their lunch break if they are alone [62]. Social activities for collaborative viewing can be encouraged by the possibility to create own videos with mobile phones. Everyday activities can be shared with other people. Reponen et al. [78] extracted two main, social motivation factors for recording videos in everyday communication: (1) users may be motivated to create videos for their own purpose (e.g., for archiving their own life) or (2) to use them as communication medium (e.g., to share one’s life with other users). Mitchell et al. [62] note that people often show personal content such as family videos to others (e.g., they want to

182

S. Buchinger et al. / Entertainment Computing 2 (2011) 175–190

show a short video to friends which they recorded during their holiday). Or mobile videos can be used as introduction for further discussions within groups (e.g., friends are waiting for a bus and one of them shows a new music video which can induce a discussion about it). The usage of mobile phones in a social context encourages to design new interfaces for better support of social activities between users. Vatavu [96] points out that new technologies (e.g., sensing technology, computer vision, and artificial intelligence techniques) open new ways to design new interfaces for controlling TV sets (e.g., with the help of the use of speech, gesture, tangible objects). Consequently, mobile user’s behavior and their usage of mobile multimedia applications are influenced. Mitchell et al. [62] also state that the design of mobile devices needs further improvements, e.g., docking solutions with speakers/largedisplays, integrated technologies to enable content exchange between mobile devices and also to allow more storage capacity. 6. Mobile services and features On one hand, traditional broadcast programs can be accompanied by supplementary features like chats, voting, SMS and quiz games, on the other hand, applications like video on demand systems eventually including navigation and search or platforms for user-generated content can be implemented. Innovative approaches are presented in [12] where the mobile device is used as a secondary screen for navigation, here previewing content in short clips while using the TV for the main program. Furthermore the content can be manipulated to make it more interesting to share it with other people. Hence interactivity plays an essential role for designing interfaces according to HCI. Bernhaupt et al. [4] discovered that the usability of services influence user ratings. The rating was higher if it was easier to use. Another study [10] noted that success of a service depends on its easy and intuitive use. Jumisko-Pyykkö et al. [46] found that users prefer a simple and intuitive navigation and search function to find the desired content easily. The study of Kaasinen et al. [47] shows that users often have problems to find available services, and according to Hussain et al. [42] it is essential for the success of Mobile TV that users can easily find what they are searching. Based on a study on interactive television [80], it has been shown that older users equally prefer a simple and easy navigation system. Furthermore, they noted that the keys should be easily distinguishable and that users liked to be guided through the various steps. In [4] it is confirmed that guided navigation was easier to follow for younger user groups also. In [68] it has been shown that the usability of an IPTV community platform is crucial for its acceptance. In addition, Jumisko-Pyykkö et al. [46] pointed out that pausing and resuming a program should be possible as well as loop streams. Interaction opens the way for new business models. A very interesting and not defined issue concerning interactivity resides in the field of digital rights management (DRM) for mobile devices. Interactivity enables the user to share, split, combine and edit the content thus creating novel issues of research. The way and the amount of interactivity proposed by a service has high impact on the content needed, therefore this section is tightly related to the next one where content parameters are discussed in more detail.

findings about preferred content types are described in this section. Of course originality will be very important and like for traditional television, it represents always an open issue. In this section is presented about what has been thought till now. In analogy to [72] content types have been assigned in two categories: The professional content that is already known from traditional television and user generated content. 7.1. Professional content Jumisko-Pyykkö et al. [46] mention that several Mobile TV studies found out that the most interesting genres are news, music, sport, and live broadcasts. This is confirmed by many other investigations [29,73,6,74]. However, it is also mentioned that the selection of channels on Mobile TV would be often too limited [47]. In the study of [10] 131 phone users were interviewed about the content they consumed. (see Fig. 5). Interestingly, results revealed a relationship between users’ age and the content type they preferred. Again news were a top content appreciated by nearly all ages. This predilection was even more pronounced for users older than 25 years. For younger people, sports and Music TV were of main interest. In addition, it is worth knowing that news, soap, quiz and sports are those genres during which participants mostly talk while watching and have the highest desire to share the viewing experience for traditional TV [30]. Generally it can be said that people expect to watch the same content as proposed by standard television and the study of Kaasinen et al. [47] shows that users do not often watch new piloted channels and they prefer the mainstream broadcast channels (e.g., MTV3). Reasons are that they know the mainstream broadcast channels and what kind of content they get [47]. For example, if they have a preferred channel for watching the news, they would also like to have it on Mobile TV [47]. However, the content for Mobile TV needs to be adapted to the small screen and differing viewing conditions. A still unsolved problem is the composition of made-for-mobile sports content, for instance soccer matches. On one hand, it is necessary to show the whole pitch to understand what is going on, on the other hand it is necessary to show very small details such as the ball, as to be able to follow the game. In their evaluation about interactive television, Bernhaupt et al. [4] noted that users were interested in local news or weather forecasts. Furthermore, Jensen [43] represents the different and new opportunities which interactive television can provide to its viewers. For example content (like text, graphics or video) can be laid over the currently displayed video

7. Content Interesting or entertaining content represents the main attraction for users. In comparison to other services, platforms like YouTube offer very bad quality but its success is due to highly entertaining content that eclipses the lack of quality. Therefore

Fig. 5. Results of Mobinet 2005 by A.T. Kearney and University of Cambridge – content type [61].

183

S. Buchinger et al. / Entertainment Computing 2 (2011) 175–190

content with further interactive functions, to provide a space for additional informations. Additionally, interactive television gives users the possibility to select content on demand, and provides space for a wide range of content types (e.g., news, movies or music). However, the study of Kaasinen et al. [47] shows that users only want to use on-demand videos if they are easy, effortless and quick enough. Furthermore, users prefer to get information about the size of the video and a descriptive title of the video before they start the download [47]. Personalization of the content becomes more important, e.g., viewers can modify a program or the system individualizes content based on viewer needs. For personalization, Chorianopoulos [16] presents two possible approaches: collaborative filtering and content-based filtering. The collaborative filtering approach makes predictions which are based on ratings from users who share the same rating patterns with the active user [16]. In contrast, the content-based filtering approach makes predictions based on the user’s previous preferences and the user gets a selection of items which may be of interesting to him/her [16]. 7.2. User generated content According to Kenton O’Hara [69], people want to possess and create content. Platforms like YouTube are showing to have success also for mobiles. Users wants to create their own content and to share it with other users in mobile communities [82]. Furthermore, the opportunity to generate content and to share it with other users allows a wide range of possible applications. For example, children and teachers can take photos or videos of the activities in kindergarten which allows parents to see their child’s daily life (e.g., with whom their children played) [66]. Mobile phones can be also used as tools for creating and editing personal multimedia content. Cesar et al. [12] presented an architecture for media content selection and organization, where users are also allowed to personally enrich the content, e.g., by adding text comments or drawings. Users enjoyed the application of these new features and adopted them with ease. Field studies performed in the UK and in Belgium confirmed these results. Furthermore, Juhlin et al. [44] mention that the trend to allow users to broadcast video from mobile devices to a web page as social medium has grown in popularity since 2005. Although users often try the technology, Juhlin et al. [44] point out that for amateurs the opportunity to broadcast video is not enough and therefore it is necessary to develop concepts for better support. Taken together, it can be stated that the sharing of user-generated content constitutes an important user expectation.

Flat fee of max 5€ per month

20.4% 45% 33.7%

No fee, short interruptions for commercials Fee for each content (0.20€ to 0.50€)

Fig. 6. Fee types for Mobile TV [29].

age €9.7 monthly and they would prefer ’’pay per view’’ payment instead of paying for packages of channels, because they also have to pay for channels which they do not watch. Furthermore, Miyauchi et al. [63] point out that reduction of noise and interruptions are important aspects to consider if users have to pay for mobile TV, because they would not pay for it if the TV reception condition is not good enough. Also more recent studies [28] reveal that even early adopters reject to pay for content (42%), they would prefer to watch adds (71%). A user study considering the future of commercials for Mobile TV has been performed by Zukunft Digital [105]. It has been found that men and women have different tastes and expectations. Short interruptions for commercials are felt as less disturbing than in traditional TV. Interactive elements should make sense and have a defined goal. Furthermore, results from [105] confirm the position presented in [72]. It has been shown that advertisements are likely to be considered as much entertaining as the other content and viewed in a similar way. Consumers of interactive services have the opportunity to watch what they want to see, but the effect of the advertisement will be much higher due to the possibility of more accurate tracking and measuring.

9. Technical parameters Many reviews agree about the fact that for Mobile TV, specific made-for-mobile content must be produced, but except for [54] we could not find any indication on how this should be done. Therefore we want to add some recent results showing the importance and influence of technical parameters.

9.1. Response time and channel switching 8. Costs and commercials Costs and commercials are a hot topics in the field of Mobile TV, since someone needs to pay for the service. If interactivity is offered instead of simple broadcasting, users can skip commercials, but commercials are needed to finance the service. In [29] test persons could choose between a flat fee of 6 5 Euros per month, a free Mobile TV service with commercial interruptions like in traditional TV and the possibility to pay a small amount of money (between 0.20 and 0.50 Euros) for single contents. The flat fee was the most attractive option for test users, it has been chosen by 44.9%, followed by 33.7% chosing interruptions through commercials, and 20.4% preferring to pay for single contents (see Fig. 6). In contrast to the commonly used flat fee model, it is possible to pay for each single content consumed. Currently this is a common practice when selling pornographic content. All in all people are not willing to pay much for Mobile TV [95,33,73]. For example, the study of Kaasinen et al. [47] shows that users would accept to pay on aver-

Especially for live sport events the immediacy of the content delivery is very important. Consider the case where there is a time delay of several seconds between the transmission of Mobile and standard TV. An audience seeing the scoring of a goal in an important championship with noticeable delay is likely to be severely annoyed, even if this delay is as low as 15 s. In such a situation, viewers of the delayed presentation may hear cheering from their neighbors, but are still waiting to perceive the goal. Response time is even higher for transmissions over satellite, yet this case is of lesser importance since its main usage scenario is in cars or in rural areas that are not crowded. A similar problem exists for channel switching delays caused by time slicing currently applied to save power consumption. People do not want to wait too long [14] for the appearance of the new channel. Providing low response times for channel hopping, being at least comparable to the ones known from standard TV, is addressed in [79,40] and still represents an important open issue for Mobile TV.

184

S. Buchinger et al. / Entertainment Computing 2 (2011) 175–190

One solution of reducing the channel switching times could consist in pre-loading content such as commercials or short clips and to play them locally while waiting for a response. The impact waiting times on the users’ satisfaction and its increase obtained by showing preloaded content has been presented in [81]. 9.2. Text quality In this section, a simple way of improving Mobile TV quality is presented. Knoche et al. has found [52] that when transcoding a video from standard TV to Mobile TV, the text size from news tickers usually becomes too small. As a consequence the perceived quality can be significantly improved by simply increasing the text size of a news ticker (see Figs. 7 and 8 as an example [52]). 9.3. Framerate When bandwidth is limited a video cannot be transmitted with both full picture quality and smooth motion. Instead, a compromise for the encoding parameters has to be made, which on the one hand satisfies the bandwidth restrictions, but on the other

hand optimizes the perceived quality of the video. This compromise may well depend on the content to be encoded. Therefore it is necessary to know for which content type the fluid motion is more important than detailed picture quality or vice versa. In [7,19] it has been found out that good picture quality with a low framerate returns generally better results than high frame rate with lower picture quality. The result of the user study performed in [7] state that for situations with limited bandwidth budget, for news and comic contents containing little motion, the frame rate should be reduced to 5 frames per second, while for a typical advertisement and soccer game, a value of 10 frames per second should be used. Fig. 9 taken from [7] shows the quality perception for different frame rates in relation to the amount of temporal activity indicating the amount of color changes over time. It can be observed that for low temporal activity (below 8.7) the optimal frame rate is 5 frames per second while for high temporal activity it is 10 frames per second. The curves representing 15 and 25 frames per second result in lower quality parameter values. Furthermore in [19] it is described that impairments after scene changes are noticed less frequently than impairments in other parts of a video. 9.4. Shot types versus resolution

Fig. 7. Recoded video with original news ticker font, Proc. of ACM Multimedia 2006 [52] Recoded video with increased font size, Proc. of ACM Multimedia 2006 [52]

Knoche et al. [51] investigated the effect of different shot types in relation to low resolutions, being typical for mobile devices. For news content the medium shot which portraits the upper half of a subject’s body should be used, but for football content a shot showing less detail is preferred. Furthermore it has been found that for a young audience extreme long shots can be used when the resolution is 240  180 or higher. By choosing a lower resolution the perceived quality may be poor. Knoche et al. considered in [53] the problem described in Fig. 7. For sports events like soccer or ice hockey matches, very large or extremely large shots need to be used, to enable the user to follow the game. These shots have the drawback of showing less detail, therefore the ball or the puck of the match cannot always be identified. In [37] it is mentioned that this happens to almost anyone watching ice hockey. Knoche proposed in [53] a zooming scheme where participants could switch between original and zoom enhanced soccer footage at three sizes with the result that zoom factors between 1.14 and 1.33 were preferred for all sizes. Furthermore it was added that the optimal zooming coefficient depends on the target display size. 9.5. Viewing distance, size and definition The viewing distance for Mobile TV is at arm length. It does not depend on the size of the screen, the content, or the resolution. In [55] it has been evaluated that preferred viewing distances are

Fig. 8. Temporal activity (TA) versus mean opinion score (MOS).

Fig. 9. Second approach of our framework for mobile multimedia applications.

S. Buchinger et al. / Entertainment Computing 2 (2011) 175–190

between 25 and 50 cm with an average of 32 cm. It is stated that ‘‘Mobile TV viewing distances might depend more on the posture of people within a given environment’’. Analysis about the dependencies between size and resolution returned the following result: A resolution requires a minimum size. More in detail: A minimum size of 19.6 mm was required for a resolution of 120  90. Preferred sizes for 120  90 and 168  126 for devices with 116 ppi display are 32.6 mm and 37 mm. 9.6. Audio-visual quality In Section 4.4 it has already been discussed that there are many scenarios where people will not use headphones [64,95]. In this case the question about Audio-visual quality is superfluous. In Section 7 it has been stated that news, soaps, quiz and sport are those genres during which participants talk mostly while watching [30], at the same time these are the contents that people want to consume via Mobile TV. This knowledge leads again to the question whether audio-visual quality investigations do make sense for Mobile TV. Nevertheless, there are situations at home or during the lunch break where audio-visual quality could matter, and in fact, according to [46] users estimated good audio-visual quality to be a positive criterion. In order to obtain a more detailed knowledge on this purpose, audio-visual attention and its influence on subjectively perceived quality has been studied for standard definition multimedia content by performing the following subjective test [56]: A video source showing a moving element (ME) as well as a audio source (AS) was processed in two different ways:  Blurring the entire image, except for the moving element.  Blurring the entire image, except for the audio source. In Table 1, examples of strongly blurred frames that have been shown to the audience of the described experiment are depicted next to their source frame. It has been found that the users’ attention is more attracted by sound emitting objects than by any others. Blurring effects on the sound emitting objects have caused perceived quality degradation, while blurring effects on the moving elements generally did not change the quality perception. It needs to be mentioned that the moving element of some video material used in this experiment contains only looped motion continuously repeated during the whole presentation. This monotony could be one of the reasons why the users’ attention results to be lower for the selected moving elements than for the

sound source. Therefore, ongoing research is focusing on more spontaneous motion elements. 9.7. Region of interest According to the audiovisual study discussed above the visual attention as well as the identification of the regions of interest (ROI) of a video are crucial aspects for subjectively perceived quality estimation. Investigations have been performed on how this information can be integrated in existing quality estimation techniques such as the SSIM [98] and a significant increase in the prediction and accuracy could be shown [20]. Table 2 shows the original image on the left, then in the middle the same image distorted in a region of interest (the face below the eyes) causing high annoyance and finally on the right the image containing a distortion in the background on the top left resulting in only low disturbances to perceived quality. Based on this knowledge and in addition to the fact that mobile devices are small and environmental conditions are far from being optimal, it can be deduced that the quality of the region of interest has even more impact on the quality perception of mobile TV content than it is estimated to be for traditional TV. Equally, background quality is estimated to be of even lower importance. Hence, it has been proposed to automatically detect the foreground an background of a video and to treat them differently by allocating more encoding bitrate to the foreground and less, e.g by using a blur function, to the background [9]. By changing the saturation value of a video frames further improvements regarding the video quality can be reached. Table 3 shows (a) the original frame, then (b) the desired segmentation and (c) the processed image with blurred background with lower saturation and higher saturation values allocated to the foreground. These settings require less encoding bitrate than the original and subjectively perceived quality is currently under study. First results indicate that some quality improvement could be assessed. 9.8. Codec and bitrate Jumisko-Pyykkö et al. [45] tested different codecs for audio/ video material for mobile devices. Six different content types have been under investigation at QCIF format and bitrates in a range from 80 to 128 kbit/s. By summarizing all results the H.264 codec returned the best quality. In order to reduce server bandwidth for a true video on demand system a new concept called Low Start has been presented recently [8]. It consists in encoding the first part of the movie with a lower bitrate than the rest. Action movies encoded with different Low

Table 1 Test material used during the experiment testing the influence of audio-visual attention on subjectively perceived quality [56].

}

185

186

S. Buchinger et al. / Entertainment Computing 2 (2011) 175–190

Table 2 Example illustrating the annoyance of distortions in different regions of an image [20].

Table 3 Improved Mobile TV video frame [9].

Start parameter values for mobile devices have been shown to an audience with the result that a short low start with high bitrate reduction is preferred to a lower bitrate reduction for a longer period of time. The result has been confirmed by performing the MSE [11], SSIM [98] and VQM [103,99] of the test sequences with the MSU Video Quality Measurement Tool [97]. In [2] the bitrate has been set to a maximum value and was then decreased step by step in order to find the minimum bitrate for mobile phone, PDA and laptop videos still rated as acceptable by the test persons. For PDAs bitrate was still acceptable by using 50% of the original bitrate of 448 kbit/s and even more for mobile phone and laptop. 9.9. Transmission schemes We identified three different types of technologies for delivering Mobile TV: satellite-based, terrestrial and telecom-based technologies. Digital Video Broadcasting for Handheld (DVB-H) [23] represents the most common format in Europe to deliver Mobile TV. It offers high downstream data rates and time slicing is implemented to reduce power consumption of small handhelds [16]. Interaction can be realized by combining DVB-H with other technologies. The satellite-based DVB version is DVB-SH as specified

in [25]. MediaFLO represents a competitive alternative to DVB-H. It has been developed in the Unites States and is currently provided by the company Qualcomm. FLO stands for Forward Link Only, therefore the transmission path is one way [75]. Qualcomm has performed a comparative study on DVB-H versus MediaFLO [76]. The Asian competitor from South Korea is called Digital Multimedia Broadcasting (DMB). S-DMB stands for operation via satellite while T-DMB uses a terrestial mode as specified in [24]. Multimedia Broadcast Multicast Service (MBMS) is a telecom-based broadcasting service specified in 3GPP technical specifications [1]. Telecom-based alternatives are Broadcast and Multicast Services (BCMCS) and Time Division-Synchronous Code Division Multiple Access (TD-SCDMA), which is a 3G standard mainly used in the People’s Republic of China. 9.10. Haptic elements Until recently, communication between devices and users was primarily focused on visual and audio channels. However, mobile devices are generally requesting a large variety of demands on users’ attentions requiring additional interaction schemes to satisfy the user. In fact, during the last few years, mobile phones equipped with a touch screen such as the IPhone, Ericsson Satio,

S. Buchinger et al. / Entertainment Computing 2 (2011) 175–190

Samsung Instinct s30, Samsung Eternerty or Nokia 5800 were becoming very popular. In this way a larger variety of communication means based on the sense of touch could be considered. According to MacLean [58], haptic feedback helps to alleviate the visual and audio overload, by providing another communication channel. Bau et al. [3] point out that the haptic channel provides useful information for interacting with the physical environment, because it is fast, needs little conscious attention and allows encoding of abstract information. In particular, Farrugia [26] states that interfaces on touch screens are perceived to be more realistic if a haptic feedback after pressing the virtual buttons is added. However, to build an effective communication through touch, it is necessary to have a deeper knowledge on the type and quantity of adequate abstract information. Hence, knowledge on perception limits and general understandability of distinguishable stimuli and their meanings is required [58]. Therefore, studies have been performed where several haptic patterns were developed and tested. For example, Turunen et al. defined some haptic patterns for mobile devices based on series of pulses with different intensity and delays between the pulses and integrated them in a multimodal media center interface [94]. Referring to the fact that mobile devices are mainly held in hands or carried close to body, such haptic feedbacks are considered to be very effective for mobile devices, e.g., in alerting or informing users via vibrations or varying pressures [85]. However, MacLean [58] underlines that haptic feedback is only useful when it is supportive for the users, otherwise it represents a further annoying distraction. According to Farrugia [26], mobile gaming and phone navigation are examples of successful haptic feedback application. Mixed feedbacks are mainly used for voice and text communication. In the particular case of interactive TV, haptic feedback increases users’ feeling to be more involved in the virtual world [71]. However, Modhrain and Oakley [71] state that haptic feedback for interactive TV requires a very high update rate representing a practical production issue for high quality and stable feedback. 9.11. Context awareness In contrast to standard television, mobile multimedia services are not consumed in a static way, e.g., lying on a sofa or sitting on a chair. It is possible that some users’ watch the same content at the bus station, at home, in a bar or in many other, different places highly differing in their environmental background conditions. Nowadays, mobile phones are equipped with sensors such as accelerometers to detect motion and light sensors to collect data representing the current viewing conditions. Then, these data sets are used to improve the quality of multimedia display, e.g., by switching between portrait and landscape mode according to the current holding position. Many other parameters could be collected to improve the knowledge on the users’ context [13] that could be used in many different ways. For example, a procedure on integrating environmental information in usability evaluation has been proposed [41] and performance improvements due to intelligent systems such as Proactive Replica Placement based on Mobility Prediction have been researched [32]. Context-awareness as well as its deployment seems to be an important and growing topic nowadays. When users are commuting the loudness changes of environmental noise revealed to be very annoying. When changing from silent to noisy areas the content can not be followed any more. Hence, it is necessary to switch to a higher loudness level. As a

187

consequence when returning to more silent places the selected loudness starts to be disturbing and needs to be readjusted. A compensation method for loudness and auditory masking has been presented in [83] and revealed to improve the quality of experience significantly.

10. Discussion In the previous sections a large amount of single aspects have been presented and analyzed. Summarizing, the following development can be observed. From 2005 to 2008 Mobile TV has been set up in several countries with high expectations. The offered services consisted mainly in broadcasting live content and video on demand, both predominantly focusing on professional content. The success was limited due to several reasons: In the beginning, a large amount of technical issues refrained the acceptance of Mobile TV. For example several devices revealed to have problems in receiving data, channel switching times were and still are unacceptably long, proposed user interfaces were far from being intuitive and pleasant to handle. A second important issue were the costs that users had to consider. For consuming Mobile TV it was necessary to buy a new, rather expensive device and to switch to a more expensive contract with the service provider. Hence, potential users were reluctant to spend a noticeable amount of extra money without being certain of the benefit to expect. Content for Mobile TV was mainly obtained by recoding the available TV content to a format of smaller size. In this way, users’ had to watch the same content as on TV just with lower quality. Due to the retarding success of Mobile TV production companies were not open to undertake significant investments for producing made for mobile content. Hence, after the first excitement, Mobile TV did not seem to be as attractive and successful as it supposed to be. However, during the last two years a large number of technical issues have been solved. The availability of improved devices with attractive user interfaces plays an essential role. More and more users are in possession of a smart phone and have signed an extended contract with their service provider, e.g., with flat rate, in order to have access to the internet. Today, additional costs for the user for consuming mobile multimedia seem to be negligible. As a consequence, a large and continuously increasing user group has already adopted several new technologic elements. The barrier for using also mobile multimedia besides other services is noticeably reduced. Furthermore, the amount of available applications has significantly increased. The free access to the iPhone and Android SDK has encouraged a large number of developers to realise their ideas and to contribute them to the community for free or for a small fee on specific platforms. As a consequence, not only the way of consuming mobile multimedia has changed as predicted by Wipperberg, but these new applications also enable the users to contribute their own content. For example the Reel Director App enables the user to edit videos directly on the mobile phone. One motivation for generating and contributing new content is based on the success of the social platforms. Hence, Mobile TV nowadays has a completely different shape than two years ago. It is strongly associated to open decentralised platforms that are formed by the users and the broadcasting television approach is only one of many concepts for Mobile TV. In Section 2.3 the selected mobile multimedia key factors have been integrated in the DIME framework [102]. At this stage, it represents the former Mobile TV status. Development of new technologies and future trends have influenced users’ watching behavior. To consider the changes in the

188

S. Buchinger et al. / Entertainment Computing 2 (2011) 175–190

Fig. 10. Second approach of our framework for mobile multimedia applications.

mobile multimedia communities, we modified our first approach and added feedback loops to the users (see orange arrows in Fig. 10). For example, new services, which assist social activities between users, can change users’ motivations and allow them to see the usage of mobile multimedia applications in a new light (e.g., to use mobile videos to build their own communities). Furthermore, the feedback loops to the users help to involve users in the design and development process of mobile multimedia applications. It can be noticed that due to recent developments in the area of mobile multimedia the Human Centered Design approach is widely applied.

Cost and commercial: the interplay of commercials and the problem of fee types are not solved yet. It will probably depend on the quality and originality of the produced commercials. Technical performance: according to sport, the quality of the presentation is not good enough. For live broadcasting and channel switching the response time is still too high. Currently there is a lot of information available for Mobile TV, but the list of open issues mentioned above shows that there are a lot of interesting and crucial questions that require further investigations.

11. Conclusion and future work References In this survey we analyzed results obtained by manifold studies on Mobile TV from the user perspective, combining the Human Centered Design with the Quality of Experience approach, because it is increasingly necessary to consider and to include user needs and expectations into the design process of a successful product. We identified differences caused by cultural aspects, user profiles and others. Furthermore we presented a wide range of technical aspects that need to be considered in order to create and propose a successful service. Finally, we deduced some important open issues that represent the focus of our future work. User: user groups can be defined by age, device, purpose of use or social context. The best type of user profile still needs to be found. Mobile device: the acceptance of headphones is not clear and further research will be done to investigate in more detail headphone acceptance and implications for the content production in case of headphone rejection. Another issue consists in defining a meaningful usage of haptic elements such as different types of vibrations or arm gestures. Context: context awareness obtained by using sensors or other measurement methods can be used to improve the offered mobile service. Mobile services and features: studies show that user priorities often differ from currently offered services and features. Content: users want to share and modify content on their mobile devices.

[1] 3GPP. Multimedia Broadcast/Multicast Service (MBMS); Architecture and functional description, 2008. Available from: . [2] F. Agboma, A. Liotta, User centric assessment of mobile contents delivery, in: Proceedings of MoMM’06, Indonesia, Yogyakarta, 2006. [3] O. Bau, U. Petrevski, W. Mackay, Bubblewrap: a textile-based electromagnetic haptic display, in: Proceedings of CHI EA’09, ACM, 2009. [4] R. Bernhaupt, M. Obrist, M. Tscheligi, Usability and usage of ITV services: lessons learned in an Austrian field trial, Computers and Entertainment 5 (2) (2007). [5] R. Bernhaupt, M. Obrist, A. Weiss, E. Beck, M. Tscheligi, Trends in the living room and beyond: results from ethnographic studies using creative and playful probing, Computers and Entertainment 6 (1) (2008) 1–23. [6] C. Breunig, Mobile Television in Germany, Media Perspektiven 11 (2006). [7] S. Buchinger, H. Hlavacs, Subjective Quality of Mobile MPEG-4 Videos with Different Frame Rates, Journal of Mobile Multimedia 1 (4) (2006). [8] S. Buchinger, H. Hlavacs, Optimal server bandwidth for mobile video on demand, Annales des Télécommunications 65 (1–2) (2010) 31–46. [9] S. Buchinger, M. Nezveda, W. Robitza, P. Hummelbrunner, H. Hlavacs, Mobile TV coding, in: Workshop on IPTV Technology and Multidisciplinary Applications, Zagreb, Croatia, 2009. [10] C. Carlsson, P. Walden, Mobile TV – to live or die by content, in: Proceedings of HICSS’07, 2007. [11] G. Casella, E. Lehmann, Theory of point estimation, Springer Texts in Statistics, 1998. [12] P. Cesar, D. Bulterman, D. Geerts, J. Jansen, H. Knoche, W. Seager, Enhancing social sharing of videos: Fragment,annotate, enrich and share, in: Proceedings of ACM Multimedia, Vancouver, British Columbia, Canada, October 26–31, 2008. [13] G. Chen, D. Kotz, A survey of context-aware mobile computing research. Technical report, Hanover, NH, USA, 2000. [14] J. Chipchase, C. Yanqing, Y. Jung, Personal television: a qualitative study of Mobile TV users in south Korea, in: Mobile HCI, Conference workshop on Mobile TV, 2006.

S. Buchinger et al. / Entertainment Computing 2 (2011) 175–190 [15] K. Chorianopoulos, Scenarios of use for sociable Mobile TV, in: A. Marcus, A.C. Roibás, R. Sala, editors, Mobile TV: Customizing Content and Experience, Human–Computer Interaction Series Part 1. [16] K. Chorianopoulos, Personalized and mobile digital TV applications, Multimedia Tools Appl. 36 (1–2) (2008) 1–10. [17] H. Crump, E. Schneiders, D. Hoss, Renaissance der terrestrik, Digitalmagazin. Info, Media Perspektiven 2 (2008). [18] K. De Moor, I. Ketyko, W. Joseph, T. Deryckere, L. De Marez, L. Martens, G. Verleye, Proposed framework for evaluating quality of experience in a mobile, testbed-oriented living lab setting, Mobile Networks and Applications 15 (2010) 378–391. [19] N. Ende, H. Hesselle, L. Meesters, Towards content-aware coding: user study, Interactive TV: A Shared Experience 4471 (2007) 185–194. [20] U. Engelke, H.-J. Zepernick, Optimal region-of-interest based visual quality assessment, Proceedings of SPIE Electronic Imaging, vol. 7240, San Jose, USA, 2009. [21] L. Eronen, Combining quantitative and qualitative data in user research on digital television, in: Proceedings of the First Panhellenic Conf. with Inter. Participant on HCI, 2001, pp. 51–56. [22] L. Eronen, User centered research methods for interactive television, in: Proceedings of EuroITV’05, 2005. [23] ETSI, Digital Video Broadcasting (DVB); Transmission System for Handheld Terminals (DVB-H), ETSI EN 302304, Nov 2004. [24] ETSI, Digital Audio Broadcasting (DAB) DMB video service User Application Specification, ETSI TS 102 428, Jun 2005. [25] ETSI, Digital Video Broadcasting (DVB); Framing structure, channel coding and modulation for satellite transmission to handheld (DVB-SH), ETSI EN 302 583, Jan 2008. [26] M. Farrugia, Future user interfaces for mobile devices, in: Proceedings of WWW’06, 2006. [27] M. Fiedler, T. Hossfeld, P. Tran-Gia, A generic quantitative relationship between quality of experience and quality of service, Network, IEEE 24 (2) (2010) 36–41. [28] I. Frank N. Magid Associates. The OMVC Mobile TV Study: Live, Local Programming will Drive Demand for Mobile TV, Technical report, 2009. [29] N. Friedrich, Nutzung? Forschung! Ergebnisse einer repräsentativen Umfrage, in: TVienna, 2008. [30] D. Geerts, P. Cesar, D. Bulterman, The implications of program genres for the design of social television systems, in: Proceedings of UXTV’08, 2008. [31] D. Geerts, K. De Moor, I. Ketykó, A. Jacobs, J. Van den Bergh, W. Joseph, L. Martens, L. De Marez, Linking an integrated framework with appropriate methods for measuring QoE, Workshop on Quality of Multimedia Experience (2010). [32] J. Gossa, A.G. Janecek, K.A. Hummel, W.N. Gansterer, J.-M. Pierson, Proactive replica placement using mobility prediction, Mobile Data Management Workshops 0 (2008) 182–189. [33] G. Graf, Mobisodes & Co – ein Praxistest. Magazin TENDENZ, Media Perspektiven, 2006. [34] Guardian. Mini-series for upwardly mobile chinese, Guardian News & Media 2008, 2005. Available from: . [35] B. Hanington, Methods in the making: a perspective on the state of human research in design, Design Issues 19 (4) (2003) 9–18. [36] G. Harboe, Introduction to social TV, in: A. Marcus, A.C. Roibás, R. Sala (Eds.), Mobile TV: Customizing Content and Experience, Human–Computer Interaction Series Part 1. [37] R. Harper, T. Regan, M. Rouncefield, Taking hold of TV: learning from the literature, in: Proceedings of OZCHI’06, Sydney, Australia, 2006. [38] M. Hassenzahl, A. Platz, M. Burmester, K. Lehner, Hedonic and ergonomic quality aspects determine a software’s appeal, in: Proceedings of CHI’00, ACM, New York, NY, USA, 2000. [39] M. Hassenzahl, N. Tractinsky, User experience – a research agenda, Behaviour & Information Technology 25 (2) (2006) 91–97. [40] C.-H. Hsu, M. Hefeeda, Bounding switching delay in Mobile TV broadcast networks, in: Proceedings of SPIE Multimedia Computing and Networking, vol. 7253, 72530A, 2009. [41] K. Hummel, A. Hess, T. Grill, Environmental Context Sensing for Usability Evaluation in Mobile HCI by Means of Small Wireless Sensor Networks, in: Proceedings of MoMM’08, 2008. [42] Z. Hussain, M. Lechner, H. Milchrahm, S. Shahzad, W. Slany, M. Umgeher, T. Vlk, P. Wolkerstorfer, User interface design for a mobile multimedia application: an iterative approach, in: Proceedings of ACHI’08, Feb 2008. [43] J.F. Jensen, Interactive television: new genres, new format, new content, in: Proceedings of IE’10, 2005. [44] O. Juhlin, A. Engström, E. Reponen, Mobile broadcasting: the whats and hows of live video as a social medium, in: Proceedings of MobileHCI’10, ACM, New York, NY, USA, 2010. [45] S. Jumisko-Pyykkö, J. Häkkinen, Evaluation of subjective video quality of mobile devices, Proceedings of ACM Multimedia, 2005. [46] S. Jumisko-Pyykkö, M. Weitzel, D. Strohmeier, Designing for user experience: what to expect from mobile 3d TV and video?, in: Proceedings of UXTV’08, 2008. [47] E. Kaasinen, M. Kulju, T. Kivinen, V. Oksman, User acceptance of Mobile TV services, in: Proceedings of MobileHCI’09, ACM, New York, NY, USA, 2009. [48] A. Kill, MobiTV, Inc. Confirms Size Does Matter, Technical report, 2010.

189

[49] H.-J. Kim, K.-H. Lee, J. Zhang, In-service feedback QoE framework, in: Proceedings of CTRQ’10, 2010. [50] H. Knoche, J. McCarthy, Mobile users needs and expectations of future multimedia services, in: Proceedings of WWRF12, 2004. [51] H. Knoche, J. McCarthy, M. Sasse, How low can you go? the effect of low resolutions on shot types in mobile TV, Personalized and Mobile Digital TV Applications in Springer Multimedia Tools and Applications Series 36 (1–2) (2008) 145–166. [52] H. Knoche, J. McCarthy, M.A. Sasse, Reading the fine print: the effect of text legibility on perceived video quality in mobile TV, in: Proceedings of ACM Multimedia, Santa Barbara, USA, October 2006. [53] H. Knoche, M. Papaleo, M. Sasse, A. Vanelli-Coralli, The kindest cut: enhancing the user experience of mobile TV through adequate zooming, in: Proceedings of ACM Multimedia, Augsburg, Germany, 2007. [54] H. Knoche, M. Sasse, Getting the Big Picture on Small Screens: Quality of Experience in Mobile TV, Idea Group, 2007. [55] H. Knoche, M. Sasse, The sweet spot: how people trade off size and definition on mobile devices, in: Proceedings of ACM Multimedia, Vancouver, Canada, 2008. [56] J.-S. Lee, F. De Simone, T. Ebrahimi. Influence of audio-visual attention on perceived quality of standard definition multimedia content, in: Proceedings of QoMEX’09, 2009. [57] B. Lievens, E. Vanhengel, J. Pierson, A. Jacobs. Does mobile television enhance a new television experience?, in: A. Marcus, A.C. Roibás, R. Sala, (Eds.), Mobile TV: Customizing Content and Experience, Human–Computer Interaction Series Part 1. [58] K.E. MacLean. Using haptics for mobile information display, in: Proceedings of PERMID’08, 2008. [59] M. Maguire, N. Bevan, User requirements analysis: a review of supporting methods, in: Proceedings of IFIP’02, 2002. [60] J. Mäki. Finnish Mobile TV results, August 2005. [61] N. Menon, M. Page, M. Watt, S. Bell. Mobile data services: a selection of key findings, in: Proceedings of Mobinet’05, 2005. [62] A.S. Mitchell, K. O’Hara, A. Vorbau. Social properties of mobile video, in: A. Marcus, A.C. Roibás, R. Sala, editors, Mobile TV: Customizing Content and Experience, Human–Computer Interaction Series Part 1. [63] K. Miyauchi, T. Sugahara, H. Oda, Different attitudes concerning the usage of live mobile tv and mobile video, in: A. Marcus, A.C. Roibás, R. Sala (Eds.), Mobile TV: Customizing Content and Experience, Human–Computer Interaction Series Part 1. [64] K. Miyauchi, T. Sugahara, H. Oda, Relax or study?: a qualitative user study on the usage of mobile tv and video, Proceedings of EuroITV’08, vol. 5066/2008, Salzburg, 2008. [65] S. Moller, K.-P. Engelbrecht, C. Kuhnel, I. Wechsung, B. Weiss. A taxonomy of quality of service and quality of experience of multimodal human–machine interaction, in: Workshop on Quality of Multimedia Experience, Jul 2009. [66] J. Näsänen, A. Oulasvirta, A. Lehmuskallio. Mobile media in the social fabric of a kindergarten, in: Proceedings of CHI’09, ACM, New York, NY, USA, 2009. [67] Nokia. DVB-H. Mobile TV Forum. Available from: (accessed 18.12.2008). [68] M. Obrist, Usability & user experience: preliminary results from evaluating an IPTV community platform, in: Proceedings of EuroITV’08, Salzburg, Austria, 2008. [69] K. O’Hara, A.S. Mitchell, A. Vorbau, Consuming video on mobile devices, in: Proceedings of CHI’07, April 2007. [70] V. Oksman, E. Noppari, A. Tammela, M. Mäkinen, V. Ollikainen. Mobile TV in everyday life contexts – individual entertainment or shared experiences?, in: Proceedings of EuroITV’07, 2007. [71] S. O’Modhrain, I. Oakley. Touch TV: adding feeling to broadcast media, in: Proceedings of EuroITV’03, 2003. [72] S. Orgad, This box was made for walking. Nokia Mobile TV report (2006). [73] O. Petrovic, M. Fallenbck, C. Kittl, A. Langl, Mobile TV in Austria 2 (2006). [74] Projektgruppe Mobiles Fernsehen, Mobiles Fernsehen: Interessen, potentielle Nutzungskontexte und Einstellungen der Bevölkerung, Media Perspektiven (2007). [75] Qualcomm, FLO Technology overview (referred 19.12.2008), 2007. Available from: . [76] Qualcomm, Technology Comparison: MediaFLO and DVB-H (referred 19.12.2008), 2007. Available from: . [77] P. Repo, K. Hyvönen, M. Pantzar, P. Timonen, Users inventing ways to enjoy new mobile services – the case of watching mobile videos, in: Proceedings of HICSS’04, 2004. [78] E. Reponen, J. Lehikoinen, and J. Impiö, Mobile video in everyday social interactions, in: A. Marcus, A.C. Roibás, Sala, (Eds.), Mobile TV: Customizing Content and Experience, Human–Computer Interaction Series Part 1. [79] M. Rezaei, I. Bouazizi, V. Vadakital, M. Gabbouj. Optimal Channel Changing Delay for Mobile TV Over DVB-H, in: Proceedings of PORTABLE’07, 2007. [80] M. Rice, N. Alm, Designing new interfaces for digital interactive television usable by older adults, Computers and Entertainment*** 6 (1) (2008) 1–20. [81] W. Robitza, S. Buchinger, H. Hlavacs, Acceptance of mobile TV channel switching delays, in: Workshop on Quality of Multimedia Experience, Trondheim, Norway, 2010. [82] A.C. Roibás, S. Johnson. Unfolding the user experience in new scenarios of pervasive interactive TV, in: Proceedings of CHI’06, 2006.

190

S. Buchinger et al. / Entertainment Computing 2 (2011) 175–190

[83] M. Sack, S. Buchinger, H. Hlavacs, Loudness and auditory masking compensation for Mobile TV, Proceedings of BMSB’10, Shanghai, China, 2010. [84] R. Schatz, L. Baillie, P. Fröhlich, S. Egger, T. Grechenig, What are you viewing? exploring the pervasive social tv experience, in: A. Marcus, A.C. Roibás, R. Sala (Eds.), Mobile TV: Customizing Content and Experience, Human–Computer Interaction Series Part 1. [85] A. Sears, J.A. Jacko, Human–Computer Interaction: Design Issues, Solutions, and Applications, Crc Pr Inc., 2009. [86] M. Siller, W.J. QoS Arbitration for Improving the QoE in Multimedia Transmission, in: Proceedings of VIE’03, 2003. [87] E. Simmons, 2010 American Mobile Consumer Report, Technical report, 2010. [88] C. Sodergard, Mobile television - technology and user experiences, vol. 506, VTT Publications, 2003. [89] L. Sørensen, H.W. Nicolajsen, Generating ideas for new mobile TV services: accepting and socializing mobile television, in: Proceedings of EuroITV’10, ACM, New York, NY, USA, 2010. [90] L. Stockbridge, Mobile TV: Experience of the UK Vodafone and Sky Service, in: Proceedings of EuroITV’06, Athens, Greece, May 2006. [91] D. Stone, C. Jarrett, M. Woodroffe, User Interface Design and Evaluation, Morgan Kaufman Publ. Inc., 2005. [92] Study Group 12. Quality of Experience Requirements for IPTV Services, ITU-T Recommendations G.1080, ITU, 2008. [93] B. Svoen, Consumers, participants, and creators: young people’s diverse use of television and new media, Comput. Entertain*** 5 (2) (2007). [94] M. Turunen, J. Hakulinen, J. Hella, J.-P. Rajaniemi, A. Melto, E. Mäkinen, J. Rantala, T. Heimonen, T. Laivo, H. Soronen, M. Hansen, P. Valkama, T. Miettinen, R. Raisamo, Multimodal interaction with speech, gestures and haptic feedback in a media center application, in: Proceedings of INTERACT’09, 2009.

[95] M. Vangenck, A. Jacobs, B. Lievens, E. Vanhengel, J. Pierson, Does mobile television challenge the dimension of viewing television?, An explorative research on time, place and social context of the use of mobile television content, Proceedings of EuroITV’08, vol. 5066, Salzburg, 2008. [96] R.-D. Vatavu, Creativity in interactive TV: personalize, share, and invent interfaces, in: A. Marcus, A.C. Roibás, R. Sala (Eds.), Mobile TV: Customizing Content and Experience, Human–Computer Interaction Series Part 1. [97] D. Vatolin, A. Moskvin, and O. Petrov. Msu video quality measurement tool, December 2008. Available from: . [98] Z. Wang, E. Simoncelli, A. Bovik. Multi-scale Structural Similarity for Image Quality Assessment, in: Proceedings of the 37th IEEE Asilomar Conf. on Signals, Systems and Computers, 2003. [99] A. Watson, Toward a perceptual video quality metric, San Jose CA, 1998. [100] L. Whitney, Smartphones to dominate PCs in Gartner forecast, Technical Report, 2010. [101] J. Wippersberg. Die Forderung nach ‘‘sehbarem Fernsehen, in: TV 3.0 Journalistische und politische Herausforderungen des Fernsehens im Digitalen Zeitalter, Germany, Berlin, Mar 2008. [102] W. Wu, A. Arefin, R. Rivas, K. Nahrstedt, R. Sheppard, Z. Yang, Quality of experience in distributed interactive multimedia environments: toward a theoretical framework, in Proceedings of MM’09, ACM, New York, NY, USA, 2009. [103] F. Xiao. Dct-based video quality evaluation, 2000. Available from: . [104] D. Zillmann, The coming of media entertainment, in: D. Zillmann, P. Vorderer (Eds.), Media entertainment: the psychology of its appeal, Lawrence Erlbaum Associates, 2000, pp. 1–20. [105] Zukunft Digital, Mobile TV Studie, Werben & Verkaufen, 36 (2008).