Subscribe Us

Get free daily email updates!

Follow us!

Tuesday, May 10, 2011

The Cognitive federated environment: Towards Personalization and Individualization of Devices and Services.


Intelligent networking and connectivity refers to the ability of a network to adapt its services to the user’s context and to support connections based on the application services attributes. These are considered the key technologies for fostering the mobility of users and the versatility of services. This is due to the fact, that when intelligence is contained in the networking protocols (e.g., information on security needs, user context and profiling, priority, and QoS issues) then the business applications can be built in a more efficient way and consequently to allow for broader functionality in the service layer.


The research towards cognitive packet networks goes in parallel with advances
towards knowledge-enabled applications and services. Topics of special interest are
issues regarding the availability of resources, the online adaptation in response to
quality of service (QoS) needs, the perceived performance of the different network
infrastructure elements, and so forth. Ad hoc scenarios require the collaboration between network nodes, which can be user terminals (UTs), base stations (BSs), sensor nodes, and so forth, depending on the type of ad hoc network. The most important is the exchange of the sensing information, but also the exchange of policies, results of (local or individual) decisions or any other type of network or user related context information is required. This in turn requires the support of efficient encoding and transport of sensing data as well as the support of generic and extensible (XML) data exchange.

The communication of sensing information or context, in general, in the course of distributed and collaborative sensing and decision-making must be suitable for both link layer and network layer transfers. Here, link layer protocols are fast and lightweight but suitable only for information dissemination in the direct vicinity, (e.g., by one-hop broadcast such as in ad-hoc scenarios). Network layer protocols can be designed for assured point-to-point and point to multipoint communication, such as in multihop communications for network wide context data exchange.
Personalization of services allows the user for deciding, in which way and when to
make use of services and how to manage these [14]. Service providers and operators,
thanks to personalization, may obtain loyal customers by meeting their expectations
in a more effective and efficient way. To be able to achieve personalization, it is important to understand the concepts of usability and user experience. The motivation behind the introduction of the ‘user experience’ concept is found in the lack of subjectivity related to using a product. The user experience includes emotional and aesthetic issues, which exceed the quality of use and put focus on the quality of experience. If a user’s expectations are met, then there is a neutral user experience, if his/her expectations are not met, then there will be a negative experience and if the user’s expectations are more than met there is a positive user experience. This understanding leads to a more relative measure of the user evaluation than many of the ones included in more traditional tests of usability.

In an ad hoc environment, the system is not only networked and mobile but also based on the covering of unlimited activities. These activities will be shifting and may include both known and unknown services and relations (to both technologies and people), which complicate the matter even further. In order to evaluate the usability and user experience for a scenario of changing rapidly different activities, different contexts, and for different networked interactions and relations, issues such as security, trust, control and navigating around the different possibilities, requests, and context become key.

Services can have a list of associated personalized parameters, (e.g., volume, QoS). During the service instantiation, the personalization returns the preferred values for the requested personalizable parameters to the instantiated service allowing it to personalize itself. Services may need to personalize again when context changes occur. When this happens, the process of personalization will notify any updated preference outcomes to the appropriate services allowing them to repersonalize at runtime, dynamically changing to meet the user’s needs [16]. Such context changes may also affect the architecture of a running service. Composite services, made from two or more services may need to be recomposed due to a context change.

Services are described by means of ontologies. Personalization ontologies are manually coded conceptual formalizations of user situations, activities, goals, and behavior patterns (e.g., “the user is at a business meeting,” or “the user is going to take a person to the airport,”) and preference outcomes and pro-active actions (e.g., “background color,” or “start service of type T”), which can be used to describe the user preferences in a compact, readable, and meaningful way and to provide useful metadata when learning them automatically the from user action histories.

Thus, context is any information that can be used to characterize the situation of an entity. An entity is a person, a place, or an object that is considered relevant to the interaction between a user and an application, including the user and the application themselves. The total set of user-related information, preferences, rules, and settings, which affects the way, in which a user experiences terminals, devices, and services, is defined as a user profile. In the general case, it is a complex undertaking to decide, which part of the entire available user profile and context information is relevant and useful for performing a service adaptation. Organizing the information in an ontology supports reasoning and decision-making, but there is a lot more research to be done on how to combine this with an intelligent application logic and policies that provide a proper protection of the user privacy.

Today, a large number of Web sites offer users or subscribers a basic level of
personalization. This can be initiated, when the user signs up for the first time, where typically a set of personal data such as name, address, e-mail address, phone number(s), may be requested, and the user chooses a user-ID and password to access the personalized services later on. Furthermore, the user is often given the option of ticking various preferences or areas of interest. More sophisticated services will collect data about the usage history and based on this perform some “intelligent” processing in order to provide relevant information or offers to the user. In an existing situation, the large number of third party profiles have to be handled independently by the user, and the personalization benefits are somewhat limited.

Another trend emerging with advances in the area of ad hoc networks is the sharing of content. Due to a proliferation of tools for content production and sharing, it has become quite easy for the average customer to become a content—in addition to being a service provider. This proliferation of information, content, products, and services provided and owned by individual users, introduces new opportunities and challenges, namely, the following:
•Service integration instead of service provisioning;
•Privacy and identity management;
•Seamless interaction/access to the telecommunications infrastructure providing a high level of experience.

The next generation network can be viewed as a plethora of autonomous domains of various sizes that intersections that would need to cooperate based on dynamic service level agreements and associated trust relationships [21]. Such a trend imposes the requirements for interoperability of solutions and segments on horizontal and vertical levels. An example of a federated environment is shown in the illustration below.

In the illustration above, it seen that third party providers seek new opportunities in terms of acquiring roles in emerging complex value chains and systems, mainly targeting the exploitation of network capabilities opened-up by the network operators. Third parties can thus become providers of enabling services (platform/middleware providers) or end-service providers (content/application providers). The wireless hot spots are rapidly becoming a dominant technology in urban environments. A major consequence is the high fragmentation of the low layers of the value chain, namely the venue (location owners) and the hot-spot operator layer. The hot spots a great opportunity for mobile network operators and Internet service providers (ISPs) to add additional services to their portfolios. The aggregated service providers serve as middlemen between the content/application providers, and the enabling services providers. These cover basic roles such as roaming between networks/services, variable end-to-end QoS provisioning, and service bundling.

To sum it, in a next generation network, the user will play an increasingly central role not only as a complex consumer, but also as a new type of provider


0 Responses to “ The Cognitive federated environment: Towards Personalization and Individualization of Devices and Services. ”

Post a Comment