Page 85 - Trust in ICT 2017

Trust in ICT 1

Figure 26 – MAPE-K feedback loops for adaptive trust agents

The system is highly dynamic which implies the need for adaptive decision making and autonomic agents
with control loops to manage resources. A promising approach to handle such dynamics is self-adaptation
that can be realized by a MAPE-K feedback loop. To provide an evidence that the system goals are satisfied,
regarding the changing conditions, state of the art advocates the use of formal methods. However, it is
important to remark that the trust agents in Figure 43 do not replace the monitoring phase of the MAPE-K,
but instead it filters out the trust information from other information while holding the required knowledge
to support the autonomic decision-making process.

The distributed nature of the trust agents assure quick responses and scalability of the solution. In Figure 43,
the monitor function aggregates, correlates and further filters the information until it determines a symptom
that needs to be analyzed. Analyze function performs complex data analysis and reasoning on the symptoms
provided by the monitor function. Analyze function would be influenced by stored knowledge data which, in
fact, virtually centralized but physically exists within the trust agents. If changes are required, a change
request is logically passed to the plan function. The plan function structures the actions needed to achieve
goals and objectives and creates or selects a procedure to enact a desired alteration in the managed resource.
At the same time it can take on many forms, ranging from a single command to a complex work-flow.
Execution phase changes the behavior of the managed resource using effectors, based on the actions
recommended by the plan function. In fact, the executors are open APIs to the trust managers’ feedback
system.
The knowledge in Figure 26 is the standard data associated with the monitor, analyze, plan and execute
functions. The knowledge here is shared among the trust agents and could be virtually centralized using cloud
techniques to facilitate decision making. This would include data such as all trust related information, context
information, topology information, historical logs, metrics, symptoms, policies, etc. This system now
becomes self-adaptive based on MAPE-K feedback loops that deal with dynamic trust issues arising due to
openness. It is important to notice that the particular focus is on adaptations that require elevating or
downgrading the level of trust in a system.

8.6 Using Blockchain as Tool

Blockchain technology can assist smart devices to become autonomous agents, independently conducting a
different of transactions. Blockchain technology in case of not existing of a centralized server brokering
messages, enhancing file storage, transmissions and deciding roles any decentralized IoT. Applying the

77