Page 55 - Trust in ICT 2017

Trust in ICT 1

A trust management concerns part or all of trust properties in different contexts for different purposes and
should achieve the following goals [30]:
(1) Trust relationship and decision: trust management provides an effective way to evaluate trust
relationships of any two entities and assist them to make a wise decision to communicate and
collaborate with each other.
(2) Data perception trust: data sensing and collection should be reliable in the trust management
system.

(3) Privacy preservation: user privacy including user data and personal information should be flexibly
preserved according to the policy and expectation of IoT users. This objective relates to the IoT
system objective properties in general.
(4) Data fusion and mining trust: the huge amount of data collected in IoT should be processed and
analyzed in a trustworthy way with regard to reliability, holographic data process, privacy
preservation and accuracy.
(5) Data transmission and communication trust: data should be transmitted and communicated
securely in the IoT system. Unauthorized system entities cannot access private data of others in data
communications and transmission.
(6) Quality of services: QoS should be ensured.
(7) System security and robustness: trust management should effectively counter system attacks to
gain sufficient confidence of system users.
(8) Generality: trust management for various systems and services is preferred to be generic that can
be widely applied, which is a system objective property.
(9) Human-Computer Trust Interaction: trust management provides sound usability and supports
human–computer interaction in a trustworthy way, thus can be easily accepted by its users.
(10) Identity trust: The identifiers of system entities are well managed for the purpose of trustworthy.
Scalable and efficient identity management in is expected.
[31] proposed a mechanism for extracting trust information from the security system of a service based on
the needs of an entity. Trust is used as a security metric between an entity and systems. [32] proposed a P2P
trust model. An adaptive trusted decision making method based on historical evidences window is used to
improve system efficiency. In Ad hoc network, an entropy theory based distributed trust model provided a
mechanism to select trusted paths [33]. The trust value of each path is obtained through multi-layer and
multi-level calculation, and someone can choose credible routes to implement the interaction. For WSN, a
cluster-based layered trust scheme is characterized as a typical model [34]. Based on the trust values, a node
assigns a trust state to other nodes. It calculates the trust value of the sensor nodes at each level, and choose
a set of nodes to participate in the transaction. From above investigated trust solutions, some elements or
attributes of trust management can be extracted:
• Service. It defines the role of the trust management. The basic idea of trust management is that the
security decision needs to rely on the additional safety information provided by a trusted third party.
Trust, as a “soft” third party, provides a service for the service requester and the service provider in
a network system.

• Decision making - the purpose of the trust management. Trust is collected to judge the credibility of
the cooperative nodes, based on which make a decision to deliver a service, select a credible routing
and transmit a data.
• Self-organizing. It depicts the way of the trust management. Based on trust decision, a series of
nodes or even sub-networks can be selected and self-organized to perform a certain task (i.e.
forwarding the packages, sensing the data) cooperatively in network scene (i.e. IoT).

In this trust management approach, service, decision making and self-organizing are the three basic essential
elements.
Trust management in MANETs is needed when participating nodes, without any previous interactions, desire
to establish a network with an acceptable level of trust relationships among themselves. Examples would be

47