Page 23 - Trust in ICT 2017

Trust in ICT 1

Figure 2 – Different views on trust

5 Use cases and explanation of trust provisioning
This section illustrates various use cases for trust provisioning based on the technical report of ITU-T CG-Trust
and materials from other SDOs (e.g., oneM2M) as well as related literature. In addition, this section also
analyses these uses cases in terms of purpose, method, actors and considerations for measuring trust.

5.1 Trust Use Cases in Networking Aspects
5.1.1 Trust-based routing protocols

5.1.1.1 Description
Secure routing is especially important in wireless networks. However there are many attacks toward wireless
network routing protocols due to their open, distributed and dynamic nature.

In ad-hoc and sensor networks, it is very important to secure each node. An adversary may overtake some
critical nodes and inject malicious behaviours, which leads to revelation of secure information and collapse
of entire network. There are two common types of misbehaving nodes: selfish nodes and malicious nodes. If
a node does not cooperate in packet forwarding due to some resource constraints, such as low memory or
battery life, it is said to be selfish node. A selfish node may not have any intention to destruct the system; an
adversary may reprogram a compromised node to behave selfishly. On the other hand, a malicious node has
an objective to destruct the system badly, even at the cost of its own resources.
The security attacks in ad-hoc and sensor networks may be compared and classified from multiple
perspectives. One way of classifying attacks is based on capabilities and resources an adversary has in his
possession. In this type of classification, attacks may be classified as outsider (external) attack and Insider
attack. In outsider attack, attacker lacks authentication and key information and such type of attack can easily
be dealt with classical security mechanism such as cryptography, encryption and authentication. In insider
attack, an adversary already has all key and cryptographic information, therefore such type of attack cannot
be dealt with traditional security measures.

Another classification is based on adversary’s intention to destruct the system. The attacks may be classified as
Trust Management related attack and network related attack. The intention of Trust Management related
attack is to degrade the performance of trust management system which leads to the inaccurate decisions. For
example, in trust aware routing mechanisms, if misbehaving nodes are not properly detected and isolated by
trust management system, then these nodes may become part of selected routing path and perform malicious
activity. In network related attack, the intention of an adversary is to destruct overall performance of network
by intentionally dropping data packets, energy drain and reporting incorrect sensed data. Such attacks can be
detected and prevented by trust management system. For example, a black-hole attack intentionally drops all
the received packets, which in results degrade the overall network performance in terms of Packet Delivery
Ratio (PDR). Yet another way to characterize attacks is based on perspective of the efficacy of countermeasure,
such as, traditional security solutions and trust based security solutions, to prevent attacks.

15