Trust Management for Defending On-Off Attacks

By Admin on

A trust management scheme can be used to aid an automated decision-making process for an access control policy. Since unintentional temporary errors are possible, the trust management solution must provide a redemption scheme to allow nodes to recover trust. However, if a malicious node tries to disguise its malicious behaviours as unintentional temporary errors, the malicious node may be given more opportunities to attack the system by disturbing the redemption scheme. Existing trust management schemes that employ redemption schemes fail to discriminate between temporary errors and disguised malicious behaviours in which the attacker cleverly behaves well and badly alternatively. In this paper, we present the vulnerabilities of existing redemption schemes, and describe a new trust management and redemption scheme that can discriminate between temporary errors and disguised malicious behaviours with a flexible design. We show the analytical results of the trust management scheme, and demonstrate the advantages of the proposed scheme with simulation conducted in a Wireless Sensor Network.

Direct Trust is established upon observations on whether the previous interactions between the subject and the agent are successful. The observation is often described by two variables: s denoting the number of successful interactions and f denoting the number of failed interactions. For example, in the beta-function based method , the direct trust value is calculated as s+1 s+f+2 . Recommendation trust is a special type of direct trust. It is for trust relationship {subject: agent, making correct recommendations}. When the subject can judge whether a recommendation is correct or not, the subject calculates the recommendation trust from sr and fr values, where sr and fr are the number of good and bad recommendations received from the agent, respectively. This judgement is often done by checking consistence between observations and recommendations, or among multiple recommendations. When using the beta-function based methods, the recommendation trust can be calculated as sr+1 sr+fr+2 .

 Indirect Trust: Trust can transit through third parties. For example, if A has established a recommendation trust relationship with B and B has established a trust relationship with Y , then A can trust Y to a certain degree if B tells A its trust opinion (i.e. recommendation) about Y . This phenomenon is called trust propagation. Indirect trust is established through trust propagations. Two key factors determine indirect trust. The first is when and from whom the subject can collect recommendations. For example, in a sensor network, a sensor may only get recommendations from its neighbour when there is a significant change in their trust records. This affects the number of available recommendations and the overhead of collecting recommendations. The second is to determine how to calculate indirect trust value based on recommendations. When node B establishes direct trust in node Y and node A establishes recommendation trust in node B, A − B − Y is one recommendation path. One recommendation path can contain more than two hops, such as A − B1 − B2 − · · · − Y , and there may exist multiple recommendation paths, such as A − B1 − Y , A − B2 − Y , · · · etc. Trust models determines how to calculate indirect trust between A and Y from trust propagation paths. There have been many trust models proposed for various applications .

for more details click here

SPIRO Google Plus