Dynamic P2P trust model based on time-window feedback mechanism

The trust models employed by the existing systems do not provide adequate support to cope with dishonest feedbacks, strategically altering behaviors of malicious peers efficiently and the incentive mechanism. The time-window based local trust model TW-Trust was presented to solve these problems. It considers the inherent connection among trust, reputation and incentive and the effect of time factor on the trust computation. Six trust parameters are introduced to compute trustworthiness of peers, named direct trust, history reputation, difference reputation, misusing reputation, similarity and incentive utility. These parameters are adjusted in time to reflect the dynamic of the trust environment using feedback control mechanism, thus, the trust evaluation has better adaptability to the dynamic of trust. Theoretical analysis and experimental results show that the TW-Trust has advantages in countering strategic altering behavior and dishonest feedbacks of malicious peers and solving the sparseness issue in the computation of similarity. It is more effective in trust computation error and transaction failure rate than the previous models.