A dual-scale trust-aware method based on auxiliary decision-making in social networks

Abstract

The present invention relates to the technical field of network security situation awareness, in particular to a social network-based dual-scale trust perception method for auxiliary decision-making, which uses the existing service satisfaction degree of the network entity to calculate the trust degree of the network entity; uses a trust measurement function to detect the trust of the network entity According to the current trust deviation degree of the network entity, the entity is classified into the trust type set by the system, and the trust degree of the network entity is adjusted; the present invention experiences trust acquisition, understanding, decision support, performance evaluation and management, and in time , referring to the Ebbinghaus forgetting rate, in terms of space, according to the correlation of network entities, establish trust and service satisfaction, and further use load balancing technology to effectively avoid network resource congestion, effectively identify and deal with malicious entity behavior, and realize Situational awareness, sensitivity and trust measurement methods for malicious entities are more realistic.

Description

technical field [0001] The invention relates to the technical field of network security situational awareness, in particular to a social network-based dual-scale trust awareness method based on auxiliary decision-making. Background technique [0002] Situational awareness can acquire, understand, and predict the state or trend of sensing elements under certain conditions. It is currently used in aircraft situational awareness, anti-missile early warning systems, space surveillance systems, smart dust, etc. Trust perception is to specifically use the trust of network entities in the trust-based network security mechanism as the perception element of situational awareness, and then perform processing such as acquisition and prediction. [0003] With the rise of ubiquitous network and network fusion technology, the network is endowed with characteristics such as frequent interaction between resources, a sharp increase in network traffic and data volume, and is also endowed with...

AI Technical Summary

Problems solved by technology

[0004] The trust-based network security mechanism can be divided into four parts: information collection, rating evaluation, entity selection, and behavioral rewards and punishments. Trust is established by referring to direct trust, indirect trust, or third-party recommendations, but a small number of mechanisms have formed a trust ring, and most There is no real-time decision-making process for malicious behavior, and the attack and malicious behavior entities cannot be visualized

The research on the trust-based security mechanism needs to be improved in several aspects: 1) In the time dimension, the weight distribution of historical trust degree adapts to the law of time decay, and the closer the event, the higher the weight; in the space dimension, the comprehensive trust degree calculation When , the weight distribution of direct trust degree and indirect trust degree is fixed, and cannot adapt to dynamic changes, and the introduction of third-party recommendation trust does not examine the availability and trust degree of the third party, and it is not known whether the third party is captured, which is likely to cause security risks; 2) Most mechanisms mainly consider how to establish a trust model, without taking decisions on malicious behavior, and there is no threat assessment; 3) The mechanism only has one evaluation standard, which does not allow the existence of entity subjectivity, and each entity does not have an independent standard, so it cannot be adapted to The special needs of each entity; 4) Attack identification and decision-making are incomplete, and there is no plan for identifying and responding to abnormal attacks; 5) There is no workload balancing mechanism, and most mechanisms choose the most credible entity to provide services, which is easy to form a monopoly and cause excess Load and resource congestion; 6) A penalty time interval should be set for entities rejected by the system to return to the system; 7) Trust is only established for one type of service, and entities provide trusted services for certain service types, but specific service types provide Untrusted services, no countermeasures

[0005] The existing technology does not use the interaction between security elements, and does not endow network entities with the ability to intelligently deal with complex environments to ensure the autonomy of security mechanisms

Images