Because mobile devices only have the limited resource in socially aware networking, some network nodes are unwilling to sacrifice their resource to forward messages to others for free, thus forming selfish nodes. However selfish nodes will cause dropping in the message delivery rate and affect the network performance, so this paper has proposed a diversity-based selfish node detection algorithm (DSNDA). Firstly, it mainly applies the node forwarding willingness mechanism to detect whether the node is selfish. If the node has the inadequate resource, that is, it cannot forward any messages for others; then it will be selfish. If the node’s resource is enough and the node has forwarded messages to other nodes, it indicates that the node is not selfish. Otherwise, DSNDA algorithm will use reputation mechanism to check the node further. If the node has broken faith, it indicates that the node is selfish. On the contrary the message confirmation mechanism will be employed to make the final check on the node. If a message does not be received or discarded after the node’s receiving it, which proves that it has selfish behaviour. If not, the node is normal. It can be proved by the simulation experiment and results that the DSNDA algorithm has significantly improved the delivery rate of messages, comparing with other algorithms. At the same time, reducing the network delay, thus verifying the feasibility of DSNDA.

由于移动设备在具有社交意识的网络中仅具有有限的资源，因此某些网络节点不愿意牺牲其资源来免费将消息转发给其他人，从而形成自私的节点。但是自私节点将导致消息传递速率下降并影响网络性能，因此本文提出了一种基于分集的自私节点检测算法（DSNDA）。首先，它主要应用节点转发意愿机制来检测节点是否自私。如果节点的资源不足，即它无法转发任何消息给其他节点；那将是自私的。如果节点的资源足够，并且该节点已将消息转发到其他节点，则表明该节点不是自私的。否则，DSNDA算法将使用信誉机制进一步检查该节点。如果该节点违反了信仰，则表明该节点是自私的。相反，将使用消息确认机制对节点进行最终检查。如果节点接收到消息后未收到或丢弃该消息，则表明该消息具有自私行为。如果不是，则表明节点正常。仿真实验和结果表明，与其他算法相比，DSNDA算法显着提高了邮件的传递速度。同时，减少了网络延迟，从而验证了DSNDA的可行性。证明它具有自私的行为。如果不是，则表明节点正常。仿真实验和结果表明，与其他算法相比，DSNDA算法显着提高了邮件的传递速度。同时，减少了网络延迟，从而验证了DSNDA的可行性。证明它具有自私的行为。如果不是，则表明节点正常。仿真实验和结果表明，与其他算法相比，DSNDA算法显着提高了邮件的传递速度。同时，减少了网络延迟，从而验证了DSNDA的可行性。