...
首页> 外文期刊>International Journal of Applied Engineering Research >Spoofing Attack Detection and Localizing Multiple Adversaries in Deep Neural Network
【24h】

Spoofing Attack Detection and Localizing Multiple Adversaries in Deep Neural Network

机译:欺骗深度神经网络中的攻击检测和定位多个对手

获取原文
获取原文并翻译 | 示例
           

摘要

In deep neural network transfer the data from source to destination through mobile nodes. It is an infrastructure less network so it communicates the mobile nodes without any access points. In the deep neural network the mobile nodes communication has no centralized controller and it has dynamic network topology so the attackers hack the communication process of mobile node. And therefore energy consumption is increased, high end to end delay and lead to loss of QOS. The neural network dynamically changed, therefore scheduling will be complicated, routing overhead and throughput will be reduced. The topology changed repeatedly scheduling the packets of mobile node will be complicated. In the proposed system Adaptive Scheduling (AS) Algorithm is used to schedule the packets using safe Q in routing protocol. By using the NS-2 simulator, there are low end to end delay then it afford QOS provisioning in network and easy to bandwidth estimate and energy consumed. The deep neural network using this proposed method reorganization and localizing the spoof attack from multiple hostile of mobile node communication process.
机译:在深神经网络中,通过移动节点将数据从源转移到目的地。它是一个基础架构较少的网络,因此它传达了没有任何接入点的移动节点。在深度神经网络中,移动节点通信没有集中控制器,它具有动态网络拓扑,因此攻击者破解了移动节点的通信过程。因此,能量消耗增加,最终延迟的高端并导致QoS丧失。神经网络动态地改变,因此调度将是复杂的,路由开销和吞吐量将减少。重复调度移动节点数据包的拓扑将变得复杂。在所提出的系统自适应调度(AS)算法中用于在路由协议中使用安全Q调度分组。通过使用NS-2模拟器,最终延迟结束延迟,然后在网络中提供QoS供应,易于带宽估计和消耗的能量。深度神经网络使用这种提出的方​​法重组并从移动节点通信过程多敌对的欺骗攻击定位。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号