高级检索
    任 伟, 刘腾红, 金 海. 移动Ad Hoc网络中针对拥塞的RoQ DDoS攻击及其防御[J]. 计算机研究与发展, 2006, 43(11): 1927-1932.
    引用本文: 任 伟, 刘腾红, 金 海. 移动Ad Hoc网络中针对拥塞的RoQ DDoS攻击及其防御[J]. 计算机研究与发展, 2006, 43(11): 1927-1932.
    Ren Wei, Liu Tenghong, Jin Hai. Congestion-Based RoQ DDoS Attacking and Defense Scheme in Mobile Ad Hoc Networks[J]. Journal of Computer Research and Development, 2006, 43(11): 1927-1932.
    Citation: Ren Wei, Liu Tenghong, Jin Hai. Congestion-Based RoQ DDoS Attacking and Defense Scheme in Mobile Ad Hoc Networks[J]. Journal of Computer Research and Development, 2006, 43(11): 1927-1932.

    移动Ad Hoc网络中针对拥塞的RoQ DDoS攻击及其防御

    Congestion-Based RoQ DDoS Attacking and Defense Scheme in Mobile Ad Hoc Networks

    • 摘要: 根据网络容量理论,移动Ad Hoc网络中存在针对拥塞的RoQ分布式拒绝服务攻击,其攻击模式包括脉冲攻击、循环攻击、自消耗攻击和泛洪攻击.防御机制包括检测和响应,检测信号包括RTS/CTS包频率、信号干涉频率以及包重传次数,响应机制依靠ECN标记和通知.NS2模拟结果显示,复杂拓扑结构将更容易受到攻击,攻击节点的分散将加大攻击效果.脉冲攻击产生明显的吞吐率和延迟抖动,当同速率攻击流增加到5个时,受害流吞吐率下降到77.42%,延迟增加110倍.

       

      Abstract: Congestion-targeted RoQ (reduction of quality of service) DDoS (distributed denial of service) attacking is discussed in details for the first time. The principle of attacking is pointed out on the basis of the analysis of network capacity. The four categories of the attacking patterns such as pulsing attacking, round robin attacking, self-whisper attacking and flooding attacking, are also described. The defense schemes are proposed, which include the detection of three signals, such as RTS/CTS packets, signal interference frequency and retransmission times, and response scheme with ECN (explicit congestion notification) marking method. The extensive NS2 simulation results show that the pulsing attacking mode leads to the great jitter of the goodput and delay. The increasing of delay and decreasing of goodput becomes obvious with the addition of attacking flows. The delay performance goes up to 110 times and goodput performance drops down to 77.42% when five attacking flows with the same rate occur. The complicated topology is more vulnerable and the distribution of attacking nodes will generate more obvious impacts. The dropping packets are also growing corresponding with the addition of attacking flows, because of IFQ (interface queue) overflow and routing overhead.

       

    /

    返回文章
    返回