Bankrupting Sybil Despite Churn
Bankrupting Sybil Despite Churn
复制标题
DOI：
10.1109/icdcs51616.2021.00048
复制
发表时间：
2020-10
期刊：
影响因子：
--
通讯作者：
Diksha Gupta;Jared Saia;Maxwell Young
中科院分区：
文献类型：
--
作者：
Diksha Gupta;Jared Saia;Maxwell Young
当攻击者伪装成多个身份（ID）时，就会发生Sybil攻击。将Sybil（坏）ID的数量限制在少数对于使用成熟的工具来容忍恶意行为至关重要，例如拜占庭协议和安全多方计算。一种流行的强制西比尔少数的技术是资源燃烧：可验证的网络资源消耗，如计算能力，带宽或内存。不幸的是，基于资源燃烧的典型防御需要非Sybil（好）ID消耗至少与对手一样多的资源。此外，即使系统成员相对稳定，它们也具有高成本。在这里，我们提出了一个新的Sybil防御，ERGO，保证（1）总是有少数的Sybil ID;（2）当系统受到重大攻击时，好的ID消耗渐近小于坏的。特别地，对于可以呈指数变化的流失率，ERGO的资源燃烧率为，其中是对手的资源燃烧率，是好ID的加入率。我们根据经验评估ERGO与之前的Sybil防御。与其他Sybil防御不同，ERGO可以与机器学习技术相结合，以保持其理论保证的方式识别Sybil ID。基于我们将ERGO与两种最先进的Sybil防御方法进行比较的实验，我们表明，在没有机器学习的情况下，ERGO的性能提高了2个数量级，而使用机器学习的情况下，ERGO的性能提高了3个数量级。
A Sybil attack occurs when an adversary pretends to be multiple identities (IDs). Limiting the number of Sybil (bad) IDs to a minority is critical to the use of well-established tools for tolerating malicious behavior, such as Byzantine agreement and secure multiparty computation. A popular technique for enforcing a Sybil minority is resource burning: verifiable consumption of a network resource, such as computational power, bandwidth, or memory. Unfortunately, typical defenses based on resource burning require non-Sybil (good) IDs to consume at least as many resources as the adversary. Additionally, they have a high cost, even when the system membership is relatively stable. Here, we present a new Sybil defense, ERGO, that guarantees (1) there is always a minority of Sybil IDs; and (2) when the system is under significant attack, the good IDs consume asymptotically less than the bad. In particular, for churn rate that can vary exponentially, the resource burning rate of ERGO is, where is the resource burning rate of the adversary, and is the join rate of good IDs. We empirically evaluate ERGO alongside prior Sybil defenses. Unlike other Sybil defense, ERGO can be combined with machine learning techniques for identifying Sybil IDs, in a way that maintains its theoretical guarantees. Based on our experiments comparing ERGO with two state-of-the-art Sybil defenses, we show that ERGO improves by up to 2 orders of magnitude without machine learning, and up to 3 orders of magnitude using machine learning.