作为一种能满足特定场景监控需求的网络覆盖类型,Sweep Coverage能够通过使用较少的节点来完成对Points of Interest(POI)的覆盖.为了改进以前的Sweep Coverage机制在覆盖间隔相差较大时的性能,本文提出了一种新的考虑了路径增量与覆盖间隔差异的插入启发式(Coverage Different Insert Heuristic)CDIH算法.实验结果表明,在相同的网络监控场景中,新的启发式算法较以往的覆盖机制均能有较好的表现.
VoIP(voice over IP)是基于UDP/IP协议族的语音通信技术,当信道环境变差时不可避免地会产生网络分组丢失,这给建立在其上的VoIP隐写的可靠传输带来了挑战。提出利用纠删码对秘密信息进行冗余预处理,再结合矩阵嵌入编码实现最小失真的隐写,从而建立了基于联合编码的嵌入和提取模型。在此基础上,分析了关键参数对联合编码性能的影响并给出了最优参数的选取算法。实验结果表明,所提联合编码能够有效提高隐写系统的抗分组丢失能力,且能减少对语音流的修改。
In this paper, we present the first ciphertext-policy attribute-based encryption (CP-ABE) scheme for polynomial-size general circuits based on bilinear maps which is more suitable for practical use and more efficient than multilinear maps. Our scheme uses a top-down secret sharing and FANOUT gate to resist the "backtracking attack" which is the main barrier expending access tree to general circuit. In the standard model, selective security of our scheme is proved. Comparing with current scheme for general circuits from bilinear maps, our work is more efficient.
AEZ is an AES-based authenticated encryption submitted to the ongoing CAESAR competition and was presented at Eurocrypt2015 with AEZ v3. There are three models for AEZ, AEZ-core, AEZ-tiny and AEZ-prf. In this paper, we consider the security of AEZprf for AEZ v4.2, the latest version of AEZ.Our major finding is a collision of any 256-bit associated data for AES-prf. Then we launch collision attacks in a quantum setting and a classical setting respectively under different assumptions. In the quantum setting, by Simon's quantum algorithm, we amount a forgery with O(n) quantum superposition queries and an overwhelming probability close to 1.In the classical setting, one with the key of AEZ-prf can also construct the forgeries. Our results show that the AEZ-prf models of AEZ v4.2 is not secure in both the quantum setting and classical world. Furthermore, our results can also be applied to AEZ v3, which has been published on Eurocrypt 2015. As far as we know, no cryptanalysis of AEZ v4.2 has been published so far.