Explainable AI encourages machine learning applications in the real world, whereas data-free model extraction attacks (DFME), in which an adversary steals a trained machine learning model by creating input queries with generative models instead of collecting training data, have attracted attention as a serious threat. In this paper, we propose MEGEX, a data-free model extraction attack against explainable AI that provides gradient-based explanations for inference results, and investigate whether the gradient-based explanations increase the vulnerability to the data-free model extraction attacks. In MEGEX, an adversary leverages explanations by Vanilla Gradient as derivative values for training a generative model. We prove that MEGEX is identical to white-box data-free knowledge distillation, whereby the adversary can train the generative model with the exact gradients. Our experiments show that the adversary in MEGEX can steal highly accurate models - 0.98×, 0.91×, and 0.96× the victim model accuracy on SVHN, Fashion-MNIST, and CIFAR-10 datasets given 1.5M, 5M, 20M queries, respectively. In addition, we also apply sophisticated gradient-based explanations, i.e., SmoothGrad and Integrated Gradients, to MEGEX. The experimental results indicate that these explanations are potential countermeasures to MEGEX. We also found that the accuracy of the model stolen by the adversary depends on the diversity of query inputs by the generative model.

Delay Tolerant Network (DTN) is developed to cope with intermittent connectivity and long delay in wireless networks. Due to limited connectivity, DTN is vulnerable to flooding attack in which malicious nodes flood the network with superfluous data to deplete the network resources. Existing works mitigate internal flooding attacks by rate-limit to constrain the number of messages that nodes can generate per time slot. However, rate-limit cannot flexibly accommodate burst traffic in which nodes may have sending demands higher than the rate-limit for a short period. In this paper, we propose FDER to detect flooding attack and yet allow legitimate burst traffic simultaneously. Nodes exchange their histories of encounter records (ER) which record the sent messages during their previous encounters. The ER history is used to infer a node's new message transmission rate over time and the number of forwarded replicas per message. The adversary nodes that send too many messages or replicas can thus be detected. Since ERs serve as useful tools for monitoring the sending behavior of nodes over a long time period, FDER could detect the burst traffic violation efficiently. We also design FP - a fairness forwarding policy to ensure fairness in the delivery performance fairness between nodes with normal traffic and those with burst traffic. FP uses ER information to observe nodes' rate of new message generation and adjust their forwarding priorities accordingly. Simulation results show that FDER can detect flooding attack at a higher accuracy and a lower delay compared to state-of-the-art scheme with affordable overhead. Moreover, FP could mitigate the smart flooding attack and still provide the performance fairness to support bursty traffic scenario.

UAVs have found their applications in numerous applications from recreational activities to business in addition to military and strategic fields. However, research on UAVs is not going on as quickly as the technology. Especially, when it comes to the security of these devices, the academia is lagging behind the industry. This gap motivates our work in this article as a stepping stone for future research in this area. A comprehensive survey on the security of UAVs and UAV-based systems can help the research community keep pace with, or even lead the industry. Although there are several reviews on UAVs or related areas, there is no recent survey broadly covering various aspects of security. Moreover, none of the existing surveys highlights current and future trends with a focus on the role of an omnipresent technology such as AI. This article endeavors to overcome these shortcomings. We conduct a comprehensive review on security challenges of UAVs as well as the related security controls. Then we develop a future roadmap for research in this area with a focus on the role of AI. The future roadmap is established based on the identified current trends, under-researched topics, and a future look-ahead.

In ad hoc networks that allow devices to dynamically configure networks via wireless communication, a secure routing protocol is a technology that guarantees the validity of routing with the use of cryptographic authentication. The secure DSR with ID-based sequential aggregate signatures (ISDSR) is a recently introduced secure routing protocol with “cryptographically compact chain,” where each device signs both the routing information and signatures generated by the previous device without increasing the size of signatures and the ID information propagated via packets can be utilized as a public key. However, ISDSR requires communication with a centralized key generation center (KGC), and thus a new device may experience difficulties in joining the protocol. Moreover, the implementation results of ISDSR have not been presented. In this work, we present ISDSR+, a new secure routing protocol without a centralized KGC that uses distributed key generation and the conventional features of ISDSR. ISDSR+ relies on a novel signature scheme where any node can receive a secret key as long as the number of available KGCs is more than a certain threshold. In other words, even if several KGCs are unavailable, the remaining available KGCs can still provide secret keys. We furthermore show promising results of ISDSR+ via a prototype implementation on Raspberry Pis. The results show the computational time of 0.1 s for both the secret key generation and the round trip time (RTT) of routing information under reasonable settings. The RTT of ISDSR+ is better than that of a naive secure routing protocol with RSA signatures.

In the use of a cloud storage, sharing of data with efficient access control is an important requirement in addition to data security and privacy. Cui et al. (IEEE Trans. on Comp. 2016) proposed key-aggregate searchable encryption (KASE), which allows a data owner to issue an aggregate key that enables a user to search in an authorized subset of encrypted files by generating an encrypted keyword called trapdoor. While the idea of KASE is elegant, to the best of our knowledge, its security has never been discussed formally. In this paper, we discuss the security of KASE formally and propose provably secure schemes. The construction of a secure KASE scheme is non-trivial, and we will show that the KASE scheme of Cui et al. is insecure under our definitions.We first introduce our provably secure scheme, named first construction, with respect to encrypted files and aggregate keys in a single-server setting. In comparison with the scheme of Cui et al., the first construction is secure without increased computational costs. Then, we introduce another provably secure scheme, named main construction, with respect to trapdoors in a twoserver setting. The main construction guarantees the privacy of a search, encrypted files, and aggregate keys. Considering 5,000 encrypted keywords, the first construction can finish search within three seconds and the main construction can finish search within six seconds.