The security of deep neural networks has triggered extensive research on the adversarial example. The gradient or optimization-based adversarial example generation algorithm has poor practicality and cannot combine high success rate and high efficiency; using GAN to generate adversarial examples has the problems of gradient disappearance and unstable generation. In this paper, we propose a novel adversarial example generation algorithm based on WGAN-Unet, which uses the structure of WGAN and Unet to form a generative adversarial network to improve the stability of network training, and uses the cosine loss function to measure the category loss and improve the success rate of adversarial attacks. The adversarial example generation using WGAN-Unet is compared with other algorithms in terms of human eye perception, time consumption, attack success rate, and image quality, proving our scheme's superiority.
In recent years, with the rapid development of deep neural networks in the field of artificial intelligence, some of the security issues involved have gradually attracted attention in the industry, one of which is adversarial sample attacks. The attacker inputs carefully designed adversarial samples to the deep learning model, causing the attacked model to output misclassification results with high confidence, which seriously threatens the robustness of the deep learning model. Based on the commonly used deep learning network model, combined with the interpretability research of neural networks, we propose an effective region generation algorithm (ERGA) for adversarial sample generation, which can overcome the defects of the current commonly used algorithms. In our approach, the effective region selection step is added in the adversarial sample generation process, which overcomes the limitation of common adversarial sample generation algorithms that are limited to global pixel perturbation. We try to limit the number of pixels to be changed while maintaining a higher attack success. The algorithm also optimizes the process of counter disturbance generation, solves the uncertainty of the gradient update direction and amplitude in the iterative process. In addition, it also introduces the interpretability research of counter samples, which can be used to a certain extent in the deep learning network. At the same time, ERGA can improve the ability of supervision and self-examination of the classification results.
Virtualization technology becomes a hot IT technology with the popu-larity of CloudComputing.However, new security issues arise with it.Specifically, the resources sharing and data communication in virtual machines are most concerned.In this paper an access control model is proposed which combines the Chinese Wall and BLP model.BLP multi-level security model is introduced with corresponding improvement based on PCW (Prioritized Chinese Wall) security model.This model can be used to safely control the resources and event behaviors in virtual machines.Experimental results show its effectiveness and safety.
Cloud Computing has become a well-known primitive nowadays; many researchers and companies are embracing this fascinating technology with feverish haste. In the meantime, security and privacy challenges are brought forward while the number of cloud storage user increases expeditiously. In this work, we conduct an in-depth survey on recent research activities of cloud storage security in association with cloud computing. After an overview of the cloud storage system and its security problem, we focus on the key security requirement triad, i.e., data integrity, data confidentiality, and availability. For each of the three security objectives, we discuss the new unique challenges faced by the cloud storage services, summarize key issues discussed in the current literature, examine, and compare the existing and emerging approaches proposed to meet those new challenges, and point out possible extensions and futuristic research opportunities. The goal of our paper is to provide a state-of-the-art knowledge to new researchers who would like to join this exciting new field.
Plagiognathops microlepis,which has a wide range of food resouse,humus,for example,organic debris and algal,is an important economical species and well-known as scavenger.By using vertical polyacrylamide gel electrophoresis and specific dye measures,LDH isozymes in 6 tissues(brain,kidney,liver,heart,muscle and eyes)was investigated.The results showed that the electrophoretograms of LDH isoenzyme patterns exhibited remarkable tissue-specificities: 7 LDH isoenzyme bands,encoded by Ldh-a,Ldh-b and Ldh-c three loci,were detected from all the 6 tissues,3 bandes were found in brain,kidney,heart and eyes,while in liver and muscle the bands were 7 and 1,respectively,the only one band found in muscle was predominant in expressing,4 specific bands encoded by Ldh-c were detected on the cathode side of PAGE hectograph withing liver.The sequence of isoenzyme activity in brain,kidney,liver,heart and eyes are LDH-1LDH-2LDH-3,LDH-3LDH-1LDH-2,LDH-3LDH-5LDH-2LDH-6LDH-1LDH-4LDH-7,LDH-1LDH-2LDH-3 and LDH-3LDH-1LDH-2,respectively,while the isoenzyme activity of the only one band(LDH-3)detected in muscle was 100%;The rates of flow(Rf)for isoenzymes LDH-A4 was inferior than LDH-B4,indicating that the electrophoretic style of LDH isozymes in Plagiognathops microlepis is BA;Consistent results were acquired by scanning and analyzing the sub unit in LDH isoenzymes,namely,Ldh-a was preponderantly in expressing in kidney,liver,muscle and eyes,While Ldh-b was preponderant in heart and brain.
In this paper, a secure and efficient access authentication scheme is designed between edge devices in the smart grid, to achieve secure communication between devices in the smart grid. Considering the characteristics of power grid equipment, identity based cryptography is introduced to embed the equipment identity, timestamp, etc., into the process of identity recognition and private key update. Combined with the characteristics of private key update in the cryptographic system, bilinear pairing is used to negotiate the communication key in each time interval, and only hash, XOR operation and concatenation operation are used to achieve lightweight authentication scheme. The scheme can not only meet the security requirements, but also minimize the burden on edge device resources. Experimental results show that the communication cost and computation cost of our scheme are less than the existing schemes.
Adversarial sample attacks seriously threaten the security and robustness of deep learning models. There are three problems in state-of-the-art adversarial sample generation schemes: the gradient update step size needs to be manually selected, inaccurate gradient update direction and uncontrollable times of iterations. In order to solve these problems, Root Mean Square Prop optimization algorithm (RMSProp) is proposed, which is integrated with IFGSM and IFGM. This algorithm can be easily extended to other attacks, and to a certain extent it can alleviate the trade-off between white box attacks and deliverability. The algorithm proposed in this paper can generate non-targeted adversarial samples more efficiently and quickly. Experiments show that it can generate effective and robust adversarial samples against current mainstream convolutional neural network (CNN).
In this work, we conduct an in-depth survey on recent multimedia storage security research activities in association with cloud computing. After an overview of the cloud storage system and its security problem, we focus on four hot research topics. They are data integrity, data confidentiality, access control, and data manipulation in the encrypted domain. We describe several key ideas and solutions proposed in the current literature and point out possible extensions and futuristic research opportunities. Our research objective is to offer a state-of-the-art knowledge to new researchers who would like to enter this exciting new field.
In recent years, smart phone technology is becoming increasingly popular. The dangers of mobile phone malwares are becoming more and more serious. In this paper we present a new mobile smartphone malware detection scheme based on Hidden Markov Model (HMM) which is different from the traditional signature scanning methods. Firstly, we monitor the key press and system function call sequence, and take the key press as hidden state. After decoding HMM model, abnormal process can be detected using the matching rate of HMM output to the actual key press sequence. The experimental results demonstrate that the proposed method can effectively detect mobile malwares.
Embedded SIM enables SIM provisioning to be performed after end users taken their equipments, and the security of the remote SIM provisioning protocol is of great significance to the security of whole mobile networks. This paper conducts formal analysis and verification for the session key agreement protocol, which has the highest security requirements. After modeling with Petri net, we deduce that the Petri net is 1-bounded. Afterwards we describe the protocol in Promela language. We use SPIN protocol verification tool to verify the security of the protocol. The results show that the protocol has passed all tests and is currently safe.
'/%3e%3cuse xlink:href='%23a' transform='rotate(23.036 .093 25.536)'/%3e%3cuse xlink:href='%23a' transform='rotate(45.87 1.273 16.18)'/%3e%3cuse xlink:href='%23a' transform='rotate(69.945 .996 12.078)'/%3e%3cuse xlink:href='%23a' transform='rotate(20.66 -19.689 31.932)'/%3e%3c/svg%3e)