Biblio

Click-through rate prediction is an essential task in industrial applications, such as online advertising. Recently deep learning based models have been proposed, which follow a similar Embedding&MLP paradigm. In these methods large scale sparse input features are first mapped into low dimensional embedding vectors, and then transformed into fixed-length vectors in a group-wise manner, finally concatenated together to fed into a multilayer perceptron (MLP) to learn the nonlinear relations among features. In this way, user features are compressed into a fixed-length representation vector, in regardless of what candidate ads are. The use of fixed-length vector will be a bottleneck, which brings difficulty for Embedding&MLP methods to capture user's diverse interests effectively from rich historical behaviors. In this paper, we propose a novel model: Deep Interest Network (DIN) which tackles this challenge by designing a local activation unit to adaptively learn the representation of user interests from historical behaviors with respect to a certain ad. This representation vector varies over different ads, improving the expressive ability of model greatly. Besides, we develop two techniques: mini-batch aware regularization and data adaptive activation function which can help training industrial deep networks with hundreds of millions of parameters. Experiments on two public datasets as well as an Alibaba real production dataset with over 2 billion samples demonstrate the effectiveness of proposed approaches, which achieve superior performance compared with state-of-the-art methods. DIN now has been successfully deployed in the online display advertising system in Alibaba, serving the main traffic.

The power outages of the last couple of years around the world introduce the indispensability of technological development to improve the traditional power grids. Early warnings of imminent failures represent one of the major required improvements. Costly blackouts throughout the world caused by the different severe incidents in traditional power grids have motivated researchers to diagnose and investigate previous blackouts and propose a prediction model that enables to prevent power outages. Although, in the new generation of power grid, the smart grid's (SG) real time data can be used from smart meters (SMs) and phasor measurement unit sensors (PMU) to prevent blackout, it demands high reliability and stability against power outages. This paper implements a proactive prediction model based on deep-belief networks that can predict imminent blackout. The proposed model is evaluated on a real smart grid dataset. Promising results are reported in the case study.

6LoWPAN technology realizes the IPv6 packet transmission in the IEEE 802.15.4 based WSN. And 6LoWPAN is regarded as one of the ideal technologies to realize the interconnection between WSN and Internet, which is the key to build the IoT. Contiki is an open source and highly portable multitasking operating system, in which the 6LoWPAN has been implemented. In contiki, only several K Bytes of code and a few hundred bytes of memory are required to provide a multitasking environment and built-in TCP/IP support. This makes it especially suitable for memory constrained embedded platforms. In this paper, a lightweight 6LoWPAN gateway based on Contiki is designed and its designs of hardware and software are described. A complex experiment environment is presented, in which the gateway's capability of accessing the Internet is verified, and its performance about the average network delay and jitter are analyzed. The experimental results show that the gateway designed in this paper can not only realize the interconnection between 6LoWPAN networks and Internet, but also have good network adaptability and stability.

A new kind of Square Lattice Photonic Crystal Fiber (SLPCF) is proposed, the first ring is formed by elliptical holes filled with ethanol. To regulate the dispersion and the confinement loss we put a circular air-holes with small diameters into the third ring of the cladding area. The diameter of the core is arranged as d2=2*A-d, where A is the pitch and d diameter of the air-holes. After simulations, we got a dispersion low as 0.0494 (ps/Km. nm) and a confinement loss also low as 2.6×10-7(dB/m) at a wavelength of 1.55 $μ$m. At 0.8 $μ$m we obtained a nonlinearity high as 60.95 (1/km. w) and a strong guiding light. Also, we compare the filled ethanol elliptical holes with the air filled elliptical holes of our proposed square lattice photonic crystal fiber. We use as a simulation method in this manuscript the two-dimensional FDTD method. The utilization of the proposed fiber is in the telecommunication transmission because of its low dispersion and low loss at the c-band and in the nonlinear applications.

Over the past decade, the reliance on Unmanned Aerial Systems (UAS) to carry out critical missions has grown drastically. With an increased reliance on UAS as mission assets and the dependency of UAS on cyber resources, cyber security of UAS must be improved by adopting sound security principles and relevant technologies from the computing community. On the other hand, the traditional avionics community, being aware of the importance of cyber security, is looking at new architecture and designs that can accommodate both the traditional safety oriented principles as well as the cyber security principles and techniques. It is with the effective and timely convergence of these domains that a holistic approach and co-design can meet the unique requirements of modern systems and operations. In this paper, authors from both the cyber security and avionics domains describe our joint effort and insights obtained during the course of designing secure and resilient embedded avionics systems.

Shielding systems such as AMD's Secure Encrypted Virtualization aim to protect a virtual machine from a higher privileged entity such as the hypervisor. A cornerstone of these systems is the ability to protect the memory from unauthorized accesses. Despite this protection mechanism, previous attacks leveraged the control over memory resources to infer control flow of applications running in a shielded system. While previous works focused on a specific target application, there has been no general analysis on how the control flow of a protected application can be inferred. This paper tries to overcome this gap by providing a detailed analysis on the detectability of control flow using memory access patterns. To that end, we do not focus on a specific shielding system or a specific target application, but present a framework which can be applied to different types of shielding systems as well as to different types of attackers. By training a random forest classifier on the memory accesses emitted by syscalls of a shielded entity, we show that it is possible to infer the control flow of shielded entities with a high degree of accuracy.

With the rapid development of network and communication technologies, everything is able to be connected to the Internet. IoT devices, which include home routers, IP cameras, wireless printers and so on, are crucial parts facilitating to build pervasive and ubiquitous networks. As the number of IoT devices around the world increases, the security issues become more and more serious. To handle with the security issues and protect the IoT devices from being compromised, the firmware of devices needs to be strengthened by discovering and repairing vulnerabilities. Current vulnerability detection tools can only help strengthening traditional software, nevertheless these tools are not practical enough for IoT device firmware, because of the peculiarity in firmware's structure and embedded device's architecture. Therefore, new vulnerability detection framework is required for analyzing IoT device firmware. This paper reviews related works on vulnerability detection in IoT firmware, proposes and implements a framework to automatically detect authentication-bypass flaws in a large scale of Linux-based firmware. The proposed framework is evaluated with a data set of 2351 firmware images from several target vendors, which is proved to be capable of performing large-scale and automated analysis on firmware, and 1 known and 10 unknown authentication-bypass flaws are found by the analysis.

Cross-site scripting (XSS) is a scripting attack targeting web applications by injecting malicious scripts into web pages. Blind XSS is a subset of stored XSS, where an attacker blindly deploys malicious payloads in web pages that are stored in a persistent manner on target servers. Most of the XSS detection techniques used to detect the XSS vulnerabilities are inadequate to detect blind XSS attacks. In this research, we present machine learning based approach to detect blind XSS attacks. Testing results help to identify malicious payloads that are likely to get stored in databases through web applications.

When implemented on real systems, cryptographic algorithms are vulnerable to attacks observing their execution behavior, such as cache-timing attacks. Designing protected implementations must be done with knowledge and validation tools as early as possible in the development cycle. In this article we propose a methodology to assess the robustness of the candidates for the NIST post-quantum standardization project to cache-timing attacks. To this end we have developed a dedicated vulnerability research tool. It performs a static analysis with tainting propagation of sensitive variables across the source code and detects leakage patterns. We use it to assess the security of the NIST post-quantum cryptography project submissions. Our results show that more than 80% of the analyzed implementations have at least one potential flaw, and three submissions total more than 1000 reported flaws each. Finally, this comprehensive study of the competitors security allows us to identify the most frequent weaknesses amongst candidates and how they might be fixed.

Now a days, Cloud computing has brought a unbelievable change in companies, organizations, firm and institutions etc. IT industries is advantage with low investment in infrastructure and maintenance with the growth of cloud computing. The Virtualization technique is examine as the big thing in cloud computing. Even though, cloud computing has more benefits; the disadvantage of the cloud computing environment is ensuring security. Security means, the Cloud Service Provider to ensure the basic integrity, availability, privacy, confidentiality, authentication and authorization in data storage, virtual machine security etc. In this paper, we presented a Local outlier factors mechanism, which may be helpful for the detection of Distributed Denial of Service attack in a cloud computing environment. As DDoS attack becomes strong with the passing of time, and then the attack may be reduced, if it is detected at first. So we fully focused on detecting DDoS attack to secure the cloud environment. In addition, our scheme is able to identify their possible sources, giving important clues for cloud computing administrators to spot the outliers. By using WEKA (Waikato Environment for Knowledge Analysis) we have analyzed our scheme with other clustering algorithm on the basis of higher detection rates and lower false alarm rate. DR-LOF would serve as a better DDoS detection tool, which helps to improve security framework in cloud computing.

Microsoft's PowerShell is a command-line shell and scripting language that is installed by default on Windows machines. Based on Microsoft's .NET framework, it includes an interface that allows programmers to access operating system services. While PowerShell can be configured by administrators for restricting access and reducing vulnerabilities, these restrictions can be bypassed. Moreover, PowerShell commands can be easily generated dynamically, executed from memory, encoded and obfuscated, thus making the logging and forensic analysis of code executed by PowerShell challenging. For all these reasons, PowerShell is increasingly used by cybercriminals as part of their attacks' tool chain, mainly for downloading malicious contents and for lateral movement. Indeed, a recent comprehensive technical report by Symantec dedicated to PowerShell's abuse by cybercrimials [52] reported on a sharp increase in the number of malicious PowerShell samples they received and in the number of penetration tools and frameworks that use PowerShell. This highlights the urgent need of developing effective methods for detecting malicious PowerShell commands. In this work, we address this challenge by implementing several novel detectors of malicious PowerShell commands and evaluating their performance. We implemented both "traditional" natural language processing (NLP) based detectors and detectors based on character-level convolutional neural networks (CNNs). Detectors' performance was evaluated using a large real-world dataset. Our evaluation results show that, although our detectors (and especially the traditional NLP-based ones) individually yield high performance, an ensemble detector that combines an NLP-based classifier with a CNN-based classifier provides the best performance, since the latter classifier is able to detect malicious commands that succeed in evading the former. Our analysis of these evasive commands reveals that some obfuscation patterns automatically detected by the CNN classifier are intrinsically difficult to detect using the NLP techniques we applied. Our detectors provide high recall values while maintaining a very low false positive rate, making us cautiously optimistic that they can be of practical value.

The article considers the approach to identifying potentially unsafe data in program code of embedded systems which can lead to errors and fails in the functioning of equipment. The sources of invalid data are revealed and the process of changing the status of this data in process of static code analysis is shown. The mechanism for annotating functions that operate on unsafe data is described, which allows to control the entire process of using them and thus it will improve the quality of the output code.

Cybersecurity in control systems has been actively discussed in recent years. In particular, networked control systems (NCSs) over the Internet are exposed to various types of cyberattacks such as false data injection attacks. This paper proposes a detection and mitigation method of the false data injection attacks in interactive NCSs, i.e., bilateral teleoperation systems. A bilateral teleoperation system exchanges position and force information through the Internet between the master and slave robots. The proposed method utilizes two redundant communication channels for both the master-to-slave and slave-to-master paths. The attacks are detected by a tamper detection observer (TDO) on each of the master and slave sides. The TDO compares the position responses of actual robots and robot models. A path selector on each side chooses the appropriate position and force responses from the responses received through the two communication channels, based on the outputs of the TDO. The proposed method is validated by simulations with attack models.

Mobile Ad Hoc Networks are dynamic in nature and have no rigid or reliable network infrastructure by their very definition. They are expected to be self-governed and have dynamic wireless links which are not entirely reliable in terms of connectivity and security. Several factors could cause their degradation, such as attacks by malicious and selfish nodes which result in data carrying packets being dropped which in turn could cause breaks in communication between nodes in the network. This paper aims to address the issue of remedy and mitigation of the damage caused by packet drops. We proposed an improvement on the EAACK protocol to reduce the network overhead packet delivery ratio by using hybrid cryptography techniques DES due to its higher efficiency in block encryption, and RSA due to its management in key cipher. Comparing to the existing approaches, our simulated results show that hybrid cryptography techniques provide higher malicious behavior detection rates, and improve the performance. This research can also lead to more future efforts in using hybrid encryption based authentication techniques for attack detection/prevention in MANETs.

Internet of things (IoT) is the smart network which connects smart objects over the Internet. The Internet is untrusted and unreliable network and thus IoT network is vulnerable to different kind of attacks. Conventional encryption and authentication techniques sometimes fail on IoT based network and intrusion may succeed to destroy the network. So, it is necessary to design intrusion detection system for such network. In our paper, we detect routing attacks such as sinkhole and selective forwarding. We have also tried to prevent our network from these attacks. We designed detection and prevention algorithm, i.e., KMA (Key Match Algorithm) and CBA (Cluster- Based Algorithm) in MatLab simulation environment. We gave two intrusion detection mechanisms and compared their results as well. True positive intrusion detection rate for our work is between 50% to 80% with KMA and 76% to 96% with CBA algorithm.

This research proposes a system for detecting known and unknown Distributed Denial of Service (DDoS) Attacks. The proposed system applies two different intrusion detection approaches anomaly-based distributed artificial neural networks(ANNs) and signature-based approach. The Amazon public cloud was used for running Spark as the fast cluster engine with varying cores of machines. The experiment results achieved the highest detection accuracy and detection rate comparing to signature based or neural networks-based approach.

Wireless Sensor Network (WSN) is often to consist of adhoc devices that have low power, limited memory and computational power. WSN is deployed in hostile environment, due to which attacker can inject false data easily. Due to distributed nature of WSN, adversary can easily inject the bogus data into the network because sensor nodes don't ensure data integrity and not have strong authentication mechanism. This paper reviews and analyze the performance of some of the existing false data filtering schemes and propose new scheme to identify the false data injected by adversary or compromised node. Proposed schemes shown better and efficiently filtrate the false data in comparison with existing schemes.

The evolution of the microelectronics manufacturing industry is characterized by increased complexity, analysis, integration, distribution, data sharing and collaboration, all of which is enabled by the big data explosion. This evolution affords a number of opportunities in improved productivity and quality, and reduced cost, however it also brings with it a number of risks associated with maintaining security of data systems. The International Roadmap for Devices and System Factory Integration International Focus Team (IRDS FI IFT) determined that a security technology roadmap for the industry is needed to better understand the needs, challenges and potential solutions for security in the microelectronics industry and its supply chain. As a first step in providing this roadmap, the IFT conducted a security survey, soliciting input from users, suppliers and OEMs. Preliminary results indicate that data partitioning with IP protection is the number one topic of concern, with the need for industry-wide standards as the second most important topic. Further, the "fear" of security breach is considered to be a significant hindrance to Advanced Process Control efforts as well as use of cloud-based solutions. The IRDS FI IFT will endeavor to provide components of a security roadmap for the industry in the 2018 FI chapter, leveraging the output of the survey effort combined with follow-up discussions with users and consultations with experts.

Advent of Cyber has converted the entire World into a Global village. But, due to vurneabilites in SCADA architecture [1] national assests are more prone to cyber attacks.. Cyber invasions have a catastrophic effect in the minds of the civilian population, in terms of states security system. A robust cyber security is need of the hour to protect the critical information infastructrue & critical infrastructure of a country. Here, in this paper we scrutinize cyber terrorism, vurneabilites in SCADA network systems [1], [2] and concept of cyber resilience to combat cyber attacks.

Wireless Internet of Things (IoT) devices share several features such as limited energy supply, low computing power, limited memory size, and vulnerable radio communication network. IETF proposed the Constrained Application Protocol (CoAP) for this type of network. This paper presents implementation of CoAP into an embedded IoT device used for smart Energy Storage System (ESS) under microgrid environment. Confirmable message type was adopted to provide reliable communication. Since the frame size of IEEE 802.15.4 physical layer was limited to 127 bytes, the header of 6LoWPAN and UDP was compressed to reduce fragmentation and reassembly overhead. Performance of the communication service was tested by measuring round trip time between two end nodes of developed system.

Additive manufacturing (AM, or 3D printing) is a novel manufacturing technology that has been adopted in industrial and consumer settings. However, the reliance of this technology on computerization has raised various security concerns. In this paper, we address issues associated with sabotage via tampering during the 3D printing process by presenting an approach that can verify the integrity of a 3D printed object. Our approach operates on acoustic side-channel emanations generated by the 3D printer’s stepper motors, which results in a non-intrusive and real-time validation process that is difficult to compromise. The proposed approach constitutes two algorithms. The first algorithm is used to generate a master audio fingerprint for the verifiable unaltered printing process. The second algorithm is applied when the same 3D object is printed again, and this algorithm validates the monitored 3D printing process by assessing the similarity of its audio signature with the master audio fingerprint. To evaluate the quality of the proposed thresholds, we identify the detectability thresholds for the following minimal tampering primitives: insertion, deletion, replacement, and modification of a single tool path command. By detecting the deviation at the time of occurrence, we can stop the printing process for compromised objects, thus saving time and preventing material waste. We discuss various factors that impact the method, such as background noise, audio device changes and different audio recorder positions.

Distributed denial of service (DDoS) attacks is a serious cyberattack that exhausts target machine's processing capacity by sending a huge number of packets from hijacked machines. To minimize resource consumption caused by DDoS attacks, filtering attack packets at source machines is the best approach. Although many studies have explored the detection of DDoS attacks, few studies have proposed DDoS attack prevention schemes that work at source machines. We propose a reliable, lightweight, transparent, and flexible DDoS attack prevention scheme that works at source machines. In this scheme, we employ a hypervisor with a packet filtering mechanism on each managed machine to allow the administrator to easily and reliably suppress packet transmissions. To make the proposed scheme lightweight and transparent, we exploit a thin hypervisor that allows pass-through access to hardware (except for network devices) from the operating system, thereby reducing virtualization overhead and avoiding compromising user experience. To make the proposed scheme flexible, we exploit a configurable packet filtering mechanism with a guaranteed safe code execution mechanism that allows the administrator to provide a filtering policy as executable code. In this study, we implemented the proposed scheme using BitVisor and the Berkeley Packet Filter. Experimental results show that the proposed scheme can suppress arbitrary packet transmissions with negligible latency and throughput overhead compared to a bare metal system without filtering mechanisms.

Mobile ad hoc networks (MANETs) are self-configuring, dynamic networks in which nodes are free to move. These nodes are susceptible to various malicious attacks. In this paper, we propose a distributed trust-based security scheme to prevent multiple attacks such as Probe, Denial-of-Service (DoS), Vampire, User-to-Root (U2R) occurring simultaneously. We report above 95% accuracy in data transmission and reception by applying the proposed scheme. The simulation has been carried out using network simulator ns-2 in a AODV routing protocol environment. To the best of the authors' knowledge, this is the first work reporting a distributed trust-based prevention scheme for preventing multiple attacks. We also check the scalability of the technique using variable node densities in the network.

Inference based techniques are one of the major approaches to analyze DNS data and detect malicious domains. The key idea of inference techniques is to first define associations between domains based on features extracted from DNS data. Then, an inference algorithm is deployed to infer potential malicious domains based on their direct/indirect associations with known malicious ones. The way associations are defined is key to the effectiveness of an inference technique. It is desirable to be both accurate (i.e., avoid falsely associating domains with no meaningful connections) and with good coverage (i.e., identify all associations between domains with meaningful connections). Due to the limited scope of information provided by DNS data, it becomes a challenge to design an association scheme that achieves both high accuracy and good coverage. In this paper, we propose a new approach to identify domains controlled by the same entity. Our key idea is an in-depth analysis of active DNS data to accurately separate public IPs from dedicated ones, which enables us to build high-quality associations between domains. Our scheme avoids the pitfall of naive approaches that rely on weak "co-IP" relationship of domains (i.e., two domains are resolved to the same IP) that results in low detection accuracy, and, meanwhile, identifies many meaningful connections between domains that are discarded by existing state-of-the-art approaches. Our experimental results show that the proposed approach not only significantly improves the domain coverage compared to existing approaches but also achieves better detection accuracy. Existing path-based inference algorithms are specifically designed for DNS data analysis. They are effective but computationally expensive. To further demonstrate the strength of our domain association scheme as well as improve the inference efficiency, we construct a new domain-IP graph that can work well with the generic belief propagation algorithm. Through comprehensive experiments, we show that this approach offers significant efficiency and scalability improvement with only a minor impact to detection accuracy, which suggests that such a combination could offer a good tradeoff for malicious domain detection in practice.

The rise of social networks during the last 10 years has created a situation in which up to 100 million new images and photographs are uploaded and shared by users every day. This environment poses a ideal background for those who wish to communicate covertly by the use of steganography. It also creates a new set of challenges for steganalysts, who have to shift their field of work away from a purely scientific laboratory environment and into a diverse real-world scenario, while at the same time having to deal with entirely new problems, such as the detection of steganographic channels or the impact that even a low false positive rate has when investigating the millions of images which are shared every day on social networks. We evaluate how to address these challenges with traditional steganographic and statistical methods, rather then using high performance computing and machine learning. By the double embedding attack on the well-known F5 steganographic algorithm we achieve a false positive rate well below known attacks.