Biblio

How to develop AI systems that can explain how they made decisions is one of the important and hot topics today. Inspired by the dual-process theory in psychology, this paper proposes a human-in-the-loop approach to develop System-2 AI that makes an inference logically and outputs interpretable explanation. Our proposed method first asks crowd workers to raise understandable features of objects of multiple classes and collect training data from the Internet to generate classifiers for the features. Logical decision rules with the set of generated classifiers can explain why each object is of a particular class. In our preliminary experiment, we applied our method to an image classification of Asian national flags and examined the effectiveness and issues of our method. In our future studies, we plan to combine the System-2 AI with System-1 AI (e.g., neural networks) to efficiently output decisions.

We propose a high efficiency Early-Complete Brute Force Elimination method that speeds up the analysis flow of the Camouflage Integrated Circuit (IC). The proposed method is targeted for security qualification of the Camouflaged IC netlists in Intellectual Property (IP) protection. There are two main features in the proposed method. First, the proposed method features immediate elimination of the incorrect Camouflage gates combination for the rest of computation, concentrating the resources into other potential correct Camouflage gates combination. Second, the proposed method features early complete, i.e. revealing the correct Camouflage gates once all incorrect gates combination are eliminated, increasing the computation speed for the overall security analysis. Based on the Python programming platform, we implement the algorithm of the proposed method and test it for three circuits including ISCAS’89 benchmarks. From the simulation results, our proposed method, on average, features 71% lesser number of trials and 79% shorter run time as compared to the conventional method in revealing the correct Camouflage gates from the Camouflaged IC netlist.

In the current scenario, the data owners would like to access data from anywhere and anytime. Hence, they will store their data in public or private cloud along with encryption and particular set of attributes to access control on the cloud data. While uploading the data into public or private cloud they will assign some attribute set to their data. If any authorized cloud user wants to download their data they should enter that particular attribute set to perform further actions on the data owner's data. A cloud user wants to register their details under cloud organization to access the data owner's data. Users wants to submit their details as attributes along with their designation. Based on the Users details Semi-Trusted Authority generates decryption keys to get control on owner's data. A user can perform a lot of operation over the cloud data. If the user wants to read the cloud data he needs to be entering some read related, and if he wants to write the data he needs to be entering write related attribute. For each and every action user in an organization would be verified with their unique attribute set. These attributes will be stored by the admins to the authorized users in cloud organization. These attributes will be stored in the policy files in a cloud. Along with this attribute,a rule based engine is used, to provide the access control to user. If any user leaks their decryption key to the any malicious user data owners wants to trace by sending audit request to auditor and auditor will process the data owners request and concludes that who is the convict.

Android kernel fuzzing is a research area of interest specifically for detecting kernel vulnerabilities which may allow attackers to obtain the root privilege. The number of Android mobile phones is increasing rapidly with the explosive growth of Android kernel drivers. Interface aware fuzzing is an effective technique to test the security of kernel driver. Existing researches rely on static analysis with kernel source code. However, in fact, there exist millions of Android mobile phones without public accessible source code. In this paper, we propose a hybrid interface recovery method for fuzzing kernels which can recover kernel driver interface no matter the source code is available or not. In white box condition, we employ a dynamic interface recover method that can automatically and completely identify the interface knowledge. In black box condition, we use reverse engineering to extract the key interface information and use similarity computation to infer argument types. We evaluate our hybrid algorithm on on 12 Android smartphones from 9 vendors. Empirical experimental results show that our method can effectively recover interface argument lists and find Android kernel bugs. In total, 31 vulnerabilities are reported in white and black box conditions. The vulnerabilities were responsibly disclosed to affected vendors and 9 of the reported vulnerabilities have been already assigned CVEs.

With the introduction of Cloud computing, a new era of computing has begun. Cloud has the ability to provide flexible, cost effective pay-as-you-go service. In the modern day computing, outsourcing of data/multimedia into the cloud has become an effective trend as cloud provides storage as a service, platform/software as a service, infrastructure as a service etc. Seamless exchange of data /multimedia is made possible ensuring the data available anytime, anywhere. Even though cloud based services offer many advantages, data owners are still hesitant to keep their data with the third party. Confidentiality, Integrity, Privacy and Non-repudiation are the major concerns of the outsourced data. To secure the data exchange between users and the cloud, many traditional security approaches are proposed. In this paper, a hybrid encryption technique to secure the images is proposed. The scheme uses Elliptic Curve Cryptography to generate the secret key, which in turn used for DES and AES algorithms.

Nowadays the provision of online services by government or business organizations has become a standard and necessary operation. Transferring data including the confidential or sensitive information via Internet or insecure network and exchange of them is also increased day by day. As a result, confidential information leakage and cyber threats are also heightened. Confidential information trading became one of the most profitable businesses. Encrypting the data is a solution to secure the data from being exposed. In this paper, we would like to propose a solution for the secure transfer of data using symmetric encryption, asymmetric encryption technologies and Key Generation Server as a mixed hybrid solution. A Symmetric encryption, modified AES algorithm, is used to encrypt data. Digital certificate is used both for data encryption and digital signing to assure data integrity. Key generation server is used to generate the second secret key from the publicly recognized information of a person and this key is used as a second secret key in the modified AES. The proposed hybrid solution can be utilized in any applications that require high confidentiality, integrity of data and non-repudiation.

It is a well-known fact that the use of Cloud Computing is becoming very common all over the world for data storage and analysis. But the proliferation of the threats in cloud is also their; threats like Information breaches, Data thrashing, Cloud account or Service traffic hijacking, Insecure APIs, Denial of Service, Malicious Insiders, Abuse of Cloud services, Insufficient due Diligence and Shared Technology Vulnerable. This paper tries to come up with the solution for the threat (Denial of Service) in cloud. We attempt to give our newly proposed model by the hybridization of Genetic algorithm and extension of Diffie Hellman algorithm and tries to make cloud transmission secure from upcoming intruders.

Internet of Things (IoT) is experiencing significant growth in the safety-critical applications which have caused new security challenges. These devices are becoming targets for different types of physical attacks, which are exacerbated by their diversity and accessibility. Therefore, there is a strict necessity to support embedded software developers to identify and remediate the vulnerabilities and create resilient applications against such attacks. In this paper, we propose a hardware security vulnerability assessment based on fault injection of an embedded application. In our security assessment, we apply a fault injection attack by using our clock glitch generator on a critical medical IoT device. Furthermore, we analyze the potential risks of ignoring these attacks in this embedded application. The results will inform the embedded software developers of various security risks and the required steps to improve the security of similar MCU-based applications. Our hardware security assessment approach is easy to apply and can lead to secure embedded IoT applications against fault attacks.

Embedded Systems (ES) development has been historically focused on functionality rather than security, and today it still applies in many sectors and applications. However, there is an increasing number of security threats over ES, and a successful attack could have economical, physical or even human consequences, since many of them are used to control critical applications. A standardized and general accepted security testing framework is needed to provide guidance, common reporting forms and the possibility to compare the results along the time. This can be achieved by introducing security metrics into the evaluation or assessment process. If carefully designed and chosen, metrics could provide a quantitative, repeatable and reproducible value that would reflect the level of security protection of the ES. This paper analyzes the features that a good security metric should exhibit, introduces a taxonomy for classifying them, and finally, it carries out a literature survey on security metrics for the security evaluation of ES. In this review, more than 500 metrics were collected and analyzed. Then, they were reduced to 169 metrics that have the potential to be applied to ES security evaluation. As expected, the 77.5% of them is related exclusively to software, and only the 0.6% of them addresses exclusively hardware security. This work aims to lay the foundations for constructing a security evaluation methodology that uses metrics so as to quantify the security level of an ES.

Mobile and IoT operating systems–and their ensuing software updates–are usually distributed as binary files. Given that these binary files are commonly closed source, users or businesses who want to assess the security of the software need to rely on reverse engineering. Further, verifying the correct application of the latest software patches in a given binary is an open problem. The regular application of software patches is a central pillar for improving mobile and IoT device security. This requires developers, integrators, and vendors to propagate patches to all affected devices in a timely and coordinated fashion. In practice, vendors follow different and sometimes improper security update agendas for both mobile and IoT products. Moreover, previous studies revealed the existence of a hidden patch gap: several vendors falsely reported that they patched vulnerabilities. Therefore, techniques to verify whether vulnerabilities have been patched or not in a given binary are essential. Deep learning approaches have shown to be promising for static binary analyses with respect to inferring binary similarity as well as vulnerability detection. However, these approaches fail to capture the dynamic behavior of these systems, and, as a result, they may inundate the analysis with false positives when performing vulnerability discovery in the wild. In particular, they cannot capture the fine-grained characteristics necessary to distinguish whether a vulnerability has been patched or not. In this paper, we present PATCHECKO, a vulnerability and patch presence detection framework for executable binaries. PATCHECKO relies on a hybrid, cross-platform binary code similarity analysis that combines deep learning-based static binary analysis with dynamic binary analysis. PATCHECKO does not require access to the source code of the target binary nor that of vulnerable functions. We evaluate PATCHECKO on the most recent Google Pixel 2 smartphone and the Android Things IoT firmware images, within which 25 known CVE vulnerabilities have been previously reported and patched. Our deep learning model shows a vulnerability detection accuracy of over 93%. We further prune the candidates found by the deep learning stage–which includes false positives–via dynamic binary analysis. Consequently, PATCHECKO successfully identifies the correct matches among the candidate functions in the top 3 ranked outcomes 100% of the time. Furthermore, PATCHECKO's differential engine distinguishes between functions that are still vulnerable and those that are patched with an accuracy of 96%.

The core operation of public-key cryptosystem e.g. elliptic curve cryptography (ECC) is the modular multiplication. It is the heavy computational block and the most costly cryptographic operation. Area-Efficient hardware architecture of 256-bit modified interleaved modular multiplication (IMM) is represented in this research. The novelty of this work is the device area minimization with keeping computational time as minimum as possible i.e., 2.09 μs for ECC with Koblitz Curve. In this research, IMM is implemented using a fewer number of resources such as 421 slices, 514 FF pairs, 522 registers, 1770 LUTs, and 1463 LUT-FF pairs. This hardware implementation provides a maximum frequency of 122.883 MHz and area-time (AT) product 0.879 and throughput rate 122.49 Mbps on Virtex-7 FPGA technology which is better than the other related recent works. The proposed design saves approximately 61.75% to 93.16% slice LUTs, 95.76% to 133.69% LUT-FF pairs, and 103.8% to 168.65% occupied slices on the Virtex-7 FPGA for the 256-bit prime field. This proposed hardware implementation design also keeps less AT product which is the most crucial parameter for ECC operation. To our best knowledge, this design provides better performance than the recently available designs for IMM for ECC operation.

Packet classification is a key function in network security systems in SDN, which detect potential threats by matching the packet header bits and a given rule set. It needs to support multi-dimensional fields, large rule sets, and high throughput. Bit Vector-based packet classification methods can support multi-field matching and achieve a very high throughput, However, the range matching is still challenging. To address issue, this paper proposes a Range Supported Bit Vector (RSBV) algorithm for processing the range fields. RSBV uses specially designed codes to store the pre-computed results in memory, and the result of range matching is derived through pipelined Boolean operations. Through a two-dimensional modular architecture, the RSBV can operate at a high clock frequency and line-rate processing can be guaranteed. Experimental results show that for a 1K and 512-bit OpenFlow rule set, the RSBV can sustain a throughput of 520 Million Packets Per Second.

Recognising user's risky behaviours in real-time is an important element of providing appropriate solutions and recommending suitable actions for responding to cybersecurity threats. Employing user modelling and machine learning can make this process automated by requires high-performance intelligent agent to create the user security profile. User profiling is the process of producing a profile of the user from historical information and past details. This research tries to identify the monitoring factors and suggests a novel observation solution to create high-performance sensors to generate the user security profile for a home user concerning the user's privacy. This observer agent helps to create a decision-making model that influences the user's decision following real-time threats or risky behaviours.

Bitcoin was the first successful decentralized cryptocurrency and remains the most popular of its kind to this day. Despite the benefits of its blockchain, Bitcoin still faces serious scalability issues, most importantly its ever-increasing blockchain size. While alternative designs introduced schemes to periodically create snapshots and thereafter prune older blocks, already-deployed systems such as Bitcoin are often considered incapable of adopting corresponding approaches. In this work, we revise this popular belief and present CoinPrune, a snapshot-based pruning scheme that is fully compatible with Bitcoin. CoinPrune can be deployed through an opt-in velvet fork, i.e., without impeding the established Bitcoin network. By requiring miners to publicly announce and jointly reaffirm recent snapshots on the blockchain, CoinPrune establishes trust into the snapshots' correctness even in the presence of powerful adversaries. Our evaluation shows that CoinPrune reduces the storage requirements of Bitcoin already by two orders of magnitude today, with further relative savings as the blockchain grows. In our experiments, nodes only have to fetch and process 5GiB instead of 230GiB of data when joining the network, reducing the synchronization time on powerful devices from currently 5h to 46min, with even more savings for less powerful devices.

Anonymous communication networks (ACNs) are intended to protect the metadata during communication. As classic ACNs, onion mix-nets are famous for strong anonymity, in which the source defines a static path and wraps the message multi-times with the public keys of nodes on the path, through which the message is relayed to the destination. However, onion mix-nets lacks in resilience when the static on-path mixes fail. Mix failure easily results in message loss, communication failure, and even specific attacks. Therefore, it is desirable to achieve resilient routing in onion mix-nets, providing persistent routing capability even though node failure. The state-of-theart solutions mainly adopt mix groups and thus need to share secret keys among all the group members which may cause single point of failure. To address this problem, in this work we propose a hybrid routing approach, which embeds the onion mix-net with hop-by-hop routing to increase routing resilience. Furthermore, we propose the threshold hybrid routing to achieve better key management and avoid single point of failure. As for experimental evaluations, we conduct quantitative analysis of the resilience and realize a local T-hybrid routing prototype to test performance. The experimental results show that our proposed routing strategy increases routing resilience effectively, at the expense of acceptable latency.

Big data provides a way to handle and analyze large amount of data or complex set. It provides a systematic extraction also. In this paper a hybrid security analysis based on intelligent adaptive learning in big data has been discussed with the current trends. This paper also explores the possibility of cloud computing collaboration with big data. The advantages along with the impact for the overall platform evaluation has been discussed with the traditional trends. It has been useful in the analysis and the exploration of future research. This discussion also covers the computational variability and the connotation in terms of data reliability, availability and management in big data with data security aspects.

Humans are a key part of software development, including customers, designers, coders, testers and end users. In this keynote talk I explain why incorporating human-centric issues into software engineering for next-generation applications is critical. I use several examples from our recent and current work on handling human-centric issues when engineering various `smart living' cloud- and edge-based software systems. This includes using human-centric, domain-specific visual models for non-technical experts to specify and generate data analysis applications; personality impact on aspects of software activities; incorporating end user emotions into software requirements engineering for smart homes; incorporating human usage patterns into emerging edge computing applications; visualising smart city-related data; reporting diverse software usability defects; and human-centric security and privacy requirements for smart living systems. I assess the usefulness of these approaches, highlight some outstanding research challenges, and briefly discuss our current work on new human-centric approaches to software engineering for smart living applications.

Logic encryption is a popular Design-for-Security(DfS) solution that offers protection against the potential adversaries in the third-party fab labs and end-users. However, over the years, logic encryption has been a target of several attacks, especially Boolean satisfiability attacks. This paper exploits SAT attack's inability of deobfuscating sequential circuits as a defense against it. We propose several strategies capable of preventing the SAT attack by obfuscating the scan-based Design-for-Testability (DfT) infrastructure. Unlike the existing SAT-resilient schemes, the proposed techniques do not suffer from poor output corruption for wrong keys. This paper also offers various probable solutions for inserting the key-gates into the circuit that ensures protection against numerous other attacks, which exploit weak key-gate locations. Along with several gate-level obfuscation strategies, this paper also presents a Cellular Automata (CA) guided FSM obfuscation strategy to offer protection at a higher abstraction level, that is, RTL-level. For all the proposed schemes, rigorous security analysis against various attacks evaluates their strengths and limitations. Testability analysis also ensures that none of the proposed techniques hamper the basic testing properties of the ICs. We also present a CA-based FSM watermarking strategy that helps to detect potential theft of the designer's IP by any adversary.

Adaptation of e-commerce in third world countries requires more secure computing facilities. Online data is vulnerable and susceptible to active attacks. Hundreds of security mechanisms and services have been proposed to curb this challenge. However, available security mechanisms, sufficiently strong, are heavy for the machines used. To secure online data where machines' processing power and memory are deficient, a Hybrid Encryption Standard (HES) is proposed. The HES is built on the Data Encryption Standard (DES) algorithm and its siblings. The component units of the DES are redesigned towards reduced demands for processing power and memory. Precisely, white box designs of IP tables, PC tables, Expansion tables, Rotation tables, S-boxes and P-boxes are proposed, all aimed at reducing the processing time and memory demands. Evaluation of the performance of the HES algorithm against the performance of the traditional DES algorithm reveal that the HES out-performs the DES with regards to speed, memory demands, and general acceptance by novice practitioners in the cryptography field. In addition, reproducibility and flexibility are attractive features of the HES over the DES.

Distributed training is a novel approach to accelerating training of Deep Neural Networks (DNN), but common training libraries fall short of addressing the distributed nature of heterogeneous processors or interruption by other workloads on the shared processing nodes. This paper describes distributed training of DNN on computational storage devices (CSD), which are NAND flash-based, high-capacity data storage with internal processing engines. A CSD-based distributed architecture incorporates the advantages of federated learning in terms of performance scalability, resiliency, and data privacy by eliminating the unnecessary data movement between the storage device and the host processor. The paper also describes Stannis, a DNN training framework that improves on the shortcomings of existing distributed training frameworks by dynamically tuning the training hyperparameters in heterogeneous systems to maintain the maximum overall processing speed in term of processed images per second and energy efficiency. Experimental results on image classification training benchmarks show up to 3.1x improvement in performance and 2.45x reduction in energy consumption when using Stannis plus CSD compare to the generic systems.

An ideal lattice is defined over a ring learning with errors (Ring-LWE) problem. Polynomial multiplication over the ring is the most computational and time-consuming block in lattice-based cryptography. This paper presents the first hardware design of the polynomial multiplication for LAC, one of the Round-2 candidates of the NIST PQC Standardization Process, which has byte-level modulus p=251. The proposed architecture supports polynomial multiplications for different degree n (n=512/1024/2048). For designing the scheme, we used the Vivado HLS compiler, a high-level synthesis based hardware design methodology, which is able to optimize software algorithms into actual hardware products. The design of the scheme takes 274/280/291 FFs and 204/217/208 LUTs on the Xilinx Artix-7 family FPGA, requested by NIST PQC competition for hardware implementation. Multiplication core uses only 1/1/2 pieces of 18Kb BRAMs, 1/1/1 DSPs, and 90/94/95 slices on the board. Our timing result achieved in an alternative degree n with 5.052/4.3985/5.133ns.

In this first paper of a two-part series, the harmonic plane decomposition is introduced, which is an extension of the vector-space decomposition. In multiphase electrical machines with variable phase-pole configurations, the vector-space decomposition leads to a varying numbers of vector spaces when changing the configuration. Consequently, the model and current control become discontinuous. The method in this paper is based on samples of each single slot currents, similarly to a discrete Fourier transformation in the space domain that accounts for the winding configuration. It unifies the Clarke transformation for all possible phase-pole configurations such that a fixed number of orthogonal harmonic planes are created, which facilitates the current control during reconfigurations. The presented method is not only limited to the modeling of multiphase electrical machines but all kinds of existing machines can be modeled. In the second part of this series, the harmonic plane decomposition will be completed for all types of machine configurations.

The Representational State Transfer (REST) is a distributed application architecture style which adopted on providing various network services. The identity authentication protocol Kerberos has been used to guarantee the security identity authentication of many service platforms. However, the deployment of Kerberos protocol is limited by the defects such as password guessing attacks, data tampering, and replay attacks. In this paper, an optimized Kerberos protocol is proposed and applied in a REST-style Cloud Storage Architecture. Firstly, we propose a Lately Used Newly (LUN) key replacement method to resist the password guessing attacks in Kerberos protocol. Secondly, we propose a formatted signature algorithm and a combination of signature string and time stamp method to cope with the problems of tampering and replay attacks which in deploying Kerberos. Finally, we build a security protection module using the optimized Kerberos protocol to guarantee a secure identity authentication and the reliable data communication between the client and the server. Analyses show that the module significantly improves the security of Kerberos protocol in REST-style cloud storage services.

This is an innovative practice full paper. In past projects, we have successfully used a private TOR (anonymity network) platform that enabled our students to explore the end-to-end inner workings of the TOR anonymity network through a number of controlled hands-on lab assignments. These have saisfied the needs of curriculum focusing on networking functions and algorithms. To be able to extend the use and application of the private TOR platform into cryptography courses, there is a desperate need to enhance the platform to allow the development of hands-on lab assignments on the cryptographic algorithms and methods utilized in the creation of TOR secure connections and end-to-end circuits for anonymity.In tackling this challenge, and since TOR is open source software, we identify the cryptographic functions called by the TOR algorithms in the process of establishing TLS connections and creating end-to-end TOR circuits as well tearing them down. We instrumented these functions with the appropriate code to log the cryptographic keys dynamically created at all nodes involved in the creation of the end to end circuit between the Client and the exit relay (connected to the target server).We implemented a set of pedagogical lab assignments on a private TOR platform and present them in this paper. Using these assignments, students are able to investigate and validate the cryptographic procedures applied in the establishment of the initial TLS connection, the creation of the first leg of a TOR circuit, as well as extending the circuit through additional relays (at least two relays). More advanced assignments are created to challenge the students to unwrap the traffic sent from the Client to the exit relay at all onion skin layers and compare it with the actual traffic delivered to the target server.

Supervisory Control and Data Acquisition (SCADA) systems have been a frequent target of cyberattacks in Industrial Control Systems (ICS). As such systems are a frequent target of highly motivated attackers, researchers often resort to intrusion detection through machine learning techniques to detect new kinds of threats. However, current research initiatives, in general, pursue higher detection accuracies, neglecting the detection of new kind of threats and their proposal detection scope. This paper proposes a novel, reliable host-based intrusion detection for SCADA systems through the Operating System (OS) diversity. Our proposal evaluates, at the OS level, the SCADA communication over time and, opportunistically, detects, and chooses the most appropriate OS to be used in intrusion detection for reliability purposes. Experiments, performed through a variety of SCADA OSs front-end, shows that OS diversity provides higher intrusion detection scope, improving detection accuracy by up to 8 new attack categories. Besides, our proposal can opportunistically detect the most reliable OS that should be used for the current environment behavior, improving by up to 8%, on average, the system accuracy when compared to a single OS approach, in the best case.