Biblio

Data poisoning is a type of adversarial attack on training data where an attacker manipulates a fraction of data to degrade the performance of machine learning model. There are several known defensive mechanisms for handling offline attacks, however defensive measures for online learning, where data points arrive sequentially, have not garnered similar interest. In this work, we propose a defense mechanism to minimize the degradation caused by the poisoned training data on a learner's model in an online setup. Our proposed method utilizes an influence function which is a classic technique in robust statistics. Further, we supplement it with the existing data sanitization methods for filtering out some of the poisoned data points. We study the effectiveness of our defense mechanism on multiple datasets and across multiple attack strategies against an online learner.

One major tool of Energy Management Systems for monitoring the status of the power grid is State Estimation (SE). Since the results of state estimation are used within the energy management system, the security of the power system state estimation tool is most important. The research in this area is targeting detection of False Data Injection attacks on measurements. Though this aspect is crucial, SE also depends on database that are used to describe the relationship between measurements and systems' states. This paper presents a two-stage optimization framework to not only detect, but also correct cyber-attacks pertaining the measurements' model parameters used by the SE routine. In the first stage, an estimate of the line parameters ratios are obtained. In the second stage, the estimated ratios from stage I are used in a Bi-Level model for obtaining a final estimate of the measurements' model parameters. Hence, the presented framework does not only unify the detection and correction in a single optimization run, but also provide a monitoring scheme for the SE database that is typically considered static. In addition, in the two stages, linear programming framework is preserved. For validation, the IEEE 118 bus system is used for implementation. The results illustrate the effectiveness of the proposed model for detecting attacks in the database used in the state estimation process.

The Internet of Things (IoT) is advancing technology by creating smart surroundings that make it easier for humans to do their work. This technological advancement not only improves human life and expands economic opportunities, but also allows intruders or attackers to discover and exploit numerous methods in order to circumvent the security of IoT networks. Hence, security and privacy are the key concerns to the IoT networks. It is vital to protect computer and IoT networks from many sorts of anomalies and attacks. Traditional intrusion detection systems (IDS) collect and employ large amounts of data with irrelevant and inappropriate attributes to train machine learning models, resulting in long detection times and a high rate of misclassification. This research presents an advance approach for the design of IDS for IoT networks based on the Particle Swarm Optimization Algorithm (PSO) for feature selection and the Extreme Gradient Boosting (XGB) model for PSO fitness function. The classifier utilized in the intrusion detection process is Random Forest (RF). The IoTID20 is being utilized to evaluate the efficacy and robustness of our suggested strategy. The proposed system attains the following level of accuracy on the IoTID20 dataset for different levels of classification: Binary classification 98 %, multiclass classification 83 %. The results indicate that the proposed framework effectively detects cyber threats and improves the security of IoT networks.

As a core problem of multi-agent systems, multiagent pathfinding has an important impact on the efficiency of multi-agent systems. Because of this, many novel multi-agent pathfinding methods have been proposed over the years. However, these methods have focused on different agents with different goals for research, and less research has been done on scenarios where different agents have the same goal. We propose a multiagent pathfinding method incorporating a multi-objective gray wolf optimization algorithm to solve the multi-agent pathfinding problem with the same objective. First, constrained optimization modeling is performed to obtain objective functions about agent wholeness and security. Then, the multi-objective gray wolf optimization algorithm is improved for solving the constrained optimization problem and further optimized for scenarios with insufficient computational resources. To verify the effectiveness of the multi-objective gray wolf optimization algorithm, we conduct experiments in a series of simulation environments and compare the improved multi-objective grey wolf optimization algorithm with some classical swarm intelligence optimization algorithms. The results show that the multi-agent pathfinding method incorporating the multi-objective gray wolf optimization algorithm is more efficient in handling multi-agent pathfinding problems with the same objective.

With the future 6G era, spiking neural networks (SNNs) can be powerful processing tools in various areas due to their strong artificial intelligence (AI) processing capabilities, such as biometric recognition, AI robotics, autonomous drive, and healthcare. However, within Cyber Physical System (CPS), SNNs are surprisingly vulnerable to adversarial examples generated by benign samples with human-imperceptible noise, this will lead to serious consequences such as face recognition anomalies, autonomous drive-out of control, and wrong medical diagnosis. Only by fully understanding the principles of adversarial attacks with adversarial samples can we defend against them. Nowadays, most existing adversarial attacks result in a severe accuracy degradation to trained SNNs. Still, the critical issue is that they only generate adversarial samples by randomly adding, deleting, and flipping spike trains, making them easy to identify by filters, even by human eyes. Besides, the attack performance and speed also can be improved further. Hence, Spike Probabilistic Attack (SPA) is presented in this paper and aims to generate adversarial samples with more minor perturbations, greater model accuracy degradation, and faster iteration. SPA uses Poisson coding to generate spikes as probabilities, directly converting input data into spikes for faster speed and generating uniformly distributed perturbation for better attack performance. Moreover, an objective function is constructed for minor perturbations and keeping attack success rate, which speeds up the convergence by adjusting parameters. Both white-box and black-box settings are conducted to evaluate the merits of SPA. Experimental results show the model's accuracy under white-box attack decreases by 9.2S% 31.1S% better than others, and average success rates are 74.87% under the black-box setting. The experimental results indicate that SPA has better attack performance than other existing attacks in the white-box and better transferability performance in the black-box setting,

The most widely used protocol for routing across the 6LoWPAN stack is the Routing Protocol for Low Power and Lossy (RPL) Network. However, the RPL lacks adequate security solutions, resulting in numerous internal and external security vulnerabilities. There is still much research work left to uncover RPL's shortcomings. As a result, we first implement the worst parent selection (WPS) attack in this paper. Second, we offer an intrusion detection system (IDS) to identify the WPS attack. The WPS attack modifies the victim node's objective function, causing it to choose the worst node as its preferred parent. Consequently, the network does not achieve optimal convergence, and nodes form the loop; a lower rank node selects a higher rank node as a parent, effectively isolating many nodes from the network. In addition, we propose DWA-IDS as an IDS for detecting WPS attacks. We use the Contiki-cooja simulator for simulation purposes. According to the simulation results, the WPS attack reduces system performance by increasing packet transmission time. The DWA-IDS simulation results show that our IDS detects all malicious nodes that launch the WPS attack. The true positive rate of the proposed DWA-IDS is more than 95%, and the detection rate is 100%. We also deliberate the theoretical proof for the false-positive case as our DWA-IDS do not have any false-positive case. The overhead of DWA-IDS is modest enough to be set up with low-power and memory-constrained devices.

We propose a novel theoretical approach for solving a Stochastic Security Game using augmented Markov Decison Processes and an experimental evaluation. Most of the previous works mentioned in the literature focus on Linear Programming techniques seeking Strong Stackelberg Equilibria through the defender and attacker’s strategy spaces. Although effective, these techniques are computationally expensive and tend to not scale well to very large problems. By fixing the set of the possible defense strategies, our approach is able to use the well-known augmented MDP formalism to compute an optimal policy for an attacker facing a defender patrolling. Experimental results on fully observable cases validate our approach and show good performances in comparison with optimistic and pessimistic approaches. However, these results also highlight the need of scalability improvements and of handling the partial observability cases.

Based on the analysis of cluster supply chain risk characteristics, starting from the analysis of technical risk dimensions, information risk dimensions, human risk dimensions, and capital risk dimensions, a cluster supply chain risk severity assessment index system is designed. The fuzzy C-means clustering algorithm based on data flow is used to cluster each supply chain, analyze the risk severity of the supply chain, and evaluate the decision of the supply chain risk severity level based on the cluster weights and cluster center range. Based on the analytic hierarchy process, the risk severity of the entire clustered supply chain is made an early warning decision, and the clustered supply chain risk severity early warning level is obtained. The results of simulation experiments verify the feasibility of the decision method for cluster supply chain risk severity, and improve the theoretical support for cluster supply chain risk severity prediction.

There are problems in which the use of linear regression is not sufficiently justified. In these cases, fuzzy linear regression can be used as a modeling tool. The problem of constructing a fuzzy linear regression can usually be reduced to a linear programming problem. One of the features of the resulting linear programming problem is that it uses a relatively large number of constraints in the form of inequalities with a relatively small number of variables. It is known that the problem of constructing a fuzzy linear regression is reduced to the problem of linear programming. If the user does not have enough computing power the resulting problem can be transferred to the cloud server. Two approaches are used for the confidential transfer of the problem to the server: the approach based on cryptographic encryption, and the transformational approach. The paper describes a protocol based on the transformational approach that allows for secure outsourcing of fuzzy linear regression.

Evolutionary computation (EC) is a kind of advanced computational intelligence (CI) algorithm and advanced artificial intelligence (AI) algorithm. EC algorithms have been widely studied for solving optimization and scheduling problems in various real-world applications, which act as one of the Big Three in CI and AI, together with fuzzy systems and neural networks. Even though EC has been fast developed in recent years, there is an assumption that the algorithm designer can obtain the objective function of the optimization problem so that they can calculate the fitness values of the individuals to follow the “survival of the fittest” principle in natural selection. However, in a real-world application scenario, there is a kind of problem that the objective function is privacy so that the algorithm designer can not obtain the fitness values of the individuals directly. This is the privacy-preserving optimization problem (PPOP) where the assumption of available objective function does not check out. How to solve the PPOP is a new emerging frontier with seldom study but is also a challenging research topic in the EC community. This paper proposes a rank-based cryptographic function (RCF) to protect the fitness value information. Especially, the RCF is adopted by the algorithm user to encrypt the fitness values of all the individuals as rank so that the algorithm designer does not know the exact fitness information but only the rank information. Nevertheless, the RCF can protect the privacy of the algorithm user but still can provide sufficient information to the algorithm designer to drive the EC algorithm. We have applied the RCF privacy-preserving method to two typical EC algorithms including particle swarm optimization (PSO) and differential evolution (DE). Experimental results show that the RCF-based privacy-preserving PSO and DE can solve the PPOP without performance loss.

Deep learning has made remarkable achievements in various domains. Active learning, which aims to reduce the budget for training a machine-learning model, is especially useful for the Deep learning tasks with the demand of a large number of labeled samples. Unfortunately, our empirical study finds that many of the active learning heuristics are not effective when applied to Deep learning models in batch settings. To tackle these limitations, we propose a density weighted diversity based query strategy (DWDS), which makes use of the geometry of the samples. Within a limited labeling budget, DWDS enhances model performance by querying labels for the new training samples with the maximum informativeness and representativeness. Furthermore, we propose a beam-search based method to obtain a good approximation to the optimum of such samples. Our experiments show that DWDS outperforms existing algorithms in Deep learning tasks.

Enabled by the industrial Internet, intelligent transportation has made remarkable achievements such as autonomous vehicles by carnegie mellon university (CMU) Navlab, Google Cars, Tesla, etc. Autonomous vehicles benefit, in various aspects, from the cooperation of the industrial Internet and cyber-physical systems. In this process, users in autonomous vehicles submit query contents, such as service interests or user locations, to service providers. However, privacy concerns arise since the query contents are exposed when the users are enjoying the services queried. Existing works on privacy preservation of query contents rely on location perturbation or k-anonymity, and they suffer from insufficient protection of privacy or low query utility incurred by processing multiple queries for a single query content. To achieve sufficient privacy preservation and satisfactory query utility for autonomous vehicles querying services in cyber-physical systems, this article proposes a novel privacy notion of client-based personalized k-anonymity (CPkA). To measure the performance of CPkA, we present a privacy metric and a utility metric, based on which, we formulate two problems to achieve the optimal CPkA in term of privacy and utility. An approach, including two modules, to establish mechanisms which achieve the optimal CPkA is presented. The first module is to build in-group mechanisms for achieving the optimal privacy within each content group. The second module includes linear programming-based methods to compute the optimal grouping strategies. The in-group mechanisms and the grouping strategies are combined to establish optimal CPkA mechanisms, which achieve the optimal privacy or the optimal utility. We employ real-life datasets and synthetic prior distributions to evaluate the CPkA mechanisms established by our approach. The evaluation results illustrate the effectiveness and efficiency of the established mechanisms.

With the maturity of Internet of things technology, massive data transmission has become the focus of research. In order to solve the problem of low speed of traditional hybrid data fast distribution algorithm for wireless sensor networks, a hybrid data fast distribution algorithm for wireless sensor networks based on visual Internet of things is designed. The logic structure of mixed data input gate in wireless sensor network is designed through the visual Internet of things. The objective function of fast distribution of mixed data in wireless sensor network is proposed. The number of copies of data to be distributed is dynamically calculated and the message deletion strategy is determined. Then the distribution parameters are calibrated, and the fitness ranking is performed according to the distribution quantity to complete the algorithm design. The experimental results show that the distribution rate of the designed algorithm is significantly higher than that of the control group, which can solve the problem of low speed of traditional data fast distribution algorithm.

Internet of Things (IoT) is flourishing in several application areas, such as smart cities, smart factories, smart homes, smart healthcare, etc. With the adoption of IoT in critical scenarios, it is crucial to investigate its security aspects. All the layers of IoT are vulnerable to severely disruptive attacks. However, the attacks in IoT Network layer have a high impact on communication between the connected objects. Routing in most of the IoT networks is carried out by IPv6 Routing Protocol for Low-Power and Lossy Networks (RPL). RPL-based IoT offers limited protection against routing attacks. A trust-based approach for routing security is suitable to be integrated with IoT systems due to the resource-constrained nature of devices. This research proposes a trust-based secure routing protocol to provide security against packet dropping attacks in RPL-based IoT networks. IoT networks are dynamic and consist of both static and mobile nodes. Hence the chosen trust metrics in the proposed method also include the mobility-based metrics for trust evaluation. The proposed solution is integrated into RPL as a modified objective function, and the results are compared with the default RPL objective function, MRHOF. The analysis and evaluation of the proposed protocol indicate its efficacy and adaptability in a mobile IoT environment.

In a cloud data center, the client requests are catered by placing the services in its servers. Such services are deployed through a sandboxing platform to ensure proper isolation among services from different users. Due to the lightweight nature, containers have become increasingly popular to support such sandboxing. However, for supporting effective and efficient data center resource usage with minimum resource footprints, improving the containers' consolidation ratio is significant for the cloud service providers. Towards this end, in this paper, we propose an exciting direction to significantly boost up the consolidation ratio of a data-center environment by effectively managing the containers' states. We observe that many cloud-based application services are event-triggered, so they remain inactive unless some external service request comes. We exploit the fact that the containers remain in an idle state when the underlying service is not active, and thus such idle containers can be checkpointed unless an external service request comes. However, the challenge here is to design an efficient mechanism such that an idle container can be resumed quickly to prevent the loss of the application's quality of service (QoS). We have implemented the system, and the evaluation is performed in Amazon Elastic Compute Cloud. The experimental results have shown that the proposed algorithm can manage the containers' states, ensuring the increase of consolidation ratio.

Web caching is one strategy that can be used to speed up response times by storing frequently accessed data in the cache server. Given the cache server limited capacity, it is necessary to determine the priority of cached data that can enter the cache server. This study simulated cached data prioritization based on an objective function as a characteristic of problem-solving using an optimization approach. The objective function of web caching is formulated based on the variable data size, count access, and frequency-time access. Then we use the knapsack problem method to find the optimal solution. The Simulations run three swarm intelligence algorithms Ant Colony Optimization (ACO), Genetic Algorithm (GA), and Binary Particle Swarm Optimization (BPSO), divided into several scenarios. The simulation results show that the GA algorithm relatively stable and fast to convergence. The ACO algorithm has the advantage of a non-random initial solution but has followed the pheromone trail. The BPSO algorithm is the fastest, but the resulting solution quality is not as good as ACO and GA.

Web caching is a strategy that can be used to speed up website access on the client-side. This strategy is implemented by storing as many popular web objects as possible on the cache server. All web objects stored on a cache server are called cached data. Requests for cached web data on the cache server are much faster than requests directly to the origin server. Not all web objects can fit on the cache server due to their limited capacity. Therefore, optimizing cached data in a web caching strategy will determine which web objects can enter the cache server to have maximum profit. This paper simulates a web caching strategy optimization with a knapsack problem approach using the Ant Colony optimization (ACO), Genetic Algorithm (GA), and a combination of the two. Knapsack profit is seen from the number of web objects that can be entered into the cache server but with the minimum objective function value. The simulation results show that the combination of ACO and GA is faster to produce an optimal solution and is not easily trapped by the local optimum.

Model predictive control (MPC) has become one of the most attractive control techniques due to its outstanding dynamic performance for motor drives. Besides, MPC with constant switching frequency (CSF-MPC) maintains the advantages of MPC as well as constant frequency but the selection of weighting factors in the cost function is difficult for CSF-MPC. Fortunately, the application of artificial neural networks (ANN) can accelerate the selection without any additional computation burden. Therefore, this paper designs a specific artificial neural network optimized by genetic algorithm (GA-ANN) to select the optimal weighting factors of CSF-MPC for permanent magnet synchronous motor (PMSM) drives fed by three-level T-type inverter. The key performance metrics like THD and switching frequencies error (ferr) are extracted from simulation and this data are utilized to train and evaluate GA-ANN. The trained GA-ANN model can automatically and precisely select the optimal weighting factors for minimizing THD and ferr under different working conditions of PMSM. Furthermore, the experimental results demonstrate the validation of GA-ANN and robustness of optimal weighting factors under different torque loads. Accordingly, any arbitrary user-defined working conditions which combine THD and ferr can be defined and the optimum weighting factors can be fast and explicitly determined via the trained GA-ANN model.

Demand response (DR) is one of the most economical methods for peak demand reduction, renewable energy integration and ancillary service support. Residential electrical energy consumption takes approximately 33% of the total electricity usage and hence has great potentials in DR applications. However, residential DR encounters various challenges such as small individual magnitude, stochastic consuming patterns and privacy issues. In this study, we propose a stochastic optimal mechanism to tackle these issues and try to reveal the benefits from residential DR implementation. Stochastic residential load (SRL) models, a generation cost prediction (GCP) model and a stochastic optimal load aggregation (SOLA) model are developed. A set of uniformly distributed scalers is introduced into the SOLA model to efficiently avoid the peak demand rebound problem in DR applications. The SOLA model is further transformed into a deterministic LP model. Time-of-Use (TOU) tariff is adopted as the price structure because of its similarity and popularity. Case studies show that the proposed mechanism can significantly reduce the peak-to-average power ratio (PAPR) of the load profile as well as the electrical energy cost. Furthermore, the impacts of consumers' participation levels in the DR program are investigated. Simulation results show that the 50% participation level appears as the best case in terms system stability. With the participation level of 80%, consumers' electrical energy cost is minimized. The proposed mechanism can be used by a residential load aggregator (LA) or a utility to plan a DR program, predict its impacts, and aggregate residential loads to minimize the electrical energy cost.

The standard routing technique that was developed for satisfying low power IoT application needs is RPL which is a protocol in compliance with 6LoWPAN specification. RPL was created for addressing the issues and challenges of constrained and lossy network routing. However, RPL does not accomplish efficiency with respect to power and reliability altogether which are definitely needed in IoT applications. RPL runs on routing metrics and objective function which determines the optimal path in routing. This paper focuses on contributing a comprehensive survey on the improved objective functions proposed by several researchers for RPL. In addition, the paper concentrates on highlighting the strengths and shortcomings of the different approaches in designing the objective function. The approaches built on Fuzzy logic are found to be more efficient and the relevant works related to these are compared. Furthermore, we present the insights drawn from the survey and summarize the challenges which can be effectively utilized for future works.

Increased penetration of distributed energy resources (DERs) creates challenges in formulating the security constrained optimal power flow (SCOPF) problem as the number of models for these resources proliferate. Specifically, the number of devices with different mathematical models is large and their integration into the SCOPF becomes tedious. Henceforth, a process that seamlessly models and integrates such new devices into the SCOPF problem is needed. We propose an object-oriented modeling approach that leads to the autonomous formation of the SCOPF problem. All device models in the system are cast into a universal syntax. We have also introduced a quadratization method which makes the models consisting of linear and quadratic equations, if nonlinear. We refer to this model as the State and Control Quadratized Device Model (SCQDM). The SCQDM includes a number of equations and a number of inequalities expressing the operating limits of the device. The SCOPF problem is then formed in a seamless manner by operating only on the SCQDM device objects. The SCOPF problem, formed this way, is also quadratic (i.e. consists of linear and quadratic equations), and of the same form and syntax as the SCQDM for an individual device. For this reason, we named it security constrained quadratic optimal power flow (SCQOPF). We solve the SCQOPF problem using a sequential linear programming (SLP) algorithm and compare the results with those obtained from the commercial solver Knitro on the IEEE 57 bus system.

The paper is devoted to the development and research of the redundancy cyber resiliency technique based on fast rerouting under security metric with the implementation of the basic schemes for network elements protection, namely node, link, path, and bandwidth. Within the model, the secure fast rerouting task is formulated as an optimization problem of nonlinear programming. The model is configured in order to calculate primary and backup paths that contain links with the minimum values of the probability of compromise that is achieved by using the appropriate weights in the objective function, the value of which is minimized. Numerical research has been conducted, results of which proved the proposed model efficiency and adequacy for the practical application.

Directional Overcurrent Relays (DOCRs) play an essential role in the power system protection to guarantee the reliability, speed of relay operation and avoiding mal-trip in the primary and backup relays when unintentional fault conditions occur in the system. Moreover, the dual setting protection scheme is more efficient protection schemes for offering fast response protection and providing flexibility in the coordination of relay. In this paper, the Adaptive Modified Firefly Algorithm (AMFA) is used to determine the optimal coordination of dual setting DOCRs in the ring distribution system. The AMFA is completed by choosing the minimum value of pickup current (\textbackslashtextbackslashpmbI\textbackslashtextbackslashpmbP) and time dial setting (TDS). On the other hand, dual setting DOCRs protection scheme also proposed for operating in both forward and reverse directions that consisted of individual time current characteristics (TCC) curve for each direction. The previous method is applied to the ring distribution system network of PT. Pupuk Sriwidjaja by considering the fault on each bus. The result illustration that the AMFA within dual setting protection scheme is significantly reaching the optimized coordination and the relay coordination is certain for all simulation scenarios with the minimum operation. The AMFA has been successfully implemented in MATLAB software programming.

Underwater Acoustic Sensor Networks (UASNs) are often deployed in hostile environments, and they face many security threats. Moreover, due to the harsh characteristics of the underwater environment, UASNs are vulnerable to malicious attacks. One of the most dangerous security threats is the eavesdropping attack, where an adversary silently collects the information exchanged between the sensor nodes. Although careful assignment of transmission power levels and optimization of data flow paths help alleviate the extent of eavesdropping attacks, the network lifetime can be negatively affected since routing could be established using sub-optimal paths in terms of energy efficiency. In this work, two optimization models are proposed where the first model minimizes the potential eavesdropping risks in the network while the second model maximizes the network lifetime under a certain level of an eavesdropping risk. The results show that network lifetimes obtained when the eavesdropping risks are minimized significantly shorter than the network lifetimes obtained without considering any eavesdropping risks. Furthermore, as the countermeasures against the eavesdropping risks are relaxed, UASN lifetime is shown to be prolonged, significantly.

The Global Demand for agriculture and dairy products is rising. Demand is expected to double by 2050. This will challenge agriculture markets in a way we have not seen before. For instance, unprecedented demand to increase in dairy farm productivity of already shrinking farms, untethered perpetual access to veterinarians by small dairy farms, economic engines of the developing countries, for animal husbandry and, finally, unprecedented need to increase productivity of veterinarians who're already understaffed, over-stressed, resource constrained to meet the current global dairy demands. The lack of innovative solutions to address the challenge would result in a major obstacle to achieve sustainable food future and a colossal roadblock ending economic disparities. The paper proposes a novel innovative data driven framework cropped by data generated using dairy Sensors and by mathematical formulations using Solvers to generate an exclusive veterinarian daily farms prioritized visit list so as to have a greater coverage of the most needed farms performed in-time and improve small farmers access to veterinarians, a precious and highly shortage & stressed resource.