Single Adversary Research Articles

RAP Vol: Robust Adversary Populations With Volume Diversity Measure.

Deep reinforcement learning (DRL) algorithms have made remarkable achievements in various fields, but they are vulnerable to changes in environment dynamics. This vulnerability easily leads to poor generalization, low performance, and catastrophic failures in unseen environments, which severely hinders the application of DRL in real-world scenarios. The robustness via adversary populations (RAP) algorithm addresses this issue by introducing a population of adversaries that perturb the protagonist. However, the low data utilization efficiency and lack of population diversity greatly limit the generalization performance. This article proposes robust adversary populations with volume diversity measure (RAP Vol) to address these drawbacks. In the proposed joint adversarial training framework, we use the training data to update all adversaries rather than only a single adversary, leading to a higher data utilization efficiency and a fast convergence speed. In the proposed population diversity iterative improvement mechanism, the vectors representing adversaries span a high-dimensional region. The volume of this region is utilized to measure and enhance population diversity via its square. The ablation experiments have verified the effectiveness of our proposed method in improving the robustness against variations in environment dynamics. Also, the influence of various factors (such as adversary population size and diversity weight) on the robustness has been investigated.

FedSIGN: A sign-based federated learning framework with privacy and robustness guarantees

Federated learning enables clients to train a global model jointly without sharing their private local datasets. Despite its benefits, due to the untrustworthiness of clients and the server, traditional federated learning faces the risk of privacy leakage and poisoning attacks. Privacy-preserving methods change the original model parameters whereas robust aggregation algorithms required accurate parameters. To solve such a dilemma, we propose a new framework named FedSIGN. On the one hand, it utilizes the sign of local model update to update the global model to protect privacy and improve efficiency. On the other hand, focus on the Sybil-based poisoning attack: malicious client controlled by a single adversary who directs those clients to launch a poisoning attack, we design a Poisoning Attack Detector to identify malicious clients based on the similarity between sign vectors. Experimental results show that FedSIGN resists privacy and poisoning attacks while maintaining better global model performance. Especially, FedSIGN is not affected by the number of malicious clients and is effective in both the IID and non-IID scenarios.

The development of EASI-based multi-path analysis code for nuclear security system with variability extension

The Physical Protection System (PPS) plays an important role and must effectively deal with various adversary attacks in nuclear security. In specific single adversary path scenarios, we can calculate the PPS effectiveness by EASI (Estimated Adversary Sequence Interruption) through Probability of Interruption (PI) calculation. EASI uses a single value of the probability of detection (PD) and the probability of alarm communications (PC) in the PPS. In this study, we develop a multi-path analysis code based on EASI to evaluate the effectiveness of PPS. Our quantification method for PI considers the variability and uncertainty of PD and PC value by Monte Carlo simulation. We converted the 2-D scheme of the nuclear facility into an Adversary Sequence Diagram (ASD). We used ASD to find the adversary path with the lowest probability of interruption as the most vulnerable paths (MVP). We examined a hypothetical facility (Hypothetical National Nuclear Research Facility - HNNRF) to confirm our code compared with EASI. The results show that implementing the variability extension can estimate the PI value and its associated uncertainty. The multi-path analysis code allows the analyst to make it easier to assess PPS with more extensive facilities with more complex adversary paths. However, the variability of the PD value in each protection element allows a significant decrease in the PI value. The possibility of this decrease needs to be an important concern for PPS designers to determine the PD value correctly or set a higher standard for PPS performance that remains reliable.

Perfectly Secure Message Transmission Against Rational Adversaries

Secure Message Transmission (SMT) is a two-party cryptographic protocol by which the sender can securely and reliably transmit messages to the receiver using multiple channels. An adversary can corrupt a subset of the channels and commit eavesdropping and tampering attacks over the channels. In this work, we introduce a game-theoretic security model for SMT in which adversaries have some preferences for protocol execution. We define rational “timid” adversaries who prefer to violate security requirements but do not prefer the tampering to be detected. First, we consider the basic setting where a single adversary attacks the protocol. We construct perfect SMT protocols against any rational adversary corrupting all but one of the channels. Since minority corruption is required in the traditional setting, our results demonstrate a way of circumventing the cryptographic impossibility results by a game-theoretic approach. Next, we study the setting in which all the channels can be corrupted by multiple adversaries who do not cooperate. Since we cannot hope for any security if a single adversary corrupts all the channels or multiple adversaries cooperate maliciously, the scenario can arise from a game-theoretic model. We also study the scenario in which both malicious and rational adversaries exist.

Secure Consensus of Multiagent Systems With Input Saturation and Distributed Multiple DoS Attacks

The secure consensus problem of multiagent systems with input saturation and denial-of-service (DoS) attacks constraints is an interesting research problem. To solve this problem, the assumptions that all the agents own the same saturation level and suffer the same DoS attacks from a single adversary are usually made. This brief removes the above assumptions and investigates the secure consensus problem of multiagent systems with different saturation levels and multiple DoS attacks. The studied multiagent systems have different layers, in which different saturation levels are studied for different layers. Moreover, the DoS attacks are launched from different adversaries and will cause different effects for different agents. To achieve the desired objective, two different consensus protocols for agents in different levels are first designed using the low-gain technique. Then, the investigated dynamic system under DoS attacks is modeled by a switched system by categorizing the DoS attacks into several types. The controller is proposed by solving parametric algebraic Riccati equation (ARE). Sufficient conditions for the DoS attack duration on each channel are derived.

Federated learning for malware detection in IoT devices

Billions of IoT devices lacking proper security mechanisms have been manufactured and deployed for the last years, and more will come with the development of Beyond 5G technologies. Their vulnerability to malware has motivated the need for efficient techniques to detect infected IoT devices inside networks. With data privacy and integrity becoming a major concern in recent years, increasing with the arrival of 5G and Beyond networks, new technologies such as federated learning and blockchain emerged. They allow training machine learning models with decentralized data while preserving its privacy by design. This work investigates the possibilities enabled by federated learning concerning IoT malware detection and studies security issues inherent to this new learning paradigm. In this context, a framework that uses federated learning to detect malware affecting IoT devices is presented. N-BaIoT, a dataset modeling network traffic of several real IoT devices while affected by malware, has been used to evaluate the proposed framework. Both supervised and unsupervised federated models (multi-layer perceptron and autoencoder) able to detect malware affecting seen and unseen IoT devices of N-BaIoT have been trained and evaluated. Furthermore, their performance has been compared to two traditional approaches. The first one lets each participant locally train a model using only its own data, while the second consists of making the participants share their data with a central entity in charge of training a global model. This comparison has shown that the use of more diverse and large data, as done in the federated and centralized methods, has a considerable positive impact on the model performance. Besides, the federated models, while preserving the participant’s privacy, show similar results as the centralized ones. As an additional contribution and to measure the robustness of the federated approach, an adversarial setup with several malicious participants poisoning the federated model has been considered. The baseline model aggregation averaging step used in most federated learning algorithms appears highly vulnerable to different attacks, even with a single adversary. The performance of other model aggregation functions acting as countermeasures is thus evaluated under the same attack scenarios. These functions provide a significant improvement against malicious participants, but more efforts are still needed to make federated approaches robust.

A Real-Time Game Theoretic Planner for Autonomous Two-Player Drone Racing

In this article, we propose an online 3-D planning algorithm for a drone to race competitively against a single adversary drone. The algorithm computes an approximation of the Nash equilibrium in the joint space of trajectories of the two drones at each time step, and proceeds in a receding horizon fashion. The algorithm uses a novel sensitivity term, within an iterative best response computational scheme, to approximate the amount by which the adversary will yield to the ego drone to avoid a collision. This leads to racing trajectories that are more competitive than without the sensitivity term. We prove that the fixed point of this sensitivity enhanced iterative best response satisfies the first-order optimality conditions of a Nash equilibrium. We present results of a simulation study of races with 2-D and 3-D race courses, showing that our game theoretic planner significantly outperforms a model predictive control (MPC) racing algorithm. We also present results of multiple drone racing experiments on a 3-D track in which drones sense each others’ relative position with onboard vision. The proposed game theoretic planner again outperforms the MPC opponent in these experiments where drones reach speeds up to ${1.25\,}\mathrm{{m}}/\mathrm{{s}}$ .

A Lightweight Intrusion Detection for Sybil Attack Under Mobile RPL in the Internet of Things

The routing protocol for low-power and lossy networks (RPLs) is a standard routing protocol for resource-constrained devices in the Internet of Things (IoT) networks. Primarily, RPL can support a dynamic range of mobility among the nodes in the network, which becomes a great demand now for real-time applications. At the same time, RPL is much vulnerable to various security attacks because of its resource-constrained nature. Such security attacks might cause severe threats and destructive behavior inside the network. In this article, we primarily focus on the Sybil attack, where an attacker claims multiple illegitimate identities either by fabricating or compromising the nodes. Also, in this type of attack, a single adversary is required to control multiple legitimate nodes in the network, and thereby, the adversary node saves the physical resources. In this article, we propose a novel artificial bee colony (ABC)-inspired mobile Sybil attack modeling and lightweight intrusion detection algorithm for the Sybil attack in mobile RPL. Moreover, we considered three different categories of Sybil attack based on its behavior, and we analyze the performance of the RPL under the Sybil attack in terms of packet delivery ratio, control traffic overhead, and energy consumption. Also, we examine the performance of the proposed algorithm in terms of accuracy, sensitivity, and specificity.

Of Gods and Kings: Ashur Imagery in Nahum

Abstract Despite recognizing allusions to Ishtar in Nahum, commentators have not discussed the presence of Ashur imagery in the book. This article contends that Nah 3:18 identifies Yhwh’s ambiguous single male adversary as Assyria’s war god, Ashur. The phrase (3:18) is multivalent, referring to both the “king of Assyria” and “King Ashur.” When the verse is read in light of Assyrian inscriptions, it taunts Ashur that the human kings of Assyria no longer perform the duties assigned to them: shepherding and gathering the god’s people. The prophet, therefore, prophesied more than just Assyria’s demise; he also vilified the religion behind it.

Computing the Team–maxmin Equilibrium in Single–Team Single–Adversary Team Games

Computing the Team–maxmin Equilibrium in Single–Team Single–Adversary Team Games

NFD: Non-technical loss fraud detection in Smart Grid

Utility companies consistently suffer from the harassing of Non-Technical Loss (NTL) frauds globally. In the traditional power grid, electricity theft is the main form of NTL frauds. In Smart Grid, smart meters thwart electricity theft in some ways but cause more problems, e.g., intrusions, hacking, and malicious manipulation. Various detectors have been proposed to detect NTL frauds, but they either rely on user behavior analysis which requires a large amount of historical data or needs a lot of extra devices which are expensive. In this paper, a detector named NFD (NTL Fraud Detection), is proposed to detect NTL frauds with only a small amount of data and one additional device. NFD is based on Lagrange polynomial interpolation to model an adversary's behaviors, and detect a tampered meter by comparing the difference between the results. Different from existing detectors, our detector knows adversaries better than adversaries themselves. By building mathematical models of these adversaries, we can predict their behaviors which they may not be aware of by themselves. NFD makes it practical to detect NTL frauds both online and offline in Smart Grid. NFD can facilitate real-world applications before Smart Grid is fully deployed since it can serve the traditional power grid and Smart Grid simultaneously. Experimental results show the effectiveness of NFD. It can detect multiple tampered meters and multiple adversaries, as well as a single tampered meter and a single adversary. We also study how to tune the parameters used in NFD to further guide its practical usage.

Path-Disruption Games: Bribery and a Probabilistic Model

Path-disruption games, recently introduced by Bachrach and Porat, are coalitional games played on graphs where one or multiple adversaries each seeks to reach a given target vertex from a given source vertex, while a coalition of agents seeks to prevent that from happening by blocking every path from the source to the target for each adversary. These coalitional games model, for instance, security issues in computer networks. Inspired by bribery in voting, we introduce the notion of bribery for path-disruption games. We analyze the question of how hard it is to decide whether the adversaries can bribe some of the agents such that no coalition will form that blocks all paths for them. We show that this problem is NP-complete for a single adversary and complete for Σ2P=NPNP${\Sigma }^{\mathrm {P}}_{2} = \text {NP}^{\text {NP}}$, the second level of the polynomial hierarchy, for the case of multiple adversaries. We also expand the model by allowing uncertainty about the targets: In probabilistic path-disruption games, we assign to each vertex the probability that an adversary wants to reach it, and we study the complexity of problems related to common solution concepts (such as the core and the ź-core) and other properties of such games.

Los Zetas: operational assessment

Today, analysts of the postmodern era recognize that worldwide conflicts are increasingly influenced by the interaction between terrorists, criminals, gangs, and private armies and that this interaction is a threat to the nation state. Now, a related threat is coming into play – one that involves all of these types of groups being represented at once in a single adversary. One such multifaceted group that is in the forefront is Los Zetas, a band of Mexican cartel enforcers that cannot be easily categorized, assessed, or targeted. Within broad categories of a multitude of irregular groups, Los Zetas embodies such capabilities as extensive compartmentalized networking, pervasive intelligence and counterintelligence capabilities, amassing of advanced weaponry, brutal tactics, top level military and police training, and the ability to undermine state governments and control large swaths of territory. Los Zetas, if left unchecked and unexamined, could potentially become a great security problem for Mexico, the US, and Central America. This essay provides an operational assessment that explores Los Zetas using various criteria traditionally used by nation state militaries, and more recently by Terrorism Early Warning Groups, to assess opposing forces (OPFOR). The purpose of this operational assessment is to provide a baseline understanding of Los Zetas that would make them less imposing and more targetable.

Multi-layered Security Solution for Ad hoc Networks

Security has become a primary concern in order to provide protected communication between mobile nodes in a hostile environment. These challenges clearly make a case for building multi-layered, security solutions that achieve both broad protection and desirable network performance. In this paper the fundamental security problem of protecting the multihop network connectivity between mobile nodes in a MANET is focused. The paper identifies the security issues related to this problem, discusses the challenges to security design, and reviews the state-of the-art security proposals that protect the MANET link and network layer. There are a number of attacks that can be used to manipulate the routing in an ad hoc network. In this paper, we describe these threats, and provide solutions to overcome them. To secure the network layer, we include modifications to the existing ARAN protocol and also present a distributed probing scheme to detect malicious nodes for secure packet forwarding. To enhance the link layer security, we present methods to overcome the single adversary attack and colluding adversary attack by minimally modifying the IEEE 802.11 MAC protocol.

On MAC-layer denial of service attacks in IEEE 802.11 ad hoc networks: analysis and counter measures

Network survivability is fundamental to information security. Adversaries could compromise network functionality by attacking the physical layer, the Medium Access Control (MAC) layer or the network layer. Even though security mechanisms for the network layer have been extensively discussed, MAC layer security has not been deeply explored. Leveraging security flaws in the IEEE 802.11 MAC protocol and the salient features of a Mobile Ad Hoc Network (MANET), even an average person with limited knowledge of wireless networks can launch MAC-layer Denial of Service (DOS) attacks. In this paper, we study two types of MAC layer DOS attacks that can be easily employed by an average person: attacks launched from a single adversary by injecting enormous data flows into the network and attacks launched from two colluding adversaries by sending enormous data flows directly to each other. To counter the first attacks, we propose a packet-by-packet authentication scheme so that legitimate nodes can reject data transmission requests from unauthenticated adversaries. To mitigate the second attacks, we propose several counter measures such as a fair MAC protocol, using protecting traffic flows and distance adjustment. The contribution of this paper is that it reveals the attacks that could be launched by an average person, which is a potential threat to MANETs.

Security Analysis of DoS Vulnerability in Stream Authentication Schemes Using Hash Chaining

SUMMARY In this letter, we show that some stream authenticationschemes using hash chaining are highly vulnerable to denial of service(DoS) attacks. An adversary can disrupt all receivers of group by makinguse of modifying a few packets in those schemes. key words: stream authentication schemes, hash chaining, denial of ser-vice(DoS) attack 1. Introduction Hash chaining provides a cost effective means with a real-time streaming application which needs to support data ori-gin authentication and integrity protection in multicast. Toreduce the cost of digital signature including both the com-putation and the communication overhead in some streamauthentication schemes [1]–[4], the source generates anamortizing signature over a set of packets called a block.As a result, a signature can be generated and transmittedonce a block at the beginning or at the end of a block, whichcontains the hash value of the first or the last packet in ablock. In addition, each packet, which contains only a fewhash values of its previous or succeed packets as an authen-tication tag, is transmitted continuously, thus this is calledhash chaining.From the viewpoint of efficiency, previous work usinghash chaining may be appropriate but can be suffered fromthe denial of service attack. In particular, the DoS attackon the receiver side is much serious problem in multicastthan those in the case of unicast since a single adversary candisrupt thousands of receivers at a time. In this letter, weshow that just one or a few packet’s modifications can causethe failed verification of all packets in the block.1.1 NotationsWe use the following notations throughout this letter:

Team-Maxmin Equilibria

In a noncooperative game, a team is a set of players that have identical payoffs. We investigate zero-sum games where a team of several players plays against a single adversary. The team is not regarded as a single player because the team members might not be able to coordinate their actions. In such a game, a certain equilibrium can be selected naturally: theteam-maxmin equilibrium. It assures the team players the best payoff they can hope for, given their inability to coordinate. A team-maxmin equilibrium exists, and in a generic game it is unique.Journal of Economic LiteratureClassification Number: C72.

Hoplites and heresies: a note

A. J. Holladay has effectively reasserted the traditional view of the hoplite phalanx—that it was a dense mass of men, relying on the weight and cohesion of the whole rather than on the prowess of individuals in order to break the enemy's line.Further evidence in his support is provided by Plato'sLaches, where Nicias is made to praise the art of fighting (that is, single combat) in hoplite armour, as a fitting part of a liberal education. But when it comes to its utility in warfare he is less enthusiastic. ‘This science will help somewhat even on the actual battlefield, whenever one has to fight ranged in order with many others. But its chief benefit will be when the ranks are broken, and one has to fight singlehanded against a single adversary, and either, in pursuit, attack someone who is defending himself, or else, in retreat, protect oneself from the attack of another.’ Nicias clearly has in mind a situation like that from which Socrates extracted himself so handsomely at the Battle of Delium, as the speakers in the dialogue have just recalled; though Socrates of course made his retreat without benefit of the newfangled art of fencing.

One Man against the Empire: The Faqir of Ipi and the British in Central Asia on the Eve of and during the Second World War

Among the enemies of the British Empire on the eve of the Second World War the Faqir of Ipi was unique. He was the most determined, implacable single adversary the British Raj in India had to face amongst its own subjects, though he also utterly disregarded the logics of the international situation and unwittingly started a campaign when he had the least chance of attracting international support against the British. As a guerrilla leader he was uncompromising, unyielding, obstinate and unscrupulous in the choice of combat methods against his opponents. These included traditional methods of tribal warfare such as ambush, kidnapping and mutilation. His hatred of the British bore no relation to raison d'etat, though he was usually supported with money and military hardr ware by the Afghan authorities, especially so after the Partition, when he became the symbol of independent Pukhtunistan. The decision to attack was always his own; like the truces which he decided when his casualties had passed the accepted norm and it became necessary for him to retreat once again into the inaccessible hideouts of Waziristan. There he would wait for another opportunity to open hostilities, thus keeping the British army on the North-West Frontier fully mobilized. At one point nearly 40,000 British and Indian troops were reported to be in the field trying to capture him, while he remained elusive as ever, always succeeding

Economic Models of Arms Races

Some problems with the current models of arms races are pointed out, and some extensions and reformulations are proposed. Models of strategic exchange serve as building blocks for a more comprehensive variety that attempt to explain decisions about the division of national wealth into civilian and military components. The models take as given essential technological and preferential parameters and derive implied patterns of choice. The source and implications of discontinuity in the function relating postattack stocks of military wealth and the plausibility and results of successive strikes by a single adversary are examined. Bench marks for purely military preference patterns are established and then are modified to account for the desirability of civilian wealth. Portrayal of preferences via reaction curves is developed, and solutions to the problems of their internal and external inconsistency are proposed. Major problems remain, but economic concepts and principles can contribute at least to the statement of the vital issue of arms races.