We consider a Network Operator (NO) that owns Edge Computing (EC) resources, virtualizes them and lets third party Service Providers (SPs) run their services, using the allocated slice of resources. We focus on one specific resource, i.e., cache space, and on the problem of how to allocate it among several SPs in order to minimize the backhaul traffic. Due to confidentiality guarantees, the NO cannot observe the nature of the traffic of SPs, which is encrypted. Allocation decisions are thus challenging, since they must be taken solely based on observed monitoring information. Another challenge is that not all the traffic is cacheable. We propose a data-driven cache allocation strategy, based on Reinforcement Learning (RL). Unlike most RL applications, in which the decision policy is learned offline on a simulator, we assume no previous knowledge is available to build such a simulator. We thus apply RL in an online fashion, i.e., the model and the policy are learned by directly perturbing and monitoring the actual system. Since perturbations generate spurious traffic, we thus need to limit perturbations. This requires learning to be extremely efficient. To this aim, we devise a strategy that learns an approximation of the cost function, while interacting with the system. We then use such an approximation in a Model-Based RL (MB-RL) to speed up convergence. We prove analytically that our strategy brings cache allocation boundedly close to the optimum and stably remains in such an allocation. We show in simulations that such convergence is obtained within few minutes. We also study its fairness, its sensitivity to several scenario characteristics and compare it with a method from the state-of-the-art.

The commercial adoption of Edge Computingn (EC) will necessitate pricing schemes that align with the economic interests of Network Operator (NO), Service Providers (SPs) and mobile users. Pricing in EC is particularly challenging as it has to take into account resource constraints and contention among different workloads, which can lead to performance degradation and potentially to latency violations. We propose to model pricing subject to latency and resource constraints as an extensive game played between a NO that charges SP based on resource use, SP that charge their users a service specific price, and users that can decide to use the SP's services. We show that equilibria do exist, but computing an equilibrium is NP-hard even under complete information. We then propose an approximation algorithm for computing equilibrium prices for given application placement and we propose a novel, model-assisted Bayesian Optimization (BO) scheme combining probabilistic reparametrization and our approximation scheme. Extensive simulations show that our proposed approach achieves up to an order of magnitude higher revenue compared to state-of-the-art approaches, with a small computational overhead, and highlight the importance of taking into account contention among different workloads.

Eco-friendly freight operations are crucial for decarbonizing the transportation sector. Systematic analysis of policy measures requires a principled modeling approach. While the commonly used model referred to as routing game considers the congestible nature of transportation facilities, exiting models fail to account for environmental factors. This paper aims at providing a mathematical framework to study strategic interaction between owners of mixed fleets comprising of both internal combustion engine vehicle (ICEV) and electric vehicle (EV) trucks. This study introduces a “green” routing game with incomplete information that models strategic interaction among multiple logistic operators. These players face a pollution tax imposed on ICEVs and a potential delayed delivery cost due to EV charging requirements with uncertainty. In contrast to existing models, this novel model captures the players' trade-off between lengthier congestion delay at charging stations as the share of EV trucks increases and higher pollution costs with increased ICEVs usage, with uncertainty determined by a latent state. We first provide equilibrium characterization and present a condition for essential uniqueness. We show that this equilibrium can be computed in a distributed manner using a gradient projection method. We then introduce a public information system that broadcasts real-time information about the latent state. Importantly, we analyze value of information for providing a condition for the public information to be beneficial. Finally, we present numerical examples to illustrate settings where environmental taxation and information dissemination can improve social welfare.

Multi-access edge computing (MEC)-enabled integrated space-air-ground (SAG) networks have drawn much attention recently, as they can provide communication and computing services to wireless devices in areas that lack terrestrial base stations (TBSs). Leveraging the ample bandwidth in the terahertz (THz) spectrum, in this paper, we propose MEC-enabled integrated SAG networks with collaboration among unmanned aerial vehicles (UAVs). We then formulate the problem of minimizing the energy consumption of devices and UAVs in the proposed MEC-enabled integrated SAG networks by optimizing tasks offloading decisions, THz sub-bands assignment, transmit power control, and UAVs deployment. The formulated problem is a mixed-integer nonlinear programming (MILP) problem with a non-convex structure, which is challenging to solve. We thus propose a block coordinate descent (BCD) approach to decompose the problem into four sub-problems: 1) device task offloading decision problem, 2) THz sub-band assignment and power control problem, 3) UAV deployment problem, and 4) UAV task offloading decision problem. We then propose to use a matching game, concave-convex procedure (CCP) method, successive convex approximation (SCA), and block successive upper-bound minimization (BSUM) approaches for solving the individual subproblems. Finally, extensive simulations are performed to demonstrate the effectiveness of our proposed algorithm.

Motivated by the potential of machine-learning- based (ML) algorithms for radio access network (RAN) control and management, we consider the problem of energy-aware O-RAN service orchestration subject to ML inference time constraints. While ML applications enable complex operations in RAN control, guaranteeing service level agreements to close RAN operations in real time is a key requirement to facilitating their wider adoption. In this paper, we focus on orchestrating ML/AI workloads as near-real-time applications in O-RAN Cloud (O-Cloud). We propose PERX, an energy-efficient and performance-aware O-RAN orchestrator that predicts the performance of diverse sets of colocated ML/AL applications by learning a pairwise characterization of application inference times via hierarchical Bayesian learning. We formulate a latency-constrained integer optimization problem for application orchestration and propose an iterative procedure to solve the problem. In line with industry standards, we adopt Kubernetes as the orchestration framework to develop a latency-aware O-Cloud orchestrator. Experimental results reveal up to 50 % increase in profit with guaranteed service level agreements, compared to state of the art benchmarks.

With the emergence of time-critical applications in modern communication networks such as vehicle-to-everything (V2X) systems, there is a growing demand for proactive network adaptation and quality of service (QoS) prediction. However, a fundamental question remains largely unexplored: How can we quantify and achieve more predictable communication systems in terms of performance? To address this gap, this paper introduces a theoretical framework for defining and analyzing predictability in communication systems, with a focus on the impact of observations for performance forecasting. We establish a mathematical definition of predictability based on the total variation distance between the forecast and marginal performance distributions. A system is deemed unpredictable when the forecast distribution, providing the most comprehensive characterization of future states using all accessible information, is indistinguishable from the marginal distribution, which depicts the system's behavior without any observational input. This framework is applied to multi-hop systems under Markovian conditions, with a detailed analysis of Geo/Geo/1/K queuing models in both single-hop and multi-hop scenarios. Additionally, we apply the framework to a random-walk-based model of QoS for connected vehicles experiencing changing channel conditions. We derive exact and approximate expressions for predictability in these systems, as well as upper bounds based on spectral analysis of the underlying Markov chains. Our results have implications for the design of efficient monitoring and prediction mechanisms in future communication networks aiming to provide dependable services.

We consider correlated equilibria in strategic games in an adversarial environment, where an adversary can compromise the public signal used by the players for choosing their strategies, while players aim at detecting a potential attack as soon as possible to avoid loss of utility. We model the interaction between the adversary and the players as a zero-sum game and we derive the maxmin strategies for both the defender and the attacker using the framework of quickest change detection. We define a class of adversarial strategies that achieve the optimal trade-off between attack impact and attack detectability and show that a generalized CUSUM scheme is asymptotically optimal for the detection of the attacks. Our numerical results on the Sioux-Falls benchmark traffic routing game show that the proposed detection scheme can effectively limit the utility loss by a potential adversary. Code - https://github.com/kiarashkaz/Detection-of-Adversarial-Attacks-against-CE

State-of-the-art convolutional neural network models for object detection and image classification are vulnerable to physically realizable adversarial perturbations, such as patch attacks. Existing defenses have focused, implicitly or explicitly, on single-patch attacks, leaving their sensitivity to the number of patches as an open question or rendering them computationally infeasible or inefficient against attacks consisting of multiple patches in the worst cases. In this work, we propose SpaNN, an attack detector whose computational complexity is independent of the expected number of adversarial patches. The key novelty of the proposed detector is that it builds an ensemble of binarized feature maps by applying a set of saliency thresholds to the neural activations of the first convolutional layer of the victim model. It then performs clustering on the ensemble and uses the cluster features as the input to a classifier for attack detection. Contrary to existing detectors, SpaNN does not rely on a fixed saliency threshold for identifying adversarial regions, which makes it robust against white box adversarial attacks. We evaluate SpaNN on four widely used data sets for object detection and classification, and our results show that SpaNN outperforms state-of-the-art defenses by up to 11 and 27 percentage points in the case of object detection and the case of image classification, respectively. Our code is available at https://github.com/gerkbyrd/SpaNN.

We present a moving target defense strategy to reduce the impact of stealthy sensor attacks on feedback systems. The defender periodically and randomly switches between thresholds from a discrete set to increase the uncertainty for the attacker and make stealthy attacks detectable. However, the defender does not know the exact goal of the attacker but only the prior of the possible attacker goals. Here, we model one period with a constant threshold as a Bayesian game and use the Bayesian Nash equilibrium concept to find the distribution for the choice of the threshold in that period, which takes the defender's uncertainty about the attacker into account. To obtain the equilibrium distribution, the defender minimizes its cost consisting of the cost for false alarms and the cost induced by the attack. We present a necessary and sufficient condition for the existence of a moving target defense and formulate a linear program to determine the moving target defense. Furthermore, we present a closed-form solution for the special case when the defender knows the attacker's goals. The results are numerically evaluated on a four-tank process.

Automatic generation control (AGC) is an essential functionality for ensuring the stability of power systems, and its secure operation is thus of utmost importance to power system operators. In this paper, we investigate the vulnerability of AGC to false data injection attacks that could remain undetected by traditional detection methods based on the area control error (ACE) and the recently proposed unknown input observer (UIO). We formulate the problem of computing undetectable attacks as a multi-objective partially observable Markov decision process. We propose a flexible reward function that allows to explore the trade-off between attack impact and detectability, and use the proximal policy optimization (PPO) algorithm for learning efficient attack policies. Through extensive simulations of a 3-area power system, we show that the proposed attacks can drive the frequency beyond critical limits, while remaining undetectable by state-of-the-art algorithms employed for fault and attack detection in AGC. Our results also show that detectors trained using supervised and unsupervised machine learning can both significantly outperform existing detectors.