Machine learning (ML) anomaly detectors are commonly used to identify cyber-attacks on smart power grids because they can detect new (i.e., zero-day) attacks by classifying deviations from normal patterns as anomalies. Deeplearning-based anomaly detectors offer superior performance but are highly sensitive to the selection of threshold values for defining anomalies. Conversely, traditional (or shallow-based) detectors avoid this threshold sensitivity but often underperform, particularly when dealing with complex interdependent data. Moreover, like all ML models, these detectors are vulnerable to adversarial evasion attacks, where adversaries make small and subtle manipulations to false data to evade detection. To address these issues, we propose a robust hybrid-based anomaly detector that combines the strengths of both deep and shallow-based and is trained using explanations derived from power consumption readings rather than the raw readings themselves. This hybrid approach not only mitigates threshold sensitivity and improves performance but also enhances robustness against white-box evasion attacks. Additionally, we introduce an interpretability method using occlusion sensitivity, which helps explain how a classification decision is made for an input power consumption sample, thereby increasing trust, reliability, and understanding of various attack patterns.

ABSTRACT Smart cities use information and communication technology to promote citizen welfare and economic growth within a sustainable environment. To guarantee that different urban actors, including people, devices, companies, and governments, can communicate efficiently, securely, and reliably, a robust, adaptable network infrastructure is required. However, the increasing complexity of the systems involved poses a challenge to smart city network modeling. Network topology generators produce synthetic networks that can reflect the underlying properties of real‐world networks, providing a practical approach to designing, testing, and implementing complex systems such as smart cities, yet the limited number of network topology generators for smart city applications has long prevented the proper development, investigation, and evaluation of various network configurations. In this article, a novel Smart City Network Topology Graph Generator (SCGG) is proposed to create a pseudorandom topology that mimics real smart city networks. The main goal of SCGG is to generate a network topology for smart cities that captures the interconnectivity of several communication technologies, such as wireless sensor networks (WSN), Internet of Things (IoT), and cellular networks. The SCGG system is characterized by the number of clusters, the average number of nodes, the number of layers, and the node density. The general network architecture and path‐related variables of the generated topologies are evaluated based on different graph theory measures, focusing on both global graph‐level characteristics and local node‐level features. The experimental results, demonstrating high natural connectivity and a low spectral radius value, offer a reliable tool for optimizing and strengthening the behavior and performance of smart city networks under different conditions to improve their robustness, minimize the probability of disruptions or failures, and enhance overall efficiency to ensure a resilient network.

Existing literature confirms the ability of machine learning to identify fraudulent smart grid power consumers who report false consumption readings to pay less electricity bills. Additionally, federated learning (FL) shows promise as a way to train the detection model without requiring data sharing, thereby safeguarding consumer privacy. However, malicious participants (i.e., clients) in FL training can launch adversarial attacks by training their local models with specially crafted low-consumption data to inject a Trojan into the global model. This Trojan can then be activated during the evaluation phase to evade the detection of false data. To the best of our knowledge, not enough research has been done on this topic in the context of unsupervised learning. The absence of labels in unsupervised learning exacerbates the effectiveness of Trojan attacks and renders it more challenging to design robust defense mechanisms. In this article, we first investigate the vulnerability of one-class classifiers to Trojan attacks. Then, we propose two defense approaches named layerwise close-to-median (LWCM) and Machine Unlearning to counter this attack. In LWCM, by choosing a FL client whose last layer model parameters are near to the median of all clients’ last layer parameters to update the global model, we can identify and exclude malicious updates. The idea is that the last layer parameters of honest clients should be similar, whereas those from malicious clients are different. With the majority of clients being honest, the median values are closer to the parameters of these clients, facilitating the detection of malicious clients. In Machine Unlearning, we utilize gradient ascent-based techniques to adapt models by selectively removing attacker-related data points. This is possible because honest clients generate data resembling that of malicious clients and employ a dual-component loss function to maintain model proficiency in recognizing benign power consumption patterns while eliminating malicious patterns. To show the seriousness of Trojan attacks and the effectiveness of our countermeasures, many experiments have been carried out.

In smart grids, smart meters (SMs) transmit power consumption data to utilities for billing and energy management. However, compromised SMs can report low consumption to reduce electricity bills. Deep reinforcement learning (DRL) detectors have recently been proposed to detect these attacks due to their adaptability to new attacks and changes in power consumption patterns. This article explores backdoor attacks targeting DRL detectors during training, aiming to introduce a vulnerability in the detector. These attacks make the detector misclassify false low-consumption data when trigger samples are used while maintaining normal classification accuracy otherwise. We propose a DRL-based attack model that generates stealthy and unique trigger samples using cosine similarity. Our evaluations show the attack is initially highly successful, but its success diminishes with honest data used for incremental training of the detector. To sustain high success rates, attackers must influence incremental training. We also propose defenses, including data filtration during the preparation stage, adversarial training for the defense model during the training stage, and a combined approach, with experiments validating their effectiveness.

Federated Learning (FL) enables decentralized training of machine learning (ML) models, making it a valuable approach for detecting false data in smart power grids (SGs) to enhance grid stability while protecting consumers privacy. However, FL-based ML models remain vulnerable to adversarial attacks during both training and inference phases, which can compromise data security. To address these vulnerabilities, we first investigate the robustness of a novel FL-based false data detection approach using Explainable Artificial Intelligence (XAI), referred to as XAI-based FL detection. This approach utilizes explanations of consumers power consumption data, rather than raw data, during the training process. We assess the robustness of the XAI-based FL detection compared to traditional data-driven FL detection against two types of adversarial attacks: Gradient Inversion attacks in the training phase, where adversaries reconstruct private data from shared gradients, and Evasion attacks in the inference phase, where adversaries subtly modify input data to deceive the detection model. Then, we propose a secure XAI-based FL detector with adversarial training to defend against both attack types. The key idea is that XAI helps mask model gradients during training because XAI-generated explanations remain nearly identical across different samples. Therefore, attackers struggle to accurately reconstruct the original training data, even if they obtain precise explanations using gradient inversion attacks. Additionally, XAI effectively distinguishes between benign and malicious samples. When combined with adversarial training, XAI strengthens model robustness against evasion attacks without compromising accuracy, effectively resolving the trade-off between security and performance. Our proposed detector reduced the success rate of evasion attacks from 94.99% to 29.11 explanations, and further to 0% with adding adversarial training. It also increased the mean square error for gradient inversion attacks from 0.01 to 2.60 in the most severe attack scenarios, making such attacks ineffective.

In the advanced metering infrastructure (AMI) of the smart grid, smart meters (SMs) are deployed to collect fine-grained electricity consumption data, enabling billing, load monitoring, and efficient energy management. However, some consumers engage in fraudulent behavior by hacking their meters, leading to either traditional electricity theft or more sophisticated evasion attacks. Evasion attacks aim to illegally reduce electricity bills while deceiving theft detection mechanisms. The current methods for identifying such attacks raise privacy concerns due to the need for access to consumers' detailed consumption data to train detection mechanisms. To address privacy concerns, federated learning (FL) is proposed as a collaborative training approach across multiple consumers. Adversarial training (AT) has shown promise in countering evasion threats on machine learning models. This paper, first, investigates the susceptibility of traditional electricity theft classifiers trained by FL to evasion attacks for both independent and identically distributed (IID) and Non-IID consumption data. Then, it investigates the effectiveness of AT in securing the global electricity theft detector against evasion attacks, assuming no misbehavior from the participant consumers in the FL process. After that, we introduce a novel attack, called Distillation, which can be launched during the AT process to make the global model susceptible to evasion at inference time. Finally, extensive experiments are conducted to validate the severity of the proposed attack.

In smart power grids, certain customers are motivated by financial gains to manipulate electricity consumption data, aiming to reduce their bills. Despite the development of machine learning-based detectors, these systems remain vulnerable to evasion attacks. This paper investigates the susceptibility of deep reinforcement learning (DRL)-based detectors to evasion attacks. We propose an evasion attack model that employs the double deep Q learning (DDQN) algorithm for a black-box attack scenario. Our model generates adversarial evasion samples by altering malicious consumption data, tricking detectors into classifying them as benign. Leveraging the unique attributes of reinforcement learning (RL), our model determines optimal actions for manipulating malicious data. For comparative analysis, we compare our DRL-based model with an FGSM-based attack model. Our experiments consistently demonstrate the effectiveness of our DRL-based attack model, achieving an impressive attack success rate (ASR) ranging from 92.92% to 99.96%, outperforming the FGSM-based attack model.

In smart power grids, electricity theft causes huge economic losses to electrical utility companies. Machine learning (ML), especially deep neural network (DNN) models hold state-of-the-art performance in detecting electricity theft cyberattacks. However, DNN models are vulnerable to adversarial attacks, i.e., evasion attacks. In this work, we study the vulnerability of the DNN-based electricity theft detectors against evasion attacks and the influence of the model's regularization (generalization) on robustness. We cluster the power consumers and train a detector for each cluster, and compare the performance and robustness of this detector to a global detector that is trained on all the consumers, data. The results indicate that the cluster-based detector is not only more robust against evasion attacks but also enhances normal classification accuracy because its training data has more consumption pattern similarity compared to the training data of the global detector which requires higher level of regularization. Moreover, unlike the existing solutions that sacrifice the normal accuracy of the model to improve the robustness against evasion attacks, the proposed cluster-based detector holds state-of-the-art performance in both robustness and accuracy.

In the realm of smart grids, smart meters can be hacked to report false data to lower the consumers’ electricity bills. While machine learning (ML) techniques have shown promise in detecting false data, they are also prone to adversarial attacks, such as evasion attacks. This article investigates the impact of gradient-ensemble-based evasion attacks on the smart grid ML-based false data detectors, focusing on the white-box threat model where attackers possess detailed knowledge of the defense mechanism. First, we examines the vulnerability of three detectors (consumer-based, cluster-based, and global) to gradient-based evasion attacks. The evaluation results show an inverse relationship between robustness of the detectors and regularization (i.e., generalization), where higher data set variability usually causes higher regularization. Notably, minimal regularization level is observed when electricity consumption patterns are close. Our findings also indicate that the consumer-based detector exhibits higher accuracy and robustness but remains susceptible to zero day attacks and demands substantial computational resources for training an ML model for each consumer. In contrast, the cluster-based detector improves accuracy and exhibits satisfactory robustness compared to the global detector. Subsequently, we proposes two parallel-ensemble approaches (stacking and voting) for the cluster-based false data detectors trained on the adversarial samples. The evaluation results demonstrate that integrating clustering, adversarial training, and ensemble methods, the proposed detector enhances robustness against gradient-ensemble-based evasion attacks while significantly boosting accuracy. This stands in contrast to benchmark defenses, which often face a tradeoff between accuracy and robustness, sacrificing accuracy to bolster resilience against evasion attacks.

Vehicular communication networks are increasingly being considered as a means to conserve fuel and reduce emissions within transportation systems. This paper focuses on using traffic light signals to communicate with approaching vehicles. The communication can be traffic-light-signal-to-vehicle (TLS2V) and vehicle-to-vehicle (V2V). Based on the information sent, the vehicle receiving the message adapts its speed to a recommended speed (SR), which helps the vehicle reduce fuel consumption and emissions. The key contribution of this paper is the proposal of a comprehensive optimization model that involves V2V and TLS2V communications. The objective function is to minimize fuel consumption by and emissions from vehicles. The speed that can achieve this goal is the optimum SR(SR*). We also propose efficient heuristic expressions to compute the optimum or near-optimum value of SR.

Abstract With recent advances in the development of wireless communication networks, wireless mesh networks (WMNs) have been receiving considerable research interests in recent years. The need to support integrated services and ensure quality of service (QoS) satisfaction for various applications is one of the fundamental challenges for successful WMN deployment. In order to provide differentiated services, medium access control (MAC) should have priority management at the link layer. In code division multiple access (CDMA)‐based WMNs, the interference phenomenon and simultaneous transmissions must be considered. We propose two priority schemes for MAC in a distributed CDMA‐based WMN, taking into account interference, multimedia services, QoS requirements, and simultaneous transmissions. The first priority scheme is within a node. Each node has an independent queue for each traffic class. According to QoS requirements, the queue that should be served first is determined. The second priority scheme is among neighbor nodes. It is proposed for multiple simultaneous transmissions in the CDMA network. This scheme gives a larger chance of correct transmission to high priority traffic than low priority traffic. In addition, we propose to use adaptive spreading gain and a frame structure to achieve high resource utilization. Simulation results demonstrate that the proposed schemes can achieve effective QoS guarantee. Copyright © 2009 John Wiley & Sons, Ltd.

The need to support integrated services and provide quality of service (QoS) for various applications is one of the fundamental challenges for successful wireless mesh network (WMN) deployment. In order to provide differentiated services, medium access control (MAC) should have priority management at the link layer. In code division multiple access (CDMA) based WMNs, the interference phenomenon and simultaneous transmissions must be considered. We propose two priority schemes for MAC in a distributed CDMA-based WMN, taking into account interference, multimedia services, QoS requirements, and simultaneous transmissions. In addition, we propose to use an adaptive spreading gain and a frame structure to achieve high resource utilization. Simulation results demonstrate that the proposed schemes can achieve effective QoS guarantee.