Data Falsification Attack Research Articles

Dynamic sliding window‐cooperative spectrum sensing against massive SSDF attack in interweave cognitive internet of things

AbstractTo improve spectrum utilization, cognitive radio (CR) enables unauthorized internet of thing (IoT) devices to opportunistically access the channel underutilized by the primary user (PU) in a cognitive IoT (CIoT). To this end, cooperative spectrum sensing (CSS) plays a crucial role in CR technology, which leverages the spatial diversity of corroborative IoT devices to accurately detect the PU signal. However, this open cooperative paradigm may suffer from spectrum sensing data falsification (SSDF) attack in which malicious IoT devices intentionally mislead the fusion center (FC) by providing false sensing results to make an incorrect global decision regarding the PU status. To effectively characterize the attack behavior of malicious IoT devices, we propose a massive SSDF attack model described by the attack cycle and malicious ratio within a sensing period to characterize the malicious behaviors. Additionally, we introduce a delivery evaluation mechanism and propose a dynamic sliding window‐CSS (DSW‐CSS) to mitigate the impact of massive SSDF attack. Moreover, we introduce a sequential reporting mechanism to further reduce the number of samples required by the global decision‐making of the FC. Finally, simulation results show the flexibility and aggressiveness of the proposed attack model and demonstrate the correctness and effectiveness of DSW‐CSS.

Studying the Robustness of Anti-Adversarial Federated Learning Models Detecting Cyberattacks in IoT Spectrum Sensors

Device fingerprinting combined with Machine and Deep Learning (ML/DL) report promising performance when detecting cyberattacks targeting data managed by resource-constrained spectrum sensors. However, the amount of data needed to train models and the privacy concerns of such scenarios limit the applicability of centralized ML/DL-based approaches. Federated learning (FL) addresses these limitations by creating federated and privacy-preserving models. However, FL is vulnerable to malicious participants, and the impact of adversarial attacks on federated models detecting spectrum sensing data falsification (SSDF) attacks on spectrum sensors has not been studied. To address this challenge, the first contribution of this work is the creation of a novel dataset suitable for FL and modeling the behavior (usage of CPU, memory, or file system, among others) of resource-constrained spectrum sensors affected by different SSDF attacks. The second contribution is a pool of experiments analyzing and comparing the robustness of federated models according to i) three families of spectrum sensors, ii) eight SSDF attacks, iii) four scenarios dealing with unsupervised (anomaly detection) and supervised (binary classification) federated models, iv) up to 33% of malicious participants implementing data and model poisoning attacks, and v) four aggregation functions acting as anti-adversarial mechanisms to increase the models robustness.

An attack-resistant target localization in underwater based on consensus fusion

Target localization has been regarded as one of the most important techniques of underwater sensor networks (USNs). However, the challenges posed by weak communication, inhomogeneous medium, and openness of underwater environment make localization more difficult. With consideration of the asynchronous clock, isogradient sound speed and forging attack, this paper presents a consensus fusion-based target localization solution with USNs. The localization procedure is mainly divided into two phases, i.e., internal position estimation and external information fusion. In the first phase, the received signal strength (RSS) and the time of flight (TOF) measurements are jointly employed to develop an asynchronous localization protocol. Particularly, a ray compensation strategy is incorporated to remove the localization bias from assuming the straight-line transmission. Based on this, a consensus fusion based localization estimator is designed in the second phase to mitigate data falsification attacks introduced by malicious sensor nodes. It is worth mentioning that, the proposed localization solution in this paper can fuse the RSS and TOF measurements by the consensus interaction procedure. Furthermore, it is resistant to data falsification attacks in inhomogeneous water medium. Finally, simulation and experimental results reveal that our solution can reduce the influences of data forging attacks and improve the localization accuracy as compared with the other works.

Reinforcement Learning-Based Cooperative Target Tracking Control of Unmanned Surface Vehicles Under Data Falsification Attacks

Reinforcement Learning-Based Cooperative Target Tracking Control of Unmanned Surface Vehicles Under Data Falsification Attacks

CyberSpec: Behavioral Fingerprinting for Intelligent Attacks Detection on Crowdsensing Spectrum Sensors

Integrated sensing and communication (ISAC) is a novel paradigm using crowdsensing spectrum sensors to help with the management of spectrum scarcity. However, well-known vulnerabilities of resource-constrained spectrum sensors and the possibility of being manipulated by users with physical access complicate their protection against spectrum sensing data falsification (SSDF) attacks. Most recent literature suggests using behavioral fingerprinting and Machine/Deep Learning (ML/DL) for improving similar cybersecurity issues. Nevertheless, the applicability of these techniques in resource-constrained devices, the impact of attacks affecting spectrum data integrity, and the performance and scalability of models suitable for heterogeneous sensors types are still open challenges. To improve limitations, this work presents seven SSDF attacks affecting spectrum sensors and introduces CyberSpec, an ML/DL-oriented framework using device behavioral fingerprinting to detect anomalies produced by SSDF attacks affecting resource-constrained spectrum sensors. CyberSpec has been implemented and validated in ElectroSense, a real crowdsensing RF monitoring platform where several configurations of the proposed SSDF attacks have been executed in different sensors. A pool of experiments with different unsupervised ML/DL-based models has demonstrated the suitability of CyberSpec detecting the previous attacks within an acceptable timeframe.

Massive SSDF Attackers Identification in Cognitive Radio Networks by Using Consistent Property

To resist probabilistic spectrum sensing data falsification (SSDF) attacks in cognitive radio networks (CRNs), we in this paper propose a simple and effective attacker identification scheme from the perspective of each secondary user's (SU) historical sensing data without any prior knowledge about the strategy of attackers. In the proposed algorithm, the inconsistency property of the historical data within two consecutive sensing slots during a sensing time window is extracted to characterize different attack behaviors. Further, an optimal identification threshold is obtained for each SU and the analytical expressions of identification performance are also derived. Finally, simulation results along with theoretical analysis show the validity of the proposed scheme to defend against massive probabilistic SSDF attacks with low computation cost.

Privacy-Preserving and Syscall-Based Intrusion Detection System for IoT Spectrum Sensors Affected by Data Falsification Attacks

Crowdsensing platforms collect, process, transmit, and analyze spectrum data worldwide to optimize radio frequency spectrum usage. However, Internet-of-Things (IoT) spectrum sensors, performing some of the previous tasks, are exposed to software manipulation aiming to execute spectrum sensing data falsification (SSDF) attacks to compromise data integrity and spectrum optimization. Novel intrusion detection systems (IDSs) combining device fingerprinting with Machine and Deep Learning (ML/DL) improve the limitation of traditional solutions and remove the necessity of redundant sensors and reputation mechanisms. However, they fail when detecting SSDF attacks accurately while protecting sensors privacy. This work proposes a novel host-based and federated learning-oriented IDS for IoT spectrum sensors that considers unsupervised ML/DL and fingerprints based on system calls. The framework detection performance and consumption of resources are analyzed in local and federated scenarios with six spectrum sensors deployed on Raspberry Pis. The obtained results significantly improve related work when detecting SSDF attacks while protecting sensors privacy, and consuming CPU, memory, and storage of sensors in a reduced manner.

Adaptive Trust Threshold Model Based on Reinforcement Learning in Cooperative Spectrum Sensing

In cognitive radio systems, cooperative spectrum sensing (CSS) can effectively improve the sensing performance of the system. At the same time, it also provides opportunities for malicious users (MUs) to launch spectrum-sensing data falsification (SSDF) attacks. This paper proposes an adaptive trust threshold model based on a reinforcement learning (ATTR) algorithm for ordinary SSDF attacks and intelligent SSDF attacks. By learning the attack strategies of different malicious users, different trust thresholds are set for honest and malicious users collaborating within a network. The simulation results show that our ATTR algorithm can filter out a set of trusted users, eliminate the influence of malicious users, and improve the detection performance of the system.

SpecForce: A Framework to Secure IoT Spectrum Sensors in the Internet of Battlefield Things

The battlefield has evolved into a mobile and dynamic scenario where soldiers and heterogeneous military equipment exchange information in real-time and wirelessly. This fact brings to reality the Internet of Battlefield Things (IoBT). Wireless communications are key enablers for the IoBT, and their management is critical due to the spectrum scarcity and the increasing number of IoBT devices. In this sense, IoBT spectrum sensors are deployed on the battlefield to monitor the frequency spectrum, transmit over unoccupied bands, intercept enemy transmissions, or decode valuable information. However, IoBT spectrum sensors are vulnerable to heterogeneous cyber-attacks, and their accurate detection is an open challenge in the literature. Thus, this paper presents SpecForce, a security framework for IoBT spectrum sensors based on device behavioral fingerprinting and ML/DL techniques. SpecForce considers heterogeneous data sources to detect the most dangerous and recent cyber-attacks affecting IoBT spectrum sensors, such as impersonation, malware, and spectrum sensing data falsification attacks. To evaluate the SpecForce detection performance, it has been deployed on 25 real spectrum sensors, and results show almost perfect detection for the three cyber-attack families previously mentioned.

Cognitive radio: SSDF attack and security

Cooperative spectrum sensing can improve the performance of system detection, but when there are some malicious users in sensors, they will launch spectrum sensing data falsification attack, this is to say they send false sensing result, which will have a great influence on the final decision of fusion center and the primary user. Given that, this paper proposes a basic cooperative spectrum sensing algorithm based on reputation to defend malicious users and then improve that algorithm, advance a new algorithm-reputation weighted cooperative spectrum sensing algorithm.Verified by simulation, our algorithm has achieved the expected effect. The first algorithm can effectively resist attacks especially when the attack probability of malicious users is high. When malicious users are more intelligent, their attack probabilities are different from each other and are uncertain. At this time, the second algorithm can better improve the performance of the final decision of fusion center.

Joint spectrum sensing and D2D communications in Cognitive Radio Networks using clustering and deep learning strategies under SSDF attacks

In Spectrum Sensing and Data Falsification Attacks (SSDF), malicious Secondary Users (SUs) in Cognitive Radio Networks (CRNs) intentionally try to disrupt the global Cooperative Spectrum Sensing (CSS) decision for their self-benefit. Most existing works focus on mitigating the impact of SSDF attacks on CSS decisions. However, a small piece of work jointly studies the CSS and opportunistic data transmission under SSDF attacks in a single framework, but they have some limitations. The present work proposes joint CSS and SU data transmissions in Energy Harvesting-enabled CRNs under SSDF attacks. The present work uses the clustering strategy to isolate the malicious SUs from the honest set of SUs using reputation value and other attributes. The identified malicious and unfit SUs are restricted from opportunistic Device-to-Device (D2D) communication in CRNs. An Ensemble Learning strategy is proposed, which enhances the CSS reliability over the existing works by ∼36.91%, ∼25.00%, and ∼19.04%. Several network constraints guide the reliable SU transmissions in a hybrid framework under SSDF attacks to support the Quality-of-Services for both primary and secondary networks.

TrustSys: Trusted Decision Making Scheme for Collaborative Artificial Intelligence of Things

Many IoT-based applications have inherited the artificial intelligence of things (AIoT) techniques to explore new services and benefits of smart recording and monitoring generated information. However, hundreds of hacking incidents caused by highly sophisticated attackers have generated serious risks, where they compromised various IoT sensors for their benefits, impeding the growth of AIoT. Various security schemes have been proposed in the literature; however, it is critical to determine the legitimacy of AIoT devices in real-time scenarios during the initial deployment of the network. Therefore, this article aims to provide a secure, reliable, and trusted decision-making scheme using multiattribute methods in collaborative AIoT. The proposed system uses backpropagation and Bayesian’s rule to ensure a fast and accurate decision. In addition, agent-based modeling and population-based modeling trust schemes are used to compute the legitimacy of the communicating model. Further, the proposed system is validated over various security measures against the various decision-based conventional methods such as Fuzzy c-means, REPTree, and random tree in terms of time, accuracy, replay attack, data falsification attack, recall, region of convergence, and F-Measure. The proposed mechanism achieves 93% improvement over accuracy and attack identification against existing mechanisms.

Efficient Ordered-Transmission Based Distributed Detection Under Data Falsification Attacks

In distributed detection systems, energy-efficient ordered transmission (EEOT) schemes are able to reduce the number of transmissions required to make a final decision. In this work, we investigate the effect of data falsification attacks on the performance of EEOT-based systems. We derive the probability of error for an EEOT-based system under attack and find an upper bound (UB) on the expected number of transmissions required to make the final decision. Moreover, we tighten this UB by solving an optimization problem via integer programming (IP). We also obtain the FC's optimal threshold which guarantees the optimal detection performance of the EEOT-based system. Numerical and simulation results indicate that it is possible to reduce transmissions while still ensuring the quality of the decision with an appropriately designed threshold.

Security and Energy Management in Cooperative Spectrum Sensing

The revolutionary concept of cognitive radio (CR) has been developed to address the issue of spectrum scarcity in wireless networks. Cooperative spectrum sensing is seen as a promising strategy to dramatically increase spectrum sensing performance in cognitive radio networks, but it is vulnerable to Spectrum sensing data falsification (SSDF) attacks. Existing security references have concentrated on ways to lessen the impact of a SSDF attacking, but these approaches have relied heavily on assumptions such as the attackers being a numerical minority and there being a trustworthy node for data fusion. This realisation inspires us to conduct a thorough examination of the SSDF attack and fusion centre (FC) techniques outside of these constraints. In particular, analysing complex harmful actions allows us to think about a generic SSDF attack model that is more general than the existing methods. According to this overview of attacks, the resultant condition renders the FC clear to attack. Based on these considerations, the ideal attack approach to maximise Bayes risk is examined for both the unknown and known fusion rule scenarios. Second, by casting it as a combinatorial optimisation issue, energy efficiency of cooperative sensing is discussed. In order to enhance the overall energy efficiency of system, a deep learning framework is built based on this formulation by combining graph neural network with reinforcement learning. The efficacy of this method has been demonstrated through simulation tests at varying network scales.

Misbehavior Detection in Vehicular Ad Hoc Networks Based on Privacy-Preserving Federated Learning and Blockchain

As an irreversible trend, connected vehicles have become increasingly more popular. They depend on the generation and sharing of data between vehicles to improve safety and efficiency of the transportation system. Due to the open feature of the vehicular ad hoc network (VANET), it is possible for dishonest and misbehaving vehicles to disrupt traffic by transmitting false information. In recent years, misbehavior detection systems have been developed to detect the malicious behaviour, and machine learning methods have been employed to make the detection more accurately. However, existing misbehavior detection systems typically require a single entity (e.g., a central server) for centralized data collection and training. Model updates are restricted due to data privacy and high overhead of data communication, which reduces the defensive capability of misbehavior detection systems. In this paper, we propose a blockchain-based federated learning scheme to detect misbehavior, which is trained collaboratively by coordinating multiple distributed edge devices while ensuring data security and privacy. In addition, to further protect the privacy of the model on the blockchain, differential privacy with the Gaussian mechanism is leveraged to provide strict privacy protection. Common data falsification attacks are studied in this paper. The experimental results show that our proposed scheme is feasible and effective, and demonstrate that our scheme achieves satisfied accuracy and efficiency.

Defense against SSDF Attack and PUE Attack in CR-Internet of Vehicles (IoVs) for Millimeter Wave Massive MIMO Beamforming Systems

The Internet of Vehicles (IoV) is witnessed to play the leading role in the future of Intelligent Transportation Systems (ITS). Though many works have focused on IoV improvement, there is still a lack of performance due to insufficient spectrum availability, lower data rates, and the involvement of attackers. This paper considers all three issues by developing a novel mmWave-assisted Cognitive Radio based IoV (CR-IoV) model. The integration of CR in IoV resolves the issue of spectrum management, while mmWave technology ensures symmetry in acquiring higher data rates for Secondary Users (SUs). With the proposed mmWave-assisted CR-IoV model, symmetric improvements in network performance were achieved in three main areas: security, beamforming, and routing. Optimum detection mechanisms isolate malicious Secondary Users (SUs) in the overall network. First, Spectrum Sensing Data Falsification (SSDF) attack is detected by a Hybrid Kernel-based Support Vector Machine (HK-SVM), which is the lightweight Machine Learning (ML) technique. Then, the Primary User Emulation (PUE) attack is detected by a hybrid approach, namely the Fang Algorithm-based Time Difference of Arrival (FA-TDoA) method. Further, security is assured by validating the legitimacy of each SU through a Lightweight ID-based Certificate Validation mechanism. To accomplish this, we employed the Four Q-curve asymmetric cryptographic algorithm. Overall, the proposed dual-step security provisioning approach assures that the network is free from attackers. Next, beamforming is performed for legitimate SUs by a 3D-Beamforming algorithm that relies on Array Factor (AF) and Beampattern Function. Finally, routing is enabled by formulating Forwarding Zone (FZ) based on the forwarding angle. In the forwarding zone, optimal forwarders are selected by the Multi-Objective Whale Optimization (MOWO) algorithm. Here, a new potential score is formulated for fitness evaluation. Finally, the proposed mmWave-assisted CR-IoV model is validated through extensive simulations in the ns-3.26 simulation tool. The evaluation shows better performance in terms of throughput, packet delivery ratio, delay, bit error rate, and detection accuracy.

Secure and collision-free multi-platoon control of automated vehicles under data falsification attacks

This paper addresses the secure and safe distributed cooperative control problem of multiple platoons of automated vehicles under unknown data falsification attacks on driving commands. First, a general multi-platoon control framework is developed, which accommodates longitudinal and lateral vehicle dynamics, inter- and intra-platoon information exchanges, falsified driving commands, and unknown external disturbances. To deal with the unknown falsified driving commands on the platoon performance, a neural-network-based adaptive control strategy is developed to compensate their adverse effects. In order to avert both longitudinal and lateral collisions under various maneuvering scenarios, a built-in avoidance mechanism is then designed for each platoon vehicle. Furthermore, a secure and anti-collision multi-platoon control design approach is proposed to ensure the desired inter- and intra-platoon tracking performance with a collision-free guarantee. It is formally proved that the inter- and intra-platoon tracking errors converge to small neighborhood around zero. Finally, several comparative simulation cases are presented to verify the effectiveness and merits of the proposed multi-platoon control approach.

An Online Learning Approach for Cooperator Selection in CSS Under SSDF Attack

Cooperative spectrum sensing (CSS) can greatly improve sensing accuracy by combining the individual sensing results. However, it also faces spectrum sensing data falsification (SSDF) attack which degrades the cooperative sensing performance in practice. To deal with SSDF attack and reduce total energy consumption, it is recommended to select reliable secondary users (SUs) and SUs which have the best detection performance for cooperation. In this letter, an online learning based user selection algorithm is proposed according to the approximate ground truth about the licensed channel state, which addresses the practical problem that different users have different sensing capability and may launch SSDF attacks. In addition, the cooperator’s trust degree is updated in the online learning process. Simulation results indicate that the proposed SUs selection algorithm can quickly converge to a stable state in terms of collision probability and spectrum waste probability under highly unreliable conditions.

Deep Reinforcement Learning based reliable spectrum sensing under SSDF attacks in Cognitive Radio networks

In Spectrum Sensing Data Falsification (SSDF) attacks, malicious Secondary Users (SUs) actively send erroneous local sensing results to the Fusion Centre (FC) that influence the actual outcomes of Cooperative Spectrum Sensing (CSS). Existing trust value-based algorithms are partially successful as SUs can quickly change their characteristics from honest to malicious and vice versa. The present work explores the Deep Reinforcement Learning algorithm (DRL) for adapting the behavioural changes of SUs during CSS to overcome the issue mentioned earlier. The agent in the DRL algorithm wisely avoids the malicious data from the received sensed-energy values at FC and reduces the sensing error effectively. Simulation results show that the proposed work outperforms the conventional and Support Vector Machine-based approaches in CSS reliabilities under SSDF attacks.

Comprehensive Performance Analysis of Soft Data Fusion Schemes under SSDF Attacks in Cognitive Radio Networks

Background: Trust and security are the biggest challenges facing the Cooperative Spectrum Sensing (CSS) process in Cognitive Radio Networks (CRNs). The Spectrum Sensing Data Falsification (SSDF) attack is considered the biggest threat menacing CSS. Methods: This paper investigates the performance of different soft data combining rules such as Maximal Ratio Combining (MRC), Square Law Selection (SLS), Square Law Combining (SLC), and Selection Combining (SC), in the presence of Always Yes and Always No Malicious User (AYMU and ANMU). Results: This comparative study aims to assess the impact of such malicious users on the reliability of various soft data fusion schemes in terms of miss detection and false alarm probabilities. Furthermore, computer simulations are performed to show that the soft data fusion scheme using MRC is the best in the field of soft data computing. Conclusion: ANMU has a slight impact on CSS. Yet, AYMU affects the cooperative detection performance.