Large Language Models (LLMs) have revolutionized numerous domains by demonstrating exceptional reasoning, problem-solving, and response-generation capabilities. Despite their impressive performance, concerns persist regarding their alignment with human values and the potential for unintended behaviors when optimizing objectives. Traditional alignment methods like Reinforcement Learning from Human Feedback (RLHF) face computational and stability challenges, necessitating alternative solutions. Inspired by cognitive science, this paper introduces a Declarative Memory Module (DMM) to guide LLM decision-making. It comprises Semantic Memory, which provides ethical principles and strategies, and Episodic Memory, which recalls past experiences to reinforce or discourage actions. By explicitly integrating these elements, our approach enhances AI alignment with predefined human preferences. Experiments in game-theoretic contexts using multiple LLM models demonstrate that DMM significantly improves their alignment with human preferences, offering a promising step toward more controllable and trustworthy AI systems.

Traditional methods for streaming video content have typically been optimized for stable network connections between servers and users, but they often struggle to perform effectively in mobile environments. The fluctuating nature of wireless links poses a challenge, requiring careful consideration to ensure acceptable quality of video streaming services. Leveraging cloud infrastructure introduces additional complexities, such as increased delays resulting from the distance between the infrastructure and the user. To address these challenges, adaptive transmission techniques are crucial for minimizing packet loss and delays, thereby enhancing the overall quality of video streaming services. In this paper, we propose a solution aimed at maintaining user experience at a predefined level as per service level agreements (SLAs). This solution involves real-time detection of quality of experience (QoE) degradation and identification of potential causes based on selected criteria, thus providing adaptive video streaming service based on QoE estimation.

Recent advances in blockchain technology are enabling a data-centric revolution in the ICT ecosystem. With the intrinsic need to enforce transparency and distributed storage of the smart contracts in blockchain architectures, a number of challenges and hurdles are often raised about the use of blockchain in contexts where privacy, confidentiality, and secrecy are a concern. This talk discusses recent researches aiming at introducing homomorphic encryption as a solution to cope with this issue based on their ability of encrypted data processing. Such approaches extend the use of blockchain to modern distributed applications where data is collected, processed, and transmitted by disparate devices across multiple physical locations. The contribution brought by homomorphic encryption to blockchain-enabled applications is highlighted and the potential evolution and the underlying open issues are also explored.

The rapid technological advancements in wireless communications, ubiquitous sensing and mobile networking have paved the way for the emergence of the Internet of Things (IoT) era, where “anything” can be connected “anywhere” at “anytime”. However, the flourish of IoT still faces various security and privacy preserving challenges that need to be addressed. In such pervasive and heterogeneous environment where the context conditions dynamically and frequently change, efficient and context-aware mechanisms are required to meet the users' changing needs. Therefore, it seems crucial to design an adaptive access control scheme in order to remotely control smart things while considering the dynamic context changes. In this paper, we propose a Context-Aware Attribute-Based Access Control (CAABAC) approach that incorporates the contextual information with the Ciphertext-Policy Attribute-based Encryption (CP-ABE) to ensure data security and provide an adaptive contextual privacy. From a security perspective, the proposed scheme satisfies the security requirements such as confidentiality, context-aware privacy, and resilience against key escrow problem. Performance analysis proves the efficiency and the effectiveness of the proposed scheme compared to benchmark schemes in terms of storage, communication and computational cost.

Blackboxes are being increasingly used in the vehicular context to store and transmit information related to safety, security and many other applications. The plethora of sensors available at the different parts of the vehicle can provide enriched gathering of the data related to these applications. Nonetheless, to support multiple use cases, the blackbox must be accessible by various actors (e.g. vehicle owner, insurance company, law enforcement authorities). This raises significant challenges regarding the privacy of the data collected and stored in the blackbox. In fact, these data can often lead to tracing back accurate facts about the behaviour of the owner of the vehicle. To cope with this problem, we propose a new blackbox architecture supporting searchable encryption. This feature allows multiple users who are not able to decipher the content of the blackbox to validate properties such as path traceback and velocity. To illustrate the implementation of the proposed technique in practice, we discuss a case study related to post-accident processing by insurance companies.

Over the last few years, the rapid development of smart grids and the deployment of smart metering infrastructures on a much larger scale have raised a crucial privacy concern related to the collection of real-time energy consumption data. For these purpose, data aggregation is a widely used technique for privacy preservation. Securing the data aggregation process is a challenge since smart grids are susceptible to many cyber security threats and attacks. Therefore, several approaches have been proposed to address this issue. However, many of them are vulnerable to various types of attacks and suffer from high communication and computation costs in addition to the use of complex key management schemes. To overcome this problem, we propose a secure and privacy-preserving data aggregation scheme based on homomorphic encryption. In this solution, we take advantage of the opportunities offered by ECEL-Gamal and its homomorphic property as well as the ECDSA and ECDHE algorithms to use them in order to aggregate privacy-preserving data. Through detailed analyses, we show that our proposed scheme can meet various security requirements and preserves user privacy. In addition, experiment results demonstrate that our scheme is more efficient than existing competing approaches in terms of communication and computation overhead.

Various research efforts have focused on the problem of customer privacy protection in the smart grid arising from the large deployment of smart energy meters. In fact, the deployed smart meters distribute accurate profiles of home energy use, which can reflect the consumers' behaviour. This paper proposes a privacy-preserving lattice-based homomorphic aggregation scheme. In this approach, the smart household appliances perform the data aggregation while the smart meter works as relay node. Its role is to authenticate the exchanged messages between the home area network appliances and the related gateway. Security analysis show that our scheme guarantees consumer privacy and messages confidentiality and integrity in addition to its robustness against several attacks. Experimental results demonstrate the efficiency of our proposed approach in terms of communication complexity.

Using Internet of Things (IoT) in the health domain is one of the most promising approaches which offer a ubiquitous healthcare where sensors are used, in real time, for constant monitoring of patient's symptoms and needs wherever he is. Wireless body area network (WBAN) is a highly suitable communication tool for the medical IoT devices. However, the conception of WBAN applications is still a challenging job that should take into consideration many technical requirements such as network lifetime, security level, network throughput and data criticality and prioritization. As a consequence, a trade-off between security effectiveness, energy efficiency, and QoS requirements can be perceived as a major performance objective. In this paper, we propose a stochastic game to balance the tradeoff between network performance and security level while taking into account the context dynamics. Simulation results show that the proposed approach can achieve an acceptable security level and is more efficient than benchmark algorithms in terms of network lifetime and throughput.

Internet of Things (IoT) based pervasive healthcare systems have revolutionized the healthcare industry while enabling remote patient monitoring to improve patient care delivery and provide a highly reliable ubiquitous healthcare monitoring. In this context, Virtual Private Network (VPN) is a promising network layer technology that is used for a secure, reliable and remote access to patient health information. It is an overlay network that creates secure and dynamic tunnels between various devices across a public network such as the Internet. However, it is vulnerable to several attacks such as spoofing, snipping, and hacking as the data is transmitted through the Internet connection. In addition, due to the IoT device's energy constraints and the application requirements, the main purpose is to adaptively select the most appropriate encryption and authentication algorithms to secure the communication tunnel between these resource constrained devices and the remote user. Therefore, adaptive risk-aware secure tunnel negotiation is perceived as a major performance objective. In this paper, we propose a Stackelberg game for the security requirements negotiation between the communication peers in order to ensure a trade-off between security effectiveness and network performance while considering the dynamic context changes. Simulation results prove that the proposed approach can reduce the cost of security policy implementation in terms of performance degradation. For instance, the latency is improved by around 25% compared to static security policy.

With the pervasiveness of the Internet of Things (IoT) and the rapid progress of wireless communications, Wireless Body Area Networks (WBANs) have attracted significant interest from the research community in recent years. As a promising networking paradigm, it is adopted to improve the healthcare services and create a highly reliable ubiquitous healthcare system. However, the flourish of WBANs still faces many challenges related to security and privacy preserving. In such pervasive environment where the context conditions dynamically and frequently change, context-aware solutions are needed to satisfy the users' changing needs. Therefore, it is essential to design an adaptive access control scheme that can simultaneously authorize and authenticate users while considering the dynamic context changes. In this paper, we propose a context-aware access control and anonymous authentication approach based on a secure and efficient Hybrid Certificateless Signcryption (H-CLSC) scheme. The proposed scheme combines the merits of Ciphertext-Policy Attribute-Based Signcryption (CP-ABSC) and Identity-Based Broadcast Signcryption (IBBSC) in order to satisfy the security requirements and provide an adaptive contextual privacy. From a security perspective, it achieves confidentiality, integrity, anonymity, context-aware privacy, public verifiability, and ciphertext authenticity. Moreover, the key escrow and public key certificate problems are solved through this mechanism. Performance analysis demonstrates the efficiency and the effectiveness of the proposed scheme compared to benchmark schemes in terms of functional security, storage, communication and computational cost.

With the expansion of the Internet of Things (IoT) era and the tremendous progress of wireless communications, Wireless Body Area Networks (WBANs) have been introduced as a promising technology for remote health monitoring. However, the heterogeneity of IoT technologies, the open nature of wireless networks, and the existence of resource-constrained devices are the main challenges to design burdensome security protocols for WBANs. In this context, scalability, reliability, interoperability, and security requirements need risk-aware and adaptive security solutions that take into account the dynamic context changes. Therefore, a trade-off between risk mitigation actions and network performance can be perceived as a major performance objective. The main purpose of this work is to propose an adaptive risk management framework that considers the WBAN context changes to dynamically select the most appropriate security countermeasure and assess its implementation cost in terms of performance degradation. Furthermore, a case study is discussed to show the impact of adaptive risk assessment on network performance.

Wireless body area networks (WBANs) are perceived as an emerging key technology for the next generation ubiquitous healthcare systems. However, the openness and mobility of wireless sensor technologies make the sensor-controller communication vulnerable to be eavesdropped and linked to the sensors in transmission of patient's information. Furthermore, in such resource-constrained environment, authenticating sensor nodes anonymously with the controller node while considering their limited capabilities is a paramount security requirement. In this paper, we propose a lightweight and adaptive anonymous authentication and key agreement scheme for the two-tier WBAN. The proposed protocol enables an anonymous mutual authentication and a session key establishment between the controller node and the body sensor nodes while taking into account the dynamic context changes. From a security perspective, we demonstrate that the proposed key agreement scheme achieves the desired security properties such as anonymity, unlinkability, perfect forward secrecy, etc. Performance analysis proves that the proposed protocol outperforms benchmark schemes in terms of communication and computational overhead.

Wireless Body Area Networks (WBANs), are becoming more and more popular. They are taking a significant proportion of the E-health market. WBANs rely on medical sensors, such as electrocardiography (ECG) and blood glucose, to continuously monitor the patient's physiological parameters. These signals are then forwarded, via a network infrastructure, to an analysis center. Despite their large deployment, security issues, mainly related to data privacy, are often considered as potential obstacles that might limit the extent of such solutions. This paper introduces a new approach to cope with data privacy in WBANs. Our major contribution is two-fold: (1) to reduce a noise and obtain correct data by improving Kalman filter; and (2) to propose a new secure privacy-preserving model based on homomorphic properties and chaotic cryptosystems. The proposed model conserves a real-time analytics for monitoring healthcare and preserves data confidentiality, data privacy, and data integrity.

With the proliferation of sophisticated attacks against communication and information systems, existing protection mechanisms are becoming inefficient to guarantee an acceptable security level. Hence, new security techniques are being developed to keep up with the evolution of threats and exploits. One of the security issues that is being investigated by the research community is per-packet authentication, where security mechanisms should be implemented to enforce the authentication of a specific network flow. This paper mainly focuses on the use of biometric techniques since they offer a promising potential to enforce per-packet authentication policies in highly dynamic contexts. We provide a review of how biometrics have been proposed in the literature as an alternative to purely cryptographic mechanisms for per-packet authentication. The drawbacks of the existing are pointed out and discussed. We finally explore the research niches that will shape the evolution of this field in the near future.

The main goal of the Internet of Things (IoT) in supply chains is to provide accurate and real-time tracking of items. But, in modern business management, new perspectives are opened and new tasks are invented. In this network involving a myriad of businesses and relationships, the supply chain management (SCM) process should assure global interoperability between trading entities and partners. The traditional technologies are still unable to offer a cost efficient and trusted system for information sharing and provenance recording between these entities and partners having different clearance and priorities. Therefore, security and scalability become an important consideration for the design of supply chain architecture (SCA). In 1979, Ralph Merkle has proposed the Merkle tree which is a key management scheme based on tree architecture. In this paper, we present a secured distributed supply chain architecture based on Merkle tree. The proposed design aims at replacing the current Peer to Peer architecture using a scalable and secure key management scheme.

Cloud Computing is an environment offering a wide range of on-demand services. It allows the Cloud users to outsource their data offering them resources to store their data and documents, process their applications, etc. However, although these resources seem to be infinite, they are in reality limited and the data centers have a limited capacity to host data and applications. Therefore, Cloud users are always competing trying to have all of the necessary resources for applications processing and data storage. So, it is compulsory that Cloud providers find a mechanism allowing them to share in a transparent way the various resources between users. In this context, our research work consists in studying and proposing approaches for resources allocation in a Cloud Computing environment where various users share the resources of several Cloud providers while respecting QoS and security criteria defined via SLAs. The work undertaken in this paper handle mainly two points which are guaranteeing resources availability using game-theoretic approaches and providing fair and secure resource allocation between the various users of Cloud using DRF mechanism.

Nowadays, the evolving threats of terrorist and criminal attacks against public places have urged the need for active video surveillance systems. Such systems rely on video analytics modules that identify and pre-alarm abnormal events occurring in sensed visual scenes in real-time. By implementing these smart features in the digital cameras, the design of distributed and adaptive physical security policies is made possible. Yet, within these systems, data security and privacy are a major concern. In this paper, we introduce a video surveillance system whose architecture is based on the IoT model. The device layer of this architecture includes smart cameras that communicate in machine-to-machine (M2M) manner. Then, we propose a tunneling framework that ensures the security of the M2M communications. The novelty of this framework is that it uses a shared secret in order to generate a cipher key and the video sensor fingerprints as the main authentication parameter. The security evaluation of our framework has also shown its resistance to well known attacks.

The emergence of the Social Internet of Things (SIoT) paradigm brought new opportunities in terms of service provisioning and collaboration in distributed Internet of Things environments. However, while augmenting smart devices with more autonomy, this trend deepened the privacy hole of the Internet of Things (IoT). As a mitigation, we propose in this paper a SIoT profile matching framework allowing collaboration between devices interested in the same services. The services to be included in the SIoT profile are selected by the owner of the IoT devices. The profile is then published in a privacy preserving fashion. Since trust is critical in social networks, we propose a trust model fitting with the distributed nature of IoT interactions. This trust model is based on the decentralized trustless Blockchain concept. Security analysis and performance evaluation of our scheme showed a promising tradeoff between security and efficiency.

Most of the traditional approaches for gathering data in the information sharing platforms of manufacturing supply chains are outdated and can negatively impact the speed of real-time sharing of information. In this technology-driven world, the supply chain management operations based on the Internet of Things (IoT) are bringing day after day a great business potential. Therefore, connecting physical objects together in support of intelligent decision making, consists an enabler of real-time quality management and control in the supply chain. The connected objects should meet market demands correctly, rapidly, and profitably. The supply chains should maintain and sustain technology-based and quality-driven capabilities in order to decrease network wide costs, reduce lead duration and transit duration and improve user experience. This work tackles the implications of the roll-out of the internet of things in supply chain management tasks. We focus on the quality of experience and the quality assurance arising from data consumption as a result of connecting objects to the Internet and how this can be translated into traces and evidences for some selected properties. The proposed model will be illustrated with a test scenario to verify the selected requirements.

Cloud computing is an emerging economic model that provides a broad network access to services with many benefits to many tenants at the same time. Although it creates a large potential to develop new online services the evolution of cloud computing has been accompanied by the proliferation of various attacks against the services on the cloud infrastructures. The importance of risk management in cloud computing is mainly motivated by the dynamic context in which the services and applications are implemented. In fact, the intrinsic characteristics of cloud infrastructures prevent the use of previously developed process of traditional frameworks. The goal of this paper is to present a new risk management framework for better considering context parameters of cloud computing environment. An example is considered in which the risk treatment process in this framework proposes how to dynamically balance between the security and performance degradation due to the implementation of security mechanism.

Supply-chain and inventory management are the main areas in the internet of things (IoT) requiring low cost and secure batch mode privacy and authentication of radio frequency identification (RFID) tags. Some additional security and quality of service requirements are also necessary for a secure RFID based supply chain application. In this paper, our contribution is twofold: first, we propose a new lightweight homomorphic signature scheme that supports the aggregation. Second, based on it, we design a new efficient architecture for Supply Chain Management (SCM). Indeed, by introducing an aggregate function for reader-to-server communications, our SCM architecture significantly reduces the communication costs. Our protocol also captures more security properties such as privacy and authentication. Finally, we show that our protocol is compliant with the Gen2 version2 standard for low-cost RFID infrastructures.

Cloud computing has the benefit of offering scalability and efficiency, as well as cost-effectiveness. However, the existence of security breaches exacerbates the reluctance of potential users to host their sensitive data and services on the cloud. Indeed, the intrinsic characteristics of cloud infrastructures prevent the use of traditional security policy engineering frameworks. The dynamic context in which services and applications are implemented on the cloud does not support previously developed security risk management process. This paper proposes a novel game-theoretic model that models the trade-off between the effectiveness of the adaptive risk treatment and the cost resulting from the execution of the selected security mechanisms. The objective is to implement dynamic security policies that adapt to the dynamic nature of the cloud.

Radio frequency identification (RFID) technology is paramount for the automation and optimization of supply-chain activities. RFID tags can drastically improve the efficiency of inventory management, especially in the pharmacy supply chain. Thus, the drugs tracking can be enhanced in the pharmaceutical manufacturing plants all across the world paving the way for the wide deployment of the Internet of Things (IoT). The connected things in manufacturing processes have gained prominence owing to the growing number of unqualified pharmacies and counterfeit drugs. These sensors continuously generate a large amount of contextual information like locations and time points so as to control and manage drugs quality and distribution. Based on these contextual information, business logic can be implemented efficiently. Collection, modeling, reasoning, and distribution of context in relation to sensor data plays a critical role. In this paper, we propose a context-aware middleware for RFID based pharmacy supply chain, which aims to offer a deeper intelligence for the objects monitoring. A variant of Fosstrack middleware was proposed to offer a deeper intelligence in the RFID based supply chain.

The evolution of cloud gaming systems is substantially the security requirements for computer games. Although online game development often utilizes artificial intelligence and human computer interaction, game developers and providers often do not pay much attention to security techniques. In cloud gaming, location-based games are augmented reality games which take the original principals of the game and applies them to the real world. In other terms, it uses the real world to impact the game experience. Because the execution of such games is distributed in cloud computing, users cannot be certain where their input and output data are managed. This introduces the possibility to input incorrect data in the exchange between the gamer's terminal and the gaming platform. In this context, we propose a new gaming concept for augmented reality and location-based games in order to solve the aforementioned cheating scenario problem. The merit of our approach is to establish an accurate and verifiable proof that the gamer reached the goal or found the target. The major novelty in our method is that it allows the gamer to submit an authenticated proof related to the game result without altering the privacy of positioning data.

Cloud gaming addresses multiple challenges related to the constraints on the resources of the players computer as well as the compatibility between the online game platform and the computers. The abundant CPU, memory and storage resources available in the cloud infrastructures enable new interaction schemes between the game and the player. For instance, through the development of new video coding strategies, the user experience can be improved, especially in the case where the access to the game is made through resource-impoverished mobile terminals. The objective of this paper is to introduce a new video coding strategy for cloud gaming. The proposed strategy takes into consideration the QoS parameters and the energy-consumption of the decoder. This allows balancing the graphic quality to the lifetime of the mobile terminal. Our strategy combines the use of the Kalman filter and the Adaptive Control Feedback Control Loop. This allows the implementation of interpolation and prediction mapping functionalities in order to compensate the loss of significant information from several frames. Through an experimental comparison of our scheme with respect to the most popular cloud gaming techniques, we show that we improve the trade-off between QoS parameters (i.e., processing delay, network bandwidth) and the lifetime of the mobile terminals.

Cloud computing is an innovative model that is gaining attention and wide popularity among IT professionals and users. It provides a new mode of use of IT resources by offering everything as a service (storage, servers, network, etc.) and provisioning resources on-demand. In cloud computing environments, users outsource their data to distributed data centers in the cloud and entrust cloud providers with its storage and processing. However, data outsourcing and multi-tenants introduce new security issues that have to be addressed. In fact, being stored externally, data are no longer under the control of their owners. The storage and processing mechanisms may not be fully trustworthy and thus data confidentiality and privacy may be at risk. Therefore, in this paper, we will deal with securing stored data in shared cloud computing environment. First, we will present an overview of the main security issues addressed in the literature. Then we will focus on providing a game-theoretic model to secure the stored data from malicious users. To that purpose, we have proposed a sequential revocation game where cloud users have to choose the best strategy from a set of three strategies in order to revoke and eliminate the malicious user while minimizing their costs. At the end, we will discuss the different outcomes of this game using the subgame perfect Nash equilibrium concept.

Cloud computing is a paradigm that is redrawing the information technology landscape by outsourcing the computation and data storage services to public cloud service providers. Over the last years, cloud storage services revealed an unprecedented opportunity for Internet users to profit from online storage services. Thanks to their enriched toolbox for file sharing and syncing, cloud storage platforms provide organizations and individuals with a reliable and cost-effective collaborative workspace. However, in addition to the traditional security issues, cloud storage services introduce new security concerns that are mainly related to the insecure state in which the files are while being synchronized. Recent publications highlighted the significant impact of the security flaws that exist in the syncing protocols used by the most popular cloud storage application. In this paper, we consider the Man in the Cloud (MitC) attack demonstrated in 2015 which allows accessing the files stored in a private repository without the possession of the authentication and authorization credentials. To address this issue, we propose a biometric-based framework for cloud storage services aiming to impede intruders from launching MitC attacks. Our framework is based on our previously published technique to combine chaotic maps and fuzzy extractors. The experiments performed on real biometric features confirm the potential brought by our framework to implement strong authentication in cloud storage applications.

Due to the limited resources of mobile terminals, the service providers are adopting more cloud infrastructure to provide a powerful service for limited devices. This use needs the implementation of strong security mechanisms to guarantee the protection of information and data traffic. Since mobile devices have limited resources, it is still a major challenge to optimize and provide security simultaneously. In this work, we consider real time multimedia service, specifically video streaming, as a highly requiring service. We propose adaptive encryption scheme that takes into consideration user's mobile resources and energy consumption. Only symmetric bloc encryption algorithms are considered, since they suit the best for real time multimedia services. Furthermore, the proposed solution should guarantee an acceptable level of Quality of Experience, which is the service quality as perceived by the end user, according to a previously set SLA.

Video streaming services over the Internet are being widely used in mobile networks. Currently, service providers are relying more and more on securing data transferred between the service provider, namely the video streaming server, and the end user. However, securing video content includes at least the use of video encryption algorithm requiring processing and memory capabilities, provoking more delay and energy consumption. In any other context, energy consumption may not be a very important issue, but when using a mobile device, it will certainly become one. In this paper, we are dressing a comparison between different symmetric encryption algorithms, in terms of energy consumption, while decrypting and viewing a video stream on mobile device.

The Internet of Things is a disruptive paradigm based on the cooperation of a plethora of heterogeneous smart things to collect, transmit, and analyze data from the ambient environment. To this end, many monitored variables are combined by a data analysis module in order to implement efficient context-aware decision mechanisms. To ensure resource efficiency, aggregation is a long established solution, however it is applicable only in the case of one sensed variable. We extend the use of aggregation to the complex context of IoT by proposing a novel approach for secure cooperation of smart things while granting confidentiality and integrity. Traditional solutions for data concealment in resource constrained devices rely on hop-by-hop or end-to-end encryption, which are shown to be inefficient in our context. We use a more sophisticated scheme relying on homomorphic encryption which is not compromise resilient. We combine fully additive encryption with fully additive secret sharing to fulfill the required properties. Thorough security analysis and performance evaluation show a viable tradeoff between security and efficiency for our scheme.

The Internet of Things (IoT) is one of the revolutionary technologies for healthcare. However, this flourishing still faces too many challenges including security and privacy preservation information. To address these problems, our contribution is to present a scenario of diabetes disease assume a matching between the situation of the patient and the relevant data from monitoring point view. This scenario describes how a patient can collect enormous of vital sign and activity. Wireless Body Sensor Networks (WBSN) is one of the IoT building blocks in which a patient can be monitored using a collection of sensor nodes to improve the patients quality of life. This technology in healthcare applications should not ignore security requirements. The aim of this paper is, (1) to design a healthcare architecture for a patient monitoring system, and (2) to explore the major security requirements in WBSN.

Vehicular Networks are an ideal target for various types of rational and non-rational attackers. To cope with, reaction methods against these attacks must exist. In this direction, game theory applied to the security of wireless networks (also known as security games) can be a good way of modeling. In this paper we propose a reaction method against DOS attacks in VANET. In this method we have the choice between two proposed reaction games. Design methodology and defined metrics are inspired from game theory models. To the best of our knowledge, no similar games have been proposed before. The simulations showed the efficacy of our proposal measured by the performance of the obtained results.

Vehicular Networks must be sufficiently secured against the wide variety of security challenges to which they are exposed. Cryptography provides tools to solve many of these security problems. In this paper, we address the secure group communications problem in VANETs. By generating a secret group key that can be used to encrypt or authenticate, the members of a VANET platoon can thus guarantee secure communication between them. To address this need, we propose a new secure variant of the Diffie-Hellman algorithm for groups that we fortified by a pre-shared secret to withstand the famous Man in the Middle attack. As necessary, our proposal can be used by several types of authentication and encryption VANET applications. We demonstrate the unforgeability and the privacy of our scheme and we study its secure-efficiency.

The cloud computing paradigm has changed the way of managing identity. Domain based identity management systems are no more suitable. The alternative of a universal identifier globally handled by an all knowing identity provider proved its failure. The best opproach is to follow the cloud paradigm by offering an Identity as a Service layer. Many projects and standards were proposed to reach this goal but the lack of interoperability and the increasing number of digital identities have hampered the task. We propose in this work an Identity as a Service framework based on the Automated IDentity Finder (AIDF) system which associates one service provider or more with the suitable identity provider after user consent, thus enabling Single Sign On. Additional functionalities are claims transform between different standards and semantic mapping among heterogenous attributes in the same identity context.

Recent technological advances have given rise to the popularity and success of cloud. This new paradigm is gaining an expanding interest, since it provides cost efficient architectures that support the transmission, storage, and intensive computing of data. However, the prospect of outsourcing an increasing amount of data storage and management, the abstract nature of the cloud and the diversity of the delivered services increase their vulnerability to security incidents and attacks. The objective of this paper is to provide a comprehensive review of the existing security requirements in cloud storage environments. In addition, it explores ID-based Cryptography power to address the protection of cloud storage infrastructures. As a main contribution, a new authentication scheme for Cloud-based Storage applications is proposed.

As vehicles become increasingly intelligent, it is expected that in the near future they will be equipped with radio interfaces. This will enable the formation of vehicular networks, commonly referred to as VANETs, an instance of mobile ad hoc networks with cars as the mobile nodes. As VANETs exhibit several unique features (e.g. high mobility of nodes, geographic extension) traditional security mechanisms are not always applicable. Consequently, a plethora of research contributions have been presented to cope with the intrinsic characteristics of vehicular communication. This paper outlines the communication architecture of VANETs and discusses the security and privacy challenges that need to be overcome to make such networks practically viable. It compares the various cryptographic schemes that were suggested for VANETs and explores some future trends that will shape the research in cryptographic protocols for intelligent transportation systems.

Due to the characteristics of resource-constrained in wireless sensor networks (WSNs), energy consumption is a major concern. Data aggregation is an important method to reduce the energy consumption and communication overhead ; however, it suffers from the security problems of data integrity. In fact, designing integrity schemes for data aggregation is an imperative problem for securing WSNs. In this paper, we propose a secure data aggregation based on homomorphic MAC scheme that provide efficient secure data integrity with different trade-offs between security assumptions, computation cost, and communication payload. The scheme uses two types of homomorphic MAC (i.e., MACs and EMACs). Each MAC provides integrity of the transmitted message and each EMAC provides integrity of the MACs. Our experiments show that our scheme requires less computation overheads and energy consumption than previously methods and can effectively preserve data integrity.

Mobile Cloud Computing (MCC) is a new paradigm that extends the performance of mobile terminals and enables the deployment of applications, previously not feasible, to find their way to mobile users gaining fully from cloud processing power. Unfortunately, mobile networks are characterized by limited and varying bandwidth and large latency. These wireless network features affect the whole MCC experience and substantially affect mobile cloud-based rich media applications. An important issue that should be discussed at this level; How to ensure acceptable Quality of Experience (QoE) for end users despite mobile network condition. In this paper, we will go through research proposals for QoE enhancement, modeling and management for different mobile cloud services.

The Internet of Things (IoT) is enabling the deployment of distributed applications based on ubiquitous computing and convergent networks. Due to its heterogeneous structure, it introduces new security challenges and requirements. Particularly, the security mechanisms implemented in the IoT should adapt to the dynamic context. This paper proposes a game-based model for adaptive security in the IoT, with an emphasis on eHealth applications. We use the trade-off between security-effectiveness and energy-efficiency to evaluate adaptive security strategies. We also present the results of simulation experiments to assess the performance of the proposed model. We show that our model allows extending the lifetime of the smart things by 47% compared to existing models.

Cloud computing is changing the way we use mobile application by offering a powerful, scalable and on-demand computational resources to mobile users. However, this new paradigm is a challenging issue for forensic investigators since it combines two different environments. The adoption of Mobile Cloud computing solutions is increasing rapidly, so it is necessary for digital investigation applications and procedure to be adapted to this new environment. This paper provides an overview of mobile cloud forensics including challenging issues and some existing proposals in order to overcome these challenges.

This paper proposes a simple and robust approach to apply the space vector modulation on multilevel inverters. The identification of the adequate triangle in space vector modulation algorithms becomes more complex as the inverter's level increases. Therefore, the problem can be reduced to the selection of a two-level hexagon among many in any sector of the space vector diagram of N-level inverter. The criterion used to select the appropriate hexagon is the minimum distance between the tip point of the reference voltage vector and the center of the hexagon. Once the appropriate two-level hexagon is selected, all the remaining procedures necessary for N-level SVPWM method are thereafter done like conventional two-level inverter. Numerical simulation carried out for three, five and seven level cascaded H-bridge inverters confirms the effectiveness of the proposed method.

Wireless Sensor Networks (WSNs) are being used for many applications ranging from mobile target surveillance to intelligent home networking. The data transmitted by those applications are sensitive. Then due to the weakness of the radio links, appropriate protection mechanisms are needed to prevent attackers from exploiting those lacks of security. In this paper we present a solution to authenticate the nodes and protect the sent data. We investigate the use of secure tunnels as a solution to improve the protection of WSNs. We propose a tunneling scheme that conforms to the security requirements of WSNs while having less computational and network overhead. A set of experiments has been conducted to assess the performance of the proposed scheme with regard to traditional tunnels built using the IPSec protocol. We found that our protocol considerably reduces the number of transmitted messages as well as the computational load, which makes it suitable for WSNs.

The proliferation of fourth generation (4G) networks has motivated the development of new services characterised by ubiquity and convergence. This means that the end-user can access such services from everywhere through the use of multiple networking infrastructures. Despite the benefits resulting from such services, they are intrinsically vulnerable to many spoofing attacks. Thus, a user needs to be identified and authenticated to different serving networks and service providers. The definition of different identities for the same user would be hardly manageable since it would introduce identity confusion and heavy management procedures, especially in the overlapping zones. This study proposes a unique identifer assignment and management techniques in a heterogeneous 4G environment. In this study, the authors introduce a new syntax for the temporal mobile subscriber identity and develop the associated secure identity management process. The proposed scheme, the so-called advanced identity management guarantees mutual authentication, privacy and tracking avoidance, which are of primary importance in 4G networks.

Wireless Sensor Networks (WSNs) have become an attractive technology to support various applications. In many situations, the region where the sensor nodes are deployed needs to be uniformly covered. This paper proposes a new technique to implement self-organization strategies for WSNs. We show that the process of iterating the computation of the Voronoi diagram generated by the sensors has a maximum fixed point which corresponds to the Nash equilibrium of a strategic game where sensor nodes move iteratively to the centers of the Voronoi cells. We have shown that this fixed-point represents the best deployment that can be achieved, from the coverage point of view. We also performed a set of experiments to show that the proposed technique outperform the existing approaches.

Research in the wireless sensor networks field has been plagued by difficulties in realistic simulations. These difficulties are often the result of non-realistic assumptions which need to be removed from the mathematical models. Especially, radio propagation models are often simplified and do not take into consideration irregularities such as fading, shadowing, and temporal non-stationarity. In this paper, we investigate the effect of radio temporal irregularities on sensor deployment and sleep scheduling. We show that existing scheduling algorithms suppose that the sensor coverage range does not vary according to time, which is not true because the available energy decreases. We analyze the impact of this shortcoming on the performance of sleep scheduling algorithms and propose an alternative scheme that guarantees a uniform density distribution despite of the presence of radio temporal irregularities.

Wireless Sensor Networks (WSNs) have inspired tremendous research interest in since the mid-1990s. Advancement in wireless communication and miniature electromechanical systems (MEMSs) have enabled the development of low-cost, low power, multi-functional, tiny sensor nodes that can sense the environment, perform data processing, and communicate with each other untethered over short distances. Most of the applications deployed over WSNs require strong coverage requirements, especially those related to the detection and tracking of distributed events. Moreover, these events are forwarded to the analysis center(s) through a set of sink nodes that locally gather data emanating from the elementary sensors. This paper proposes a coverage control scheme that adapts to the situation where multiple sink nodes are deployed within the monitored area. On the opposite to traditional coverage approaches that aim at guaranteeing a uniform density distribution, we place the sensor nodes in a manner that increases the coverage degree according to their proximity to a sink node. To reduce the complexity of the optimization process, we consider a discrete search space by structuring the monitored into a uniform grid. An evolutionary algorithm is then used to choose whether to activate or not sensor nodes within every cell of the grid. We conducted a set of simulations in order to evaluate the performance of the proposed strategy, mainly in ensuring multi-target tracking.

Attacks against wireless sensor networks (WSNs) are becoming more and more frequent. Due to the sensitive nature of these networks, appropriate security mechanisms should therefore be developed. This paper introduces a fault-tolerant approach for WSNs. The basic idea to detect malicious nodes is to proceed to intermediate verification of the signed messages within the communication paths. A tolerance state is defined in which a set of tests are performed on the malicious node in order to definitely reject it or rehabilitate it.

With the increasing sophistication of attack techniques and scenarios, appropriate automated decision-making systems should be developed. This paper defines a new security language allowing to cope with attack scenarios through the representation of both attacks and security solutions in a single syntactic framework. A subsequent semantic analysis has also been introduced. To implement this reasoning, we introduce a security compiler-like architecture that comes up with substantial novelties with regard to traditional compilers (used in software engineering). The most important innovations are the computation of abstract attack/counter measure specifications and the resolution of the fundamental security equation (FSE). Unlike existing compilation schemes, our approach aims at building a relational specification of the attack through a traversal of its semantic tree. The security solution(s) corresponding the attack of interest is (are) then found by solving the FSE, in the relational algebra of attacks and decisions. Concrete examples have been analyzed in order to highlight the potential of the proposed relational algebra-based security language, called ReAlSec.

Optical burst switching (OBS) is one promising method for data transfer in photonic networks based on a WDM (Wavelength Division Multiplexing) technology. In the OBS scheme, the wavelength is exclusively reserved along the source and destination nodes, when the burst data is generated at the source. Then, efficient data transfer is expected. However, its performance is heavily dependent on the number of links that the lightpath goes through. TCP-based applications account for a majority of data traffic in the Internet; thus understanding and improving the performance of TCP over OBS networks is critical. In this paper, we present a new parallel wavelength reservation method for optical burst switching (OBS) networks based on adapting the set of potential wavelengths with the number of hops in the path. The uniqueness of this work when compared to existing works is that buffering resources, which consist of Optical Delay Lines (ODLs), are considered in the reservation mechanism. The consequence of this is that the time made by the segments in the various buffers across the selected path is taken into consideration. Various simulation experiments have been conducted to evaluate the performance of the scheme. It is found that our approach has better performance than previous related works reported in the literature.

The intrinsic vulnerabilities of wireless networks place a great demand on efficient encryption techniques.With the proliferation of video streaming applications, the need for multimedia encryption techniques is becoming of utmost importance. This paper presents a scalable coding scheme for multimedia transmission over wireless networks. The proposed approach unifies the compression and encryption functionalities using the wavelet transform and chaotic functions to build up random permutations. The major advantage of our chaotic multimedia encryption process is that a client can access to multiple resolutions of the streamed video. These resolutions vary according to the security level of the client as well as the networking and processing capabilities. A set of simulations have also been carried out to assess the performance of the proposed coding scheme in terms of security, compression ratio, and error resilience.

Target tracking based restrictively on acoustic and electromagnetic (EM) sensors may not provide adequate information regarding the mobile target. Hence, imaging sensors can be used to provide visual information. This paper develops an image-based tracking approach based on epipolar geometry and Kalman filtering. A corner detection technique is used to identify the prominent features in the frame sequence. By applying epipolar transforms, the trajectory of the target can be reconstructed along the image sequence. A recursive Kalman filter is then developed to enhance the robustness of the approach with regard to noise. Experimental results show that the proposed approach outperforms the existing image-based tracking approaches, especially for multi-target tracking.

This paper proposes a security architecture for optical code division multiple access networks based on lattice cryptography. Unlike existing approaches, which have mainly focused on optical components such as phase masks and delay lines, our cryptosystem encompasses a secure code construction process. In fact, given a set of optical orthogonal codewords, we associate a pair of bases (a public basis and a private basis) to every user and we show that the projection of a codeword on the public basis of the receiver and the addition of a random error enhances the confidentiality performance of the code. The proposed public key cryptosystem is based on lattice cryptography. The security of this scheme relies mainly on the complexity of the closest vector-problem in an integer lattice. We found that our technique performs better than the existing approaches in terms of robustness to cryptanalysis. We also study the security of our lattice cryptosystem with regard to the properties of the bases pairs and the error vectors.

This paper deals with a combination of work in the fields of artificial intelligence and computer security. It describes a decision model based on a new genetic algorithm approach for intrusion response system (NGAA-IRS). A brief survey of intrusion detection and response system (IDRS), genetic algorithm (GA), and its application to IDRS are presented. Then, the proposed model, parameters and evolution process for GA are discussed in details. The model is characterized by a new implementation of individual structure based on a matrix of response-resource entries and a fitness function based on cost benefit approach for selecting the appropriate solution. These features are specific to NGAA-IRS model and do not be used in other implementations beforehand.

Research in the field of wireless sensor networks (WSNs) has been plagued by difficulties in performing realistic simulations. For instance, most of the existing coverage optimization techniques presuppose that the region covered by a sensor node is a disc characterized by the radio transmission range. This assumption is rigorously false because of the propagation phenomena including fading and shadowing. This paper investigates the impact of these radio propagation phenomena on the coverage optimization strategy in a WSN. We rely on the log-normal shadowing model to represent the effect of the environmental features on the WSN performance. We propose a coverage model taking into account irregular radio propagation. We carry out a mathematical analysis to compute the average number of targets a sensor would detect under both perfect and irregular propagation conditions. We also conduct simulations to assess our random deployment model with respect to the existing strategies.

Existing mobility models for wireless sensor networks generally do not preserve a uniform scattering of the sensor nodes within the monitored area. This paper proposes a coverage-preserving random mobility model called DPRMM. Direction and velocity are randomly chosen according to the local information about sensor density. Our mobility model is devised in a manner that sensors move towards the least covered regions within their neighborhood. We show that this guarantees a rapid convergence to the steady state while preserving a uniform coverage degree on the monitored region. Throughout the simulations we have carried out, we find that the analytical study is corroborated by practical experiments. Our experiments show that the average distance made by a target without being detected is approximately enhanced at least by a factor of 2 using DPRMM.

The paper proposes a progressive access control scheme for JPEG2000 image codestreams based on chaotic maps. The access control scheme provides the possibility to access partially decrypted images from a single encrypted JPEG 2000 stream. The underlying encryption algorithm is based on chaotic permutations of pixel gray levels, positions, and wavelet filter coefficients. Depending on the key provided to the user, the decryption process will stop at a given resolution. The two major characteristics of the access control scheme is that it provides a unifying encryption-compression approach and is fully compliant with the JPEG 2000 encoding process.

This paper proposes a chaotic encryption scheme for image data transmission and multimedia communication. To this end, we first extend the definition of the well-known Arnold 2-D chaotic map to the four-dimensional context. We demonstrate that the chaotic behaviour of the Arnold function is enhanced through our extension to four dimensions. Second, we develop a cryptosystem based on the use of chaotic maps and wavelet decomposition. In fact, by involving the wavelet decomposition filters in the encryption process in addition to the pixel positions and the image gray levels, the 4-D chaotic map is defined in a way that it extends the 3-D chaotic map presented in the literature. This allows to adapt the efficiency of the encryption algorithm to the security needs and the available bandwidth through progressive encryption. According to the length of the symmetric key, different resolutions of the image can be obtained at the decryption level.

Wireless sensor networks (WSNs) are being used in many applications in order to gather sensitive information and forward it to an analysis center. Since a WSN consists of resource- impoverished sensor nodes, the packet forwarding process should be energy efficient. Therefore, resource limitations should be taken into consideration when designing a WSN infrastructure. This paper proposes a random routing strategy for WSNs. The approach relies on the flooding technique, which has the advantage to possess minimal routing overhead (in the sense that no routing table information is exchanged) and maximal delivery rate. We introduce an enhancement that allows reducing energy consumption and extending network lifetime by randomly forwarding packets at every node. The forward probability is a decreasing function of the number of hops made by the packet. An analytical model is developed in order to illustrate the functionalities of the proposed strategy. Finally, simulations are conducted to assess the performance of the probabilistic routing protocols.

With the advance of sensing technologies and its applications, sophisticated (i.e., mobile, wireless) sensor networks are gaining increasing interest. Although such networks allow to implement interesting tracking functionalities, their performance is severely affected by energy limitations. This paper introduces an activity scheduling protocol that extends the network lifetime while the area coverage is kept unchanged. A cost-benefit analysis is conducted in order to assess the cost of the additional number of sensors that should be deployed to implement our protocol. Simulation results show that the proposed approach outperforms the existing scheduling techniques.

Mobile sensor networks (MSNs) consist of large number of small and computationally impoverished devices deployed over an area to track mobile objects. Mobility is becoming an important feature of MSNs. Recently, sensors have began to be deployed on mobile platforms such as robots. In this paper, we propose an evaluation of the mobility impact on MSN tracking efficiency. More precisely, we consider the Doppler effect on the results of several target location approaches. Two radar-based angle estimation techniques have been considered: Frequency-Modulated Continuous Wave (FMCW) Radar, and monopulse angle estimation. We also analyze the control of the uncertainty due to Doppler shift through a manipulation of the area coverage (i.e., number of sensors per area coverage).

This paper proposes to manage data structures manipulated by the proposed correlation function based on a new concept called multi-violation detectors (MvD). The MvD-based correlation function manipulates events based on metrics that evaluate them. The detection and the correlation processes are managed using a MLP (multi-layer perceptron)-like network architecture. The manipulated data structures in the MLP-like network mainly include attacks, metrics and MvD values. To add more efficiency to the implemented processes, these values are managed through a structured database where an SQL-like language for MvD retrieval is introduced. The proposed approach is based on a relational calculus. Therefore, its completeness is mathematically proved.

Security policies constitute the core of protecting information systems. Validation tools should therefore be developed to check whether a version of a specific security policy conforms to the required security properties. This paper proposes a validation framework for security policies where: (1) algebraic specifications are used to build abstract views about the security policy, (2) an executable security policy can be extracted from the algebraic representation, and (3) syntactical (resp. sematic) verification of the executable (resp. algebraic) security policy is performed.

This paper presents a relational approach for structuring security decision databases to support advanced automated functionalities. We represent attacks and countermeasures in a unifying environment using relational algebra. Both are viewed as actions that behave in opposite manners. We introduce an ordering that allows to rank security decisions according to their efficiency with respect a given attack. Furthermore, we prove that this ordering defines a lattice-structure that can serve to manage the decision database.

Mobile wireless sensor networking is characterized by a set of challenging issues including sensor and target mobility management, sensing continuity, target tracking and sensor density management. We address in this paper various issues related to the localization and tracking of targets. Our approach provides novel schemes for effective location, robustness and sensor density management.

This paper presents an approach for decision making under security risks in a computer network environment. The proposed method relies on a many sorted algebraic signature and on a rewriting system. This latter is shown to be terminating and yielding a normal form, called the risk analysis equation, that models the cost-benefit balance. Furthermore, a gradual algebraic resolution of the risk analysis equation is described. This formalism helps security analysts to automate the selection of the optimal security solutions that minimize the residual risk.

In this paper, a nonlinear subband decomposition scheme with perfect reconstruction is proposed for lossless and progressive coding of multispectral images. The merit of this new scheme is to exploit efficiently the spatial and the spectral redundancies contained in the multispectral images related to a scene of interest. Besides, the proposed method is suitable for telebrowsing applications. Experiments carried out on real scenes allow to assess its performances. The simulation results demonstrate that our approach leads to improved compression performances compared with currently used lossless coders.

A nonlinear subband decomposition scheme with perfect reconstruction is proposed for lossless coding of multispectral images. The merit of this new scheme is to exploit efficiently both the spatial and the spectral redundancies contained in a multispectral image sequence. Besides, it is suitable for progressive coding, which constitutes a desirable feature for telebrowsing applications. Simulation tests performed on real scenes allow assessment of the performances of this new multiresolution coding algorithm. The achieved compression ratios are higher than those obtained with currently used lossless coders.