Federated Learning (FL) has become a crucial framework for collaboratively training Machine Learning (ML) models while ensuring data privacy. Traditional synchronous FL approaches, however, suffer from delays caused by slower clients (called stragglers), which hinder the overall training process. Specifically, in a synchronous setting, model aggregation happens once all the intended clients have submitted their local updates to the server. To address these inefficiencies, Buffered Asynchronous FL (BAsyncFL) was introduced, allowing clients to update the global model as soon as they complete local training. In such a setting, the new global model is obtained once the buffer is full, thus removing synchronization bottlenecks. Despite these advantages, existing Secure Aggregation (SA) techniques-designed to protect client updates from inference attacks-rely on synchronized rounds, making them unsuitable for asynchronous settings.

International audience

In 5G and beyond networks, the radio communication between a User Equipment (UE) and a base station (gNodeB or gNB), also known as the air interface, is a critical component of network access and connectivity. During the connection establishment procedure, the Radio Resource Control (RRC) layer can be vulnerable to signaling storms, which threaten the availability of the radio access control plane. These attacks may occur when one or more UEs send a large number of connection requests to the gNB, preventing new UEs from establishing connections. In this paper, we investigate the detection of such threats and propose an adaptive threshold-based detection system based on Extreme Value Theory (EVT). The proposed solution is evaluated numerically by applying simulated attack scenarios based on a realistic threat model on top of real-world RRC traffic data from an operator network. We show that, by leveraging features from the RRC layer only, the detection system can not only identify the attacks but also differentiate them from legitimate high-traffic situations. The adaptive threshold calculated using EVT ensures that the system works well under diverse threat scenarios. The results show high accuracy, precision, and recall values (above 93%), and a low detection latency even under complex conditions.

International audience

Vertical Federated Graph Learning (VFGL) is a novel privacy-preserving technology that enables entities to collaborate on training Machine Learning (ML) models without exchanging their raw data. In VFGL, some of the entities hold a graph dataset capturing sensitive user relations, as in the case of social networks. This collaborative effort aims to leverage diverse features from each entity about shared users to enhance predictive models or recommendation systems, while safeguarding data privacy in the process. Despite these advantages, recent studies have revealed a critical vulnerability that appears in intermediate data representations, which may inadvertently expose link information in the graph. This work proposes a novel Link Inference Attack (LIA) that exploits gradients as a new source of link information leakage. Assuming a semi-honest adversary, we demonstrate through extensive experiments on seven real-world datasets that our LIA outperforms state-of-the-art attacks, achieving over 10% higher Area Under the Curve (AUC) in some instances, thereby highlighting a significant risk of link information leakage through gradients. Our attack’s effectiveness primarily stems from label information embedded in gradients, as evidenced by comparison with a label-only LIA. We analytically derive our Label-based LIA’s accuracy using graph characteristics, assessing target graph vulnerability. To address these vulnerabilities, we evaluate two types of defenses: edge perturbation based on differential privacy and a novel label perturbation approach, demonstrating that our proposed label perturbation defense is more effective against all attack types across all datasets examined, offering a more favorable privacy-utility trade-off. Our comprehensive analysis shows why LIAs are effective and identifies potential defenses, highlighting the need for further research to improve the security of VFGL systems against link information leakage.

Computing the distance between two non-normalized vectors x and y, represented by Δ (x, y) and comparing it to a predefined public threshold τ is an essential functionality used in privacy-sensitive applications such as biometric authentication, identification, machine learning algorithms (e.g., linear regression, k-nearest neighbors, etc.), and typo-tolerant password-based authentication. Tackling a widely used distance metric, Nomadic studies the privacy-preserving evaluation of cosine similarity in a two-party (2PC) distributed setting. We illustrate this setting in a scenario where a client uses biometrics to authenticate to a service provider, outsourcing the distance calculation to two computing servers. In this setting, we propose two novel 2PC protocols to evaluate the normalising cosine similarity between non-normalised two vectors followed by comparison to a public threshold, one in the semi-honest and one in the malicious setting. Our protocols combine additive secret sharing with function secret sharing, saving one communication round by employing a new building block to compute the composition of a function f yielding a binary result with a subsequent binary gate. Overall, our protocols outperform all prior works, requiring only two communication rounds under a strong threat model that also deals with malicious inputs via normalisation. We evaluate our protocols in the setting of biometric authentication using voice, and the obtained results reveal a notable efficiency improvement compared to existing state-of-the-art works.

This paper introduces a novel protocol for privacy-preserving biometric identification, named Monchi, that combines the use of homomorphic encryption for the computation of the identification score with function secret sharing to obliviously compare this score with a given threshold and finally output the binary result. Given the cost of homomorphic encryption, BFV in this solution, we study and evaluate the integration of two packing solutions that enable the regrouping of multiple templates in one ciphertext to improve efficiency meaningfully. We propose an end-to-end protocol, prove it secure and implement it. Our experimental results attest to Monchi's applicability to the real-life use case of an airplane boarding scenario with 1000 passengers,taking less than one second to authorize/deny access to the plane to each passenger via biometric identification while maintaining the privacy of all passengers.

52

Image registration is a key task in medical imaging applications, allowing to represent medical images in a common spatial reference frame. Current approaches to image registration are generally based on the assumption that the content of the images is usually accessible in clear form, from which the spatial transformation is subsequently estimated. This common assumption may not be met in practical applications, since the sensitive nature of medical images may ultimately require their analysis under privacy constraints, preventing to openly share the image content . In this work, we formulate the problem of image registration under a privacy preserving regime, where images are assumed to be confidential and cannot be disclosed in clear. We derive our privacy preserving image registration framework by extending classical registration paradigms to account for advanced cryptographic tools, such as secure multi-party computation and homomorphic encryption, that enable the execution of operations without leaking the underlying data. To overcome the problem of performance and scalability of cryptographic tools in high dimensions, we propose several techniques to optimize the image registration operations by using gradient approximations, and by revisiting the use of homomorphic encryption trough packing, to allow the efficient encryption and multiplication of large matrices. We focus on registration methods of increasing complexity, including rigid, affine, and non-linear registration based on cubic splines or diffeomorphisms parameterized by time-varying velocity fields . In all these settings, we demonstrate how the registration problem can be naturally adapted for accounting to privacy-preserving operations, and illustrate the effectiveness of PPIR on a variety of registration tasks. • Image registration under a privacy preserving regime. • This work extends classic registration paradigms and integrates cryptographic tools. • Privacy preserving linear & non-linear registration with sum squared differences. • Privacy preserving linear registration with mutual information. • Privacy preserving non-linear registration with cross correlration. • Privacy preserving rigid point cloud registration. • Methods to optimize image registration operations with cryptographic tools. • Results demonstrate the proposed work’s applicability in 2D-3D medical imaging tasks.

This paper proposes a novel method to perform privacy-preserving face identification based on the notion of group testing, and applies it to a solution using the Cheon-Kim-Kim-Song (CKKS) homomorphic encryption scheme. Securely computing the closest reference template to a given live template requires K comparisons, as many as there are identities in a biometric database. Our solution, named Grote, replaces element-wise testing by group testing to drastically reduce the number of such costly, non-linear operations in the encrypted domain from K to up to 2\sqrtK . More specifically, we approximate the max of the coordinates of a large vector by raising to the α-th power and cumulative sum in a 2D layout, incurring a small impact in the accuracy of the system while greatly speeding up its execution. We implement Grote and evaluate its performance.

Over the last decade, the use of Automatic Speaker Verification (ASV) systems has become increasingly widespread in response to the growing need for secure and efficient identity verification methods. The voice data encompasses a wealth of personal information, which includes but is not limited to gender, age, health condition, stress levels, and geographical and socio-cultural origins. These attributes, known as soft biometrics, are private and the user may wish to keep them confidential. However, with the advancement of machine learning algorithms, soft biometrics can be inferred automatically, creating the potential for unauthorized use. As such, it is crucial to ensure the protection of these personal data that are inherent within the voice while retaining the utility of identity recognition. In this paper, we present an adversarial Auto-Encoder-based approach to hide gender-related information in speaker embeddings, while preserving their effectiveness for speaker verification. We use an adversarial procedure against a gender classifier and incorporate a layer based on the Laplace mechanism into the Auto-Encoder architecture. This layer adds Laplace noise for more robust gender concealment and ensures differential privacy guarantees during inference for the output speaker embeddings. Experiments conducted on the VoxCeleb dataset demonstrate that speaker verification tasks can be effectively carried out while concealing speaker gender and ensuring differential privacy guarantees; moreover, the intensity of the Laplace noise can be tuned to select the desired trade-off between privacy and utility.

We propose a novel privacy-preserving, two-party computation of various distance metrics (e.g., Hamming distance, Scalar Product) followed by a comparison with a fixed threshold, which is known as one of the most useful and popular building blocks for many different applications including machine learning, biometric matching, etc. Our solution builds upon recent advances in function secret sharing and makes use of an optimized version of arithmetic secret sharing. Thanks to this combination, our new solution named Funshade is the first to require only one round of communication and two ring elements of communication in the online phase, outperforming all prior state-of-the-art schemes while relying on lightweight cryptographic primitives. Lastly, we implement our solution from scratch in portable C and expose it in Python, testifying its high performance by running secure biometric identification against a database of 1 million records in ~10 seconds with full correctness and 32-bit precision, without parallelization.

Secure aggregation consists of computing the sum of data collected from multiple sources without disclosing these individual inputs. Secure aggregation has been found useful for various applications ranging from electronic voting to smart grid measurements. Recently, federated learning emerged as a new collaborative machine learning technology to train machine learning models. In this work, we study the suitability of secure aggregation based on cryptographic schemes to federated learning. We first provide a formal definition of the problem and suggest a systematic categorization of existing solutions. We further investigate the specific challenges raised by federated learning and analyze the recent dedicated secure aggregation solutions based on cryptographic schemes. We finally share some takeaway messages that would help a secure design of federated learning and identify open research directions in this topic. Based on the takeaway messages, we propose an improved definition of secure aggregation that better fits federated learning.

Federated learning allows multiple parties to collaboratively train a global machine learning (ML) model without sharing their private datasets. To make sure that these local datasets are not leaked, existing works propose to rely on a secure aggregation scheme that allows parties to encrypt their model updates before sending them to the central server that aggregates the encrypted inputs. In this work, we design and evaluate a new secure and fault-tolerant aggregation scheme for federated learning that is robust against client failures. We first develop a threshold-variant of the secure aggregation scheme proposed by Joye and Libert. Using this new building block together with a dedicated decentralized key management scheme and an input encoding solution, we design a privacy-preserving federated learning protocol that, when executed among n clients, can recover from up to failures. Our solution is secure against a malicious aggregator who can manipulate messages to learn clients’ individual inputs. We show that our solution outperforms the state-of-the-art fault-tolerant secure aggregation schemes in terms of computation cost on the client. For example, with an ML model of 100,000 parameters, trained with 600 clients, our protocol is 5.5x faster (1.6x faster in case of 180 clients drop).

Recent advances in Fully Homomorphic Encryption (FHE) allow for a practical evaluation of non-trivial functions over encrypted data. In particular, novel approaches for combining ciphertexts broadened the scope of prospective applications. However, for arithmetic circuits, the overall complexity grows with the desired precision and there is only a limited space for parallelization. In this paper, we put forward several methods for fully parallel addition of multi-digit integers encrypted with the TFHE scheme. Since these methods handle integers in a special representation, we also revisit the signum function, firstly addressed by Bourse et al., and we propose a method for the maximum of two numbers; both with particular respect to parallelization. On top of that, we outline an approach for multiplication by a known integer. According to our experiments, the fastest approach for parallel addition of 31-bit encrypted integers in an idealized setting with 32 threads is estimated to be more than 6x faster than the fastest sequential approach. Finally, we demonstrate our algorithms on an evaluation of a practical neural network.

This paper proposes a novel collaborative decryption protocol for the Brakerski-Fan-Vercauteren (BFV) homomorphic encryption scheme in a multiparty distributed setting, and puts it to use in designing a leakage-resilient biometric identification solution. Allowing the computation of standard homomorphic operations over encrypted data, our protocol reveals only one least significant bit (LSB) of a scalar/vectorized result resorting to a pool of N parties. By employing additively shared masking, our solution preserves the privacy of all the remaining bits in the result as long as one party remains honest. We formalize the protocol, prove it secure in several adversarial models, implement it on top of the open-source library Lattigo and showcase its applicability as part of a biometric access control scenario.

With the increasing development of machine learning models in daily businesses, a strong need for intellectual property protection arised. For this purpose, current works suggest to leverage backdoor techniques to embed a watermark into the model, by overfitting to a set of particularly crafted and secret input-output pairs called triggers. By sending verification queries containing triggers, the model owner can analyse the behavior of any suspect model on the queries to claim its ownership. However, when it comes to scenarios where frequent monitoring is needed, the computational overhead of these verification queries in terms of volume demonstrates that backdoor-based watermarking appears to be too sensitive to outlier detection attacks and cannot guarantee the secrecy of the triggers.

Clustering is an unsupervised machine learning technique that outputs clusters containing similar data items. In this work, we investigate privacy-preserving density-based clustering which is, for example, used in financial analytics and medical diagnosis. When (multiple) data owners collaborate or outsource the computation, privacy concerns arise. To address this problem, we design, implement, and evaluate the first practical and fully private density-based clustering scheme based on secure two-party computation. Our protocol privately executes the DBSCAN algorithm without disclosing any information (including the number and size of clusters). It can be used for private clustering between two parties as well as for private outsourcing of an arbitrary number of data owners to two non-colluding servers. Our implementation of the DBSCAN algorithm privately clusters data sets with 400 elements in 7 minutes on commodity hardware. Thereby, it flexibly determines the number of required clusters and is insensitive to outliers, while being only factor 19x slower than today's fastest private K-means protocol (Mohassel et al., PETS'20) which can only be used for specific data sets. We then show how to transfer our newly designed protocol to related clustering algorithms by introducing a private approximation of the TRACLUS algorithm for trajectory clustering which has interesting real-world applications like financial time series forecasts and the investigation of the spread of a disease like COVID-19.

Since machine learning models have become a valuable asset for companies, watermarking techniques have been developed to protect the intellectual property of these models and prevent model theft. We observe that current watermarking frameworks solely target image classification tasks, neglecting a considerable part of machine learning techniques. In this paper, we propose to address this lack and study the watermarking process of various machine learning techniques such as machine translation, regression, binary image classification and reinforcement learning models. We adapt current definitions to each specific technique and we evaluate the main characteristics of the watermarking process, in particular the robustness of the models against a rational adversary. We show that watermarking models beyond classification is possible while preserving their overall performance. We further investigate various attacks and discuss the importance of the performance metric in the verification process and its impact on the success of the adversary.

International audience

International audience

This paper studies the statistical characterization of detecting an adversary who wants to harm some computation such as machine learning models or aggregation by altering the output of a differentially private mechanism in addition to discovering some information about the underlying dataset. An adversary who is able to modify the published information from a differentially private mechanism aims to maximize the possible damage to the system while remaining undetected. We present a trade-off between the privacy parameter of the system, the sensitivity and the attacker’s advantage (the bias) through determining the threshold for the best critical region of the hypothesis testing problem for deciding whether or not the adversary’s attack is detected. Such tradeoffs are provided for Laplace mechanisms using one-sided and two-sided hypothesis tests. Corresponding error probabilities are analytically derived and ROC curves are presented for various levels of the sensitivity, the absolute mean of the attack and the privacy parameter. Subsequently, we provide an interval for the bias induced by the adversary so that the defender detects the attack. Finally, we adapt the Kullback-Leibler differential privacy to adversarial classification.

Secure aggregation protocols allow anaggregator to compute the sum of multiple users' data in a privacy-preserving manner. Existing protocols assume that users from whom the data is collected, are fully trusted on the correctness of their individual inputs. We believe that this assumption is too strong, for example when such protocols are used for federated learning whereby the aggregator receives all users' contributions and aggregate them to train and obtain the joint model. A malicious user contributing with incorrect inputs can generate model poisoning or backdoor injection attacks without being detected. In this paper, we propose the first secure aggregation protocol that considers users as potentially malicious. This new protocol enables the correct computation of the aggregate result, in a privacy preserving manner, only if individual inputs belong to a legitimate interval. To this aim, the solution uses a newly designed oblivious programmable pseudo-random function. We validate our solution as a proof of concept under a federated learning scenario whereby potential backdoor injection attacks exist.

Internet of Things (IoT) technology promises to bring new value creation opportunities across all major industrial sectors. This will yield industries to deploy more devices into their networks. A key pillar to ensure the safety and security of the running services on these devices is remote attestation. Unfortunately,existing solutions fail to cope with the recent challenges raised by large IoT networks. In particular, the heterogeneity of the devices used in the network affects the performance of a remote attestation protocol. Another challenge in these networks is their dynamic nature: More IoT devices may be added gradually over time. This poses a problem in terms of key management in remote attestation.

The Internet of Things (IoT) technology has ex- panded widely across the world, promising new data management opportunities for industries, companies and individuals in different sectors, such as health services or transport logistics. The exponentially increasing number of IoT devices, their origin diversity, their limited capabilities in terms of resources, as well as the ever-increasing amount of data, raise new challenges for security and privacy protection, precluding traditional access control solutions to be integrated to this new environment. In this paper, we propose a reliable, server-aided, policy-based access control mechanism, named CHARIOT, that enables an IoT platform to verify credentials of different devices requesting access to the data stored within it. CHARIOT enables IoT devices to authenticate themselves to the platform without compromising their privacy by using attribute-based signatures. Our solution also allows secure delegation of costly computational operations to a cloud server, hence relieving the workload at IoT devices' side.

Abstract Searchable Encryption (SE) allows a user to upload data to the cloud and to search it in a remote fashion while preserving the privacy of both the data and the queries. Recent research results describe attacks on SE schemes using the access pattern, denoting the ids of documents matching search queries, which most SE schemes reveal during query processing. However SE schemes usually leak more than just the access pattern, and this extra leakage can lead to attacks (much) more harmful than the ones using basic access pattern leakage only. We remark that in the special case of Multi-User Searchable Encryption (MUSE), where many users upload and search data in a cloud-based infrastructure, a large number of existing solutions have a common leakage in addition to the well-studied access pattern leakage. We show that this seemingly small extra leakage allows a very simple yet powerful attack, and that the privacy degree of the affected schemes have been overestimated. We also show that this new vulnerability affects existing software. Finally we formalize the newly identified leakage profile and show how it relates to previously defined ones.

With the advent of cloud computing, individuals and companies alike are looking for opportunities to leverage cloud resources not only for storage but also for computation. Nevertheless, the reliance on the cloud to perform computation raises the unavoidable challenge of how to assure the correctness of the delegated computation. In this regard, we introduce two cryptographic protocols for publicly verifiable computation that allow a lightweight client to securely outsource to a cloud server the evaluation of high-degree univariate polynomials and the multiplication of large matrices. Similarly to existing work, our protocols follow the amortized verifiable computation approach. Furthermore, by exploiting the mathematical properties of polynomials and matrices, they are more efficient and give way to public delegatability. Finally, besides their efficiency, our protocols are provably secure under well-studied assumptions.

Nowadays we are witnessing the democratization of cloud services. As a result, more and more end-users (individuals and businesses) are using these services for achieving their electronic transactions (shopping, administrative procedures, B2B transactions, etc.). In such scenarios, personal data is generally flowed between several entities and end-users need (i) to be aware of the management, processing, storage and retention of personal data, and (ii) to have necessary means to hold service providers accountable for the usage of their data. In fact, dealing with personal data raises several privacy and accountability issues that must be considered before to promote the use of cloud services. In this paper, we propose a framework for the representation of cloud accountability policies. Such policies offer to end-users a clear view of the privacy and accountability obligations asserted by the entities they interact with, as well as means to represent their preferences. This framework comes with two novel accountability policy languages. An abstract one devoted for the representation of preferences/obligations in an human readable fashion. And a concrete one for the mapping to concrete enforceable policies. We motivate our solution with concrete use case scenarios.

In this paper, we address the problem of privacy preserving delegated word search in the cloud. We consider a scenario where a data owner outsources its data to a cloud server and delegates the search capabilities to a set of third party users. In the face of semi-honest cloud servers, the data owner does not want to disclose any information about the outsourced data; yet it still wants to benefit from the highly parallel cloud environment. In addition, the data owner wants to ensure that delegating the search functionality to third parties does not allow these third parties to jeopardize the confidentiality of the outsourced data, neither does it prevent the data owner from efficiently revoking the access of these authorized parties. To these ends, we propose a word search protocol that builds upon techniques of keyed hash functions, oblivious pseudo-random functions and Cuckoo hashing to construct a searchable index for the outsourced data, and uses private information retrieval of short information to guarantee that word search queries do not reveal any information about the data to the cloud server. Moreover, we combine attribute-based encryption and oblivious pseudo-random functions to achieve an efficient revocation of authorized third parties. The proposed scheme is suitable for the cloud as it can be easily parallelized.

Existing privacy controls based on access control techniques do not prevent massive dissemination of private data by unauthorized users. We suggest a usage control enforcement scheme that allows users to gain control over their data during its entire lifetime. The scheme is based on a peer-to-peer architecture whereby a different set of peers is randomly selected for data assignment. Usage control is achieved based on the assumption that at least t out of any set of n peers will not behave maliciously. Such a system would still suffer from re-injection attacks whereby attackers can gain ownership of data and the usage policy thereof by simply re-storing data after slight modification of the content. In order to cope with re-injection attacks the scheme relies on a similarity detection mechanism. The robustness of the scheme has been evaluated in an experimental setting using a variety of re-injection attacks.

With the continuous and exponential increase of the number of users and the size of their data, data deduplication becomes more and more a necessity for cloud storage providers. By storing a unique copy of duplicate data, cloud providers greatly reduce their storage and data transfer costs. The advantages of deduplication unfortunately come with a high cost in terms of new security and privacy challenges. We propose ClouDedup, a secure and efficient storage service which assures block-level deduplication and data confidentiality at the same time. Although based on convergent encryption, ClouDedup remains secure thanks to the definition of a component that implements an additional encryption operation and an access control mechanism. Furthermore, as the requirement for deduplication at block-level raises an issue with respect to key management, we suggest to include a new component in order to implement the key management for each block together with the actual deduplication operation. We show that the overhead introduced by these new components is minimal and does not impact the overall storage and computational costs.

Many incentive mechanisms have been proposed to foster cooperation among nodes in Peer-to-Peer (P2P) networks. Unfortunately, most of existing solutions rely on the existence of an online centralized authority that is in charge of a fair distribution and transaction of credits (incentives) between peers. Such centralized mechanisms mainly suffer from privacy leakage and single point of failure problems. To cope with these problems, we propose to take advantage of the distributed nature of P2P networks in order for the peers to take care of credit-based operations. Cheating and other DoS attacks are prevented thanks to a threshold security mechanism where the operation should be approved by a predefined certain number of peers. The main novelty of the proposed mechanism is the fact that a “credit” is assigned to some peers using distributed hash tables, hence, peers can follow and control the history of operations with respect to this credit, only. Thanks to this new approach, a malicious node cannot easily keep track of all operations originating from a single node and the impact of cheating or similar attacks would be strongly reduced.

The extremely widespread adoption of Online Social Networks (OSNs) raises many questions on privacy and access control. Regardless of the particular centralized or de-centralized nature of the OSN, the achievable security and privacy degree strongly depends on the graph-theoretical properties of the social graph representing the real friendship relations between the users. In this paper, we analyze the relationship between the social network graph topology and the achievable privacy. We observe three metrics, namely degree distribution, clustering coefficient and mixing time, and show that they give fundamental insights on the privacy degree of the OSN. We propose how to exploit these insight for the design of future privacy-friendly OSN.

Jamming attacks can severely affect the performance of Wireless Sensor Networks (WSNs) due to their broadcast nature. The most reliable solution to reduce the impact of such attacks is to detect and localize the source of the attack. In this paper, we investigate the feasibility of localizing an omni-antenna jammer. We propose Catch the Jammer (CJ), an efficient jammer localization scheme whereby victim nodes at the border of the jammed region share their location information with their one-hop neighbor nodes which further collaborate to find the position of the jammer. This new localization technique first computes a convex hull for the set of victim nodes and further extracts the corresponding minimum covering circle. Simulation results show that CJ outperforms most of the existing localization algorithms depending on the variation of the jammer's transmission range and the position of the jammer.

Online Social Network (OSN) applications and services such as picture sharing, wall posting, and the like, nowadays have a strong impact on the way users interact with each other. Catering for a broad range of users of all ages, and a vast difference in social, educational, and national background, these applications and services allow even users with limited technical skills to share a wide range of personal information with a theoretically unlimited number of partners. This advantage comes at the cost of increased security and privacy exposures for users for two main reasons: first of all, users tend to disclose private personal information with little guard, and secondly, existing OSN applications severely suffer from vulnerabilities in their privacy protection or the lack thereof. The exploitation of these vulnerabilities can lead a malicious user to launch many different types of attacks such as Id theft, profile cloning or secondary data collection. Furthermore, even assuming a perfect protection from such malicious users, legitimate users are still exposed to a major orthogonal privacy threat, since in all existing OSN applications, the service provider has access to all the data including some private information stored and managed by the application itself and can misuse such information easily. Since the access to users' private data is the underpinning of a promising business model, current OSN services are not likely to address this problem in the near future. Researchers recently proposed to design the OSN application based on a peer-to-peer architecture in order to avoid centralized control over users' data. While in one hand a peer-to-peer model seems to be a good candidate to build a privacy preserving solution that avoids centralized control, on the other hand it lacks any a priori trust relationships among parties.

Autonomic and opportunistic communications require specific routing algorithms, like replication-based algorithms or context-based forwarding. Privacy is a major concern for protocols which disseminate the context of their destination. In this paper, we focus on the privacy issue inherent to context-based protocols, in the framework of an original epidemic forwarding scheme, which uses context as a heuristic to limit the replication of messages. We define the achievable privacy level with respect to the trusted communities assumption, and the security implications. Indeed, privacy in such an environment raises challenging problems, which lead us to a solution based on two refinements of identity-based encryption, namely searchable encryption and policy-based encryption. This new solution enables forwarding while preserving privacy by allowing secure partial matches in the header and by enforcing confidentiality of the payload.

Several cooperation enforcement schemes based on rewarding mechanisms such as electronic cash or online credits have lately been proposed to prevent selfish behavior in ad-hoc networks. However, these schemes suffer from the lack of fairness guarantees or the reliance on costly mechanisms such as tamper- proof hardware or the requirement for trusted third parties (TTPs) that are not suitable for ad-hoc networks. In this paper, we present a new cooperation-enforcement scheme that is perfectly suitable for ad-hoc delay-tolerant networks. The protocol is based on a simple technique called hot- potato forwarding whereby in order to receive a packet, potential recipients must first deliver an advance reward to the sender prior to the transmission of the packet. Thanks to this technique cooperation among nodes becomes mandatory and poisoning attacks and cheating actions are inherently prevented. The second contribution in our scheme is an optimistic fair exchange protocol that solves the fairness problem that is inherent to peer rewarding schemes. The protocol achieves total fairness with the help of a TTP and is optimistic in that the TTP is only involved in case of conflict between peer nodes. Correct execution of the protocol does not require any access to the TTP, so fairness is achieved without any impact on well-behaving nodes. The fairness of the protocol is validated through the exhaustive analysis of all possible protocol traces.

Most of the actual group rekeying solutions only deal with security and scalability issues and are severely lacking with respect to reliability and customer satisfaction. We suggested a reliable group rekeying protocol whereby the key server first partitions members with respect to their membership duration and offers a strongly reliable delivery for long-lived members. In this paper, we review this protocol that combines proactive FEC and replication techniques, describe a user-oriented key assignment scheme in order to define FEC blocks. We then analyze the efficiency of the protocol based on different simulations and show that thanks to this protocol, the number of long-lived members losing their keying material is significantly reduced while the communication overhead has slightly increased.

Even though group rekeying is one of the most visited areas in network security, solutions still are severely lacking with respect to reliability and real customer expectations. We first classify secure multicast applications with regard to these expectations and suggest a new approach that defines different recipient categories based on their "loyalty" and that treats each category differently by offering better service to more loyal recipients. We propose to restructure the logical key hierarchy (LKH) scheme by separately regrouping members based on their membership duration, aiming at preserving members with long duration membership from the impact of rekeying operations caused by arrivals or departures of short-lived members. We then describe an extensive method for computing system parameters, like rekeying intervals, based on customer satisfaction criteria.

Networks integrating satellite shared access, such as DVB-S/RCS, are particularly exposed to denial of service (DoS) attacks through which a satellite terminal can maliciously use the Network Control Center (NCC) resources by submitting a high number of bogus requests. Anti-clogging techniques used in terrestrial mesh networks to thwart DoS attacks fall short of solving DoS problems in the case of satellite networks due to the inherent broadcast capability of the communication medium. We suggest a generic mechanism to prevent DoS attacks in control plane protocols. The principle behind this mechanism is an efficient identification technique that allows the satellite servers to quickly discard bogus requests. Thanks to this technique, intruders controlling the total bandwidth of the up-link can only consume a very small fraction of the NCC's resources.