Ismaïl Biskri

Decision support protocol for intrusion detection in VANETs

Vehicular Ad hoc Networks (VANETs) are so difficult to secure due to the wireless technology and its several known security holes. To protect against attacks, methods and techniques have been developed. The Intrusion Detection System (IDS) can detect malicious actions made to the system. In vehicular ad hoc networks, IDSs are in charge of analyzing incoming and outgoing packets to identify malicious signatures. However, without a decision making mechanism, they are useless. This paper designs a decision making protocol for security information in VANETs. Our study is based on two IDS approaches. In the first one, the IDS are installed on vehicles, while in the second one they are installed on the Road Side Units (RSU). In both approaches, vehicles are grouped according to their speed. Corroboration of an attack is based on a probabilistic model of ratio computation between vehicles or RSUs having answered to the signature of the attack. Our aim is to design a decision support mechanism. The dynamic topology of VANET allows a strong prevention by broadcasting the information. So when an attack occurs, the protocol allows the corroboration of the latter and alert neighboring clusters.

Protocol of change pseudonyms for VANETs

We propose in this paper a security protocol based on periodic change of pseudonyms. The idea is to avoid illegal traceability of vehicles during their communications and preserve their privacy and confidential information. Two different approaches are proposed. In the first approach, each vehicle asks the central authority a new communication pseudonym after a time t. While in the second approach, each vehicle generates itself after a time t, a new communication pseudonym. Our objective is to permit at least two vehicles to change their pseudonym in the same time interval. We evaluate in this work, the bandwidth used by considering the vehicles speed in each approach. The proposed protocol is based on equidistant distribution of the road side unit and uses the average of speed permitted on the road to evaluate lifetime t of the communications pseudonyms and certificates. The exchange of information is based on asymmetric and symmetric cryptography scheme and it uses hash function. Our protocol provides authentication, non-repudiation and privacy.

A lightweight key management scheme based on an Adelson-Velskii and Landis tree and elliptic curve cryptography for wireless sensor networks

Wireless sensor networks are increasingly used in most varied fields such as environment, health, and military. Often, information transmitted on these networks requires encryption to maintain confidentiality, integrity, and non‐repudiation. But encryption techniques used to encrypt data on wired networks are not suitable for sensor networks that consist of small nodes equipped with limited resources. In this paper; we propose a security method for wireless sensor networks that provides good protection while taking into account the limited resources of the sensors. This method is based on an effective key management scheme with a minimum storage of keys. It is based on the combination and improvement of two approaches already proposed by the research community: cryptography based on elliptic curves and key management based on an Adelson‐Velskii and Landis tree. Compared with RECC ‘a routing‐driven elliptic curve cryptography based key management scheme for heterogeneous sensor networks’ and CECKM ‘high‐effect key management associated with secure data transmission approaches in sensor networks using a hierarchical‐based cluster elliptic curve key agreement’, two methods based on Diffie–Hellman elliptic curve cryptography method, our method reduces energy consumption, storage memory, and extends the lifetime of the sensor network. Our simulation results illustrate that our approach saves significant time and memory and reduces the number of exchanged packets during keys installation phase. Also, it requires fewer processing operations and maintains the scalability of the network. Concurrency and Computation: Practice and Experience, 2013.© 2013 Wiley Periodicals, Inc.

Extraction of Strong Associations in Classes of Similarities

Several algorithms are proposed to support the process of automated classification of textual documents. Each of these algorithms has characteristics that influence the classification result. Depending on the amount and nature of the data submitted, the quality of results may vary considerably from one algorithm to another. The generated classes are often noisy. In addition, the number of classes created can be significant. These constraints can easily become barriers to data analysis. This paper presents a method that exploits the notion of association rules to extract regularities in the classes of similarities produced by classifiers.

Proof of concept of a security based on lifetime of communication's pseudonyms for the VANETs

To make Vehicular Ad Hoc Network (VANET) applications useful to the users, the security problem must be solved. Recent researches have suggested the use of a set of anonymous keys certified by the issuing CA (Central authority) to preserve privacy, authentication, and confidentiality of the communicating entities. But how to determine the right time for the vehicles to exchange their communication pseudonyms and what would be the impacts on the network resources such as time processing and memory. In this paper, we propose a protocol that preserves authentication, non repudiation, and location privacy and helps vehicles to exchange their pseudonyms at roughly the same time. It is based on calculating the Euclidean distance and the average of the speed permitted on the path to evaluate the lifetime of the communications pseudonyms. The exchange of the information is based on asymmetric and symmetric cryptography scheme and it uses hash function. The protocol permits to determine the expiration time of pseudonyms and how to make the distribution of all the road side units along the road in order to establish a good communication between them and the vehicles.

Text classification: A preferred tool for audio file classification

This paper presents a general outline related to the numerical classification generalization applied to sound documents. The approach put forward combines concepts pertaining to natural language processing and frequency analysis. The raw data of the audio documents are converted into strings of alphanumerical characters that are well suited to numerical analysis. Thus, processing similar to that employed to classify text documents can be applied. Combining text and audio analyses makes it possible to generate a system that can categorize large corpuses of heterogeneous documents.

APPLICATIVE AND COMBINATORY CATEGORIAL GRAMMAR AND SUBORDINATE CONSTRUCTIONS IN FRENCH

In this article we will present a classification and an analysis, by means of Applicative and Combinatory Categorial Grammar (ACCG), of relative, completive and indirect interrogative propositions ...

COORDINATION AND APPLICATIVE CATEGORIAL TYPE LOGIC

Despite extensive theoretical work carried out on coordination during the last three decades, much research is still devoted to proving the merits of one model over another. In our article, we will show how the coordination process in French (with using the conjunction et (and)) can be explained through Applicative and Combinatory Categorial Grammar (ACCG).

CLASSIFICATION AND CATEGORIZATION IN COMPUTER-ASSISTED READING AND TEXT ANALYSIS

Automated computer classification and categorization have been successfully applied in the field of document retrieval and text mining. These methods are increasingly used in the field of computer-assisted reading and analysis of texts (CARAT) in the field of the humanities and social sciences. Here, the primary aim is not to tag the document for better recall, but to help a specialized reader and analyst to navigate through the themes, concepts, and content of a textual corpus according to a particular research objective. This chapter presents some of the methods and applications of these techniques.

A Flexible Approach for Text Processing Engineering

In this paper, we present a flexible, modular, consistent, and coherent approach for text processing engineering. Each, processing chain dedicated to text processing would be regarded as a serial or parallel assembly of modules underlying particular tasks that a user wants to apply to a text. A user, according to his/her needs and perspectives, might want build and validate his/her own processing chain by assembling a set of modules in a certain configuration. We propose, in this paper, a theoretical formal system based on the model of the typed applicative grammars and the combinatory logic. This approach allow to provide a general framework in which a user would be able to build multiple text analysis processes according to his/her determined objectives. It will also systematize the verification of the logical consistency of the sequence of modules in the assembly that characterizes a given processing chain.