Raimir Holanda Filho

A New Collaborative Approach for Intrusion Detection System on Wireless Sensor Networks

Due to its characteristics, the Wireless Sensor Networks (WSN) are useful in a variety of applications such as environmental monitoring, surveillance, tracking, among others. Given the nature of such applications, where many are mission critical, the WSN become targets of possible attackers interested in harming the network. Among the defense mechanisms, the intrusion detection systems play important role in detecting attacks that can overcome the preventing techniques. This paper proposes a new collaborative and decentralized approach for intrusion detection system. Special nodes, called monitors, will be responsible for monitoring the behavior of neighbor nodes. The malicious activities evidences discovered by each monitor will be shared and correlated with the purpose of increasing the accuracy in detection of intruders. Experiment conducted by simulation show that our solution is effective in reducing the false positives.

Network traffic prediction using PCA and K-means

The growing demand for link bandwidth and node capacity is a frequent phenomenon in IP network backbones. Within this context, traffic prediction is essential for the network operator. Traffic prediction can be undertaken based on link traffic or on origin-destination (OD) traffic which presents better results. This work investigates a methodology for traffic prediction based on multidimensional OD traffic, focusing on the stage of short-term traffic prediction using Principal Components Analysis as a technique for dimensionality reduction and a Local Linear Model based on K-means as a technique for prediction and trend analysis. The results validated with data on a real network present a satisfactory margin of error for use in practical situations.

Sensor Data Security Level Estimation Scheme for Wireless Sensor Networks

Due to their increasing dissemination, wireless sensor networks (WSNs) have become the target of more and more sophisticated attacks, even capable of circumventing both attack detection and prevention mechanisms. This may cause WSN users, who totally trust these security mechanisms, to think that a sensor reading is secure, even when an adversary has corrupted it. For that reason, a scheme capable of estimating the security level (SL) that these mechanisms provide to sensor data is needed, so that users can be aware of the actual security state of this data and can make better decisions on its use. However, existing security estimation schemes proposed for WSNs fully ignore detection mechanisms and analyze solely the security provided by prevention mechanisms. In this context, this work presents the sensor data security estimator (SDSE), a new comprehensive security estimation scheme for WSNs. SDSE is designed for estimating the sensor data security level based on security metrics that analyze both attack prevention and detection mechanisms. In order to validate our proposed scheme, we have carried out extensive simulations that show the high accuracy of SDSE estimates.

Internet traffic classification using a Hidden Markov Model

This paper examines the performance of a new Hidden Markov Model (HMM) structure used as the core of an Internet traffic classsifier and compares the results against other models present in the literature. Traffic modeling and classification find importance in many areas such as bandwidth management, traffic analysis, prediction and engineering, network planning, Quality of Service provisioning and anomalous traffic detection. The new HMM structure, which takes into account the packet payload size (PS) and the inter-packet times (IPT) sequences, is obtained by concatenation of a first part which is framed with a HMM profile with another part whose structure is that of a fully-connected HMM. The first part captures the specific properties of the initial protocol packets while the second part captures the statistical properties of the whole sequence present in the flow. Models generated are found to increase the accurate in classifying different traffic classes in the analysed dataset. The average accuracy obtained by the classifier is 62.5% having seen only five packets, 80.0% after examining 13 packets and 95.5% after seeing the unidirectional entire flow.

WSN Routing: An Geocast Approach for Reducing Consumption Energy

Wireless Sensor Networks are ad hoc wireless networks that present hardware and energy reserve restrictions. In general, this kind of network can be utilized as powerful tools in order to monitor and, eventually, control an environment. The routing protocols, for the WSN, must have self-configuration characteristics aimed at discovering which is the best way for transferring the information, with delivery assurance and with minimum energy consumption, among the nodes that compose the network. This article proposes a modification of the Directed Diffusion routing protocol so as to reduce the energy consumption in the network when there is an occurrence of failures. The proposal utilizes a Geocast approach to repair broken paths by constructing a new routing tree. The performed simulations demonstrated that our proposal is more efficient from the energy viewpoint.

A time and financial loss estimation using a highly parallel scheduling model for IT change management

The IT governance is defined as a set of rules, activities and processes that fits with the IT company strategy, ensuring the return on various aspects provided by IT in terms of services and maintainability of the organization. Currently, change management appears as a sensible point into the IT governance. The main activity of change management is related to scheduling definition. This scheduling definition consists in allocate a set of changes to each change window. The current literature has some works that argues about a sequential and parallel implementation of changes allocated into each change window. This work proposes a new model that presents two important contributions: The first contribution is related to scheduling optimization with a model of parallel implementation of changes in agreement with BDIM concepts, resulting in a higher number of changes in each window and consequently reducing the time to implement the change. The second contribution consists in determine the total implementation time of each family of changes, considering the concept of indirect dependency presented in this work.

An integration of fuzzy inference systems and Genetic Algorithms for Wireless Sensor Networks

Wireless sensor networks WSNs consist of a set of sensor nodes in order to detect and transmit environmental characteristics, such as, temperature, humidity or lightness. These sensor nodes, after capturing an event, should communicate with a special node, named sink node. However, the use of a single sink node implies a bottleneck in the sensor network, specially for real time applications. To overcome this problem, researches are focusing on studies for the selection of routes in sensor networks with multiple sink nodes. The approach proposed by this paper presents the application of Genetic Fuzzy Systems GFSs to estimate the quality of routes in WSNs, in order to ensure communication between multiple sensor nodes and multiple sink nodes. A Mamdani Fuzzy Inference System is used to select the best sink node for communication at a given moment, based on network features such as the available energy of the route and the number of hops to reaches the sink node. Genetic Algorithms GAs are used to obtain the optimal setting of design parameters of the Mamdani fuzzy inference system. The proposed route classification was implemented by computer simulations to demonstrate its feasibility and the results showed a sensor network with longer lifetime, based on the appropriate selection of the sink and the route used to send packets through the network.

Online traffic classification based on sub-flows

Traffic classification by application class provides useful information for various tasks of network engineering and administration. However, offline classification of flows has limited its practical application to auditing tasks, long-term planning and other analytical issues. Therefore, research on traffic classification now moves towards the search for accurate and efficient methods of classification in order to meet online tasks such as traffic monitoring and shaping and other specific-application operations. In this work we apply the One-Against-All Approach (OAA) for two online classification strategies based on statistical features of TCP sub-flows. One uses the first N packets of the bi-directional TCP session and the other applies to sub-flows of the N packets starting at a random position in the flow. In our variant of the OAA approach, the problem of classifying an object in one of M classes is reduced to M binary classification problems with an associated decision rule, with each of them possibly using a different subset of features and sub-flow size. We investigated the effect of variation in the amount of N on the results of classification and the smaller set of variables in each of the above problems. This study used the Naïve Bayes classifier.

A hybrid approach based on genetic fuzzy systems for Wireless Sensor Networks

Wireless sensor networks (WSNs) are composed of sensor nodes in order to detect and transmit features from the physical environment. Generally, the sensor nodes transmit informations to a special node, called sink. The use of an unique sink represents a bottleneck in a network, especially for applications in real time. In this sense, some researches have directed studies to the use of multiple sinks. The approach proposed by this paper presents the application of Genetic Fuzzy System (GFS) for the selection of routes in WNSs, in order to make the communication between multiple sensor nodes and multiple sink nodes. Fuzzy Inference System of Mamdani are used to determine the most appropriate sink node through consideration of some characteristics of the sensors network, such as energy and number of hops. Genetic Algorithms are employed to obtain the optimal adjustment of Mamdanis fuzzy inference system parameters. By applying GAs, we intend to achieve both a fuzzy database and a fuzzy rules base to maximize performance of the application of Mamdanis inference system in the selection of routes in Wireless Sensor Networks. The proposed route selection was applied by means of computer simulations to demonstrate the feasibility of the approach implemented. The results obtained through simulations demonstrated a sensor network with a longer lifetime, through the choice of the adequate sink used for sending packets through the network in order to find the best routes.

Automated design of fuzzy rule base using ant colony optimization for improving the performance in Wireless Sensor Networks

The Wireless Sensor Networks (WSNs) are composed of small sensor nodes capable of sensing (collecting), processing and transmitting data related to some phenomenon in the environment. The sensor nodes have severe constraints, such as: limited power supply, low network bandwidth, short wireless communication range, and limited CPU processing and memory storage. Communication in WSN consumes more energy than sensing and processing performed by the network nodes. Therefore, as the sensor nodes are battery-powered and recharging or replacing batteries, in most cases, is infeasible, maximizing the benefits of limited resources in WSNs have become one relevant and challenging issue. The WSN routing protocols must have autoconfiguration features in order to find out which is the best route for communication, thus increasing delivery assurance and decreasing the energy consumption between nodes that comprise the network. This paper presents a proposal for estimating the quality of routes using fuzzy systems to assist the Directed Diffusion routing protocol. The fuzzy system is used to estimate the degree of the route quality, based on the number of hops and the lowest energy level among the nodes that form the route. An Ant Colony Optimization (ACO) algorithm is used to adjust in an automatic way the rule base of the fuzzy system in order to improve the classification strategy of routes, hence increasing the energy efficiency of the network. The simulations showed that the proposal is effective from the point of view of the packet loss rate, the necessary time to send a specific number of messages to the sink node and the lifetime of the first sensor node, which is defined as the period that the first sensor node die due to the battery depletion.