Natalija Vlajic

Wireless sensor networks: to cluster or not to cluster?

The key challenge in the design and operation of wireless sensor networks (WSNs) is the maximization of system lifetime. Node clustering is commonly considered as one of the most promising techniques for dealing with the given challenge, and as such has been referred to by many researchers. It is interesting to observe, however, that very few, if any, published research works provide explicit analysis of node clustering in WSNs and/or manage to prove its actual effectiveness. In this paper we take a closer analytical look at WSNs of clustered organization. We prove that these networks do not necessarily outperform non-clustered WSNs. The condition that ensures superior performance of clustered WSNs, with absolute certainty, is that the formed clusters lie within the isoclusters of the monitored phenomenon. We also show that in clustered WSNs which satisfy the given condition, cluster sizes do not need to match the sizes of their respective underlying isoclusters. Instead, simple 5-hop clusters can provide near-optimal network performance under a wide range of cluster-to-sink and cluster-to-isocluster spatial arrangements

Vector quantization of images using modified adaptive resonance algorithm for hierarchical clustering

A modified adaptive resonance theory (ART2) learning algorithm, which we employ in this paper, belongs to the family of NN algorithms whose main goal is the discovery of input data clusters, without considering their actual size. This feature makes the modified ART2 algorithm very convenient for image compression tasks, particularly when dealing with images with large background areas containing few details. Moreover, due to the ability to produce hierarchical quantization (clustering), the modified ART2 algorithm is proved to significantly reduce the computation time required for coding, and therefore enhance the overall compression process. Examples of the results obtained are presented, suggesting the benefits of using this algorithm for the purpose of VQ, i.e., image compression, over the other NN learning algorithms.

Strategies for improving performance of IEEE 802.15.4/ZigBee WSNs with path-constrained mobile sink(s)

Most of the existing works on the topic of sink mobility in wireless sensor networks (WSNs) are of purely theoretical nature. The aim of this paper is to discuss the challenges as well as potential benefits associated with the use of mobile sinks in WSNs that operate in space-constrained environments and employ real-world technology. Specifically, we examine the pros and cons of deploying path-constrained sink mobility in the framework of IEEE 802.15.4/ZigBee enabled sensor networks. The main contributions of this paper are as follows: First, we demonstrate that the advantages of deploying path-constrained sink mobility, as identified in one of our earlier works [4], are not fully applicable to ZigBee WSNs. Specifically, our OPNET-based simulation study shows that in ZigBee WSNs the findings from [4] hold only conceptually, at the highest level of user-data routing. However, once all of the mobility-related overhead is accounted for, no actual benefit of deploying a mobile-over deploying a static-sink can be observed. Subsequently, we propose the use of three mechanisms for control of mobility-related overhead in ZigBee WSNs: Suppressed Route Discover, Node Association Based on Residual Energy, and Footprint Chaining. The most complex of the three mechanisms (Footprint Chaining) is studied in detail, and conditions under which this technique achieves optimal performance are precisely identified. The presented simulation results prove that with the three proposed mechanisms in place the benefits of mobile-over static-sink deployment can be regained, almost to the same extent as theoretically identified in [4]. To our knowledge, this paper is one of the first attempts to bring the topics of path-constrained sink mobility and ZigBee standard together. It is also the first published work to propose improvements to the current ZigBee standard specifically targeted for WSNs that involve the use of mobile sinks.

Unsupervised Clustering of Web Sessions to Detect Malicious and Non-malicious Website Users

Abstract Denial of Service (DoS) attacks are recognized as one of the most damaging attacks on the Internet security today. Recently, malicious web crawlers have been used to execute automated DoS attacks on web sites across the WWW. In this study, we examine the use of two unsupervised neural network (NN) learning algorithms for the purpose web-log analysis: the Self- Organizing Map (SOM) and Modified Adaptive Resonance Theory 2 (Modified ART2). In particular, through the use of SOM and Modified ART2, our work aims to obtain a better insight into the types and distribution of visitors to a public web-site based on their link-traversal behaviour, as well as to investigate the relative differences and/or similarities between malicious web crawlers and other non-malicious visitor groups. The results of our study show that, even though there is a pretty clear separation between malicious web-crawlers and other visitor groups, around 8% of malicious crawlers exhibit very ‘human-like’ browsing behaviour and as such pose a particular challenge for future web-site security systems.

Performance of IEEE 802.15.4 in wireless sensor networks with a mobile sink implementing various mobility strategies

In this work, we investigate the advantages and challenges of deploying a single mobile sink in IEEE 802.15.4/ZigBee wireless sensor networks (WSNs). The first part of the paper provides an overview of the most recent research on sink mobility in WSNs, placing a special emphasis on different types of sink mobility (random, predictable and controlled) and discussing the application scenarios most suitable for their respective deployment. In the second part of the paper, our OPNET model for simulation of large-scale and ZigBee-based wireless sensor networks is presented. The model enables effective evaluation of random and predictable sink mobility under varying conditions and forms of routing in the underlying ZigBee WSN. The results obtained using this model show that in terms of energy efficiency ZigBeepsilas tree-based routing outperforms ZigBeepsilas mesh routing, both in the case of random and predictable sink mobility. At the same time, under both mobility models, tree-based routing generates longer delays in the delivery of data reporting packets. Furthermore, when compared against each other assuming identical network conditions, random mobility is shown to achieve higher energy efficiency and shorter packet delays than predictable mobility.

Sink mobility in wireless sensor networks: When theory meets reality

The use of sink mobility in wireless sensor networks (WSN) is commonly recognized as one of the most effective means of load balancing, ultimately leading to fewer failed nodes and longer network lifetime.

Performance Analysis of ZigBee-Based Wireless Sensor Networks with Path-Constrained Mobile Sink(s)

In the wireless sensor network (WSN) literature, the use of a mobile sink is commonly viewed as one of the most successful means of load balancing as well as an effective defense against the so-called hot-spot phenomenon. The aim of this paper is to investigate the real-world applicability of the known theoretical benefits associated with the use of mobile sink(s). In particular, we examine the pros and cons of deploying path-constrained mobile sink(s) in IEEE 802.15.4 / ZigBee-based WSNs.The main contributions of this paper are as follows: First, analytically and through simulation, we demonstrate that in idealistic (zero-overhead) networks the use of a mobile sink does result in a more even distribution of routing load and longer network lifetime, as earlier suggested in the WSN literature. Moreover, in small- to medium- size zero-overhead WSNs, the outer-peripheral and the diagonal-cross trajectory appear to be more effective than other types of mobile-sink trajectories. Unfortunately, real-world networks, including ZigBee WSNs, are not zero-overhead – these networks employ special mechanisms that generate additional (overhead) traffic in order to manage congestion and node mobility. The results of our OPNET-base simulation study demonstrate that in IEEE 802.15.4 / ZigBee WSNs, once all of the overhead traffic is accounted for, the theoretical advantage of deploying a mobile- vs. deploying a static- sink completely disappears. Hence, for anybody contemplating the use of a mobile sink in ZigBee sensor networks, the minimization of protocol overhead may have to be the first course of action. In the last part of the paper, we introduce two simple mechanisms for reduction of mobility-related overhead in ZigBee WSNs. The presented simulation results suggest that with these mechanisms in place the superiority of mobile- over static- sink deployment can be regained.

Self-Healing Wireless Sensor Networks: Results That May Surprise

According to a recent DARPA announcement, self-healing - the ability of a network to effectively combat coverage and routing holes and network disconnection - represents one of the most desired operational properties of wireless sensor networks (WSNs) for military applications. Although previously considered in the WSN literature, the concept of network self- healing, and specifically self-healing by means of mobile nodes, still remains a greatly understudied research area. hi this paper, by focusing on one particular type of holes - routing holes, the energy aspect of combating these holes through the deployment of a single mobile (super) node is discussed. The specific contributions of the paper include: 1) It is proven that although bridging a routing hole by means of a mobile node may seems very intuitive, the deployment of the mobile is often hard to formally justify. For instance, the use of the mobile turns out to be completely energy unjustifiable in all circle- and square- like shaped holes, regardless of their actual size or number of boundary nodes actively involved in routing. Accordingly, the need to consider other parameters, such as overall transmission delay or static-node failure, when deciding whether/where to deploy the mobile, is demonstrated. 2) Building on the results of 1), we propose OPlaMoN - a simple distributed algorithm for determining the Optimal Placement of a Mobile Node within a routing hole of any arbitrary topology. As the name implies, the algorithm solves a rather complex optimization problem by breaking it into smaller fragments which are, then, partially solved by individual nodes. The final solution is reached through a cooperative decision-making process, assuming a minimum exchange of information among the effected nodes. The algorithm has excellent energy conserving properties and, as such, is highly suited for WSN environments. We believe the findings of this paper can serve as a good starting point and encourage further research on the deployment of mobile nodes for the purpose of self-healing in wireless sensor networks.

Categorizing Web pages using modified ART

This paper describes a modified term-frequency/inverse document-frequency model together with a modified adaptive resonance theory algorithm, incorporated into an autonomous agent for rapid and accurate hypertext clustering. The agent has been implemented in Java, and tested on various Web data collections.

An adaptive neural network approach to hypertext clustering

The WWW is an online hypertextual collection, and a more sophisticated algorithm for Web page clustering may have to be based on combined term-similarity and hyperlink-similarity measures. It has been observed that nearly all currently employed techniques for document classification on the Web make use of textual information only. In addition, most of these techniques are incapable of discovering the real nature of the collection to which they are applied due to rather inefficient clustering algorithms employed. This paper describes a novel technique for hypertext clustering, called an adaptive hypertext clustering (AHC) algorithm. This algorithm has been derived from a modified neural network algorithm, and adjusted to the problem of combined term-similarity and hyperlink-similarity measures. The results presented in the paper show that AHC can be easily adapted to enable the most appropriate Web page classification within collections of various thematic and functional profiles, suggesting its main benefits over the traditional techniques.