This PDF file contains the front matter associated with SPIE Proceedings Volume 6773, including the Title Page, Copyright information, Table of Contents, Introduction (if any), and the Conference Committee listing.

Handover (or handoff) management is the set of actions that allows a wireless network to maintain a mobile user's connection as it moves through different cells during a call. The handover is typically done during a cell boundary crossing and all of the mobile terminal's connection information must be transferred into the new cell (or new base station). So, it is important to count the number of handovers made by a user during a call for proper network resources dimensioning and performance prediction. Location management is the set of actions aimed to find the current location of an inactive mobile user for call delivery. It implies the transmission of signaling messages for dynamic databases updates, paging, and so on. This problem is formulated in terms of cost optimization and it is mathematically related to handover counting. In this paper, we summarize many of the proposed model to deal with these problems, and we will be mainly focussed on the renewal process approach.

A wireless ad hoc sensor network is a configuration for area surveillance that affords rapid, flexible deployment in arbitrary threat environments. There is no infrastructure support and sensor nodes communicate with each other only when they are in transmission range. The nodes are severely resource-constrained, with limited processing, memory and power capacities and must operate cooperatively to fulfill a common mission in typically unattended modes. In a wireless sensor network (WSN), each sensor at a node can observe locally some underlying physical phenomenon and sends a quantized version of the observation to sink (destination) nodes via wireless links. Since the wireless medium can be easily eavesdropped, links can be compromised by intrusion attacks from nodes that may mount denial-of-service attacks or insert spurious information into routing packets, leading to routing loops, long timeouts, impersonation, and node exhaustion. A cross-layer design based on protocol-layer interactions is proposed for detection and identification of various intrusion attacks on WSN operation. A feature set is formed from selected cross-layer parameters of the WSN protocol to detect and identify security threats due to intrusion attacks. A separate protocol is not constructed from the cross-layer design; instead, security attributes and quantified trust levels at and among nodes established during data exchanges complement customary WSN metrics of energy usage, reliability, route availability, and end-to-end quality-of-service (QoS) provisioning. Statistical pattern recognition algorithms are applied that use observed feature-set patterns observed during network operations, viewed as security audit logs. These algorithms provide the "best" network global performance in the presence of various intrusion attacks. A set of mobile (software) agents distributed at the nodes implement the algorithms, by moving among the layers involved in the network response at each active node and trust neighborhood, collecting parametric information and executing assigned decision tasks. The communications overhead due to security mechanisms and the latency in network response are thus minimized by reducing the need to move large amounts of audit data through resource-limited nodes and by locating detection/identification programs closer to audit data. If network partitioning occurs due to uncoordinated node exhaustion, data compromise or other effects of the attacks, the mobile agents can continue to operate, thereby increasing fault tolerance in the network response to intrusions. Since the mobile agents behave like an ant colony in securing the WSN, published ant colony optimization (ACO) routines and other evolutionary algorithms are adapted to protect network security, using data at and through nodes to create audit records to detect and respond to denial-of-service attacks. Performance evaluations of algorithms are performed by simulation of a few intrusion attacks, such as black hole, flooding, Sybil and others, to validate the ability of the cross-layer algorithms to enable WSNs to survive the attacks. Results are compared for the different algorithms.

The paper describes new method for routing classification of the packets in the networks with static routing. Static routing is mostly used in the stable and well-controlled networks, which are sensitive to the additional complexity introduced by the dynamic routing schemes. For example, it is a straightforward choice for the embedded networks on terminal. The available solutions for the static routing require full-size or longest prefix-based aggregated routing table, which is inefficient taking into account static nature of the performed routing decisions. The standard approach is based on the full implementation of the route lookup procedure, which for every packet performs search for the longest prefix match in the routing table. As a result, it increases implementation complexity of the network end points and switches, requires high speed memory for storing routing tables, and leads to an additional processing delay and energy consumption for each transmitted packet. This work is based on the observation that clever planning of the networks with static routing allows applying new principles of route lookup. As a consequence it allows significantly reduce complexity of the packet classification and forwarding procedures and minimize amount of consumed resources. Complexity reduction of the packets classification and forwarding procedures allows simplifying implementation of the networking part of the protocol stack, which results in reduction of the device components cost, decrease of the power and memory consumption, as well as the packet processing delay. The paper contains a description of the proposed route classification method and discusses its applicability for broad range of networks with static routing.

In this paper we proposes a novel Passive Optical Network (PON) based broadband wireless access network architecture to provide multimedia services (video telephony, video streaming, mobile TV, mobile emails etc) to mobile users. In the conventional wireless access networks, the base stations (Node B) and Radio Network Controllers (RNC) are connected by point to point T1/E1 lines (Iub interface). The T1/E1 lines are expensive and add up to operating costs. Also the resources (transceivers and T1/E1) are designed for peak hours traffic, so most of the time the dedicated resources are idle and wasted. Further more the T1/E1 lines are not capable of supporting bandwidth (BW) required by next generation wireless multimedia services proposed by High Speed Packet Access (HSPA, Rel.5) for Universal Mobile Telecommunications System (UMTS) and Evolution Data only (EV-DO) for Code Division Multiple Access 2000 (CDMA2000). The proposed PON based back haul can provide Giga bit data rates and Iub interface can be dynamically shared by Node Bs. The BW is dynamically allocated and the unused BW from lightly loaded Node Bs is assigned to heavily loaded Node Bs. We also propose a novel algorithm to provide end to end Quality of Service (QoS) (between RNC and user equipment).The algorithm provides QoS bounds in the wired domain as well as in wireless domain with compensation for wireless link errors. Because of the air interface there can be certain times when the user equipment (UE) is unable to communicate with Node B (usually referred to as link error). Since the link errors are bursty and location dependent. For a proposed approach, the scheduler at the Node B maps priorities and weights for QoS into wireless MAC. The compensations for errored links is provided by the swapping of services between the active users and the user data is divided into flows, with flows allowed to lag or lead. The algorithm guarantees (1)delay and throughput for error-free flows,(2)short term fairness among error-free flows,(3)long term fairness among errored and error-free flows,(4)graceful degradation for leading flows and graceful compensation for lagging flows.

Service providers today are facing the challenge of operating and maintaining multiple networks, based on multiple technologies. Network Management System (NMS) solutions are being used to manage these networks. However the NMS is tightly coupled with Element or the Core network components. Hence there are multiple NMS solutions for heterogeneous networks. Current network management solutions are targeted at a variety of independent networks. The wide spread popularity of IP Multimedia Subsystem (IMS) is a clear indication that all of these independent networks will be integrated into a single IP-based infrastructure referred to as Next Generation Networks (NGN) in the near future. The services, network architectures and traffic pattern in NGN will dramatically differ from the current networks. The heterogeneity and complexity in NGN including concepts like Fixed Mobile Convergence will bring a number of challenges to network management. The high degree of complexity accompanying the network element technology necessitates network management systems (NMS) which can utilize this technology to provide more service interfaces while hiding the inherent complexity. As operators begin to add new networks and expand existing networks to support new technologies and products, the necessity of scalable, flexible and functionally rich NMS systems arises. Another important factor influencing NMS architecture is mergers and acquisitions among the key vendors. Ease of integration is a key impediment in the traditional hierarchical NMS architecture. These requirements trigger the need for an architectural framework that will address the NGNM (Next Generation Network Management) issues seamlessly. This paper presents a unique perspective of bringing service orientated architecture (SOA) to legacy network management systems (NMS). It advocates a staged approach in transforming a legacy NMS to SOA. The architecture at each stage is detailed along with the technical advantages and business benefits achieved at each stage of transformation.

In this paper, we compare the power consumption of a "line-of-sight" free space optical (FSO) link and a radio frequency (RF) data link. We investigate a 2.5 Gbps line-of-sight FSO interconnection, which consists of a vertical cavity surface emitting laser (VCSEL) driver (MAX3795), a VCSEL laser diode (LD), a PIN photodiode (PD), a transimpedance amplifier (MAX3864), and a limiting amplifier (MAX3746). It is shown that the total power consumption is about 370.6 mW in simple NRZ on-off keying (OOK) modulation format. Different lens configurations are discussed in terms of the integration/setup efforts and the beam controlling effects. A 250 Kbps commercial radio frequency (RF) link comprised in Tmote sky module (Moteiv Corporation) is explored to compare with the FSO link. The average power supplied to the radio transceiver is about 50.76 mw. The estimated energy consumption for the aforementioned RF link is 2.03×10^-4 mJ/bit, while the end-to-end FSO consumes 1.48×10^-7 mJ/bit. This paper provides design outlines from the aspect of the power consumption of FSO and RF wireless communication technologies for distributed sensor network applications.

Near Field communication (NFC) technology enables a flexible short range communication. It has large amount of envisaged applications in consumer, welfare and industrial sector. Compared with other short range communication technologies such as Bluetooth or Wibree it provides advantages that we will introduce in this paper. In this paper, we present an example of applying NFC technology to industrial application where simple tasks can be automatized and industrial assembly process can be improved radically by replacing manual paperwork and increasing trace of the products during the production.

Load distribution across multiple parallel paths is an important consideration. In many practical contexts, the aggregate traffic from source to sink may be such that no single link can carry the load. In an MPLS domain, this problem can be addressed by instantiating multiple paths. The main objective of this paper balances traffic at the flow level among the parallel Label Switched Paths (LSPs) in MPLS networks. Different from other proposals, our new framework is based on the distributable traffic (DT), where cross-traffic in real networks is considered, and each LSP is modeled as an M/G/1 processor-sharing queue. We define a flow to be a sequence of packet having the same identifier, and dispatch packet belonging to one flow to the same path, so the packet disorder problem is avoided effectively. This mechanism only needs to be implemented in the ingress LSRs and the egress LSRs. A new defined cost function is being used to distribute traffic to path. We computer the cost function based on the delay and packet loss of each LSPs, and minimize the cost function. The minimized cost function is inverse ratio to DT. If the cost function of a certain LSP is smaller, it means that more traffic can be distributed on this LSP. Extensive simulations using NS2 are performed with MPLS modules. Simulation results show that our approach so effective that the throughput is increased significantly and reduces the end-to-end delay and the packet drop rate, and it can distribute the traffic onto parallel LSPs more evenly and fairly.

The energy-constrained nature of wireless ad hoc networks calls for the protocols that use the energy efficiency as its primary design goal. In order to evaluate and compare the energy-aware protocols in terms of their energy efficiency, an energy-consumption model which can accurately compute the energy consumed by the data communication activities is crucial. In this paper, we firstly give a comprehensive summary of the existing energy-consumption models. All energyconsumption models are categorized into three types, and the characteristics of each type are discussed in detail. Secondly, we propose a new efficient energy-consumption model which is a collection of functions of the packet size and the RF (Radio Frequency) power level. The main contribution of our model is that all nodes in the model can change their RF power-rate and radio states (e.g. transmitting, receiving, idle and sleep) according to the communication requirement, and their energy cost can be calculated correctly. Finally, by comparing DSDV against dPAMEEL, we show that our energy model can effectively calculate the energy consumption for different energy-aware protocols.

In this paper, we employ real wireless data that draw from well known archives of network traffic traces and discusses the characteristic of the aggregate WLAN traffic. We test the nonstationary of aggregate WLAN traffic by using the Augmented Dickey-Fuller (ADF) based unit root test. The results show that we cannot reject the unit root for either of all downstream traffic. We also repeat the analysis for the first difference of these series. The findings show that we can strongly reject the unit root. These test reveal that the downstream traffic is nonstationary and integrated of order one. Likewise, the same testing procedure for upstream traffic concluded that some of aggregate upstream traffic series are stationary at the 5% level of significance, and some traffic series are nonstationary at the each level of significance. Our result also showed that aggregate upstream traffic series are also integrated of order one.

Today, the main traffic in peer-to-peer (P2P) network is still multimedia files including large numbers of music files. The study of Music Information Retrieval (MIR) brings out many encouraging achievements in music search area. Nevertheless, the research of music search based on MIR in P2P network is still insufficient. Query by Humming (QBH) is one MIR technology studied for years. In this paper, we present a server based P2P music sharing system which is based on QBH and integrated with a Hierarchical Index Structure (HIS) to enhance the relation between surface data and potential information. HIS automatically evolving depends on the music related items carried by each peer such as midi files, lyrics and so forth. Instead of adding large amount of redundancy, the system generates a bit of index for multiple search input which improves the traditional keyword-based text search mode largely. When network bandwidth, speed, etc. are no longer a bottleneck of internet serve, the accessibility and accuracy of information provided by internet are being more concerned by end users.

Although IPv4 is still working, IPv6 is considered as the backbone and characteristic of the NGI. With the development of Internet, new protocols and network equipments are required to develop. It is necessary to test the new protocols and network equipments extensively before deployment. This paper proposes the design and implementation of RENEW, a useable and accurate network emulator which supports both IPv4 and IPv6 protocols. Besides, it also works on Windows platform. In our IPv6 testbed, we use RENEW to emulate various network characteristics and conditions including bandwidth, delay packet loss and jitter. Compared with the expected values, results are acceptable. Through implementation and experimentation study, we have shown that RENEW does provide the real-time control and change on the parameters of IPv6 network conditions effectively and expediently on Windows. It also gives enough accuracy and more satisfactory convenience to the development and test work for the new protocols.

DHT (Distributed Hash Table) is the key technology in structured P2P. DHT sets up logic topology with certain structure among the nodes and builds relations between nodes and data resources according to some rules. Routing table structure based on B⁺ tree for DHT is presented which builds up the balanced multilevel ordered tree-indexes to the routing tables so as to form levels indexes and ordered link between nodes. In term of this method, key search performs along the ordered indexes in B⁺ tree and the problem of resources location which is the core problem in DHT network is resolved efficiently. This Routing table structure is of advantaged to manage numerous routing information in network and makes the ruleless information well-regulated. It not only enhances the lookup efficiency and achieves the range search but also can control the lookup length in the height of B⁺ tree. When nodes join and depart, index structure can make the maintained routing information few in each node by updating linked message only. At the same time, for linked information of successor node being mainly stored in every node, the storage cost is decreased. It is effective and scalable routing table structure.

Since most ad hoc mobile devices today operate on batteries, energy-aware routings and power control techniques in wireless networks have drawn considerable research interests recently. This paper presents a reliable energy-aware routing algorithm (REARP) for unreliable ad hoc networks. The lifetime of the whole network and the energy cost for each packet are considered simultaneously in the routing processes of the REARP. Different from the formers, the energy cost of the link layer retransmission is also computed into the total energy consumption. Moreover, the REARP appropriately adjusts the transmission power by systematically integrating the reliability and power control techniques. We conducted extensive simulations to evaluate the performance of the new routing algorithms compared to a number of existing routing algorithms.

We propose a congestion controller based on the Proportional-Integral-Differential Neuron Network (PIDNN). As existing controllers, our controller employs the queue size in bottleneck link router as a congestion indicator to trigger packet dropping. The target queue length and the feedback, actual queue length, act as the controller's two input signals. The packet dropping probability is computed by PIDNN controller with its simple embedded algorithm in term of the predefined state function and output function. Thus, the dropping probability decides to drop or to accept an incoming packet so that the queue length is kept at (or near) the target level. This controller's performance is examined under various network configurations, and compared to proposed congestion algorithms, including PI and RED. Our simulation results show that, with comparable simple implementation, this scheme has short response time, better robustness, and more adaptability, especially under highly dynamic network and heavy traffic load.

A robust PID controller for active queue management (AQM) based on modern H_∞ optimal control theory is presented in this paper. Taken both robustness and closed loop performance into consideration, most desirable parameters value can be gotten through some straightforward analytical formulas. Our robust PID controller is determined only by one parameter, other than traditional PID controller is by three or more. Additionally, this new parameters determining method can not only be extended to other AQM controller based on classical control theory or optimal control theory, but also be easily understood and implementation. We evaluate the performances of the controller extensively. The results show that the robust PID congestion controller outperform the existing controller, such as PI, RED, on keeping the router queue size at the target value. The most obvious property of the controller is that it takes on robustness such that it can adapt the network dynamic.

The scheme of dynamic coordination rules in peer-to-peer database uses rule caching and forwarding to successfully solve the dependence tree break problem in the situation that peers can join and leave freely. But there are still problems that weaken the performance of query processing in this scheme. Coordination rules in cache are merged in run time when bypassing break points. If dependence trees can be optimized into a form robust against peer absence beforehand, the query process will be more efficient. This paper proposes such mechanism by doing coordination rule combinations when new peer joins the dependence tree and new forwarded coordination rule arrives in cache. When some peers leave, queries take one existing bypass rule for reformulation, instead of concatenating cached ones from scratch. In effect, this mechanism optimizes dependence tree into a more robust topology whenever new peer joins. Even when there is no peer absence, bypass rules can make query processing more efficient without following through many mediating peers, especially when data are updated frequently and frequent queries are needed. At the same time, the original dependence tree are maintained for data cache query when the target peer is absent. Since dynamic coordination rules are expressed in XSLT, we try to find a way to form one XSLT whose function is equal to a chain of XSLTs, similar to the XML reasoning. The protocol also needs to be improved to inform to launch topology optimization when new peer join or rule changes.

Peer-to-Peer (P2P) systems are currently receiving considerable interest. However, as experience with P2P networks shows, the selfish behaviors of peers may lead to serious problems of P2P network, such as free-riding and white-washing. In order to solve these problems, there are increasing considerations on reputation system design in the study of P2P networks. Most of the existing works is concerning probabilistic estimation or social networks to evaluate the trustworthiness for a peer to others. However, these models can not be efficient all the time. In this paper, our aim is to provide a general mechanism that can maximize P2P networks social welfare in a way of Vickrey-Clarke-Groves family, while assuming every peer in P2P networks is rational and selfish, which means they only concern about their own outcome. This mechanism has some desirable properties using an O(n) algorithm: (1) incentive compatibility, every peer truly report its connection type; (2) individually rationality; and (3) fully decentralized, we design a multiple-principal multiple-agent model, concerning about the service provider and service requester individually.

An important goal in P2P networks is that all peers provide resources. However, free riding and tragedy of common are real issues in P2P networks. To resolve these problems, most of the existing work is concerning probabilistic estimation to evaluate the trustworthiness or mechanism design to provide incentive. Instead of design a protocol to solve free riding, we build a micro-payment architecture for these existing protocols using virtual currency which can be more precisely measured and easily be replaced by reputation or other tokens. Our system can avoid from long-term trust learning interactions and high cost of collecting and analyzing reputation information. It can also provide peers incentive to truly report their connection type and security to malicious attacks.

Efficient organization of the nodes in decentralized peer-to-peer (P2P) networks is a challenging problem, especially in the absence of a global schema. Node clustering is an available way to optimize infrastructure and decrease traffic cost in P2P networks. This paper proposes a Density-based Distributed Node Clustering (DDNC) approach to discovering clusters in P2P networks. This approach is completely distributed, in which each node only depends on the knowledge of its neighbors for node clustering. Unlike other graph based algorithms, the DDNC approach utilizes density of node's neighbor for discovering clusters. For a given node, the DDNC determines its neighbor density by computing the link time with its neighbors, which not only considers the node connectivity but also connection quality. The DDNC scheme can also dynamically adapt its clusters according to the participation and departure of nodes. Experimental results have shown ours scheme's feasibility and efficiency.

Building a semantic search system on top of peer-to-peer (P2P) networks is becoming an attractive and promising alternative scheme for the reason of scalability, Data freshness and search cost. In this paper, we present a Suffix Arrays based algorithm for Semantic Search (SASS) in P2P systems, which generates a distributed Semantic Overlay Network (SONs) construction for full-text search in P2P networks. For each node through the P2P network, SASS distributes document indices based on a set of suffix arrays, by which clusters are created depending on words or phrases shared between documents, therefore, the search cost for a given query is decreased by only scanning semantically related documents. In contrast to recently announced SONs scheme designed by using metadata or predefined-class, SASS is an unsupervised approach for decentralized generation of SONs. SASS is also an incremental, linear time algorithm, which efficiently handle the problem of nodes update in P2P networks. Our simulation results demonstrate that SASS yields high search efficiency in dynamic environments.

Chord which used in structured P2P network is a successful routing algorithm based on DHT (Distributed Hash Table). In Chord nodes locate along the Chord ring by being assigned the node identifiers and data store in corresponding nodes with key identifier. Finger table is built in each node and is maintained when nodes join and depart. The problems in Chord are unidirectional clockwise routing along ring and information redundancy in routing table. An improved method is presented which is bidirectional routing table. Routing can perform along clockwise and anticlockwise according to the locations of the current node and destination node. The next hop direction is optimization that the next node is the nearest one apart from the destination node. This strategy limits the search area in half-ring, reduces the average hops and enhances the search efficiency. The redundancy routing information is deleted in order to decrease the added storage space in bidirectional routing table, thus the items share the routing information.

Clustering or grouping of similar objects is one of the most widely used procedures in data mining, which has received enormous attentions and many methods have been proposed in these recent decades. However these traditional clustering algorithms require all the data objects to be located at one single site where it is analyzed. And such limitation cannot face the challenge as nowadays monstrous sizes of data sets are often stored on different independently working computers connected to each other via local or wide area networks instead of one single site. Therefore in this paper, we propose a fully distributed clustering algorithm, called a fully distributed clustering based on fractal dimension (FDCFD), which enables each site to collaborate in forming a global clustering model with low communication cost. The main idea behind FDCFD is via calculating fractal dimension to group points in a cluster in such a way that none of the points in the cluster changes the cluster's fractal dimension radically. In our theoretical analysis, we will demonstrate that our approach can work very well for clustering data that is inherently distributed, collect information spread over several local sites to form a global clustering meanwhile without communication costs and delays for transmitting.

Today, cyber attacks such as worms, scanning, active attackers are pervasive in Internet. A number of security approaches are proposed to address this problem, among which the intrusion detection system (IDS) appears to be one of the major and most effective solutions for defending against malicious users. Essentially, intrusion detection problem can be generalized as a classification problem, whose goal is to distinguish normal behaviors and anomalies. There are many well-known pattern recognition algorithms for classification purpose. In this paper we describe the details of applying pattern recognition methods to the intrusion detection research field. Experimenting on the KDDCUP 99 data set, we first use information gain metric to reduce the dimensionality of the original feature space. Two supervised methods, the support vector machine as well as the multi-layer neural network have been tested and the results display high detection rate and low false alarm rate, which is promising for real world applications. In addition, three unsupervised methods, Single-Linkage, K-Means, and CLIQUE, are also implemented and evaluated in the paper. The low computational complexity reveals their application in initial data reduction process.

Clustering plays an important role in data mining. It helps to reveal intrinsic structure in data sets with little or no prior knowledge. The approaches of clustering have received great attention in recent years. However many published algorithms fail to do well in determining the number of cluster, finding arbitrary shapes of clusters or identifying the presence of noise. In this paper we present an efficient clustering algorithm which employs the theory of grid, density and fractal that can partition points in the same cluster with minimum change of fractal dimension meanwhile maximizing the self-similarity in the clusters. We show via experiments that FDC can quickly deal with multidimensional large data sets, identify the number of clusters, be capable of recognizing clusters of arbitrary shape and furthermore explore some qualitative information from data sets.