Mobile Ad-hoc Networks face challenges in achieving efficient multicasting due to dynamic topology changes and unreliable links. Existing multicast approaches either suffer from low packet delivery ratio or high overhead. These approaches rely on simple metrics like hop count to find the optimal path to the destination. Once the path is selected, all packets are sent over the same path as long as it remains available. However, a path that is deemed optimal at a specific instance of time may not retain its optimality at a subsequent moment due to node mobility. Moreover, using a metric like hop count that does not consider link quality can lead to poor packet delivery ratio, as it can favor an unreliable path over a reliable one just because it is the shortest. To tackle these concerns, a Q-Learning Adaptive Multicast Ad-hoc On-Demand Distance Vector routing protocol is proposed. It is an adaptive and bandwidth-efficient solution that utilizes link reliability as a routing metric instead of hop count, aiming to build a more stable multicast tree. By leveraging Q-learning principles, the proposed protocol continuously updates path costs to detect any deterioration. Additionally, the protocol dynamically explores the network using periodic group hello messages, enabling the identification of alternative paths and proactively switches to them if path costs deteriorate. Simulations conducted in Network Simulator 3 demonstrate the superiority of the proposed protocol over the traditional Multicast Ad-hoc On-Demand Distance Vector protocol. Furthermore, it outperforms a modified version, called Multicast Ad-hoc On-Demand Distance Vector-Route Reliability, that uses link reliability as a metric, demonstrating enhanced packet delivery ratio and reduced multicast-related overhead.

Resolving network failures after they occur through human investigation is a costly and time-consuming process. Predicting upcoming failures could mitigate this to a large extent. In this work, we collect data from a large intercontinental network and study the problem of flapping links, which are indicative of link failures. Such flapping links have their routing metric increased to divert traffic away; this is followed by corrective actions, and eventually their routing metric is lowered again to carry traffic. Using the collected data, primarily metrics reported from Internet Protocol (IP) and optical layers of the network, we develop ML models to predict upcoming link failures. Exploring a sequence of increasingly complex models, we study the relevance of optical metrics, the underlying temporal relations, and the topological relations in improving the predictive model performance. We discovered that optical features such as optical maximum and minimum power or unavailable and errored seconds increased the model’s performance (measured in average precision) by about 9 percentage points while temporal and spatial features improved it further by 8 and 7 percentage points respectively for a total improvement of 24 percentage points. • Optical metrics, in addition to IP metrics, are important when building predictive link failure models. • Capturing time-based relations improves the predictive performance. • Similarly, aggregating data from neighboring links via GNNs is beneficial. • Combining spatial and temporal dimensions, we can develop predictive link failure models with high performance.

Current research on Network Digital Twins (NDT) focuses on building models that predict the behavior of the network on simulations rather than building a complete NDT that automatically monitors a physical network and optimizes it according to its current state and traffic flows. In this paper, we implement an online learning NDT for Interior Gateway Protocol (IGP) weight optimization that monitors a physical network to find IGP weight configurations based on a selected objective, such as minimizing average traffic delay, and automatically applies them to the network. We addressed the practical issues of implementing a NDT on a real network, such as detecting and dealing with an improperly trained digital model of the network. Our results indicate that our NDT can effectively optimize a physical testbed and that fine-tuning its digital model on high traffic scenarios improves the performance of the NDT on a small network.

Current research on Network Digital Twins (NDT) focuses on building models that predict the behavior of the network on simulations rather than building a complete NDT that automatically monitors a physical network and optimizes it according to its current state and traffic flows. In this proposal, we demo a NDT for Interior Gateway Protocol (IGP) weight optimization on a physical network of 7 Cisco routers and an IXIA generator connected via Ethernet links. We will show how a link failure in this network results in a large packet loss, and how our NDT automatically optimizes the IGP weights of the network to reduce packet loss. Our demo presents a practical implementation of a NDT, where the audience can see a self-adapting network without human interference.

Being able to predict link failures in advance would be of great benefit to network operators. We use Machine Learning (ML) techniques to extract temporal and spatial relations from real network data and use them to predict link failures. We use Interior Gateway Protocol (IGP) configuration changes as a guide to achieve this. We predict link failures in the next five days based on data collected from the previous five days. We propose a modified Variational Auto Encoder (VAE) model to compress the higher dimensional dataset into a latent space that captures time-based relations in the data. We demonstrate that five days is the smallest look-back window of time required to get satisfactory prediction results. Using feature importance plots, we learned that the VAE model was able to capture intricate time-based dependencies in the error counter features to achieve good performance. In addition, using a Graph Convolutional Network (GCN), we were able to aggregate data from neighboring links to improve the model's performance. Neighbors up to two hops away carried relevant information in IGP metric settings and in traffic metric counter features. The relevance of the correlation of the features in time and space is confirmed using standard feature importance wrapper methods. Finally, by combining the VAE and GCN components, we were able to extract spatial and temporal features in conjunction, leading to further improvements. These ML approaches significantly improve existing manual methods of tracking metrics in time and space currently followed by the operator.

Industry 4.0 applications require communication networks with deterministic and bounded latency. Ethernet, augmented with the IEEE Time-Sensitive Networking (TSN) standards, can meet these requirements. One of the TSN standards, Time Aware Shaper (TAS) is ideal for periodic time-sensitive traffic in wired TSN networks. However, it relies on network-wide synchronization to a common reference time. Existing solutions for developing TAS schedules assume perfect synchronization or consider a worst-case error. This study highlights the need for using network-derived time synchronization parameters and errors to optimize TAS schedules. It raises the importance of accurate time synchronization information and proposes a novel methodology for efficient band-width utilization while ensuring deadline requirements are met for time-sensitive traffic in realistic networks. The effectiveness of the proposed framework is validated by simulating a TSN network in OMNeT++ and comparing the devised TAS schedules with schedules which use existing assumptions.

This paper proposes a reliable multicast routing protocol based on Q-learning for wireless ad-hoc networks. The proposed protocol has two goals: 1) enhance the reliability of data delivery and 2) reduce the overhead caused by multicast routing. To achieve these goals, the protocol uses link reliability as a routing metric. The protocol chooses the most reliable path for data transmission based on its Q-value. In addition, it continuously updates the Q-value of active paths and proactively switches to another path if the current path becomes less reliable. To evaluate the performance of the proposed protocol, simulations were conducted using Network Simulator 3 (NS-3). The performance of the proposed protocol was compared with the Multicast Ad-hoc On-demand Distance Vector (MAODV) protocol. The simulation results show that the proposed protocol effectively enhances reliability as it outperforms the MAODV routing protocol in terms of Packet Delivery Ratio (PDR). Moreover, it reduces the control overhead caused by multicast routing.

Hierarchical network architectures are widely deployed to reduce routing overheads and increase scalability. In our work, we are interested in large-scale Mobile Ad-Hoc Networks (MANETs) which are formed by interconnecting smaller clusters through a backbone. To support end-to-end routing in such networks, we employ a hierarchical approach as follows. The clusters are MANETs, running OLSR locally. Each cluster has a gateway, and the gateways are interconnected through a backbone. In this paper, we study three distinct solutions to provide end-to-end connectivity through the backbone: flooding all data packets through the backbone, modifying a pro-active MANET routing protocol, and modifying an on-demand MANET routing protocol. Running extensive simulations in OMNeT++, our results highlight the strengths and weaknesses of each approach.

We study LoRaWAN's performance when multiple applications are concurrently running over the same LoRaWAN network. We consider applications that generate data packets using a Poisson process, a random distribution, and at periodic intervals. The LoRa PHY layer supports a number of communication settings. However, here we focus on two specific settings: the setting recommended by LoRaWAN and the setting that yields the highest possible data rate in LoRa. Our results demonstrate the following: (i) LoRaWAN favours applications that generate packets at a higher periodic rate, (ii) LoRAWAN does not favour applications that generate packets at a higher rate under Poisson and uniform random distribution, (iii) LoRaWAN's recommended PHY setting demonstrates poor performance, (iv) LoRa's fastest data rate setting outperforms the LoRaWAN recommended setting, and (v) LoRaWAN favours applications that generate packets using uniform random and Poisson distributions over application that generates packet at periodic interval. Our results also hint that using multi-hop communication along with the LoRa's fastest data rate setting can not only increase the setting's coverage, but it may still deliver better performance relative to the LoRaWAN's recommended setting.

SDN technology has been applied to a range of different networks, ranging from Ethernet services to large cloud environments. More recently, interest has turned towards extending programmability of Optical Transport Networks (OTN). In the SDN architecture, SBIs are used to communicate between the SDN controller and the switches or routers in the network. In this paper, we deploy OpenFlow and NETCONF as SBIs in managing BoD across interconnected data centers over OTN. More specifically, we use these protocols to communicate between an OpenDayLight controller and two BTI7800 network elements that interconnect the data centers. We present experimental results for both BTI's YANG-based NETCONF implementation and our port of OpenFlow for a number of use cases. Our results show that NETCONF is faster and requires fewer control message. However, OpenFlow offers better bandwidth utilization over the interconnecting link.

We study the problem of jointly solving the contention and congestion distributed control problem in a bounded queue multihop mobile ad-hoc networks. Unlike the majority of the published work in this area, we focus on the feasibility of the proposed solution in case of random static and dynamic networks considering the signaling and overheads. In recent years a number of papers have presented solutions to this problem that are based on network utility maximization algorithms. However, this work typically necessitates either complex computations, heavy signaling/control overhead, and/or approximated suboptimal results. In this paper, we combine a specific network utility maximization problem with a simple and efficient Active Queue Management (AQM) mechanism that we believe is appropriate for mobile ad-hoc networks. Using IEEE 802.11 protocol as MAC layer protocol, we show via NS-3 simulations that the proposed Cross-Layer Design (CLD) significantly outperforms standard protocols such as TFRC in static and dynamic networks.

Internet Engineering Task Force (IETF) standardized RPL as a routing framework for the Internet of Things (IoT). RPL supports many-to-one, many-to-many, and one-to-many communications. Recent studies demonstrate that RPL performs poorly in many-to-many and one-to-many communications due to the additional control overhead associated with discovering many-to-many and one-to-many forwarding paths and its non storing mode of operation (MoP). In this paper, we present a routing framework for IoT (IoT-RF) that supports many-to-one, many-to-many, and one-to-many communications. The framework works with the storing MoP, and suggests the use of multiple gateways in a network to overcome memory limitations of the storing MoP. Moreover, the framework proposes two protocols to discover many-to-many and one-to-many forwarding paths. One of the proposed protocols discovers many-to-many and one-to-many forwarding paths through many-to-one communication, hence there is no additional control overhead associated with the protocol. The other protocol uses empty control messages to discover many-to-many and one-to-many forwarding paths. We implement our IoT-RF in the Contiki operating system. We evaluate our IoT-RF against a de facto routing framework for IoT using Cooja-based emulation experiments. Our results demonstrate that in many-to-one, many-to-many, and one-to-many communications our IoT-RF significantly outperforms the de facto routing framework in terms of the packet delivery ratio, end-to-end packet delivery delay, and control overhead.

Since Internet of Things (IoT) applications are deployed within open physical environments, their executions suffer from a wide spectrum of uncertain factors (e.g., network delay, sensor inputs). Although ThingML is a promising IoT modeling and specification language which enables the fast development of resource-constrained IoT applications, it lacks the capability to model such uncertainties and quantify their effects. Consequently, within uncertain environments the quality and performance of IoT applications generated from ThingML designs cannot be guaranteed. To explore the overall runtime performance variations caused by environmental uncertainties, this paper proposes a quantitative uncertainty evaluation framework for ThingML-based IoT designs. By adopting network of priced timed automata as the model of computation and statistical model checking as the evaluation engine, our approach can model uncertainties caused by external environments as well as support various kinds of performance queries on the extended ThingML designs. Experimental results of two comprehensive case studies demonstrate the efficacy of our approach.

In the context of the IEEE 1588 Precision Time Protocol (PTP), estimating the delay's bias is a problem that appears in both one-way (using transparent devices) or two-way message exchange mechanisms. For estimating the offset via the two-way message exchange mechanism, it is usually assumed that the expected value of delays in forward and reverse directions are equal. However, this is not a realistic assumption for packet-based wide area networks, where delays in down-link and up-link directions may have a significant difference. In this work, we propose a solution to estimate the random delay's bias and improve the synchronization accuracy of IEEE 1588. Our method is easy to implement and is compatible with the current version of the protocol. We compared our results to no bias correction and the Boot-strap method. In addition to the improvement in synchronization accuracy, our method allows us to update the slave clock recursively. The proposed method works well even in the presence of large frequency offsets and can also be implemented by using different filters.

In this paper, we propose solutions for the minimum-energy broadcasting problem for cross networks, where N nodes are located on two perpendicular lines. Our solutions consist of an algorithm which finds the optimal assignment in polynomial time, a near-optimal algorithm with less complexity (O(N)), and a distributed algorithm with complexity O(1) that gives acceptable results. To the best of our knowledge, this is the first study presenting an optimal solution for the minimum-energy broadcasting problem for a 2-D network (with cross configuration). We compare our algorithms with the broadcast incremental power (BIP) algorithm, one of the most commonly used methods for solving this problem with complexity O(N2). The results show that while the proposed optimal algorithm finds the best solution, our near-optimal algorithm performs better than BIP in all cases, and the distributed algorithm performs close to it. The performance of our non-optimal algorithms tend to be closer to the optimal solution for larger networks. We prove that all the algorithms perform the same in the asymptotic regime.

In this paper, we study the hop-count properties of one-dimensional wireless ad-hoc networks, where the nodes are placed independently and identically according to a Poisson distribution with an arbitrary density function. We derive exact equations to calculate the probability mass function of the number of hops needed for a node located at an arbitrary location in the network to receive a message from the source (located at one end of the linear network). Based on the derived formulas, we then propose localization methods. Through simulations, we show that our best proposed localization method not only has a competitive performance for a range-free method, but also outperforms range-based methods with a local distance measurement error of 10% or more. An important feature of our methods is that they are applicable to arbitrary densities. This is unlike the existing methods that are limited only to the case of uniform node densities. Moreover, the hop-count equations derived in this work can be used in analyzing other aspects of broadcasting protocols such as location verification, quality of service, and delay.

Recent studies demonstrate that the performance of a wireless sensor network (WSN) can be improved by deploying multiple sinks in the network. Therefore, in this paper we present different routing protocols for multi-sink WSNs based on the routing protocol for low-power and lossy networks (RPL). Our protocols use different routing metrics and objective functions (OFs). We use the available bandwidth, delay, MAC layer queue occupancy, and expected transmission count (ETX) as the tie-breaking metrics in conjunction with the shortest hop-count metric. Our OFs use the tie-breaking metrics on a greedy or end-to-end basis. Our simulation results demonstrate that the protocols based on the delay, buffer occupancy, and ETX metrics demonstrate best performance, increasing the packet delivery ratio by up to 25% and decreasing the number of retransmissions by up to 65%, compared to a version of the RPL protocol that only uses the hop-count metric. Another key insight is that, using the tie-breaking metrics on a greedy basis demonstrates a slight performance improvement compared to using the metrics on an end-to-end basis. Finally, our results also demonstrate that multiple sinks inside a WSN improve the RPL-based protocol performance.

In this paper, we study the hop-count properties of one-dimensional wireless ad-hoc networks, where the nodes are placed independently and identically according to a Poisson distribution with an arbitrary density function. We derive exact equations to calculate the probability mass function of two hop-count random variables: the number of hops needed for a node located at an arbitrary location in the network to receive a message from a node located at one end of the linear network, and the number of hops needed for a node located at one end of the network to receive a message from a node at an arbitrary location. Based on the derived formulas, we then propose localization and location verification methods. Through simulations, we show that our proposed localization method not only has a competitive performance for a range-free method, but also outperforms range-based methods with a local distance measurement error of 10% or more. Furthermore, the proposed location verification protocol is shown to have better results compared to the existing verification systems that also use the hop-count information. An important feature of our methods is that they are applicable to arbitrary densities. This is unlike the existing methods that are limited only to the case of uniform node densities. Using simulations, we also evaluate the proposed schemes in the presence of Rician fading and show that their performance is rather robust with respect to the change in the fading parameter. Moreover, the hop-count equations derived in this work can be used in analyzing other aspects of broadcasting protocols such as quality of service and delay.

We discriminate between different SiLabs IEEE 802.15.4 2.4GHz RF sources using the Ettus Labs USRP1 Software-Defined Radio. The wireless fingerprinting method implemented on the USRP1 device exploits differences in the phase attributes of demodulated data samples. The method does not require the use of expensive spectrum analyzer equipment and the associated high sampling and processing rates with such equipment. Instead, data sample inputs are used, sampled at a rate of 4MHz. This makes implementation using real Wireless Sensor Network nodes feasible and allows wireless fingerprints to be gathered inside each node in a network. This is important since wireless fingerprints degrade over distance, making distributed implementations more attractive. With our method, the USRP1 classifies accurately over a wide range of network conditions, including time and transmission distance. Performance is also stable for different receiving devices. We achieve average classification accuracies of 99.6% at short range, 95.3% at medium range, and 81.9% at long range when classifying a limited sample of five devices from the same manufacturer.

In this paper, we present an available-bandwidth-based proactive routing protocol for IEEE 802.15.4-based single-sink and multi-sink ad-hoc networks. The available-bandwidth-based proactive routing protocol maintains the best forwarding path in terms of the end-to-end available bandwidth towards each sink node present in a network. Moreover, a node can maintain more than one data forwarding path towards the same sink node. We performed extensive simulations, and compared our proactive routing protocol with a state-of-the-art opportunistic routing protocol. The simulation results demonstrate that the opportunistic routing protocol can distribute data load unevenly (in case of multiple sink nodes), hence results in high end-to-end delay and low Packet Delivery Ratio (PDR). In case of our proactive routing protocol, selecting forwarding paths by only considering the end-to-end available bandwidth invariably results in lengthy data forwarding paths. Lengthy data forwarding paths results in higher intra-flow contention, hence PDR and end-to-end delay are impacted. One of the simulation scenarios, using multiple sink nodes, demonstrates that in case of our proactive routing protocol, carefully selecting the data forwarding path(s) that are not too long compared to the shortest available data forwarding path(s), but have better end-to-end available bandwidth significantly improves the performance of the proactive routing protocol. Therefore, our results hint that, in general, trading off end-to-end available bandwidth and the length of a data forwarding path may improve end-to-end PDR and delay.

This paper proposes a Delay Asymmetry Correction (DAC) Model to enhance clock synchronization protocols based on a Master-Slave arrangement such as the IEEE 1588 PTP protocol. The purpose of this work is to mitigate the effects of unpredictable packet delay variations (PDV), which may aggravate asymmetric link delays on timing packets, in order to improve the synchronization accuracy of the slave clock with respect to the master clock. The key idea in our work is to filter clock updates derived from the master-slave message exchange, using only good samples for slave clock updates. The proposed solution is implemented in Network Simulator 2 (NS-2.34). NS-2 test cases are implemented according to the ITU-T G.8261 document covering various network loads and network conditions. Our simulation results indicate that the proposed solution improves the slave accuracy significantly, achieving almost perfect synchronization accuracy in the presence of a wide range of network traffic loads, network congestions, and temporary network outage. Furthermore, when there is a routing path change due to the failure in the network, the solution also maintains high slave clock accuracy with respect to the master clock.

We propose and evaluate a routing solution for a large scale tactical inter-domain network containing multiple Mobile Ad-hoc Networks (MANETs) and fixed networks. The proposed solution uses Open Shortest Path First - MANET Designated Router (OSPF-MDR); which is a modification of OSPFv3 to support MANETs as well as OSPFv3 in the fixed networks. We then propose a new border gateway routing protocol which supports the dynamic split and merger of MANETs, having routing abilities similar to the Border Gateway Protocol (BGP). This new protocol, named BGP - MANET Routing (BGP-MR)m, provides effective communication across separate and highly dynamic domains. BGP-MR introduces a dynamic election of gateways and recursive purging of routes from the BGP table as BGP peers are lost due to network mobility. The proposed solution is tested in CORE to show BGP-MR's effectiveness as a gateway protocol in mobile scenarios. Our results demonstrate BGP-MR's ability to provide seamless routing as well as reduced overhead when compared to running BGP in interconnecting highly dynamic MANETs.

Unmanned Aeronautical Ad Hoc Networks (UAANETs) are a type of Mobile Ad Hoc Networks (MANETs) which are infrastructureless and self-organizing networks. The specificity of UAANETs is that they are formed by small and medium sized Unmanned Aerial Vehicles (UAVs) also known as drones. In UAANETs as well as in MANETs, geographic routing is widely used. Geographic routing relies on Greedy Forwarding (GF), also called Greedy Geographic Forwarding (GGF). GGF fails when a packet arrives at a node that has no neighbor closer to the destination than it is. The node in this situation is referred to as a void node. In this paper, we propose a strategy that salvages packets in void node situations. We thereafter append this strategy to a protocol that features GGF. Simulations in OPNET show an increase in packet delivery ratio of about 2% at virtually no additional cost.

The mobile tactical network is a practical implementation of the mobile ad hoc network. Formed across tactical radios operating in the military very high frequency and low ultrahigh frequency bands, the mobile tactical network has distinctive characteristics when compared with generic mobile ad hoc networks, in particular with respect to its network topological behaviors and connectivity attributes. These characteristics must be understood and considered when selecting suitable network protocols. To this end, in this paper, a network science-based systematic modeling approach is applied to analyze typical deployment scenarios and identify fundamental tactical network properties. The novel framework employs realistic scenario models as well as radio physical layer performance parameters and channel models to effectively capture the dynamic network behavior that needs to be considered for protocol design. The results provide critical insights and guidance to the development of tactical network solutions.

A two-way message exchange mechanism is at the core of many clock synchronization protocols. The basic concept of this method has been designed for symmetrical links, otherwise, depending on the level of asymmetric delays, large synchronization errors may occur. Many estimation based methods have been proposed so far for estimating the degree of asymmetry and reducing this error. However, many of them are solved for a specific distribution of link random delays. Moreover, imposing a high level of computational complexity upon network nodes is another problem of them. In this paper we introduce a recursive solution for improving the performance of the basic two-way message exchange mechanism. The advantage of this method is that it keeps the protocol simple and therefore easily implementable. The proposed method does not make any specific assumption about the distribution of the random delays. It also provides the network designer with some deployment flexibility, trading off convergence speed against stability.

Network Coding is a relatively new forwarding paradigm where intermediate nodes perform a store, code, and forward operation on incoming packets. Traditional forwarding approaches, which employed a store and forward operation, suffered from the limitations of the max-flow min-cut theorem wherein sources transmitting information over bottleneck links had to compete for access to these links. With Network Coding, multiple sources are now able to transmit packets over bottleneck links simultaneously, increasing network capacity. While the majority of the contemporary literature has focused on the performance of Network Coding from a capacity perspective, the aim of this research has taken a new direction focusing on two Quality of Service metrics, Packet Delivery Ratio (PDR) and latency, in conjunction with Network Coding protocols in Mobile Ad-Hoc Networks (MANETs). Initial simulations will be performed on static environments to determine a Quality of Service baseline comparison between Network Coding protocols and traditional ad-hoc routing protocols. Additional simulations will then be performed for mobile scenarios to determine how the Network Coding protocols will compare to that of the standard ad-hoc routing protocols in the presence of mobility.

Estimating the available bandwidth in IEEE 802.15.4-based networks is a difficult and challenging task due to the shared nature of the wireless communication medium. The MAC layer decides the sharing of a communication medium, therefore the MAC layer dictates the amount of bandwidth available to a node. Some recent solutions consider the impact of the MAC layer back-off due to collision and transmitter and receiver non-synchronization on the available bandwidth, Available Bandwidth Estimation (ABE) is one such example. None of the existing solutions pro-actively consider the impact of additional overhead of an unslotted Carrier Sense Multiple Access Collision Avoidance (CSMA-CA) protocol with an increased data load inside a network on the available bandwidth. Therefore the amount of reported available bandwidth is not fully available to a node, hence this can result in poor admission decisions. In this paper, we show that increasing data load inside the IEEE 802.15.4-based network increases MAC layer overhead. Afterwards, we enhance the ABE bandwidth estimation and admission control algorithms so that they pro-actively consider the additional MAC layer overhead associated with an increased data load inside a network. We performed a number of simulations, and our simulation results show that the proposed proactive ABE method performs better than the original ABE method.

During the route discovery process, each node receiving the route request packet (RReq) will retransmit it exactly once. A distant neighbor may accidentally receive/loose the only RReq and use it to announce a new route, although that link is inferior/superior for route reply packet (RRep) or actual message routing. Overall, the constructed route may be far from the optimal. All existing route discovery schemes (including DSR/AODV) apply retransmission during route discovery exactly once (1R). Based on a realistic physical layer model, we propose two new route discovery schemes: n-retransmission (nR, retransmitting exactly n times) and n-retransmission c-reception (ncRR), retransmitting until either a total of n own retransmissions or c copies from neighbors are heard. We compare our two new scheme with traditional one, under otherwise same conditions (same metric, same packet reception probability on each link) and same choices about possibly retransmitting again upon discovering better route (R+) or discarding it (R1), generating route reply packet for every received RRep (B*), or for first and better discovered route (B2), and retransmit ting RRep exactly once (A1), maximum three times (A3), or optimally u times decided by link quality (Au). Experimental results also show that the proposed ncRR scheme (for n=2 and c=3 or c=4) achieves the best tradeoff between quality of route, success rate and message overhead in route discovery process, followed by nR scheme, and both of them are superior to the existing traditional based schemes.

Network Coding, in particular Random Linear Network Coding, has been extensively researched for communication in multihop wireless networks, such as MANETs. A key aspect of random linear network coding is the definition of generations, the set of native data packets that can be coded together. To keep the decoding complexity and packet latency low, generations are typically small. In addition, to simplify generation management, it is usually assumed that a generation only contains packets originating from the same source. For broadcast scenarios, we show via examples, analytically derived lower bounds, and protocol simulations that the latter restriction reduces the coding gains that are achievable. Allowing packets from different sources to be combined (i.e., coded together) can improve PDR, reduce packet latency, and further reduce the number of packet transmissions at the MAC layer. We also discuss the added complexity that results from having packets from different sources in a single generation and propose our solution, embedded in a RLNC broadcast protocol called ARLNCCF.

Replay attacks and their associated risks in mobile tactical networks are analyzed and a cooperative joint detection scheme is evaluated. The scheme adopts a combination of duplicated frame detection and link likelihood verification to defend against both local and remote replays. A formal analytical framework is established employing realistic tactical radio models and network scenarios. The results obtained illustrate the network-wide risk of the attack and the effectiveness of the detection scheme.

With the increasing popularity of the cloud computing paradigm, data centers have become more important. Because current virtual machine live migration solutions require the live migration to take place within the same subnet without changing the IP address of the live migrated virtual machine, large Layer 2 networks are used in data centers, which creates scalability difficulties. In order to limit the size of Layer 2 networks, live migration can be done with IP address and subnet change. We implement a solution for live migration with IP address change, where the old host of a migrated virtual machine acts as a Layer 3 proxy for the virtual machine. We evaluate the performance benefits of using TCP-Freeze with this solution, where we freeze and thaw TCP connections as the virtual machine is live migrated.

We demonstrate empirically a physical layer authentication method that uses the physical attributes of OQPSK RF signals transmitted from IEEE 802.15.4 WSN nodes, collecting data with a USRP1 software-defined radio platform. We present simple training and classification algorithm alternatives and measure and analyze the performance of the overall combination. We improve classification performance using a simple method at the receiver that discards samples using preamble `PN chip' error information and estimates a phase offset between the transmitting node's RF clock and the receiver's sampling clock.

As a result of high mobility of Unmanned Aerial Vehicles (UAVs), designing a good routing protocol is challenging for Unmanned Aeronautical Ad-hoc Networks (UAANETs). Geographic-based routing mechanisms are seen to be an interesting option for routing in UAANETs due to the fact that location information of UAVs is readily available. In this paper, a combined routing protocol, called the Reactive-Greedy-Reactive (RGR), is presented for UAANET applications, which combines the mechanisms of the Greedy Geographic Forwarding (GGF) and reactive routing. The proposed RGR employs location information of UAVs as well as reactive end-to-end paths in the routing process. Simulation results show that RGR outperforms existing protocols such as Ad-hoc On-demand Distance Vector (AODV) in search UAANET missions in terms of delay and packet delivery ratio, yet its overhead is similar to traditional mechanisms.

Cloud storage services become increasingly interesting for users to easily backup or synchronize their data. On top of this basic functionality, these services offer functions for collaboration that allow users to share their files with selected other persons in a user-friendly way. We have identified that several cloud storage services do not verify whether the registrating customer is the real owner of the email address entered during the registration. Cloud providers omit the verification for reasons of usability. Here, user-friendliness goes too far at the cost of security. This vulnerability combined with collaboration functions allows attacks on cloud customers. In this paper, we explain which attacks are possible. Missing email verification and collaboration functions allow espionage and malware distribution attacks. Execution is very easy, i.e., they can be done without coding expertise or special tools.

This work investigates an efficient and robust broadcast / multicast scheme for the bandwidth limited tactical environment. A local neighborhood based broadcast / multicast protocol is enhanced through managing dynamic link conditions to achieve required delivery ratio. Employing a realistic tactical radio model and practical tactical deployment scenarios, the proposed mechanism is evaluated using a network simulator. Compared with one of the most efficient standard protocols, the simplified multicast forwarding, the proposed scheme demonstrates improved efficiency and robustness.

Due to the availability of location information in unmanned aerial vehicles (UAVs), we propose to use geographic routing mechanisms as a core forwarding protocol in unmanned aeronautical ad-hoc networks (UAANETs) for the purpose of reducing routing overhead. As a result, this paper investigates the performance of the core forwarding mechanism i.e. the greedy geographic part. Since the forwarding mechanism for dynamic UAANETs with many statistical inter-dependencies is complex, a closed-form model does not exist. Therefore, a quadratic polynomial estimation is proposed for computing the success probability of greedy geographic forwarding based on the results of a set of realistic Monte Carlo simulations. This mathematical model can later be used to predict and evaluate the performance of other greedy-based geographic routing protocols for UAV applications.

In trust-based models for MANETs, it is often assumed that the trust value between two nodes decays proportional to the time these nodes are out of transmission range of each other. We present in this paper an analysis of the trust decay rate for some general networking and trust computation models. Our analysis allows for the consideration of the subjective trust values assigned by present and past neighbours within a given history timeframe. The mobility models being studied in this paper include random way point (RWP) and random walk (RW). This analysis exhibits the impact of mobility on the trust decay rate. Simulations are used to confirm the accuracy of analytical results and shed light on some interesting facts such as the slow decay of globally computed trust values, even if fast subjective trust decay functions (linear and exponential) are employed, or the weak impact of the network density on the trust decay rate.

ABSTRACT Because of their specific characteristics, Unmanned Aeronautical Ad‐hoc Networks (UAANETs) can be classified as a special kind of mobilead hocnetworks. Because of the high mobility of Unmanned Aerial Vehicles, designing a good routing protocol for UAANETs is challenging. Here, we present a new protocol called Reactive‐Greedy‐Reactive (RGR) as a promising routing protocol in high mobility and density‐variable scenarios. RGR combines features of reactive MANET routing protocols such as Ad‐hoc On‐demand Distance Vector with geographic routing protocols, exploiting the unique characteristics of UAANETs. In addition to combining reactive and geographic routing, the protocol has a number of features to further improve the overall performance. We present the rationale and design of the protocol, discuss the specific performance improvements in detail and provide extensive simulation results that demonstrate that RGR outperforms purely reactive or geographic routing protocols. The results also demonstrate the impact of the various protocol modifications. Copyright © 2012 John Wiley & Sons, Ltd.

Wireless sensor networks (WSNs) are usually randomly deployed in a region of interest. As a result, algorithms that can compute the location of sensor nodes within a WSN are needed. In recent years, several localization algorithms have been proposed for stationary WSNs. However, most studies only provide simulation results and most algorithms have never been implemented on a real testbed. In this paper, we implement a localization algorithm called CCA-MAP on a real WSN testbed. To the best of our knowledge, the CCA-MAP algorithm is amongst the best performing algorithms proposed for stationary WSNs. The results obtained show that the implementation results are consistent with the simulation results.

In this paper, Wireless Personal Area Network (IEEE 802.15.4) wireless channel utilization using Memsic (formerly Crossbow) IRIS Motes is measured and compared to ns2.34 simulation and theoretical models. Results show divergence between calculated, simulated and experimental results at higher throughput. Experimental results also highlight the limitations of the MoteWorks framework with regard to channel throughput measurements. A set of mitigation strategies is outlined to improve future research efforts aiming to characterize the validity of simulation models using the Memsic wireless sensor nodes.

In this paper, we first survey cross layer architectures for Wireless Sensor Networks and Wireless Multimedia Sensor Networks (WMSNs). Afterwards, we introduce a novel framework for supporting multiple applications in WMSNs. The proposed framework supports heterogeneous flows and it classifies WMSN traffic into six classes. The framework partitions the network into hexagonal cells and it uses the seven cell frequency reuse pattern to enhance throughput and minimize interference. The framework uses a shared database to enable cross layer interactions. For supporting multiple applications on a single node, an area in the memory is reserved where each application can request its protocols parameters. Protocol parameters and interaction between the shared database and applications are done through cross layer optimization middleware. A TDMA-based distributed MAC is used to support heterogeneous traffic flows.

Integrating the power grid technology with renewable power generation technologies, Demand Response (DR) programs enabled by Advanced Metering Infrastructure (AMI) were introduced into the power grid in the interest of both utilities and residents. They help to achieve load balance and increase the grid reliability by encouraging residents to reduce their power usage during peak load periods. We compare mainstream network technologies applicable to home appliance control and propose a solution combining Power Line Communication (PLC) with wireless communication in smart homes for the purpose of energy saving. Moreover, we explore and evaluate the network performance of DR programs in smart homes in such a combined network via a simplified networking simulation model.

Network Simulator Version-2 (NS-2) is an open source simulation tool originally employed to explore the performance of a single network. However, recently many researchers expressed interest in issues related to Multiple Interfaces and Multiple Channels (MIMC), in particular how to efficiently utilize multiple channels. One such scenario is the combination of power line and wireless networking technologies in smart homes. According to the requirements of smart home networking for energy management, we established a simplified simulation model combining multiple networks. A single channel is assigned to each network to allow packets to be transmitted over various networks. The framework of our model could be further extended with existing or emerging network technologies to support more generic cases related to multiple networks.

A WSN consists of numerous nodes gathering observations and combining these observations. Often, the timing of these observations is of importance when processing sensor data. Thus, a need for clock synchronization arises in WSNs. The CS-MNS algorithm has been proposed to fulfil this role. This paper compares simulation results and testbed results for CS-MNS. The simulations were done using Matlab, the testbed implementation was done in TinyOS 2.1, running on a mix of TelosB and MCIAz motes. The results demonstrate good qualitative agreement between simulation and experimentation in most cases. Quantitatively, the testbed results converge slower and achieve less synchronization accuracy, however.

In this paper, we investigate the problem of network coding and media scheduling in wireless multihop networks. Unique characteristics of the wireless media, such as omnidirectional transmissions and destructive interference, as well as having one transceiver per wireless node, imply new code design constraints for wireless networks. Here, we formulate a linear program to solve the joint scheduling and network coding problem. Using our formulation, we demonstrate that for a large percentage of randomly generated wireless networks, the optimal scheduling time shares are unequal. All the existing network code design algorithms are based on equal scheduling time shares or the considered joint optimization problems do not have sufficient information for scheduling flows during unequal time shares. Therefore, we provide these statistics to emphasize the importance of enabling the code design algorithms to include unequal time shares. Our simulations further show that the network throughput can be significantly improved if the network code is properly designed to incorporate unequal time shares.

A WSN consists of numerous nodes gathering observations and combining these observations. Often, the timing of these observations is of importance when processing sensor data. Thus, a need for clock synchronization arises in WSNs. The CS-MNS algorithm has been proposed to fulfil this role. This paper discusses our experiences implementing CS-MNS in TinyOS on TelosB and MICAz motes and experimentally evaluating its performance. The implemented protocol performs extremely well in single-hop scenarios and also achieves good clock synchronization in different multihop scenarios. In all scenarios, CS-MNS performs noticeably better than FTSP, the clock synchronization protocol provided with TinyOS 2.1.

Applications of wireless sensor network (WSN) often expect knowledge of the precise location of the nodes. One class of localization protocols patches together relative-coordinate, local maps into a global-coordinate map. These protocols require some nodes that know their absolute coordinates, called anchor nodes. While many factors influence the node position errors, in this class of protocols, using Procrustes Analysis, the placement of the anchor nodes can significantly impact the error. Through simulation, using the Curvilinear Component Analysis (CCA-MAP) protocol, we show the impact of anchor node placement and propose a set of guidelines to ensure the best possible outcome, while using the smallest number of anchor nodes possible. Scientists and researchers using sensor networks are thus enabled to focus on the sensed data with confidence in the node localization results.

Clustering schemes offer a practical way of providing scalability when dealing with large and dense Mobile Ad hoc Networks (MANETs). The feasibility of clustering can be determined by the complexity of the cluster head selection. Optimizing the cluster head selection allows for the network to be more efficient by minimizing the signaling overhead while ensuring that the network connectivity is maintained despite topology changes. In this paper, we present some new findings on the complexity of the cluster head selection algorithms. Two variants of the cluster head selection are examined: (1) the distance-constrained selection where every node in the network must be located within a certain distance to the nearest cluster head; and (2) the size-constrained selection where each cluster is only allowed to have a limited number of members. We show that the problem of minimizing the set of cluster heads is NP-hard for both variants. We propose two distributed selection algorithms, each having logarithmic approximation ratio, for these variants. We also discuss, using NS-2 simulations, the resulting cluster size distribution and cluster head density, which impact the efficient operation of the network.

Cross-layer design has proven to be a promising approach to achieving optimal performance in wireless networks. However, care must be taken when designing cross-layer protocols and frameworks so as to avoid excessive complexity and loss of standardization. In this paper, we propose a cross-layer framework for wireless sensor networks which preserves modularity and clearly defines the inter-layer interactions allowing for ease of standardization and long-term development. This Central-Networked framework consists of the centered physical and centered application models.

Efforts to reduce peak electrical demand has led to the introduction of demand response (DR) programs for residences. The RDS network is a strong candidate for delivering DR messages due to its low-cost nature and ubiquitous coverage. However, security concerns arise due to the wireless nature of the communication channel. We present evaluations of cryptographic methods that could be employed to offer source authentication over the RDS network. Simulations are used to determine the impact on the network performance by employing three digital signature protocols (BiBa, HORSE, and ECDSA). The simulation results show that, up to a distance of 90 km, all authentication schemes do not affect message reception by the receivers. ECDSA and HORSE outperform BiBa in terms of message reception beyond 90 km. ECDSA offers higher security than HORSE and BiBa but at the cost of increased computational complexity, in particular at the receivers. In addition, ECDSA has the highest bandwidth overhead.

Wireless multihop networks, built with tactical radios, have limited bandwidth; yet often need to support bandwidth-intensive 1-to-many and many-to-many communications. Efficient routing proposals for wireless broadcasting have been suggested in the literature, and more recently the use of network coding to efficiently support such communication patterns. This paper explores the use of efficient broadcasting protocols with and without the use of network coding. Through extensive simulation studies and comparisons of protocol and coding options, the performance gains and the relative tradeoffs of different approaches are unveiled for practical tactical networking scenarios.

Fourth Generation broadband wireless technologies such as WiMAX and LTE depend heavily in the performance of their schedulers to deliver high data throughput and meet quality-of-service commitments. This paper compares four different proposed schedulers for mobile WiMAX (Proportional Fairness (PF), Multiclass Modified Largest Weighted Delay First (MLWDF), Highest Urgency First (HUF), and Weighted Fair Queuing (WFQ) )in a range of environments. The evaluation is based on five industry-defined key performance indicators: average sector throughput, application throughput, average completion time, fairness index and delay). The schedulers are evaluated under three simulated environments: controlled (with a detailed analysis of each algorithm's behavior in terms of throughput over time), stationary and mobile. The controlled environment provides interesting insights about the behavior of flows with identical QoS parameters and different RF conditions, and helps to validate subsequent results obtained in the other two environments. Our results for the stationary and mobile environments show that all algorithms meet quality-of-service requirements within system capacity. Algorithms that maximize spectral efficiency (PF and MLWDF) also achieved considerable throughput improvements. MLWDF's throughput results, while outperforming all other schedulers under stationary conditions, fall behind PF in the mobile scenario. The variability introduced by the mobile environment yields no statistically significant difference among the schedulers.

RF Fingerprints can be used to distinguish between distinct transmitter signals in a Wireless Sensor Network. We present implementation results for RF Fingerprints based on Automatic Gain Control circuitry, which is present on existing IEEE 802.15.4 WSN node hardware and in other radio systems. We analyze implementation constraints and their different effects on our results. Different noise sources affect RF Fingerprint measurement accuracy on a real hardware platform. We present a simple method to improve both accuracy and reliability.

Abstract Clock synchronization is an important component of wireless sensor networks (WSNs) both for co‐ordination of node communications and for time stamping sensor data. The previously presented clock sampling mutual network synchronization (CS‐MNS) algorithm is simple, has low communication and processing overhead, and allows fully decentralized operation. We present some simulation results that indicate the potential of CS‐MNS to achieve high clock synchronization accuracy in mobile multi‐hop wireless networks. Past work has shown clock convergence under specific conditions in single‐hop networks. We show analytically that in the absence of offset errors, the network clocks converge. In the presence of offset errors, we present conditions on the degree of clock asynchrony under which the network clock rates show convergent behavior. The analysis is applicable as long as the network topology is connected and, thus, is of interest in both single‐hop and multi‐hop environments. As a side result, we also show how a network designer can use these conditions to add a bias term to the CS‐MNS algorithm and, thus, improve the start‐up dynamics of the algorithm. Furthermore, we discuss the algorithm from a security standpoint. Finally, we propose a method for adding external reference synchronization that is compatible with our security discussion. Copyright © 2009 John Wiley & Sons, Ltd.

This paper presents a novel and efficient mobility management scheme based on a pro-active caching approach (i.e., state transfer/caching occurs prior to the subscriber's movement) to extend existing pub/sub systems to the mobile, wireless domain. This approach depends largely on a mechanism that intelligently pre-loads subscriber contexts one hop ahead of its current broker. To achieve this in an automated fashion, we introduce the notion of a neighbor graph, which is built automatically, to capture user mobility patterns. We have investigated the effectiveness of our proposed approach through testbed experiments, comparing it to the current state-of-the-art solutions, durable subscription-based and reactive, proposed in the literature. The experimental results across a broad-range of workload parameters show that our proactive approach outperforms earlier approaches by wide margins. Our approach supports fast handoff with low cost in terms of state transfer overhead. It also prevents message duplication and considerably reduces message loss.

Clock synchronization is an important component of wireless sensor networks both for co-ordination of node communications and for time stamping sensor data. The properties of the previously presented clock sampling mutual network synchronization algorithm (CS-MNS) are explored. The desirable properties of this algorithm include simplicity, low communication and processing overhead, and fully decentralized operation. Past simulation and analysis has shown clock convergence under specific conditions in single hop networks. We show that in the absence of offset errors the network clocks converge. In the presence of offset error we present conditions on the degree on clock asynchrony under which the network clock rates show convergent behavior. The analysis is applicable as long as the network topology is connected and thus is of interest in both single-hop and multi-hop environments. We show how the network designer can use these conditions to add a bias term to the CS-MNS algorithm and thus improve the start-up dynamics of the algorithm. Additionally, we discuss the algorithm from a security standpoint. Finally, we propose a method for adding external reference synchronization that is compatible with our security discussion.

Accurately determining the location of mobile wireless sensor nodes in real-time is essential for many purposes. This paper proposes a new and efficient algorithm for localization of mobile node(s) within a WSN. The proposed algorithm, called iterative CCA-MAP (iCCA-MAP), is based on the CCA-MAP algorithm which applies an efficient nonlinear data mapping technique. The latter has been shown to perform extremely well for localizing stationary nodes in WSNs. Simulation results show that the localization error results for both CCA-MAP and iCCA-MAP are similar. However, the computational time required for obtaining location results using the iterative CCA-MAP scheme is far smaller than that of the original CCA-MAP. The advantage of the proposed algorithm is that it can provide the mobile node's location information at near real-time, allowing for it to be applied at a much higher frequency in order to provide up-to-date estimations of the mobile node's position, which can result in a lower localization error.

Cooperative node localization schemes that employ nonlinear data reduction often deliver higher network node position accuracy compared to many other approaches. Other advantages of such algorithms are that they require only a minimum number of anchor nodes (if we require absolute locations) and that they can be applied under both range-based and range-free conditions. This article presents a novel cooperative node localization scheme, applying an efficient neural network nonlinear projection method called Curvilinear Component Analysis (CCA). A thorough comparative performance study of the proposed scheme in different mission-critical operational network scenarios is conducted. Compared with another leading cooperative node localization algorithm, MDS-MAP, which employs Multi-Dimensional Scaling (MDS), the proposed CCA-MAP approach significantly improves position estimate accuracy in many of the scenarios. We also propose a new local edge model for range-free distance matrix approximation that considerably enhances the performance for both MDS-MAP and CCA-MAP in certain irregular network configurations which are very challenging for node positioning.

Secure authentication in wireless sensor networks is complicated by the promiscuous nature of the wireless transmission medium and by the limited hardware and software capabilities of the nodes in such networks. We analyze the requirements for using RF fingerprints to initialize secure authenticated links in a wireless sensor network for a surveillance application that allows older people to remain in their homes longer (termed the 'aging in place' application). We survey the applicable research for RF fingerprints and other security-related and authentication-related work. We show that RF fingerprints alone are inadequate for authentication. We present a new authentication protocol based on digital credentials that binds together physical-layer RF fingerprint data with higher cryptographic protocol layer data which is practical, but based on well-known and proven security principles.

To support mobile subscribers in pub/sub systems, existing approaches are mainly based on reactive methods (i.e., the message transfer occurs just after the mobile subscriber handoffs to a new broker). In this paper, we analyze a novel and efficient mobility extension based on a pro-active approach (i.e., the context transfer/caching occurs prior to the subscriber movement). We implemented both reactive and pro-active mobility extensions for JMS and conducted a range of experiments in a small testbed to compare the relative merits of each approach. Our experimental results show that a pro-active approach can provide fast handoff support with zero duplicated messages. It also achieves relatively higher throughput results and lower message loss compared to the reactive approach.

Secure authentication in wireless sensor networks is complicated by the promiscuous nature of the wireless transmission medium and by the limited hardware and software capabilities of the nodes in such networks. We analyze the requirements for using RF fingerprints to initialize secure authenticated links in a wireless sensor network for a surveillance application that allows older people to remain in their homes longer (termed the dasiaaging in placepsila application). Based on the lessons learned from our literature survey, we show that RF fingerprints alone are inadequate for authentication. We present a new authentication protocol based on digital credentials that binds together physical-layer RF fingerprint data with higher cryptographic protocol layer data which is practical, but also based on well-known and proven security principles.

To support energy-efficient routing, accurate state information about energy level should be available. But due to bandwidth constraints, communication costs, high loss rate and the dynamic topology of MANETs, collecting and maintaining up-to-date state information is a non-trivial task. In this work, we use Optimized Link State Routing (OLSR) as the underlying routing protocol and focus on residual energy level as QoS metric, which has been used for routing decisions in many energy-efficient routing protocol proposals. Our experiments show that nodes have at best imprecise state information, especially under high traffic rates. We propose and evaluate two additional techniques to reduce inaccuracies and compare them against the basic OLSR protocol. These techniques show significant improvements in accuracy levels. In particular, a technique we call Smart Prediction achieves highly accurate perceived residual energy levels under all traffic loads, mobility scenarios, and radio models.

Mutual network synchronization is a distributed method in which geographically separated clocks align their times to one another without the need of reference or master clocks. Mutual network synchronization is attractive for wireless ad hoc and sensor networks, because there is no overhead associated with the discovery, management, and tracking of specific nodes with reference clocks. Existing mutual network synchronization methods, however, make use of physical and medium access control layers that are proprietary and not widely available. This paper presents clock sampling mutual network synchronization (CS-MNS). CS-MNS is able to achieve microsecond networkwide synchronization accuracy for single-hop or multiple-hop network topologies in mobile or static wireless ad hoc and sensor networks. Different from existing mutual network synchronization approaches, the timing information is exchanged explicitly by using periodic time stamp packets. These packets can be, for instance, the same beacons used in the IEEE 802.11 or IEEE 802.15.4 standards, which, to the best of our knowledge, makes CS-MNS the first mutual network synchronization method compatible to these popular standards. A CS-MNS node adjusts the time and frequency of its clock recursively in the time domain by multiplying the time of its clock by a factor that is updated with any newly received time stamp. Sufficient stability conditions are derived via the discrete Lyapunov direct method. Additionally, CS-MNS enables several beacon medium access approaches, which are discussed and analyzed. Thorough numerical results are presented, which demonstrate at least one and two orders of magnitude improvement in scalability and accuracy, respectively, relative to the IEEE 802.11 timing synchronization function (TSF). CS-MNS also shows better accuracy than the multihop ad hoc TSF (MATSF) and the automatic self-time-correcting procedure (ASP) methods under similar scenarios. The latter is achieved with less complexity and with fully compatible IEEE 802.11 beacons.

Mobile games are one of the primary entertainment applications at present. Limited by scarce resources, such as memory, CPU, input and output, etc, mobile game development is more difficult than desktop application development, with performance as one of the top critical requirements. As object-oriented technology is the prevalent programming paradigm, most of the current mobile games are developed with object-orientation (OO) technologies. Intuitively OO is not a perfect paradigm for embedded software. Questions remain such as how OO and to what degree OO will affect the performance, executable file size, and how optimization strategies can improve the qualities of mobile game software. These questions are investigated in this paper within the mobile Role-Playing-Game (RPG) domain using five industrial mobile games developed with OO. We analyzed them and found excessive usage of OO features used for the development of mobile device applications (but normal for usual desktop applications). We then apply some optimization strategies along the way of structural programming. The experiment shows that the total jar file size of these five optimized games decreases 71 % the lines of codes decreases 59%, and the loading time of each optimized game decreases 22.73%, 34.62% 25.79% 24.65% and 16.70% respectively. Therefore, we conclude from our experiments that 00 should be used with great care in the development of mobile games, and that structural programming can be a very competitive alternative.

Current mobile middleware is designed according to a 'one-size-fits-all' paradigm, which lacks the flexibility for customization and adaptation to different situations, and does not support user-centered application scenarios well. In this paper we describe an ongoing intelligent mobile middleware research project called PLIMM that focuses on user-centered application scenarios. PLIMM is designed based on software product line ideas which make it possible for specialized customization and optimization for different purposes and hardware/software platforms. To enable intelligence, the middleware needs access to a range of context models. We model these contexts with OWL, focusing on user-centered concepts. The basic building block of PLIMM is the enhanced BDI agent where OWL context ontology logic reasoning will add indirect beliefs to the belief sets. Our approach also addresses the handling of ontology evolutions resulting from the timely adaptation of ontology to changes and the consistent propagation of these changes to all related artifacts, using Frame based product line configuration techniques.

Service Discovery Protocols (SDP) allow users and applications to automatically locate services without prior configuration. Since Mobile Ad Hoc Networks (MANETs) constitute an integral part of the ubiquitous computing scenario, increasing efforts are geared toward providing service discovery in such environments. In this work, we adapt the widely employed Service Discovery Protocols (SDP) Version 2 to MANETs. We implement and analyze the adapted protocol, SLPManet, using Network Simulator Version 2. Consequently, we propose a caching modification that increases the discovery success by 38% for poorly performing scenarios, as well as reduces the aggregate bandwidth consumption and average lookup latency by up to 98%.

Finding and maintaining paths in wireless ad hoc networks is not trivial due to node mobility and variable environmental conditions. Robust and efficient routing algorithms are required. OLSR is a link state proactive routing algorithm for ad hoc networks that makes use of an optimized mechanism, based on Multipoint Relays (MPRs), to perform broadcasting and path computation. OLSR only makes use of its partial view of the network topology along with the Minimum Hop Count (MHC) metric to construct routing paths, therefore, it suffers from a lack of knowledge, such as full topology knowledge, nodes and links status (e.g. remaining battery, link quality) and network load in order to make better routing decisions. Also, MHC paths are usually constructed by longer links, which tend to provide lower throughput and frequent breakage. In this research the network knowledge available for OLSR is extended by gradually increasing its partial network topology view. Several TC_Redundancy strategies are combined with the MPR_Coverage feature of OLSR while maintaining its optimal broadcasting mechanism. NS-2 simulation results show the costs and benefits in terms of communication overhead and increased topology knowledge, respectively, to make OLSR a robust and QoS-conducive routing protocol.

Most existing mobile ad hoc routing protocols preset the parameters for the nodes based on some assumed network conditions. Because of the dynamic characteristics of a MANET, network conditions are changing at different points of time. Adaptively adjusting routing behavior by individual nodes will improve the routing protocol performance. In this paper, we mainly focus on the node mobility, and show the correlation between overall performance and node mobility level with simulation results for three different mobility models. We also study the effect of the mobility from point of view of individual nodes. Finally we propose the number of link breakages as a mobility metric, which is mobility model independent, that nodes can use it to monitor the environment changes

In the world of wireless ad-hoc networking, the problem of service discovery is essential and important in order to detect and access services between mobile computers. This has hence led to the advent of numerous service discovery protocols (SDPs). We recognize the need for a legitimate framework for the qualitative and quantitative evaluation of SDPs in mobile networking contexts such mobile ad hoc networks (MANETs). This paper presents benchmark for MANET (BENCHManet), an initiative development of a comprehensive and sophisticated benchmark composed of reference tests, each of which reflects the configuration found in various realistic MANET applications. The presented benchmark allows for a more practical and comprehensive evaluation, as well as a fair comparison of SDPs

This paper presents our experience in evaluating the performance of a JMS-based publish/subscribe system deployed in a wireless environment. In this paper, we identify a collection of workload factors that impact the reliability of message delivery and study their affects on the system's performance. We have mainly studied the behavior of two subscription schemes, durable and nondurable, along with a set of low/high reliability factors. The reliability cost is evaluated and compared with baseline data collected on a local-area, wired network. We have performed an extensive set of experiments to cover a broad range of factor combinations. Our results show that factors with high reliability tend to greatly affect the performance. Although the wireless scenario shows higher impact on the performance, the reliability costs are relatively low. We believe that our evaluation study provides valuable lessons to system designers and users.

In this study, we present a performance evaluation of an energy conservation technique designed for mobile ad hoc networks. This technique was designed with energy fairness central to its operation. This algorithm is not a routing algorithm. It works with existing routing protocols to complement their functionality from an energy-efficiency perspective. We show that this technique scales well with increased network traffic and population. We also compare it to the on-demand power management algorithm, a technique specifically designed to reduce idle energy consumption. Our comparison shows that our technique performs better in terms of energy savings and fairness as well as network lifetime extension

The performance of MANET routing protocol depends on the network conditions, which includes mobility, congestion and transmission condition. In this paper, we discuss the impact of mobility on the performance of the routing protocols, using three different mobility models. We further look into different parameters which change with the mobility from the point of view of individual nodes, showing that nodes experience different levels of mobility at different points of time. Finally we propose that the number of link breakages observed by a node can be used as a mobility metric so that each individual node can adjust its routing behavior based on the environment around it to improve the overall routing protocol performance

Mobile ad-hoc networks (MANETs) are infrastructure-less networks where network topology changes very frequently and routing is not a trivial task. OLSR is a link-state routing algorithm that is optimized to reduce the communication overhead by reducing the amount of topology information that is continuously advertised. This optimization constrains OLSR to partial network topology knowledge which is the only input used along with the Minimum Hop Count (MHC) metric to construct routing paths. Therefore, OLSR suffers from a lack of knowledge, such as full topology knowledge, nodes and links status (e.g. remaining battery, link quality) and network load in order to make better routing decisions. However, OLSR allows manipulating the amount of advertised topology information. This research explores mechanisms to extend the topology knowledge with the objectives of: 1) fully understanding the possible trade-offs between achievable topology knowledge and communication overhead and, 2) providing support for localization and Quality of Service (QoS). Several TC_Redundancy [7] strategies are combined with the MPR_Coverage feature [7] of OLSR while maintaining its optimal broadcasting mechanism. NS-2 simulation results show the costs and benefits of increasing the topology knowledge to make OLSR a supportive routing protocol for localization and QoS. A possible NS-2 bug is also discussed. Statistical analysis is applied to eliminate any effect that this bug may have had

The peculiar nature of the mobile ad-hoc networks (MANET) imposes several challenges when designing quality of service solutions. Traffic traveling on a MANET network can be local traffic arriving from and targeting a node within the ad-hoc network. Traffic that is not local to MANET is likely to travel over fixed topology networks that employ, typically, DiffServ. This paper proposes employing a new framework at the gateway between the ad-hoc domain and the DiffServ access domain. The proposed framework facilitates a homogeneous seamless QoS interaction between both networks built over heterogeneous components. The objective is to achieve a high level of performance and autonomy with a lightweight implementation, if any, on mobile nodes. In this paper we propose the framework solution to the cross-domain QoS problem and use the aggregate resource reservation (ARSVP) for collective resource reservations, combined with a simple sponsorship mechanism.

In this paper, we propose the clock-sampling mutual network synchronization (CSMNS) as a non-hierarchical and mutual network synchronization algorithm for wireless ad hoc networks. CSMNS shows superior performance to the IEEE 802.11 timing synchronization function in terms of accuracy, scalability and robustness. An overall view of the differences between the two approaches is presented. CSMNS is compatible with the beacon messages used in the IEEE 802.11 standard, and it is PHY transparent. CSMNS-RMN (rotating master node) is proposed in order to further reduce beacon collisions and overhead. Stability, is a factor that must be considered in CSMNS. However, values of the proportional gain below 0.3 suggest a good stability performance. The use of larger C/sup max/ values in more dense networks and/or the use of techniques that randomly prioritize the transmission of beacons can further reduce the overhead and risks of instability.

Wireless bandwidth is a limited and scarce shared resource in MANETs. A number of experimental and analytical studies have shown that the multi-hop nature results in nodes experience low bandwidth. Compounding this problem further are the routing protocols themselves: as they typically minimize either hop count or number of packet retransmissions, they favor routes in the centre of a MANET. This results in a reduction of spatial reuse, inefficiently using the scarce wireless bandwidth. In our work, we are interested in the "end-to-end capacity" of the network, which we define to be the sum of the throughputs for each flow. This represents the service a network provides to its users. In this paper, we determine the maximal MANET capacity through linear programming and evaluate the performance of AODV and DSR with respect to this maximum value. The result shows that these typical MANET routing protocols do not utilize the network resources efficiently, achieving less than 17% of that capacity even in the best possible case. New routing protocols are therefore required to make more efficient use of this limited resource.

The research efforts in the area of ad-hoc networks have evolved rapidly since the establishment of the IETF MANET workgroup. The major focal points have been the routing protocols, and the ongoing enhancements to the hardware of personal devices. The QoS research in ad-hoc networks has been getting less attention, compared to routing for instance, and therefore, evolved slower. Almost all of the QoS research efforts in ad-hoc networks have been focused so far on solutions within the ad-hoc network. We are interested in raising awareness to the end-to-end QoS problems when one of the communication endpoints is inside the ad-hoc network, and the other endpoint is outside. We present a unique view of the QoS in ad-hoc networks and we propose a QoS solution that operates on the gateway to the access network. The proposed gateway has a flexible design that links QoS models running on an ad-hoc network with QoS models running on the fixed structure access network. The proposal follows classical gateway design approaches to facilitate a lightweight implementation and to provide an integrated end-to-end QoS solution. Our results show that the use of our solution leads to substantial increase in effective bandwidth associated with relative decrease in bandwidth variations and end-to-end delays. In this paper, we show these enhancements, analyze the results and comment on the behavior of the proposed gateway in various operational scenarios.

Multicasting can efficiently support a wide variety of applications that are characterized by a close degree of collaboration, typical for many mobile ad-hoc network (MANET) applications currently envisioned. To deal with the specific characteristics of MANETs, new multicast protocols that operate in an on-demand manner are being proposed and investigated. Our results show that a tree-based on-demand protocol is not necessarily the best choice. A mesh-based protocol can outperform tree-based protocols, due to the availability of alternative paths, which allow multicast datagrams to be delivered to all or most multicast receivers even if links fail.

The implementation of an intermediary-proxy is a common approach to the problem of network heterogeneity in the Internet infrastructure. Due to the hypertext nature of the most popular Internet application-the World Wide Web, image compression is considered to be one of the fundamental functions of such a proxy. It has been observed that most images embedded into Web documents are of 'information-delivery' type, so an algorithm intended for their compression has to satisfy some specific requirements. First, in order to support network (bandwidth) constraints for an arbitrary case, the algorithm should be inherently adaptive, i.e. able to provide a wide range of compression rates. Second, as dealing with images that are integral parts of an interactive application (such as a Web browser), the algorithm should be capable of preserving a sufficient level of image semantics according to the quality standards of human perception. The vector quantization (VQ) technique, in its general form, is proven to satisfy the first requirement. On the other hand, a modified adaptive resonance (modified ART2) learning algorithm (which we employ in this paper) more properly 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 satisfy the second requirement. Thus, the discussion and results presented are intended to show that modified ART2 underlying the general VQ procedure is an appropriate techniques for image compression purposes in a bandwidth-constrained environment.

The characteristics of the data traffic generated by the use of micro-browser-enabled PCS phones to gain access to the Web is of particular interest to cellular network operators. Questions such as the frequency and length of browser sessions, and the specific characteristic of the traffic generated, need to be answer by researchers. These answers are valuable in network capacity planning as more subscribers use their cellular phones to interact with the Web.In this paper, we analyse network traces generated by a Mobile Browser application. We observe daily and weekly cycles and, some evidence of self-similarity in the network traffic produced by the application. We show that the activity factor for data generated by this application is lower than voice activity factor. This fact is significant for the design of wireless networks because it is possible to multiplex a large number of data sources onto existing channels with the appropriate protocols. We also compare and contrast the Mobile Browser traffic characteristics with results for WWW traffic published in the literature.

The objective evaluation of location management algorithms depends upon the underlying mobility model, and therefore more realistic mobility models should provide more accurate evaluations. An activity-based mobility model was developed, and used to evaluate both static and dynamic location management algorithms. We compare these results with evaluations of the same algorithms under two random mobility models. The simulation results show the significant impact of the mobility model on location management algorithm assessments.

The challenge of supporting rapidly growing numbers of mobile subscribers, while constrained by limited radio spectrum, is being met through increasingly smaller radio cells. This, however, results in increased signalling for location management procedures, which reduces the bandwidth available for user traffic. Location areas in current systems, such as GSM, consist of static and arbitrarily-defined collections of cells, which do not take into account individual subscriber mobility patterns, either in space or time. A location management algorithm is proposed which uses the mobility history of individual subscribers to dynamically create individualized location areas, based on previous movements from cell to cell. The average duration spent in each visited cell is also maintained and is used to define paging which are most likely to contain the subscriber. An activity-based mobility model was developed to test the proposed algorithm. Overall, the dynamic algorithm incurred significantly lower location management costs, in terms of signalling messages generated, for all parameters examined.

Designing and implementing a visual debugger for distributed programs is a significant challenge. Distributed applications are often large and frequently exhibit a high degree of complexity. Consequently, a debugger must address problems of complexity and scale in at least two ways. First, appropriate user interfaces should allow a user to manage the vast amount of information typically obtained from distributed executions. Second, the tool itself, in handling this information, should be implemented efficiently, providing a user with reasonable response times for interactive use. Our research efforts, concentrating on these problems, have led to the development of Poet, a tool for the collection and presentation of event-based traces of distributed executions. Poet makes as few assumptions as possible about characteristics that must be possessed by all target environments. Information describing each target environment is placed in configuration files, allowing a single set of Poet executables to be used for all target environments. Comparing Poet's performance to XPVM, the standard visualization tool for PVM executions, reveals that this target-system independence does not impose a performance penalty.

Debugging distributed applications presents many challenges in addition to those found in debugging sequential applications. This paper describes a tool, and the principles underlying it, that has been developed to assist in debugging such distributed applications. Although the tool can also be applied in other environments, this paper primarily describes its application to OSF DCE. Special attention is also given to a facility that has presently been implemented only for OSF DCE, the ability to replay an application, that is, to re-execute it with execution constrained to follow the partial order of an initial execution.

Distributed applications written in Hermes typically consist of a large number of sequential processes. The use of a hierarchy of process clusters can facilitate the debugging of such applications. Ideally, such a hierarchy should be derived automatically. This paper discusses two approaches to automatic process clustering, one analyzing runtime information with a statistical approach and one utilizing additional semantic information. Tools realizing these approaches were developed and a quantitative measure to evaluate process clusters is proposed. The results obtained under both approaches are compared, and indicate that the additional semantic information improves the cluster hierarchies derived. We demonstrate the value of automatic process clustering with an example. It is shown how appropriate process clusters reduce the complexity of the understanding process, facilitating program maintenance activities such as debugging.>

Understanding the behavior of distributed applications is a very challenging task due to the complexity of these applications. To manage complexity, the top-down use of suitable abstraction hierarchies is frequently proposed. Given the complexity of distributed applications, manually deriving such abstraction hierarchies is not realistic. The execution of distributed applications is typically analyzed using an event-based approach. This paper discusses one tool that groups more primitive events into abstract events to derive a hierarchy of abstract events automatically. Ideally, these abstractions should reveal logical units of an application and their relations. To explore the abstraction hierarchies derived, an existing prototype visualization tool was modified to provide abstract visualizations. A user can navigate through these abstraction hierarchies, displaying an execution at various levels of abstraction. Examples of such abstract visualizations are given and discussed. In general, the abstractions derived automatically represent meaningful parts of the application: they can be interpreted in terms of the application domain. While the abstraction tool does not necessarily derive the best possible abstraction hierarchies in all cases, it performs the bulk of the work and provides good initial abstractions which can subsequently be refined manually.

A task scheduler based on the concept of a stochastic learning automation, implemented on a network of Unix workstations, is described. Creating an artificial, executable workload, a number of experiments were conducted to determine the effect of different workload descriptions. These workload descriptions characterize the load at one host and determine whether a newly created task is to be executed locally or remotely. Six one-dimensional workload descriptors are examined. Two workload descriptions that are more complex are also considered. It is shown that the best single workload descriptor is the number of tasks in the run queue. The use of the worst workload descriptor, the 1-min load average, resulted in an increase of the mean response time of over 32%, compared to the best descriptor. The two best workload descriptors, the number of tasks in the run queue and the system call rate, are combined to measure a host's load. Experimental results indicate that no performance improvements over the scheduler versions using a one-dimensional workload descriptor can be obtained.>