Abstract IoT video applications such as public transportation, traffic monitoring, parking and surveillance systems in smart cities collect and transmit video over multi-hop wireless network fabrics which are known to be less reliable and responsive as their wired counterparts. While multipath routing has emerged as an effective strategy to help meet IoT video transmission’s, one of the main challenges current multipath routing approaches face is to be able to select routes that meet the requirements of the driving video applications in a timely manner. To fill this gap, we recently proposed FITPATH, an adaptive multipath selection mechanism for IoT video transmission that uses a heuristic-based iterative optimization approach to estimate in real time the conditions of the underlying network while accounting for the different bitrate requirements of the application video flows. To provide a complete understanding of FITPATH’s performance, limitations, and applicability in realistic IoT video transmission scenarios, this work presents a comprehensive experimental evaluation of FITPATH under a wide range of conditions. We demonstrate that FITPATH consistently outperforms existing multipath selection mechanisms in terms of both end-user Quality of Experience (QoE) and overall network performance. We also evaluate the performance and convergence of FITPATH’s heuristic, showing that it quickly generates feasible solutions and incrementally improves QoE as the algorithm iterates. Furthermore, we propose practical deployment strategies for FITPATH under both centralized and decentralized network control plane architectures. Additionally, we analyze the impact of control messages and network dynamics on video quality in decentralized network control plane environments.

Video applications in the Internet of Things benefit from multipath routing strategies to meet Quality of Service requirements. Although several multipath selection mechanisms have been proposed, changes in topology and/or in video services requirements can impact video quality. This work presents a wireless network architecture that enables dynamic routing, adapting these changes to new path solutions that provide a better Quality of Experience for the users. Simulation results demonstrated that the proposed architecture is capable of recovering video quality in case of degradation due to topology change, as well as also confirming that the presence of control flows, essential for maintaining the topology, reduces the final video quality.

This paper explores clustering as a technique to improve energy efficiency for a variety of current and emerging IoT application scenarios. We introduce a novel load balancing clustering algorithm based on Simulated Annealing whose main goal is to increase network lifetime while maintaining adequate sensing coverage in scenarios where sensor nodes produce non-uniform data traffic. Through simulations on the Cooja/Contiki simulation-emulation platform, we compare the proposed algorithm to leading state-of-the-art clustering approaches and show that our algorithm is able to improve both time until first clusterhead failure as well as network coverage by keeping more sensor nodes alive for longer periods of time. We also explore different clustering criteria applied to our simulated annealing clustering framework and investigated different network lifetime metrics. We show that accounting for both workload and communication cost improves network lifetime by prolonging clusterhead lifetime and that clustering solely on the basis of physical distance improves network coverage.

Vehicle platooning has been shown to significantly improve road safety and driver experience, as well as increase road capacity, reduce fuel consumption, and thus green house emission. We present the Adaptive Decentralized Emergent-based Platooning system, or ADEPT for short, an emergent behavior based autonomous platooning approach that adopts a bottom-up, biologically inspired approach where each vehicle follows a set of simple rules to execute platoon maneuvers. In this paper, we evaluate ADEPT’s network performance against centralized platooning using a platooning simulator and show that ADEPT is able to yield superior network utilization. Additionally, we study the interplay between our emergent platooning system and vehicle characteristics and dynamics such as engine performance, weight, center of mass, friction, speed, and acceleration, and evaluate using a robot simulator equipped with an open dynamics physics engine. Experimental results demonstrate that ADEPT’s controller is string stable when subject to different maneuvers and disturbances. In addition to evaluating ADEPT’s controller stability, we propose methods that enable ADEPT to handle real-world conditions such as curved roads, vehicle following and obstacle avoidance. Our experimental results show that our dynamic radius of curvature determination method helps maintain inter-vehicle gap at 99% of the desired gap including on curved roads. Even during disturbance events, the gap errors are kept within the desired deviation. Using inspiration from lane assist technology, we propose a vehicle following technique that, according to our experimental results, yields high accuracy.

In this paper, we describe a novel cross-layer fingerprinting approach (CL-FP) that aims at uniquely identifying wireless devices by extracting inherent cross-layer device features based on the requirements of the driving network application(s). We demonstrate how our system can be applied to MAC spoofing detection. Our contributions include: (1) We propose the CL-FP framework and pipeline for detecting MAC spoofing attacks, which, to our knowledge, is the first of its kind; (2) We experimentally confirm theoretical results showing that easier-to-extract FP features can reliably represent harder-to-extract intrinsic physical characteristics of devices; (3) We test and evaluate the performance of the proposed CL-FP approach through simulations in the context of the MAC spoofing detection use case. Our preliminary results show that the proposed CL-FP pipeline provides a lightweight, scalable and reliable end-to-end cross-layer device fingerprinting framework.

In this paper, we introduce a novel algorithm that uses machine learning to dynamically decide whether to enable or disable IEEE 802.11 DCF’s RTS/CTS. Our algorithm continuously learns current networking conditions, namely air time, i.e. the ratio between the size of data/control information being transmitted and transmission rate, and network contention to compare the cost between using RTS/CTS or retransmitting data, and dynamically switches RTS/CTS on and off accordingly. Simulation results using a variety of WLAN- as well as wireless multi-hop ad-hoc network scenarios, including synthetic and real traffic traces, demonstrate that the proposed approach consistently outperforms current best practices, such as never enabling RTS/CTS or using a pre-specified threshold to decide whether to switch RTS/CTS on or off.

Motivated by Smart City applications and services, this article presents a novel approach to identifyinguser communitiesin communication networks. We define user communities as groups of users that share common mobility features over spatio-temporal scales of arbitrary length, such as time spent in certain locales, mobility speed, and time between consecutive movements. We describe our user community identification framework in detail including how mobility features can be extracted from real mobility traces (as examples of unlabelled data) and synthetic mobility records (as examples of labeled data). We present results obtained when using our approach in four distinct mobility scenarios represented by both unlabeled and labeled datasets. We also introduce a new validation methodology that uses image-based similarity metrics in order to assess the quality of identified communities. Our results show that the proposed approach significantly increases similarity between users within the same community as well as dissimilarity between users in different communities. We also demonstrate that the proposed user community identification approach yields significant increase in contact time amongst users belonging to the same community when compared to the average contact time when not considering community structures.

In this paper, we introduce PIMAP, an IoT-based system for continuous, real-time patient monitoring that operates in a fully autonomous fashion, i.e. without the need for human intervention. To our knowledge, PIMAP is the first open system that integrates the basic patient monitoring workflow for continuous and autonomous operation and includes sensed data collection, storage, analysis, and real-time visualization. PIMAP’s open design allows it to integrate a variety of sensors (custom and off-the-shelf), analytics, and visualization. Other novel features of PIMAP include its deployment flexibility, i.e., its ability to be deployed in different configurations depending on the specific application needs, setting, and resources, as well as PIMAP’s self-profiling and self-tuning capabilities. While PIMAP can be applied to various patient monitoring applications and settings, in this paper we focus on the unsolved problem of preventing pressure injuries.

In this paper, we introduce a novel decentralized approach to automated vehicle platooning inspired by emergent-behavior commonly found in biological systems. The proposed emergent-behavior based platooning uses simple rules that the vehicles follow when they interact to carry out maneuvers. Platooning maneuvers such as join, exit, and lane change "emerge" as a result of vehicles following these "emergent rules". Preliminary complexity analysis as well as simulation experiments yield promising results that demonstrate emergent behavior platooning as a viable form of decentralized platooning.

In this article, we introduce a novel dynamic controller assignment algorithm targeting connected vehicle services and applications, also known as Internet of Vehicles (IoV). The proposed approach considers a hierarchically distributed control plane, decoupled from the data plane, and uses vehicle location and control traffic load to perform controller assignment dynamically. We model the dynamic controller assignment problem as a multi-agent Markov game and solve it with cooperative multi-agent deep reinforcement learning. Simulation results using real-world vehicle mobility traces show that the proposed approach outperforms existing ones by reducing control delay as well as packet loss.

Terrestrial infrastructure-based wireless networks do not always guarantee their resources will be shared uniformly by nodes in vehicular networks. This is due mainly to the uneven and dynamic geographical distribution of vehicles and path loss effects. In this paper, we leverage multiple fifth-generation (5G) unmanned aerial vehicles (UAVs) to enhance fairness in network resource allocation among vehicles by positioning UAVs on-demand as “flying communication infrastructure”. We propose a deep reinforcement learning (DRL) approach to determine UAVs’ position to improve network resource allocation fairness and efficiency while considering the UAVs’ flying range, communication range, and energy constraints. We use a parametric fairness function to attain a number of resource allocation objectives ranging from maximizing the total throughput of vehicles, maximizing minimum throughput, and achieving proportional bandwidth allocation. Simulation results show that the proposed DRL approach to UAV positioning can improve network resource allocation according to the targeted fairness objective.

In this paper, we introduce Distributed Grid Partition, a distributed greedy deployment algorithm for outdoor IoT camera networks. The proposed algorithm optimizes visual network coverage over 2.5D terrain. The main idea behind Distributed Grid Partition is that each deployment node tries to find the best vantage point in its neighborhood that will maximize the network's overall visual coverage. It does so by using information from its immediate neighbors. In order to achieve a favorable cost-performance trade-off, Distributed Grid Partition uses height as a proxy for visual coverage, or fitness, avoiding expensive fitness computations. In addition, each node's contribution to network fitness is determined without knowledge of the overall network using the concept of "Wonderful Life Utility". Our experimental results show that Distributed Grid Partition results in deployments with superior coverage-cost performance when compared to other distributed optimization algorithms as well as a centralized greedy set cover heuristic.

The healthcare industry has placed significant effort on reducing pressure injuries as they often occur while patients are seeking treatment for an unrelated condition, lengthen the stay of the patient, impact treatment options, are extremely painful, and can result in death. Current work has shown that an increase in nursing care can reduce pressure injury occurrence, but increased nursing care is not sustainable. The Braden Scale is the current method of stratifying patients by risk of forming a pressure injury, but previous work has shown that the Braden Scale is not an accurate predictor of pressure injury formation. We present an objective mobility measure of pressure injury risk that can be used in addition to the Braden Scale using a wearable pressure-sensing device. By assessing the current pressure readings in real-time we can determine when a patient is mobile and present this data to the clinic. In future we aim to use this mobility measurement in a Cloud-based system that can compare the mobility measurements to similar patients to assess risk with increased accuracy.

Motivated by the need to accommodate IoT devices that have limited power, processing, storage, and communication capabilities, this work introduces IoTUS, or IoT Unified Services (IoTUS), a novel network protocol architecture that targets energy efficiency and compact memory footprint. IoTUS uses an extensible service layer that facilitates cross-layer sharing. It promotes sharing of both network control information, (e.g., number of transmissions, receptions, collisions at the data-link layer) and functionality (e.g., neighbor discovery, aggregation) by different layers of the protocol stack. Additionally, IoTUS can be used by existing network stacks without having to modify the basic operation of their protocols. We implemented IoTUS on the Cooja-Contiki network simulator/emulator. Our experimental results show improved energy efficiency resulting in longer network lifetime, as well as more compact memory footprint resulting when compared to current IoT protocol architectures.

Opportunistic networking has been proposed to address episodic connectivity, common in so-called "challenged" or "extreme" networking environments where arbitrarily frequent and long-lived connectivity disruptions are the norm, instead of the exception. Examples of extreme networking environments and applications include interplanetary communication, disaster relief, and emergency response, (semi-)autonomous driving, to name a few. This paper proposes a novel community-based opportunistic routing approach that identifies user communities based on mobility features extracted from real traces of user mobility. The proposed Deep AutoenCoder Community-based Opportunistic Routing protocol, or DACCOR, employs deep learning to identify user communities based on data extracted from user mobility traces and uses user community information to make forwarding decisions in opportunistic networking scenarios. Through extensive simulations, we evaluate DACCOR's performance and show that it outperforms well-known opportunistic forwarding protocols in terms of delivery probability, latency, and communication overhead. We also show that DACCOR's lower communication overhead yields considerable energy efficiency, increasing mobile devices' battery lifetime.

The IEEE 802.11's binary exponential backoff (BEB) algorithm plays a critical role in the throughput performance and fair channel allocation of IEEE 802.11 networks. In particular, one of BEB algorithm's parameters, the Contention Window determines how long a node needs to wait before it (re)transmits data. Consequently, choosing adequate values of the Contention Window is crucial for IEEE 802.11's performance. In this paper, we introduce a simple, yet effective machine learning approach to adjust the value of IEEE 802.11's Contention Window based on present- as well as recent past network contention conditions. Using a wide range of network scenarios and conditions, we show that our approach outperforms both 802.11's BEB as well as an existing contention window adjustment technique that only considers the last two transmissions. Our results indicate that our contention window adaptation algorithm is able to deliver consistently higher average throughput, lower end-to-end delay, as well as improved fairness.

In this paper, we propose a network planning and provisioning framework that optimizes deployment cost in C-RAN based 5G networks. Our framework is based on a Mixed Integer Quadratically Constrained Programming (MIQCP) model which optimizes ``virtualized'' 5G service chain deployment cost while performing adequate provisioning to address user demand and performance requirements. We use two realistic scenarios to showcase that our framework can be applied to different types of deployments and discuss the computational cost and scalability of our solution.

In this paper, we build upon our prior work on D2-ITS, a flexible and extensible framework to dynamically distribute network control to enable message dissemination in Intelligent Transport Systems (ITS), and extend it with handover and load balancing capabilities. More specifically, D2-ITS' new handover feature allows a controller to automatically "delegate" control of a vehicle to another controller as the vehicle moves. Control delegation can also be used as a way to balance load among controllers and ensure that required application quality of service is maintained. We showcase D2-ITS' handover and load- balancing features using the Mininet-Wifi network simula- tor/emulator. Our preliminary experiments show D2-ITS' ability to seamlessly handover control of vehicles as they move.

This paper explores clustering as a technique to improve energy efficiency for a variety of current and emerging IoT application scenarios. We introduce a novel load balanced clustering algorithm based on Simulated Annealing whose main goal is to increase network lifetime while maintaining adequate sensing coverage in scenarios where sensor nodes produce uniform or non-uniform data traffic. To this end, we also introduce a new clustering cost function that accounts not only for sensor node traffic load but also for the cost of communicating over physical distances. Through extensive simulations comparing the proposed algorithm to leading state-of-the-art clustering approaches, we show that our algorithm is able to improve both network lifetime as well as network coverage by keeping more sensor nodes alive for longer periods of time at lower computational cost.

In this paper, we proposed a novel algorithm to dynamically enable and disable IEEE 802.11 DCF's RTS/CTS handshake. We start by conducting an experimental characterization of the performance of RTS/CTS as a function of packet size, transmission rate, and network contention, which complements existing work that evaluated RTS/CTS performance analytically and empirically. Motivated by our experimental evaluation of RTS/CTS performance, our algorithm uses current packet size and transmission rate, as well as an estimate of network contention to dynamically decide whether to use RTS/CTS or not. To the best of our knowledge, the proposed algorithm is the first to enable and disable the RTS/CTS handshake based on a set of current network conditions, and automatically adapt as these conditions change. Simulation results using a variety of WLAN scenarios, including synthetic and real traffic traces, demonstrate that the proposed approach consistently outperforms current best practices, such as never enabling RTS/CTS or setting the RTS Threshold (RT), which is used to decide whether to switch RTS/CTS on or off, to a static value.

Welcome to the IEEE JSAC special issue onAirborne Communication Networks. The goal of this special issue is to disseminate the contributions in the field of airborne communication networks.

Characterizing and modeling node mobility is of critical importance in building intelligent transportation systems and their applications. In this paper, we discuss the scale-free properties of some important human mobility characteristics, namely spatial node density and mobility degree, and show that they exhibit behavior that can be described by a power-law. Based on their power-law characteristics, we derive analytical models for the spatial node density and mobility degree and show that the data generated by the proposed analytical models closely approach the empirical data extracted from the real mobility traces. Another contribution of our work is to use the proposed analytical models to build a synthetic mobility regime that is suitable for simulations of intelligent transportation systems. Finally, through network simulations, we show that the ad-hoc network routing behavior under our mobility regime closely approximates routing behavior when the corresponding real trace is used.

We propose D2-ITS, a flexible and extensible framework to dynamically distribute network control to enable message dissemination in Intelligent Transport Systems (ITS). By decoupling the control-from the data plane, D2-ITS leverages network programmability to address ITS scalability, delay intolerance and decentralization. It uses a distributed control plane based on a hierarchy of controllers that can dynamically adjust to environment- and network conditions in order to satisfy ITS application requirements. We demonstrate the benefits of D2-ITS through a proof-of-concept prototype using the ns-3 simulation platform. Results indicate lower message delivery latency with minimal additional overhead.

In smart grids, the broad use of Distributed Energy Resources (DERs) in distribution networks introduces the need for protective relaying schemes similar to those used in high-voltage networks. The introduction of intermittent DERs, as solar panels, also requires more autonomic and dynamic SCADA networks. Hence, power systems experience an increased demand for resilience in the distribution communication network. However, resilience methods currently in use still cannot meet those protection requirements. This work proposes ARES, a framework for autonomic and resilient communication for smart grids. ARES provides resilient, robust, and flexible communication for smart grids with Software Defined Network (SDN). Our proposal also provides autonomic services for SCADA that can improve smart grid application performance and efficiency. ARES fault resilience module is implemented and tested using Mininet 2.2.1 and RYU controller and presents maximum recovery time of 610 microseconds, which is an important advance compared to other proposals. In addition, ARES is transparent to end devices, keeping compatibility with legacy measurement and actuation devices.

No one likes waiting in traffic, whether on a road or on a computer network. Stuttering audio, slow interactive feedback, and untimely pauses in video annoy everyone and cost businesses sales and productivity. An ideal network should (1) minimize latency, (2) maximize bandwidth, (3) share resources according to a desired policy, (4) enable incremental deployment, and (5) minimize administrative overhead. Many technologies have been developed, but none yet satisfactorily address all five goals. The best performing solutions developed so far require controlled environments where coordinated modification of multiple components in the network is possible, but they suffer poor performance in more complex scenarios. We present TCP Inigo, which uses independent delay-based algorithms on the sender and receiver (i.e. ambidextrously) to satisfy all five goals. In networks with single administrative domains, like those in data centers, Inigo's fairness, bandwidth, and latency indices are up to 1.3X better than the best deployable solution. When deployed in a more complex environment, such as across administrative domains, Inigo possesses latency distribution tail up to 42X better.

We present TerrainLOS, an outdoor propagation model that uses Digital Elevation Models to determine whether two nodes can communicate. We have implemented TerrainLOS in the sensor network simulator COOJA and used it to evaluate how the roughness of terrain, which we classify using Average Cumulative Visibility, affects the number of edges, connectedness, latency, and power of a network. We compare the difference in results when using TerrainLOS and a simpler propagation model to show how the performance of outdoor simulation is greatly affected by a model that takes terrain into account.

In this paper we investigate and characterize user activity in WiFi networks by analyzing and comparing the behavior of users that connect to two public WiFi networks, one of them deployed in a University campus and the other in a major urban area. We characterize WiFi network user activity based on two main features, namely: time users stay connected to Access Points and Access Point load. Overall, the main contributions of our work are as follows: (1) to the best of our knowledge, this is the first study comparing user activity in two different scenarions, i.e., a University campus WiFi network and an urban WiFi network; (2) our results validate previously observed characteristics of user behavior in WiFi networks, as well as unveil new behavior patterns, such as the fact that users on campus tend to stay connected to the network for longer periods of time when compared to users in an urban area; and (3) our work is the first study to formally test and validate the hypothesis that association times in WiFi networks follows a power law and to estimate the power-law's tail index.

In this paper, we propose a novel congestion control framework for delay- and disruption tolerant networks (DTNs). The proposed framework, called Smart-DTN-CC, adjusts its operation automatically as a function of the dynamics of the underlying net- work. It employs reinforcement learning, a machine learning technique known to be well suited to problems in which the environment, in this case the network, plays a crucial role; yet, no prior knowledge about the target environment can be assumed, i.e., the only way to acquire information about the environment is to interact with it through continuous online learning. Smart-DTN-CC nodes get input from the environment (e.g., its buffer occupancy, set of neighbors, etc), and, based on that information, choose an action to take from a set of possible actions. Depending on an action's effectiveness in controlling congestion, it will be given a reward. Smart-DTN-CC's goal is to maximize the over- all reward which translates to minimizing congestion. To our knowledge, Smart-DTN-CC is the first DTN congestion control framework that has the ability to automatically and continuously adapt to the dynamics of the target environment. As demonstrated by our experimental evaluation, Smart-DTN- CC is able to consistently outperform existing DTN congestion control mechanisms under a wide range of network conditions and characteristics.

Several network protocols, services, and applications adjust their operation dynamically based on current network conditions. Consequently, keeping accurate estimates of the network and its performance as it fluctuates over time is critical. For example, both TCP and IEEE 802.11 periodically adapt some of their key operating parameters, namely, the retransmission timeout and the contention window size based on the average round trip time and the number of collisions, respectively. In this paper, we present a novel mechanism to estimate “near-future” network performance based on past network conditions. We call our approach to network performance estimation as smart experts for network state estimation (SENSE). SENSE uses a simple, yet effective, algorithm combining a machine learning method known as fixed-share with exponentially weighted moving average (EWMA). SENSE also introduces novel techniques that improve the predictability of the fixed-share framework without increasing computational complexity. SENSE is thus able to respond to network dynamics at different time scales, i.e., long- and medium-term fluctuations as well as short-lived variations. We evaluate SENSE's performance using synthetic and real datasets. Our experimental results show that, when compared to fixed-share and EWMA, SENSE yields higher estimation accuracy for all datasets due to its ability to more closely track data fluctuations.

We propose a novel distributed deployment algorithm for sensor networks whose nodes reside upon and are obstructed by 2.5D terrain. Our algorithm optimizes area coverage by computing cumulative visibility over terrain. Through simulation and comparison to centralized algorithms, we demonstrate that our distributed algorithm achieves good results and degrades gracefully with reduced internode communication. To the best of our knowledge, our distributed deployment algorithm is the first use of a distributed simulated annealing algorithm for sensor network deployment. In addition, to our knowledge, this is the first time range-limited cumulative visibility is used to guide sensor deployment over 2.5D terrain. Our results show that a centralized Simulated Annealing algorithm outperforms Pattern Search and Gradient Ascent approaches. Results also show that our version of Distributed Simulated Annealing performs well, degrading gracefully as communication radius is reduced.

Wireless visual sensor networks find a wide range of applications such as environment and habitat monitoring, object tracking, surveillance, etc. Unlike sensor networks targeting telemetry applications, which typically use "scalar" sensors, visual sensor networks produce richer, multi-media information and therefore raise additional challenges related to efficient use of energy-, node-, and network-resources. In this paper, we introduce SCmesh, a mesh network of solar-powered wireless camera nodes. Using SCmesh, low-cost, solar-powered, visual sensing nodes are able to transfer video files as well as collaborate on tasks such as visual processing, event detection, object tracking, etc. SCmesh allows node-to-node, as well as scoped- or full broadcast communication. We deploy SCmesh in a smart visual sensor network deployment and showcase its functionality and performance.

In this paper, we propose a novel modeling framework to study congestion in delay- and disruption tolerant networks (DTNs). The proposed model is based on directed site-bond percolation where sites represent space-time positions of DTN nodes, and bonds are contact opportunities, i.e. Communication links that can be established whenever nodes come in range of each other. To the best of our knowledge, this is the first model of DTN congestion using percolation theory. The proposed modeling framework is simple yet general and can be used to evaluate different DTN congestion control mechanisms in a variety of scenarios and conditions. We validate our model by showing that its results match quite well results obtained from the ONE DTN simulation platform. We also show that our model can be used to understand how parameters like buffer management policy, buffer size, routing mechanism, and message time-to-live affect network congestion.

Motivated by the internets of the future, which will likely be considerably larger in size as well as highly heterogeneous and decentralized, we propose Decentralize-SDN, D-SDN, a framework that enables not only physical- but also logical distribution of the Software-Defined Networking (SDN) control plane. D-SDN accomplishes network control distribution by defining a hierarchy of controllers that can “match” an internet's organizational- and administrative structure. By delegating control between main controllers and secondary controllers, D-SDN is able to accommodate administrative decentralization and autonomy.It incorporates security as an integral part of the framework. This paper describes D-SDN and presents two use cases, namely network capacity sharing and public safety network services.

Several core network protocols and applications adjust their operation dynamically based on current network conditions. TCP and IEEE 802.11 are notable examples, both of which periodically adapt the retransmission timeout and the contention window size depending on the average round trip time an

In this paper, we show empirically that the spatial node density resulting from human mobility follows a power law. We also show that the number of locales visited by users also exhibit heavy-tail behavior. We develop a stochastic model that confirms our empirical observations by showing that node mobility resulting from our model closely approximates mobility recorded in real traces collected from a variety of scenarios. Besides corroborating our empirical observations, we showcase another application of our model by using it to generate mobility regimes whose spatial node density exhibit heavy-tail behavior. We validate the resulting mobility generator by comparing its output against real traces.

This paper proposes a novel approach to capacity sharing in hybrid networked environments, i.e., environments that consist of infrastructure-based as well as infrastructureless networks. The proposed framework is based on Software-Defined Networking (SDN) and provides flexible, efficient, and secure capacity sharing solutions in a variety of hybrid network scenarios. In this paper, we describe the challenges raised by capacity sharing in hybrid networks, describe our framework in detail and how it addresses these challenges, and discuss implementation issues. To the best of our knowledge, this is the first SDN-based capacity sharing solution that targets hybrid networks and that incorporates security as an integral part of the proposed approach.

User privacy on the Internet has been an increasing concern in recent years. With the proliferation and sophistication of information services, data mining, and search engines, a simple network address may be used to reveal a great deal of information about a user, including location, identity, and behavior. Existing approaches to privacy, however, make unacceptable tradeoffs between performance and anonymity. For example, Tor [5] attempts to provide strong anonymity by withholding trust from third-party relays. We believe an acceptable level of privacy can be provided to most users, with noticeably lower latency and throughput impact, by working with the network provider. In this paper, we introduce AnonyFlow, an in-network anonymization service designed to efficiently and seamlessly provide privacy to users as they communicate with other endpoints and services. We design, implement, and evaluate an OpenFlow-based prototype of AnonyFlow that achieves endpoint anonymity without compromising on throughput or latency.

This paper introduces a modeling framework to analyze spatial node density in mobile networks under “waypoint”-like mobility regimes. The proposed framework is based on a set of first order ordinary differential equations (ODEs) that take as parameters (1) the probability of going from one subregion of the mobility domain to another and (2) the rate at which a node decides to leave a given subregion. We validate our model by using it to describe the steady-state behavior of real user mobility recorded by GPS traces in different scenarios. To the best of our knowledge, this is the first node density modeling framework generic enough that can be applied to any “waypoint”-based mobility regime.

In this paper, we introduce a user mobility modeling framework that accounts for both the users' social structure as well as the geographic diversity of the region of interest. SAGA, or Socially- and Geography-Aware mobility model, captures social features through the use of communities which cluster users with similar features such as average time in a cell, average speed, and pause time. SAGA accounts for geographic diversity by considering that different communities exhibit different interests for different locales; therefore, different communities are attracted to certain physical locations with different intensities. Besides introducing SAGA, the contributions of this work include: a model calibration approach based on formal statistical procedures to extract social structures and geographical diversity from real traces and set SAGA's parameters; and validation of SAGA by applying it to real mobility traces. Our experimental results show that, when compared to existing mobility regimes such as Random-Waypoint and Preferential-Attachment based mobility, SAGA is able to preserve the desired non-uniform node spatial density present in real user mobility, creating and maintaining clusters and accounting for differential node popularity and transitivity.

In this paper we introduce a novel collision-free, schedule-based medium access control protocol for wireless networks. Our protocol, TRANSFORMA (TRAffic FORecasting Medium Access) uses traffic forecasting to significantly reduce packet delivery delays for delay-sensitive applications while maintaining delivery ratios higher than those of contention-based protocols. TRANSFORMA's novel approach to channel access uses the forecast data rate of each application flow to perform distributed probabilistic channel scheduling. We show through simulations that, thanks to its per-application traffic forecasting capabilities, TRANSFORMA yields lower average delays when compared against DYNAMMA, an existing schedule-based MAC, and against 802.11 under high load. TRANSFORMA caters to emerging high data rate, real-time services that will likely be prevalent particularly at the edges of the Internet of the future. Such services which are currently represented by applications such as Skype, Google Talk, and iChat, exhibit traffic characteristics that are fairly predictable and thus well served by TRANSFORMA's traffic forecasting abilities.

In this paper, we explore the idea of using traffic forecasting to improve the delay performance of a schedule-based medium access control protocol. Schedule-based channel access has been shown to utilize network and energy resources efficiently but is often hindered by the extra delay that scheduling introduces. We explore the use of traffic forecasting to anticipate transmission schedules instead of establishing them reactively, thereby reducing scheduling delays. We show the potential performance benefits traffic forecasting can bring to schedule-based medium access in the context of an existing MAC protocol called DYNAMMA. Preliminary results using a machine-learning based traffic forecasting technique are also presented.

In this paper, we explore a novel approach to end-to-end round-trip time (RTT) estimation using a machine-learning technique known as the Experts Framework. In our proposal, each of several "experts" guesses a fixed value. The weighted average of these guesses estimates the RTT, with the weights updated after every RTT measurement based on the difference between the estimated and actual RTT. Through extensive simulations we show that the proposed machine-learning algorithm adapts very quickly to changes in the RTT. Our results show a considerable reduction in the number of retransmitted packets and a increase in goodput, in particular on more heavily congested scenarios. We corroborate our results through "live" experiments using an implementation of the proposed algorithm in the Linux kernel. These experiments confirm the higher accuracy of the machine learning approach with more than 40% improvement, not only over the standard TCP, but also over the well known Eifel RTT estimator.

The Internet has been evolving into a more heterogeneous internet work with diverse new applications imposing more stringent bandwidth and QoS requirements. Already such new applications such as YouTube, Hulu, and Netflix are consuming a large fraction of the total bandwidth. We argue that, in order to engineer future internets such that they can adequately cater to their increasingly diverse and complex set of applications while using resources efficiently, it is critical to be able to characterize the load that emerging and future applications place on the underlying network. In this paper, we investigate entropy as a metric for characterizing per-flow network traffic complexity. While previous work has analyzed aggregated network traffic, we focus on studying isolated traffic flows. Per-application flow characterization caters to the need of network control functions such as traffic scheduling and admission control at the edges of the network. Such control functions necessitate differentiating network traffic on a per-application basis. The "entropy fingerprints" that we get from our entropy estimator summarize many characteristics of each application's network traffic. Not only can we compare applications on the basis of peak entropy, but we can also categorize them based on a number of other properties of the fingerprints.

In this paper we show that human mobility exhibits "persistent" behavior in terms of the spatial density distribution of the mobile nodes over time. Using real mobility traces, we observe that the original non-homogeneous node spatial density distribution, where some regions may be quite dense while others may be completely deserted, is maintained at different instants of time. We also show that mobility models that select the next node position based on the position of other nodes, a la "preferential attachment", do not preserve the original spatial node density distribution and lead to behavior similar to random mobility as exemplified by the Random Waypoint model. To the best of our knowledge, this is the first time that these phenomena have been reported. Based on these observations, we propose a simple mobility model that preserves the desired spatial density distribution. Moreover, when simulating the operation of a network moving according to the proposed model, we found that performance results expressed by a number of network metrics also match closely results obtained under mobility governed by real traces. We also compare our results to models whose steady-state do not preserve the original non-homogeneous density distribution and show that network performance under such regimes deviates from performance under real trace mobility.

Ubiquity of portable computing devices coupled with wide availability of wireless communication present new impor- tant opportunities for applications involving media-rich content dissemination. However, as access networks become increasingly more heterogeneous, seamless data delivery across internets consisting of a variety of network technology becomes a real challenge. In this demonstration, we showcase a system that enables content dissemination over heterogeneous internets consisting of wired, infrastructure-based and infrastructure-less wireless networks that may be prone to intermittent connectivity. Using an efficient, yet flexible buffer management scheme, we are able to address application-specific performance requirements such as average delay, delivery probability, energy efficiency, etc. Our system uses the Message Delivery in Heterogeneous, Disruption-prone Networks (MeDeHa) [2]) framework to deliver messages across a heterogeneous internet coupled with History-Based Scheduling and Drop (HBSD) buffer management [1] as a way to optimize resources provided by opportunistic networks. MeDeHa, which is described in detail in [2], provides seamless data delivery over interconnecting networks of different types, i.e., infrastructure-based and infrastructure-less networks. MeDeHa's comprehensive approach to bridging infrastructure-based and infrastructureless networks also copes with intermittent connectivity. For this demonstration, we showcase a "complete stack" solution featuring, from to top to bottom, the DTN2 "bundle" layer, HBSD as an "external router" to DTN2, and MeDeHa, which handles message delivery. We have implemented, on a Linux-based testbed, (i) the MeDeHa framework, (ii) the HBSD [3] external router for the DTN2 [4] architecture.

In this paper, we present a naming scheme for heterogeneous networks composed of infrastructure-based and infrastructure-less networks where nodes may be subject to intermittent connectivity. The proposed scheme aims at decoupling object identification from location and is designed to operate with status-quo Internet routing. We showcase the proposed naming scheme implemented on the NS-3 network simulator and demonstrate that nodes are able to receive messages in both infrastructure-based and infrastructure-less networks despite frequent disconnections and changing location identifiers (i.e., IP address), while visiting different networks.

In this paper analyzed WLAN-, GPS-, and synthetic traces that record mobility in a variety of network environments. We observe that from a macroscopic level, human mobility is symmetric. In other words, the number of users that move from point A to point B approximates the number of users that go in the opposite direction, i.e., from B to A. We show that this type of symmetry is more accentuated in synthetic mobility models, in particular, in random way-point mobility. We also study the direction of movement which also exhibits symmetric behavior in both real- as well as synthetic mobility. Additional contributions of our work include metrics to quantify mobility symmetry. We conclude the paper with a discussion of possible applications of our results in mobile networking.

NA

Some current and future pervasive data driven applications must operate in “extreme” environments where end-to-end connectivity cannot be guaranteed at all times. In fact, it is likely that in these environments partitions are, rather than exceptions, part of the normal network operation. In this paper, we introduce Cover, a suite of adaptive strategies to control the trajectory of “infrastructure” nodes, which are deployed to bridge network partitions and thus play a critical role in data delivery. In particular, we focus on applications where end (or target) nodes are mobile and their mobility is unknown. Our goal is then to deploy and manage infrastructure nodes so that application-level requirements such as reliable data delivery and latency are met while still limiting deployment cost and balancing the load among infrastructure nodes. Cover achieves these goals using a localized and adaptive approach to infrastructure management based on the observed mobility of target nodes. To this end, Cover takes advantage of contact opportunities between infrastructure nodes to exchange information about their covered zones, and thus, help monitor targets in a more efficient fashion. Through extensive simulations, we show how Cover's adaptive features yield a fair distribution of targets per infrastructure node based only on limited network knowledge.

Communication networks, whether they are wired or wireless, have traditionally been assumed to be connected at least most of the time. However, emerging applications such as emergency response, special operations, smart environments, VANETs, etc. coupled with node heterogeneity and volatile links (e.g. due to wireless propagation phenomena and node mobility) will likely change the typical conditions under which networks operate. In fact, in such scenarios, networks may be mostly disconnected, i.e., most of the time, end-to-end paths connecting every node pair do not exist. To cope with frequent, long-lived disconnections, opportunistic routing techniques have been proposed in which, at every hop, a node decides whether it should forward or store-and-carry a message. Despite a growing number of such proposals, there still exists little consensus on the most suitable routing algorithm(s) in this context. One of the reasons is the large diversity of emerging wireless applications and networks exhibiting such “episodic” connectivity. These networks often have very different characteristics and requirements, making it very difficult, if not impossible, to design a routing solution that fits all. In this paper, we first break up existing routing strategies into a small number of common and tunable routing modules (e.g. message replication, coding, etc.), and then show how and when a given routing module should be used, depending on the set of network characteristics exhibited by the wireless application. We further attempt to create a taxonomy for intermittently connected networks. We try to identify generic network characteristics that are relevant to the routing process (e.g., network density, node heterogeneity, mobility patterns) and dissect different “challenged” wireless networks or applications based on these characteristics. Our goal is to identify a set of useful design guidelines that will enable one to choose an appropriate routing protocol for the application or network in hand. Finally, to demonstrate the utility of our approach, we take up some case studies of challenged wireless networks, and validate some of our routing design principles using simulations.

In this paper, we present a collision-free asynchronous multi-channel access protocol for Ad Hoc wireless networks using a single transceiver. Our protocol, dubbed AMMAC for Asynchronous Multi-channel Medium Access Control, targets low-cost and low-power deployments where nodes are equipped with a single transceiver. Other distinguishing features of AM-MAC include its simplicity and the fact that it does not require temporal synchronization among nodes. This is accomplished through an asynchronous split phase together with an observation phase as well as an unique handshake. Nodes observe the control channel for a period of time before asynchronously switching to the negotiated channel. Protocol correctness and collision-freedom in a multi-channel environment are verified. We also provide an analytical throughput assessment for our multi-channel approach. Simulation results show that AM-MAC improves performance significantly when compared to IEEE 802.11 and exhibits comparable performance to MMAC, one of the well-known multi-channel medium access control protocols, without the need for temporal synchronization.

Communication networks are traditionally assumed to be connected. However, emerging wireless applications such as vehicular networks, pocket-switched networks, etc., coupled with volatile links, node mobility, and power outages, will require the network to operate despite frequent disconnections. To this end, opportunistic routing techniques have been proposed, where a node may store-and-carry a message for some time, until a new forwarding opportunity arises. Although a number of such algorithms exist, most focus on relatively homogeneous settings of nodes. However, in many envisioned applications, participating nodes might include handhelds, vehicles, sensors, etc. These various "classes” have diverse characteristics and mobility patterns, and will contribute quite differently to the routing process. In this paper, we address the problem of routing in intermittently connected wireless networks comprising multiple classes of nodes. We show that proposed solutions, which perform well in homogeneous scenarios, are not as competent in this setting. To this end, we propose a class of routing schemes that can identify the nodes of "highest utility” for routing, improving the delay and delivery ratio by four to five times. Additionally, we propose an analytical framework based on fluid models that can be used to analyze the performance of various opportunistic routing strategies, in heterogeneous settings.

Abstract This paper introduces an energy‐efficient data collection technique that takes advantage of spatial and temporal data correlation for monitoring continuously changing conditions such as temperature, humidity, rain fall, radiation, substance concentrations, etc. The proposed technique, isoline aggregation, generates contour maps representing the condition being monitored. The resulting contour map consist of isolines which delimit regions within which the monitored condition is within a certain value range. Energy efficiency is achieved by taking advantage of the spatial and temporal correlation of the data field. Simulation results show that isoline aggregation can reduce the amount of bytes transmitted considerably. At the same time, we are able to keep high data accuracy. Copyright © 2007 John Wiley & Sons, Ltd.

In this paper, we present an efficient message delivery mechanism that enables distribution/dissemination of messages in an internet connecting heterogeneous networks and prone to disruptions in connectivity. We call our protocol MeDeHa (pronounced "medea") for Message Delivery in Heterogeneous, Disruption-prone Networks. MeDeHa is complementary to the IRTF's Bundle Architecture: while the Bundle Architecture provides storage above the transport layer in order to enable interoperability among networks that support different types of transport layers, MeDeHa stores data at the link layer addressing heterogeneity at lower layers (e.g., when intermediate nodes do not support higher-layer protocols). MeDeHa also takes advantage of network heterogeneity (e.g., nodes supporting more than one network) to improve message delivery. For example, in the case of IEEE 802.11 networks, participating nodes may use both infrastructure- and ad hoc modes to deliver data to otherwise unavailable destinations. Another important feature of MeDeHa is that there is no need to deploy special-purpose nodes such as message ferries, data mules, or throwboxes in order to relay data to intended destinations, or to connect to the backbone network wherever infrastructure is available. The network is able to store data destined to temporarily unavailable nodes for some time depending upon existing storage as well as quality-of-service issues such as delivery delay bounds imposed by the application. We evaluate MeDeHa via simulations using indoor scenarios (e.g. convention centers, exposition halls, museums etc.) and show significant improvement in delivery ratio in the face of episodic connectivity. We also showcase MeDeHa's support for different levels of quality-of-service through traffic differentiation and message prioritization.

We introduce the multi-channel flow-aware medium access control protocol, or (MFLAMA), an energy-efficient, schedule-based, multi-channel medium-access control (MAC) protocol designed for data gathering applications in wireless sensor networks. MFLAMA improves the channel utilization by establishing collision-free transmission schedules across multiple channels. Energy efficiency is achieved by preventing packet collisions, idle listening, and transmissions to a node that is not ready to receive packets. We evaluate MFLAMA through extensive simulations and quantify the improvement in channel utilization through the use of multiple channels. Our results indicate that as we increase the number of orthogonal channels used for communication, there is significant improvement in channel utilization and queueing delay. However, we notice a ldquodiminishing returnsrdquo effect as we increase the number of channels, i.e., the performance improvements observed decrease with the number of channels beyond a certain threshold. This threshold depends on the topology and traffic flow patterns being used.

We consider the problem of routing messages through a network with episodic connectivity without a priori knowledge of node schedules or locations. We present Steward assisted routing (StAR), an efficient loop-free routing framework that can operate in networks that are well-connected as well as in networks that exhibit intermittent connectivity; the proposed protocol uses Steward nodes to deliver data to destinations that may be partitioned from the source. We also introduce a companion protocol, scoped contact and interest propagation (SCIP), which contains mechanisms with which dissemination of routing control information for destinations of interest is scoped to a well-defined region of the network. We evaluate our protocols through simulations with four distinct mobility scenarios, including two that are generated using data traces from real networks. Our results show that StAR achieves delivery rates comparable to epidemic routing with far less signaling overhead. We also show that the addition of SCIP to StAR reduces route maintenance overhead significantly without impacting delivery rates.

This paper introduces a scheduled-access, multi-channel medium access control (MAC) framework for wireless multi-hop ad hoc networks (MANETs). The proposed framework dubbed dynamic multi-channel medium access, or DYNAMMA, features: (1) ability to dynamically adapt to application-specific traffic patterns, (2) collision-free, multi-channel operation, (3) energy efficiency, and (4) minimum signaling overhead. We evaluate DYNAMMA through extensive simulations and compare its performance against scheduled-access (e.g., TRAMA) and contention-based (e.g., 802.11) MAC protocols for different application scenarios. Our results show that DYNAMMA's ability to perform collision-free transmission over multiple channels significantly increases system capacity through higher channel utilization and spatial re-use. When compared to TRAMA, DYNAMMA's efficiency in terms of signaling overhead yields considerable energy savings as well as queueing delay reduction. We also present an implementation of DYNAMMA over an Ultra-Wideband (UWB) radio testbed. Our UWB testbed results indicate that DYNAMMA can achieve both high channel utilization (close to 90% for our experiments) and high energy efficiency (nodes, on average, sleep one third of the time).

In this paper, we explore the impact of reference node, or "landmark", placement on the accuracy of the coordinate systems built using topology-based localization techniques. Such techniques employ landmarks to which each node computes its hop-count distance. A node's coordinates is given by the hop-count distance to all landmarks. To our knowledge, our paper is the first to study the impact of landmark placement on the accuracy of the resulting coordinate system. We show that placing landmarks on the periphery of the topology yields more accurate coordinate systems when compared to placing landmarks in the interior of the topology. Nevertheless, our simulation results also show that, in general, if enough landmarks are used, random landmark placement yields comparative performance to placing landmarks on the boundary randomly or equally spaced. This is an important result since boundary placement (especially at equal distances) may turn out to be infeasible and/or prohibitively expensive (in terms of power consumption as well as processing and communication overhead). This is also the first study to consider not only uniform, synthetic topologies, but also, non-uniform topologies resembling more concrete deployments.

Communication networks (wired or wireless) have traditionally been assumed to be connected at least most of the time. However, emerging applications such as emergency response, special operations, smart environments, VANETs, etc. coupled with node heterogeneity and volatile links will likely change the typical conditions under which networks operate. In fact, in such scenarios, networks may be mostly disconnected. To cope with frequent, long-lived disconnections, opportunistic routing techniques have been proposed in which, at every hop, a node decides whether it should either forward and/or store-and-carry a message. As a result, a number of message replicas may be created and routed independently ("spraying"). Most opportunistic routing schemes to-date perform greedy replication handing over a copy of a message to the first nodes encountered. Yet, in a network with heterogeneous nodes, where some nodes may be much "better" relays than others, such greedy schemes may waste valuable message replicas (and thus energy, storage space, etc.) on "useless" relays. For this reason, we propose the idea of utility-based replication, where some fitness or utility function is maintained for all nodes in a distributed fashion, and a small budget of message replicas is allocated according to this utility only to the fittest nodes. We describe a number of variations using different utility functions, and show that an improvement of up to 5-6× in delay can be achieved over greedy algorithms.

In the above titled paper (ibid., vol 5, no. 1, pp. 28-42, Jan 06), corrections were made to Figure 6.

We introduce the game-theoretic stochastic routing (GTSR) framework, a proactive alternative to today's reactive approaches to route repair. GTSR minimizes the impact of link and router failure by 1) computing multiple paths between source and destination and 2) selecting among these paths randomly to forward packets. Besides improving fault tolerance, the fact that GTSR makes packets take random paths from source to destination also improves security. In particular, it makes connection eavesdropping attacks maximally difficult as the attacker would have to listen on all possible routes. The approaches developed are suitable for network layer routing, as well as for application layer overlay routing and multipath transport protocols such as the stream control transmission protocol (SCTP). Through simulations, we validate our theoretical results and show how the resulting routing algorithms perform in terms of the security/fault-tolerant/delay/throughput trade-off. We also show that a beneficial side effect of these algorithms is an increase in throughput, as they make use of multiple paths.

Recently, it became apparent that group-oriented services are one of the primary application classes targeted by MANETs. As a result, several MANET-specific multicast routing protocols have been proposed. Although these protocols perform well under specific mobility scenarios, traffic loads, and network conditions, no single protocol has been shown to be optimal in all scenarios. The goal of this paper is to characterize the performance of multicast protocols over a wide range of MANET scenarios. To this end, we evaluate the performance of mesh and tree-based multicast routing schemes relative to flooding and recommend protocols most suitable for specific MANET scenarios. Based on the analysis and simulation results, we also propose two variations of flooding, scoped flooding and hyper flooding, as a means to reduce overhead and increase reliability, respectively. Another contribution of the paper is a simulation-based comparative study of the proposed flooding variations against plain flooding, mesh, and tree-based MANET routing. In our simulations, in addition to "synthetic" scenarios, we also used more realistic MANET settings, such as conferencing and emergency response.

Abstract Although multihop ad hoc networks (MANETs) have typically been considered as isolated, stand‐alone networks with no connection to the Internet, it is quite likely that future internetworks may interconnect numerous MANET clouds, each running different routing mechanisms. This could be primarily dictated by administrative constraints or different network functional and performance requirements and can result in scenarios where nodes belonging to different MANET clouds running different routing mechanisms may wish to communicate with one another. In this paper, we introduce techniques to allow interoperability of various multicast routing protocols in MANETs. In particular, we investigate two different interoperability techniques, that is, flooding‐based interoperability and facilitator‐assisted interoperability. Our preliminary evaluation of the two techniques indicates that the flooding‐based mechanism is best suited for scenarios involving intermittent communications between different ad hoc domains whereas the facilitator‐assisted approach is best suited for data intensive applications such as video‐conferencing. Copyright © 2006 John Wiley & Sons, Ltd.

This paper describes a technique to reduce the computational complexity of max-flow routing, based on a hierarchical decomposition of the network. The computational complexity of this hierarchical max-flow routing is comparable to that of Dijkstra's and Bellman-Ford's algorithms. It is shown that in many of today's networks, this hierarchical approach provides nearly the same performance as flat (i.e., non-hierarchical) routing, with significantly less computation.

This paper introduces robust and secure routing (RoST), an integrated and flexible approach to secure routing in wireless mobile ad hoc networks (MANETs). RoST incorporates robustness and security into the underlying MANET routing protocol. By using a layered approach, RoST offers flexibility to the target MANET application to choose the required level of security. Most secure MANET routing solutions to-date either assume that security services (e.g., key distribution) is handled externally or propose cryptographic schemes that target MANETs. RoST, on the other hand, combines both robustness and security techniques integrating them into a protocol suite. At its core, it takes advantage of the network's inherent redundancy to efficiently discover multiple routes ensuring robust delivery; additional layers provide in-built key exchange, authentication, data integrity and privacy. We implemented RoST by extending the dynamic source routing (DSR) protocol. Through simulations, we demonstrate how RoST can circumvent security attacks. We also show that, under normal operating conditions, RoST and DSR exhibit similar performance. Thus, RoST is able to provide robustness and security without significantly impacting performance.

This paper introduces a novel energy efficient data aggregation algorithm that targets spatially correlated data in sensor networks. Isolines aggregation works by detecting isolines which are the lines in a contour map. Energy efficiency is achieved by having only the nodes that detect the isoline report to the sink. Simulation results show that isoline aggregation can lead to significant energy savings (some scenarios reported that no aggregation can send close to 150% more bytes than isolines aggregation) with adequate data accuracy. We also compared isolines against polygon aggregation, our implementation of an approach representing existing spatially-correlated data aggregation mechanisms. Our results report that isolines exhibit higher accuracy with a slight advantage in energy efficiency.

We introduce FLAMA (flow-aware medium access), an energy-efficient medium-access control (MAC) protocol designed for wireless sensor networks. FLAMA achieves energy efficiency by preventing idle listening, data collisions and transmissions to a node that is not ready to receive packets. It adapts medium access schedules to the traffic flows exhibited by the application. FLAMA is simple enough so that it can be run by nodes with limited processing, memory, communication, and power capabilities. We evaluate the performance of FLAMA through simulations and test-bed experimentation. Simulation results indicate that, in terms of reliability, queuing delay and energy savings, FLAMA outperforms TRAMA, the first traffic-adaptive, schedule-based MAC proposed for sensor networks, and S-MAC, a contention-based energy-efficient MAC. FLAMA achieves significantly smaller delays (up to 75 times) when compared to TRAMA with significant improvement in energy savings and reliability, demonstrating the importance of application awareness in medium access scheduling. Our simulation and test-bed results show that FLAMA achieves better end-to-end reliability with significant energy savings compared to S-MAC.

We present the Protocol for multicasting over directional antennas (MODA) for mobile ad hoc networks (MANET). MODA is the first protocol for MANET's that uses directional antennas to reduce data packet overhead. Without increasing energy consumption, MODA increases the range of transmission as a result of which fewer nodes are involved in the forwarding process, which results in a reduction in data packet overhead. Using simulations in Qualnet 3.5, we compare MODA with PUMA and ODMRP. The results from a wide range of scenarios of varying mobility, group members, number of senders, traffic load, and number of multicast groups show that MODA attains comparable packet delivery ratios to ODMRP and PUMA, while incurring far less overhead

This paper introduces an energy-efficient data collection technique that takes advantage of spatial/temporal data correlation to generate maps for continuous monitoring (e.g., of environmental conditions such as temperature, humidity, etc.). In its essence, the proposed technique, isoline aggregation, works by detecting isolines which are the lines that make up a contour map. Energy efficiency through spatial aggregation is achieved by having only nodes that detect the isoline report to the sink. Simulation results show that isoline aggregation can reduce the amount of bytes transmitted by a factor of 11 compared to when no data aggregation is used and by up to 4 times when compared to an existing spatial-correlation based aggregation mechanism. At the same time, we are able to keep high data accuracy.

This paper evaluates the effect of timing in data aggregation algorithms. In-network aggregation achieves energy-efficient data propagation by processing data as it flows from information sources to sinks. Our goal is to show that the decision of when to "clock out" data as it is processed by nodes have significant performance impact in terms of data accuracy and freshness. Using the sensor network paradigm where all nodes produce information periodically, we compare three aggregation timing policies. Through extensive simulations we show that setting up the clock out timer based on a node's position in the aggregation tree results in a beneficial "cascading effect", yielding considerable energy efficiency, yet maintaining data accuracy and freshness.

This paper presents an analytical model to predict energy consumption in saturated IEEE 802.11 single-hop ad hoc networks under ideal channel conditions. The model we introduce takes into account the different operational modes of the IEEE 802.11 DCF MAC, and is validated against packet-level simulations. In contrast to previous works that attempted to characterize the energy consumption of IEEE 802.11 cards in isolated, contention-free channels (i.e., single sender/receiver pair), this paper investigates the extreme opposite case, i.e., when nodes need to contend for channel access under saturation conditions. In such scenarios, our main findings include: (1) contrary to what most previous results indicate, the radio's transmit mode has marginal impact on overall energy consumption, while other modes (receive, idle, etc.) are responsible for most of the energy consumed; (2) the energy cost to transmit useful data increases almost linearly with the network size; and (3) transmitting large payloads is more energy efficient under saturation conditions

Typical applications of mobile ad hoc networks (MANET) require group-oriented services. Digital battlefields and disaster relief operations make data dissemination and teleconferences a key application domain. Network-supported multicast is hence critical for efficient any-to-many communications. However, very little work has been done on "reliable" transport multicast. We propose and evaluate reliable adaptive lightweight multicast (RALM). The design choices of RALM are motivated by lessons we learned from evaluating the performance of traditional wired reliable multicast transport protocols (in particular, SRM) in ad hoc networks. We argue the two components, reliability and congestion control, are essential in designing a reliable multicast transport protocol for MANETs. RALM addresses both reliability and congestion control. It achieves reliability by guaranteeing data delivery to troubled receivers in a round-robin fashion. RALM's send-and-wait congestion control uses NACK feedback to adjust to congestion experienced by receivers. We show through simulations that RALM achieves perfect reliability while exhibiting low end-to-end delay and minimal control overhead compared against other protocols.

Most standard implementations of TCP perform poorly when packets are reordered. In this paper, we propose a new version of TCP that maintains high throughput when reordering occurs and yet, when packet reordering does not occur is friendly to other versions of TCP. The proposed TCP variant, or TCP-PR, does not rely on duplicate acknowledgments to detect a packet loss. Instead, timers are maintained to keep track of how long ago a packet was transmitted. In case the corresponding acknowledgment has not yet arrived and the elapsed time since the packet was sent is larger than a given threshold, the packet is assumed lost. Because TCP-PR does not rely on duplicate acknowledgments, packet reordering (including out-of-order acknowledgments) has no effect on TCP-PR performance. Through extensive simulations, we show that TCP-PR performs consistently better than existing mechanisms that try to make TCP more robust to packet reordering. When the case that packets are not reordered, we verify that TCP-PR maintains the same throughput as typical implementations of TCP (specifically, TCP-SACK) and shares network resources fairly.

The traffic-adaptive medium access protocol (TRAMA) is introduced for energy-efficient collision-free channel access in wireless sensor networks. TRAMA reduces energy consumption by ensuring that unicast, multicast, and broadcast transmissions have no collisions, and by allowing nodes to switch to a low-power, idle state whenever they are not transmitting or receiving. TRAMA assumes that time is slotted and uses a distributed election scheme based on information about the traffic at each node to determine which node can transmit at a particular time slot. TRAMA avoids the assignment of time slots to nodes with no traffic to send, and also allows nodes to determine when they can become idle and not listen to the channel using traffic information. TRAMA is shown to be fair and correct, in that no idle node is an intended receiver and no receiver suffers collisions. The performance of TRAMA is evaluated through extensive simulations using both synthetic- as well as sensor-network scenarios. The results indicate that TRAMA outperforms contention-based protocols (e.g., CSMA, 802.11 and S-MAC) as well as scheduling-based protocols (e.g., NAMA) with significant energy savings.

In this paper we present a general hybrid systems modeling framework to describe the flow of traffic in communication networks. To characterize network behavior, these models use averaging to continuously approximate discrete variables such as congestion window and queue size. Because averaging occurs over short time intervals, one still models discrete events such as the occurrence of a drop and the consequent reaction (e.g., congestion control). The proposed hybrid systems modeling framework fills the gap between packet-level and fluid-based models: by averaging discrete variables over a very short time scale (on the order of a round-trip time), our models are able to capture the dynamics of transient phenomena fairly accurately. This provides significant flexibility in modeling various congestion control mechanisms, different queuing policies, multicast transmission, etc. We validate our hybrid modeling methodology by comparing simulations of the hybrid models against packet-level simulations. We find that the probability density functions produced by ns-2 and our hybrid model match very closely with an L1-distance of less than 1%. We also present complexity analysis of ns-2 and the hybrid model. These tests indicate that hybrid models are considerably faster.

This paper evaluates FLIP, a flexible header protocol for power-constrained, heterogeneous networks. We show that FLIP can improve the energy efficiency of a system considerably, prolonging the system's lifetime. For example, when employing FLIP to implement an existing sensor network communication paradigm, we obtain 50% energy savings when compared to the paradigms original implementation. Further, we use a sample data gathering application that calculates the average of a sensor network's attribute (e.g., average temperature) and show FLIP's energy efficiency when compared to static header approaches.

Untethered nodes in mobile ad hoc networks strongly depend on the efficient use of their batteries. In this paper, we propose a new metric, the drain rate, to forecast the lifetime of nodes according to current traffic conditions. This metric is combined with the value of the remaining battery capacity to determine which nodes can be part of an active route. We describe new route selection mechanisms for MANET routing protocols, which we call the minimum drain rate (MDR) and the conditional minimum drain rate (CMDR). MDR extends nodal battery life and the duration of paths, while CMDR also minimizes the total transmission energy consumed per packet. Using the ns-2 simulator and the dynamic source routing (DSR) protocol, we compare MDR and CMDR against prior proposals for energy-aware routing and show that using the drain rate for energy-aware route selection offers superior performance results. Methods keywords are system design and simulations.

Shortest path routing leaves connections at risk of interception and eavesdropping since the path over which data packets travel is fairly predictable and easy to determine. To improve routing security, we propose a proactive mechanism, which we call secure stochastic routing, that explores the existence of multiple routes and forces packets to take alternative paths probabilistically. We investigate game theoretic techniques to develop routing policies which make interception and eavesdropping maximally difficult. Through simulations, we validate our theoretical results and show how the resulting routing algorithms perform in terms of the security/delay/throughput trade-off. We observe that a beneficial side-effect of these algorithms is an increase in throughput, as they make use of multiple paths. The Internet was designed to use redundancy to enhance reliability. We suggest that, through stochastic methods, redundancy be used to increase security.

Mobile ad hoc networks' (MANETs) inherent power limitation makes power-awareness a critical requirement for MANET protocols. We propose a new routing metric, the drain rate, which predicts the lifetime of a node as a function of current traffic conditions. We describe the minimum drain rate (MDR) mechanism which uses a combination of the drain rate with remaining battery capacity to establish routes. MDR can be employed by any existing MANET routing protocol to achieve a dual goal: extend both nodal battery life and connection lifetime. Using the ns-2 simulator and the dynamic source routing (DSR) protocol, we compared MDR to the minimum total transmission power routing (MTPR) scheme and the min-max battery cost routing (MM-BCR) scheme and proved that MDR is the best approach to achieve the dual goal.

The use of contention-based MAC protocols combined with hidden terminal problems make multi-hop wireless ad hoc networks much more sensitive to load and congestion than wired networks or even wireless cellular networks. In such an environment, we argue that multicast reliability cannot be achieved solely by retransmission of lost packets as is typically done in wired networks with protocols such as SRM. We contend that in order to achieve reliable multicast delivery in such networks, besides reliability mechanisms, we must also consider jointly two components: reliability and congestion control. In this paper, we propose CALM, a congestion controlled, adaptive, lightweight multicast transport protocol and show that congestion control alone can significantly improve reliable packet delivery in ad hoc networks when compared to traditional "wired" reliable multicast protocols.

The diverse nature of MANETs makes it almost impossible for a single routing protocol to perform well under a wide range of operating conditions. Therefore the solution may be to adopt an adaptive strategy to routing and the co-existence and interoperability of different routing protocols. Considering that MANETs are generally deployed in mission critical applications the adaptive protocol should be capable of providing high reliability and timeliness guarantees in the presence of mobility. To this end we develop an adaptive flooding protocol in which nodes can dynamically switch routing mechanisms based on their perspective of network conditions. We use relative velocity as the switching criterion. Each node periodically computes its velocity relative to that of its neighbor set and based on its computation switches to one of the three modes, i.e. scoped flooding, plain flooding or hyper flooding modes. Simulations using our adaptive protocol under various realistic scenarios have shown that such protocols provide impressive benefits and can be used as the basis for developing adaptive, integrated routing techniques for MANETs of the future.

Focuses on the requirements of "better than best effort" (high hop-by-hop delivery guarantee) broadcasting in highly dynamic mobile multi-hop ad-hoc networks (MANETs). Our work is motivated by mission-critical applications, such as disaster relief and military operations. This class of applications is characterized by: (1) high delivery guarantee requirements, even in the presence of high mobility, and (2) a broadcast style of communication, where all nodes are receivers. Extensive simulations conducted on two different platforms show that, as node speeds, network traffic load and number of senders increase, the performance of existing multicast protocols (exemplified by ODMRP and MAODV) degrades in terms of packet delivery and overhead. In contrast, simple flooding, while clearly not a panacea, performs comparatively well and shows promise as a foundation for more specialized protocols for highly dynamic MANETs of the future.

Interconnecting heterogeneous devices, that is, devices with varying capabilities, has raised new challenges in the design of network protocols. This paper describes the design of the Flexible Interconnecting Protocol, or FLIP, whose goal is to interconnect heterogeneous devices. FLIP is a flexible protocol that addresses the needs of heterogeneous networks: it incurs little overhead when run by simple devices, while still providing a range of functions that can be performed by more sophisticated devices. We describe a simplified implementation of FLIP under Linux. We also conducted a preliminary evaluation of FLIP's overhead and functionality in the context of IP (IPv4 and IPv6) and sensor network environments. FLIP incurs reasonably low overhead when providing IPv4 and IPv6 functionality (1 and 3 bytes respectively), yet it does particularly well in the case of small payloads. When compared to a sensor-specific protocol, FLIP incurs a small overhead increase while still providing full protocol functionality.

This paper investigates network latency metrics in the context of the server proximity problem. Using a combination of experimentation and statistical analysis, we study the correlation among number of network and administrative system (AS) hops, and round-trip time (RRT). We ran experiments involving 601 Internet sites spanning 5 continents. Our results show reasonably strong AS hop-network hop correlations of up to 70%. We also observe an average RTT-number of hop correlation close to 50%, which represents a considerable improvement over what Crovella and Clark observed in 1995. Based on our results, we argue that, when choosing a single metric for network latency, RTT should be the metric of choice when trying to reduce clients' perceived latency. However, hop counts are good indicators of network resource usage. Another factor that favors RTT is that it is the least expensive metric to measure.

This paper studies the behavior of ARDP, a request response transport protocol, when operating in a shared communication infrastructure like the Internet. Our experiments demonstrate that ARDP backs off in the presence of congestion, yet tries to take advantage of available bandwidth. We also show that ARDP is well-behaved when competing for network resources with TCP.

This paper evaluates the congestion control performance of Pseudofed, a congestion-controlled, reliable multicast transport protocol for bulk data transfer. Pseudofed's congestion control mechanism is based on the concept of representatives, a small, dynamic set of multicast group members. By reducing the congestion control problem to a bounded set of receivers, representatives allow the point-to-point congestion control model used by unicast protocols like TCP to scale to larger multicast groups. Other features that contribute to the scalability of Pseudofed's congestion control algorithm are: (1) attempting to distinguish between correlated and uncorrelated packet losses, (2) not requiring complete knowledge of the multicast group, and (3) not exchanging control communication with congestion free subtrees.

This paper evaluates a congestion control mechanism for reliable multicast applications that uses a small dynamic set of group members, or representatives, to provide timely and accurate feedback on behalf of congested subtrees of a multicast distribution tree. Through simulations, we evaluate our algorithm with and without TCP cross traffic. This study demonstrates that while our algorithm makes use of bandwidth when available, it responds to congestion in a timely fashion thereby limiting data drops. When competing for bandwidth with TCP traffic, our algorithm gracefully relinquishes bandwidth. While not as aggressive as TCP, it does grab a reasonable portion of the bandwidth yet keeps the overall multicast packet loss low.

Internet information services replicate their servers to improve availability, response time, and fault tolerance. Traditional replication algorithms do not address the scale and administrative decentralization of today's internetworks. We have proposed and implemented a scalable and efficient tool to replicate wide-area, autonomously managed services. We target replication degrees of thousands of weakly consistent replicas. The main goal of our replication tool is to make traditional replication services scale in today's exponentially growing, autonomously managed internetworks. Our tool, which we call flood-d, allows servers to be organized in multiple replication groups. For each replication group, flood-d builds a logical update topology that is resilient to server failure, and tries to minimize the communication cost and propagation time needed to transmit updates. Flood-d's logical topologies are computed based on communication latency and available network bandwidth. This paper describes flood-d, and presents simulation results obtained when using flood-d to extend existing replication algorithms. Our results show the gains of organizing service replicas into multiple, smaller replication groups, and using network-cognizant logical topologies to propagate updates. We argue that existing as well as emerging Internet information services can benefit from flood-d's services.

For a reliable, feedback dependent multicast transport protocol to scale, it must avoid the feedback implosion problem, particularly if the protocol targets arbitrarily large multicast groups communicating over lossy networks. Most existing suppression based feedback control mechanisms address the implosion problem using timers based on round-trip time (RTT) estimates between each receiver and the source. The algorithm presented has three major benefits: it does not need to compute the RTT from all receivers to the source, does not require knowledge of group membership, and provides prompt feedback. A small set of representative receivers and probabilistic suppression are used to limit feedback. We believe that this approach will perform well in real networks. Simulations show that for various multicast group sizes, a few representatives can keep the amount of feedback low while not degrading feedback timeliness.

This paper considers the interaction of HTTP with several transport protocols, including TCP, Transaction TCP, a UDP-based request-response protocol, and HTTP with persistent TCP connections. We present an analytic model for each of these protocols and use that model to evaluate network overhead carrying HTTP traffic across a variety of network characteristics. This model includes an analysis of the transient effects of TCP slow-start. We validate this model by comparing it to network packet traces measured with two protocols (HTTP and persistent HTTP) over local and wide-area networks. We show that the model is accurate within 5% of measured performance for wide-area networks, but can underestimate latency when the bandwidth is high and delay is low. We use the model to compare the connection-setup costs of these protocols, bounding the possible performance improvement. We evaluate these costs for a range of network characteristics, finding that setup optimizations are relatively unimportant for current modem, ISDN, and LAN users but can provide moderate to substantial performance improvement over high-speed WANs. We also use the model to predict performance over future network characteristics.

This paper reports the design, implementation, and performance of a scalable and efficient tool to replicate Internet information services. Our tool targets replication degrees of tens of thousands of weakly-consistent replicas scattered throughout the Internet's thousands of autonomously administered domains. The main goal of our replication tool is to make existing replication algorithms scale in today's exponentially-growing, autonomously-managed internetworks.

Over a million computers implement the Internet's Domain Name System or DNS, making it the world's most distributed database and the Internet's most significant source of wide-area RPC-like traffic. Last year, over eight percent of the packets and four percent of the bytes that traversed the NSFnet were due to DNS. We estimate that a third of this wide-area DNS traffic was destined to seven root name servers. This paper explores the performance of DNS based on two 24-hour traces of traffic destined to one of these root name servers. It considers the effectiveness of name caching and retransmission timeout calculation, shows how algorithms to increase DNS's resiliency lead to disastrous behavior when servers fail or when certain implementation faults are triggered, explains the paradoxically high fraction of wide-area DNS packets, and evaluates the impact of flaws in various implementations of DNS. It shows that negative caching would improve DNS performance only marginally in an intern...

Policy routing protocols incorporate policy related constraints into the route computation and packet forwarding functions for inter-administrative domain (AD) communication. However, this functionally exacerbates the already critical problem of routing information distribution and storage overhead in very large internets (e.g., 100000 ADs). The authors investigate the scalability of the inter-domain policy routing (IDPR) architecture. In particular, they present an informal analysis of connectivity database and update overhead. A model of the IDPR architecture is defined and tested by varying several parameters. The results obtained illustrate the scaling properties of the IDPR architecture and their dependencies upon internet configuration, connectivity among ADs, and number of policies. It is found that, under certain reasonable assumptions, global information for an internet of 5000 transit ADs (and 50000 stub ADs) will occupy on the order of 2.5 MB of storage in the route server.>