In LEO satellite handover under urban mobility, ephemeris, elevation, and field-of-view (FoV) constraints can potentially identify a reachable satellite, but they do not indicate whether the satellite's projected path in the local sky of the user terminal (UT) will remain usable after the handover decision. While present-time visibility can reject a path that is already blocked, it cannot warn that a currently reachable satellite is losing clearance near a building, canopy, or bridge boundary. The resulting gap for LEO satellite handover is thus not just candidate discovery, but candidate ranking under future local obstruction. This paper aims to address this gap based on future local-sky clearance, the path-level margin between each candidate satellite projection and the predicted terrain or skyline boundary. We propose SkyMRT, a future-skymask rollout model that estimates boundary motion from recent binary skymasks and UT motion, as well as SMASH, a deterministic selector that checks the incumbent and ranks candidates by worst-case clearance, serviceable fraction, and near-boresight priority. Our evaluations on an urban digital-twin replay with Starlink TLE (two-line element) propagation and packet-level ns-3 replay show that the resulting selector reduces blocked exposure by 5x compared with a geometry-and-FoV selection baseline. Against a same-cadence present-time visibility baseline, it reduces packet error rate by 2.2x and transition time by 10.2x. These results show that future local-sky clearance can indeed change urban LEO satellite selection from late blockage repair to early ranking of future-serviceable candidates.

This study developed an elastic wristband called Polyband for profiling the unique deformation of the skin and muscles around the wrist during hand gestures through resistance measurements. A proof-of-concept design was demonstrated to recognize a simple set of gestures with 92.9% accuracy.

By predicting the air pollutant concentration, people can take precautions to avoid overexposure to air pollutants. Consequently, accurate$\mathbf{PM}_{2.5}$prediction becomes more important. In this paper, we propose a$\mathbf{PM}_{2.5}$prediction system, which utilizes the dataset from EdiGreen Airbox and Taiwan EPA. Our$\mathbf{PM}_{2.5}$prediction system is composed of four parts: data collection, data preprocessing, prediction model construction, and Line platform. To assess the performance of the model prediction, the daily average error and the hourly average accuracy for the duration of a week are calculated. The experimental results show that LSTM based on K- means has the best performance among all methods. Therefore, LSTM based on K-means is chosen to provide real-time$\mathbf{PM}_{2.5}$prediction through the Linebot.

Adapting to Post-Quantum Cryptography (PQC) is an inevitable shift due to stable quantum computers getting closer to fruition. As our lives are getting more entangled with Internet of Things (IoT) technologies, a quantum computer attack will be even more devastating. In this work, we evaluate the performance of PQC algorithms on IoT devices. Our empirical results show that PQC algorithms, besides their tolerance against quantum computer attacks, have the potential to be on par with typical asymmetric encryption algorithms from the TLS protocol.

A complete sensor network should include sensors, data processing, and data services. However, to establish the legitimacy of sensor data for urban governance, sensor networks should go beyond simple deployment of sensors in the built environment and strive for deeper integration of data services within civil society. This paper presents the Taiwan AirBox Project as an exemplary case of practical deployment of a sensor network to discuss the topics of open data, value-added services, and joint calibration services; as well as how these services generate productive public-private partnerships.

The progress of industrial development naturally leads to the demand for more electrical power. Unfortunately, due to the fear of the safety of nuclear power plants, many countries have relied on thermal power plants, which will cause more air pollutants during the process of coal burning. This phenomenon as well as increased vehicle emissions around us, have constituted the primary factors of serious air pollution. Inhaling too much particulate air pollution may lead to respiratory diseases and even death, especially PM$_{2.5}$. By predicting the air pollutant concentration, people can take precautions to avoid overexposure to air pollutants. Consequently, accurate PM$_{2.5}$prediction becomes more important. In this study, we propose a PM$_{2.5}$prediction system, which utilizes the dataset from EdiGreen Airbox and Taiwan EPA. Autoencoder and Linear interpolation are adopted for solving the missing value problem. Spearman’s correlation coefficient is used to identify the most relevant features for PM$_{2.5}$. Two prediction models (i.e., LSTM and LSTM based on K-means) are implemented which predict PM$_{2.5}$value for each Airbox device. To assess the performance of the model prediction, the daily average error and the hourly average accuracy for the duration of a week are calculated. The experimental results show that LSTM based on K-means has the best performance among all methods. Therefore, LSTM based on K-means is chosen to provide real-time PM$_{2.5}$prediction through the Linebot.

Time-series sensor data are used to create prediction models, called city profiles, for understanding city dynamics in the smart-city sector. These city profiles are typically created and updated by the Cloud using reported raw sensor data. However, continuously reporting raw sensor data is not energy efficient for boundary computing resources of a network. Thus, this work considers edge servers that are deployed on the boundary computing resources of a network to collaborate with the Cloud for adaptively mitigate city profiling tasks (i.e., creating city profiles) across an edge server and the Cloud. By maintaining the local city profiles on the edge or the global city profiles on the Cloud, either an edge or the Cloud can dynamically respond to user queries. However, there is a trade-off between the energy consumption of an edge and the response accuracy of the city profiles. This work designs an adaptive city profiling and synchronization approach for edges to decide when, where (i.e., an edge or the Cloud), and how to update and synchronize local and global city profiles such that the energy consumption of the edge is reduced while the accuracy of a city profile can be guaranteed. Extensive simulations are conducted using a real-world temperature dataset to evaluate the performance of the proposed approach. The simulation results indicate an average energy saving of 60% of edges compared with a typical Cloud-based approach while the required accuracy is fulfilled.

Disease surveillance is essential for the control of flu and respiratory infectious diseases including the novel coronavirus disease (COVID-19). Indoor air quality monitoring has been shown effective in understanding the effectiveness of airflow and circulation indoors to reduce the risk of infectious diseases. In this project, we developed low-cost indoor air quality monitoring devices and systems to tackle the disease surveillance problem. The monitoring device consists of a set of air quality sensors. By strategic deployment and real-time data analysis, the system is able to yield insightful air circulation information indoors. The real-time data analysis is performed on air quality for the indoor ventilation using Long Short-Term Memory (LSTM) on sensed data. A series of user-friendly visualization interfaces and chatbot applications are designed to interact with users and ensure the successful delivery of infection control information. Finally, we work closely with the Taiwan Centers for Disease Control (CDC) and conduct field experiments in 15 locations including hospitals, long-term care centers, schools with total of 144 IAQ devices.

Noise exposure has been the emerging environmental factor for human health. Yet an accurate and large-scale sound monitoring network is not available due to the expense of high-quality professional sound level meters and poorly-calibrated low-cost noise sensors. In this work, we propose a participatory sound meter calibration using smartphones. The system employs a low-cost and open-sourced calibration station to conduct side-by-side sound measurements, and all the measurement data are uploaded to the open data portal to build calibration models for different phone brands and models. We show that, using our calibration models, the MAE of calibration performance can be reduced significantly from 12.4 dbA to 2.8 dbA for the same device and 3.3 dbA for the other device of the same phone model. The results of this study can benefit crowdsourcing-based large-scale sound measurements and facilitate noise exposure, public health, and smart city researches in the future.

Narrowband Internet of Things (NB-IoT) is considered as a promising wireless communications technology for Internet of Things (IoT) especially for the outdoor environment. Many outdoor IoT applications involve large numbers of homogeneous NB-IoT devices. It is tedious to specify and accommodate these devices during application development. To resolve this issue, this paper proposes a service platform for fast development of NB-IoT applications called NB-IoTtalk. This platform utilizes a tag mechanism to provide an easy-to-manipulate graphical user interface to accommodate a large number of NB-IoT devices in an application and transparently show them in a visual map. Our approach automatically creates and parses the device profile used to interpret the payload of an NB-IoT message. We then use a smart parking lot application as an example to investigate the event-triggered reporting of NB-IoT in terms of the time-to-live (TTL) report frequency and the outage detection accuracy. This paper provides the guidelines to set the TTL interval for event-triggered NB-IoT applications.

In the past few years, Internet of Things (IoT) has emerged as an outstanding technology and can be considered as the backbone of the smart world era as it connects physical devices with the internet. On the other hand there are these intelligent conversation interfaces that use a dialogue system to have a conversation with the users and are already used widely. In this paper, we focus on integration of chatbots and IoT to address a critical problem like air quality awareness. We present the architecture of the chatbot and its implementation on an instant messaging application. The chatbot not only provides air quality, temperature and humidity information to the users but also provides services like subscription to air quality monitoring nodes in a particular area or any part of the service region, alarm services, threshold settings, geoquery and recommendation based on the pollutant levels. A detailed explanation of the scenario of use, challenges identified during development process and possible future directions of this integration have also been addressed.

No abstract available.

Smart home is one of the major Internet of Things (IoT) applications. When realizing a smart home, designing an intuitive interaction for appliance control has become a crucial issue. A control intention to an appliance comprises two main meanings, identification and control command. But, previous works seldom deliver these two meanings at the same time. To fully convey the control intention, this paper proposes a scheme of intuitive appliance control that tracks and recognizes hand moving trajectory and hand gestures respectively by exploiting acoustic information. We utilize a 3D positioning method based on time difference of arrival (TDoA) of acoustic signals to detect the trajectory of a hand movement for appliance identification and design a gesture recognition method by Doppler effect for control command. We validate the effectiveness of the proposed scheme via real-life experiments. The results show that average locating errors to the targets are 5.24° and 10.35° in the horizontal and vertical planes respectively and the accuracy of gesture recognition is 96.66%. Since human's eyesight is horizontal to the ground, locating an object in the vertical view is error-prone. We also provide a scroll gesture to assist the appliance selection.

Indoor localization systems have attracted considerable attention recently. A lot of works have used wireless signals from existing base stations to track users' locations. The major challenge to such systems is the signal-drifting problem. A promising direction to conquer this problem is to fuse the tracked wireless signals with inertial sensing data. In this work, we consider location tracking in a multi-floor building, which we call a 2.5-D space, by taking wireless signals, inertial sensing data, and indoor floor plans of a 2.5-D space as inputs and building a SPF (sensor-assisted particle filter) model to fuse these data. Inertial sensors are to capture human mobility, while particles reflect our belief of the user's potential locations. Our work makes the following contributions. First, we propose a model to partition a 2.5-D space into multiple floors connected by stairs and elevators and further partition each floor, according to its floor plan, into logical units connected by passages. Second, based on the 2.5D space model, we then propose particle sampling and resampling mechanisms over the logical units using wireless signal strengths and inertial sensing data to adjust our beliefs of the user's potential locations. Third, to conquer the signal-drifting problem, we propose a weighting mechanism to control the distribution of particles based on user's activities of walking on grounds/stairs and taking elevators. A prototype has been developed and tested to verify the model and its accuracy.

As the population density continues to grow in the urban settings, air quality is degrading and becoming a serious issue. Air pollution, especially fine particulate matter (PM2.5), has raised a series of concerns for public health. As a result, a number of large-scale, low cost PM2.5 monitoring systems have been deployed in several international smart city projects. One of the major challenges for such environmental sensing systems is ensuring the data quality. In this paper, we propose an anomaly detection framework (ADF) for large-scale, real-world environmental sensing systems. The framework is composed of four modules: 1) time-sliced anomaly detection (TSAD), which detects spatial, temporal, and spatio-temporal anomalies in the real-time sensor measurement data stream; 2) real-time emission detection, which detects potential regional emission sources; 3) device ranking, which provides a ranking for each sensing device; and 4) malfunction detection, which identifies malfunctioning devices. Using real world measurement data from the AirBox project, we demonstrate that the proposed framework can effectively identify outliers in the raw measurement data as well as infer anomalous events that are perceivable by the general public and government authorities. Because of its simple design, ADF is highly extensible to other advanced applications, and it can be exploited to support various large-scale environmental sensing systems.

There is a growing concern over the issues related to online privacy due to large availability of high quality images. To tackle the privacy concerns a face swapping application is proposed. There is a library of face images, which is created by downloading images from different sources on Internet. For any given image, first of all facial landmarks are detected. The second image is rotated and scaled so that it can properly fit over the input image. To make sure that the new image looks natural, color balance adjustment is done. After that blending of features from the second image onto the input image is done. It is also shown how this system can be used to creating appealing and funny photographs for entertainment purposes. We conclude with a study that shows the high quality of images produced by this system as compared to existing face swap applications and also limitations of this system.

In this paper, we present an energy, data and cost efficient model for mobile opportunistic PM2.5 sensing via bicycles. To facilitate the implementation of such systems, we first investigate the accuracy issue of different Inertial Monitoring Unit (IMU) built into the Arduino 101 for stop detection. Then, by curve fitting and optimization calculation on system parameter tuning and modeling, each sensor could start up at the optimum time in order to achieve minimum energy consumption and maximum data usability. Also we propose a formula that can count the minimum number of required PM2.5 sensors under the condition of total experiment time spent and total expected number of sampling point. Finally, we conduct a field experiment to evaluate the proposed model in a real world setting. The results show that total time spent of PM2.5 data collection is similar to the expected time derived from the system modeling.

Wireless indoor localization is a key technology for the future Internet of things (IoT) paradigm. In this paper, we perform an experimental comparative study of machine learning-based localization schemes, such as k-nearest neighbor (k-NN) and variants of support vector machine (SVM), based on the received signal strength (RSS) measurements of the ambient frequency modulation (FM) and digital video broadcasting- terrestrial (DVB-T) signals in three real testbed environments. The consideration of readily available, ambient radio signals frees the need for dedicated radio transmitters. Noise-reduction techniques such as feature selection and ensemble learning are proposed in conjunction with SVM. Our results examine the performance comparisons between SVM and k-NN, as well as the performance comparisons of SVM-based methods incorporating different noise-reduction schemes, with noisy RSS data. Insights into the performance of learning- based localization schemes working with real database collected from real environments are provided.

Recent advances in technology have led to the rapid development of the Internet of Things (IoT) and the use of smartphones. The phenomenon has been widely influenced by the use of smart sensors, the accuracy of which is critical to the quality of service provided. Hence, it is vital that sensors are calibrated at both the device and software levels. Some applications even rely on environment-dependent sensing results. Floor detection by barometers is one example, where the pressure of a floor actually depends on the base reference pressure. This observation motivates us to study the floor detection problem by using barometers with device-to-device (D2D) communications. In this paper, we propose the Floor Calibration Protocol (FCP), which enables a mobile device to be calibrated with other mobile devices and anchor devices by considering both temporal and spatial factors. We conducted trace- based simulations to evaluate the proposed scheme and compared its performance with that of state- of-the-art approaches to validate the efficacy of our protocol.

Recently, elder care services, especially for those with chronic diseases, have become an important application in the field of IoT. Several studies have reported that the progress of chronic diseases can be slowed down effectively by moderate physical activities. In this paper, we propose an activity intensity monitoring system to support continuous monitoring of indoor calorie consumption. The system uses wireless signals to infer calorie burned in walking. In the proposed system, Bluetooth smart beacons are deployed in various spaces of a house. The elders who wear smart wristbands are continuously monitored for their locations by received wireless signals. The system has the following features. First, it utilizes wireless signals as location and activity intensity indicators; inertial sensors are only used in the training stage. Second, it is semi-training free and only needs relatively short training time to start using the monitoring services. Finally, the system is map-free, in the sense that it can be easily deployed without constructing the interior map of the house. A prototype system is implemented on a smartphone to evaluate the feasibility of the proposed system.

Indoor localization systems aim to accurately and cost-effectively locate targets. While numerous indoor localization solutions based on Wi-Fi, Bluetooth, ZigBee and other technologies have been proposed, they fail to achieve satisfactory performance. The accuracy of these solutions is often affected by obstacles such as shelf, human etc; due to the hight frequency of the signals, which weaken the ability of signals to penetrate obstacles. Moreover, high accuracy comes at the expense of more hardware, labor-expensive deployment. To overcome these limitations, we propose an indoor localization system based on Software Defined Radio (SDR), using FM and DVB-T signals. Our system achieves sub-meter accuracy at low cost with low site survey overhead. Additionally, we investigate the temporal and window effect on different sizes of training data.

With the explosion of smartphones and social network services, location-based social networks (LBSNs) are increasingly seen as tools for businesses (e.g., restaurants and hotels) to promote their products and services. In this article, we investigate the key techniques that can help businesses promote their locations by advertising wisely through the underlying LBSNs. In order to maximize the benefit of location promotion, we formalize it as an influence maximization problem in an LBSN, i.e., given a target location and an LBSN, a set of k users (called seeds) should be advertised initially such that they can successfully propagate and attract many other users to visit the target location. Existing studies have proposed different ways to calculate the information propagation probability, that is, how likely it is that a user may influence another, in the setting of a static social network. However, it is more challenging to derive the propagation probability in an LBSN since it is heavily affected by the target location and the user mobility, both of which are dynamic and query dependent. This article proposes two user mobility models, namely the Gaussian-based and distance-based mobility models, to capture the check-in behavior of individual LBSN users, based on which location-aware propagation probabilities can be derived. Extensive experiments based on two real LBSN datasets have demonstrated the superior effectiveness of our proposals compared with existing static models of propagation probabilities to truly reflect the information propagation in LBSNs.

With the explosion of smartphones and social network services, location-based social networks (LBSNs) are increasingly seen as tools for businesses (e.g., restaurants, hotels) to promote their products and services. In this paper, we investigate the key techniques that can help businesses promote their locations by advertising wisely through the underlying LBSNs. In order to maximize the benefit of location promotion, we formalize it as an influence maximization problem in an LBSN, i.e., given a target location and an LBSN, which a set of k users (called seeds) should be advertised initially such that they can successfully propagate and attract most other users to visit the target location. Existing studies have proposed different ways to calculate the information propagation probability, that is how likely a user may influence another, in the settings of static social network. However, it is more challenging to derive the propagation probability in an LBSN since it is heavily affected by the target location and the user mobility, both of which are dynamic and query dependent. This paper proposes two user mobility models, namely Gaussian-based and distance-based mobility models, to capture the check-in behavior of individual LBSN user, based on which location-aware propagation probabilities can be derived respectively. Extensive experiments based on two real LBSN datasets have demonstrated the superior effectiveness of our proposals than existing static models of propagation probabilities to truly reflect the information propagation in LBSNs.

In this paper, we target the location promotion problem in location-based social networks (LBSNs). The location promotion problem is given a location, we select a set of users as seeds to influence as many users as possible who are likely to visit a selected location. Specifically, we model the location promotion problem as an influence maximization problem on a graph and explore the independent cascading diffusion model on the graph. To determine the propagation probability of the edges of our proposed graph, the relation between users and the selected location should be detected. A property of LBSN is that the major reason of users visiting a location is based on their mobility. Therefore, we propose a mobility model DMM (Distance-based Mobility Model) to model each user's mobility. DMM exploits random walk with restart and the power law property of users' movements. Based on DMM and the selected location, the propagation probability of edges can be derived. In the evaluation, we show the performance of our proposed algorithms on two real datasets.

Long-distance Wi-Fi technology has shown promise in several network applications that cannot utilize conventional technologies effectively. Although the network performance of Wi-Fi technology is affected by a number of environmental factors, there is a dearth of long-term, continuous, and systematic studies on the correlations between those factors and the technology's performance. In this study, we deployed a long-distance Wi-Fi testbed on our campus and conducted a one-year experiment. Comprehensive data analysis of the measurement results shows that rainfall is the major weather attribute that affects the network performance of long-distance Wi-Fi links. In addition, the performance is highly correlated to human activities in the immediate vicinity. The results also demonstrate it is possible to infer people's daily routines on campus by exploiting the long-term measurement data.

How to effectively exchange information between parties in a disaster management system is one of the fundamental challenges to support timely and efficient disaster response and relief. Specifically, the timeliness, scalability, and availability are three desirable features for information exchange. We call the framework to support information exchange with the three features an Open Information Gateway (OIGY). In this paper, we present and experiment the mechanisms to recover communicaiton service during and after disasters for Open Information Gateway. The designed mechanisms adopt long distance radio to support tele-communication and computer networks, dynamic power adjustment for radio stations to maximize the radio recovery with minimal interference, and dynamic MESH routers to connect devices located in different networks. We experimented the proposed mechanims on a physical communication testbeds to measure the delay for service recovery and reliability of the proposed mechanism.

In this paper, we tackle the problem of discovering movement-based communities of users, where users in the same community have similar movement behaviors. Note that the identification of movement-based communities is beneficial to location-based services and trajectory recommendation services. Specifically, we propose a framework to mine movement-based communities which consists of three phases: 1) constructing trajectory profiles of users, 2) deriving similarity between trajectory profiles, and 3) discovering movement-based communities. In the first phase, we design a data structure, called the Sequential Probability tree (SP-tree), as a user trajectory profile. SP-trees not only derive sequential patterns, but also indicate transition probabilities of movements. Moreover, we propose two algorithms: BF (standing for breadth-first) and DF (standing for depth-first) to construct SP-tree structures as user profiles. To measure the similarity values among users’ trajectory profiles, we further develop a similarity function that takes SP-tree information into account. In light of the similarity values derived, we formulate an objective function to evaluate the quality of communities. According to the objective function derived, we propose a greedy algorithm Geo-Cluster to effectively derive communities. To evaluate our proposed algorithms, we have conducted comprehensive experiments on two real data sets. The experimental results show that our proposed framework can effectively discover movement-based user communities.

Knowing link characteristics of a network connection is essential for the efficient design, management, and usage of a network. In view of the proliferation of power line communication systems, in this work, we propose a novel measurement approach, called PLC-Probe, to concurrently measure link capacity and available bandwidth of PLC links. PLC-Probe uses `train-chirps' to probe the network, and it combines the strengths of the packet train and packet chirp approaches for estimation of link characteristics. Using testbed experiment, we evaluate the proposed approach and show that PLC-Probe is accurate, stable, and robust against cross traffic. Moreover, PLC-Probe is simple, effective, and applicable to other multi-rated PLC-like networks.

In this paper, we develop a new bus travel time prediction framework, called Historical Trajectory based Travel/Arrival Time Prediction (HTTP) for real-time prediction of travel time over future segments (and thus the arrival time at stops) of an on-going bus journey. The basic idea behind HTTP is to use a collection of historical trajectories "similar" to the current bus trajectory to predict the future segments. Specifically, the HTTP framework (1) samples a set of similar trajectories as the basis for travel time estimation instead of relying on only one historical trajectory best matching the on-going bus journey; and (2) explores different prediction schemes, namely, passed segments, temporal features, and hybrid methods, to identify the sample set of similar trajectories. We conduct a comprehensive empirical experimentation using real bus trajectory data collected from Taipei City, Taiwan to validate our ideas and to evaluate the proposed schemes. Experimental result shows that the proposed prediction schemes significantly outperforms the state-of-the-art and baseline techniques.

How to exchange information between parties in a mega-scale disaster management system is one of the fundamental challenges to support timely and efficient disaster response and relief. Specifically, the timeliness, scalability, and availability are three desirable features for information exchange. We call the framework to support information exchange with the three features an open information gateway, OIGY in short. In this paper, we present the challenges of information gateway, and the design of the communication protocols and the fundamental components and algorithms to support the aforementioned features. The efforts of this work will be divided into two major components: one is the distributed Truthful Real-time Information Publishing and Subscribing (TRIPS), and the other one is Heterogeneous And Plug-n-PlaY networks (HAPPY). The two components in OIGY collaborate to provide reliable and timely information publish and subscription service. TRIPS is responsible for logical information exchange management. Compared to modern real-time publish and subscription services, TRIPS is aimed at information responsiveness and distributed content-based filtering in an un-reliable network. To achieve better responsiveness, TRIPS will take advantage of the run-time service composition of SOA framework to select information sources. To enhance the success rate, TRIPS relies the information routing information provided by HAPPY. HAPPY will integrate heterogeneous communication networks including 3G/WiMAX telecommunication network and mesh mobile network into a coherent communication network and discovers the routes with probabilistic bandwidth guarantee.

The comfort of rides has been identified as one of the top criteria that affects passengers' satisfaction with public transportation systems. Conventional comfort measurement approaches rely on professional measure tools or interviews from passengers, which are costly, time-consuming, and not yet feasible. The concept of Internet of Things (IoT) is a new solution to answer this problem. The idea of IoT is to interconnect state-of-the-art digital products in physical world to provide more powerful applications. Vehicles equipped with GPS devices and wireless access technologies are parts of the IoT elements in traffic networks. We use the GPS data to measure the comfort level of vehicle rides, and provide a detailed comfort statistics as a value added service. Using real data collected from one of the Taipei taxi service providers, we show that over 95% taxi trajectories are viewed as comfortable. In addition, rides without passengers get higher comfort scores than with passengers. We also give the rankings of all taxi drivers according to a number of criteria, such as the comfort score and the number of loads. With the ranking results, we can track back to the trajectories and infer drives' driving behaviors, road conditions, and traffic conditions. We believe that the proposed solution has the potential to provide a representative comfort measurement service for taxi services and additional value-added services for public transportation systems.

We propose PRoPHET+, a routing scheme for opportunistic networks designed to maximize successful data delivery rate and minimize transmission delay. PRoPHET+ computes a deliverability value to determine the routing path for packets. Deliverability is calculated using a weighted function consisting of evaluations of nodes' buffer size, power, location, popularity, and the predictability value from PRoPHET. Even though the proposed PRoPHET+'s weights are chosen based on qualitative considerations, it is possible for PRoPHET+ to perform even more efficiently in various environments by shifting the weights accordingly. Our simulation illustrates that PRoPHET+ can perform better or equal to the routing protocol PRoPHET if logical choices for weights are used.

Saving DTNs' battery power to prolong the system life time has been an important research topic which scientists try to optimize. Most researches fall into two categories - Time Division Multiple Access based ones, and Low Power Listening ones by turning the listener's radio on/off intermittenly with long sending preambles. In this paper we propose solving the problem from a whole new angle - by using the motes' physical behavior, which is detectable by relatively energy-cheap accelerometer, as the judgment for turning the radio on/off. This scheme can be used in combination with any energy efficient DTN MAC protocols. We evaluate the protocol by running simulations with real hikers' traces and the result is shown in the evaluation section.

The ESP game belongs to the genre called Games with a Purpose (GWAP), which leverage people's desire to be entertained and also outsource certain steps of the computational process to humans. The productivity of ESP-like GWAP systems depends to a great extent on the puzzle selection strategy used in the system. Although traditional approaches seek to determine the optimal number of agreements reached in each puzzle, they may be affected by the equality of outcomes issue because they ignore the differences among puzzles. In this paper, using realistic game traces, we define the puzzle diversity issue and propose a novel approach, called the Adaptive Puzzle Selection Algorithm (APSA), to promote equality of opportunity in ESP-like GWAP systems. We also introduce a data structure called the Weight Sum Tree (WST) to reduce the computational complexity of the proposed scheme and facilitate its implementation in real-world systems. Using a comprehensive set of simulations, we evaluate the APSA scheme against the traditional OPSA scheme, and demonstrate that APSA can better accommodate the differences among puzzles in ESP-like GWAP systems.

Abstract TheGames With A Purpose(GWAP) genre is a type ofHuman Computationthat outsources certain steps of the computational process to humans. By taking advantage of people's desire to be entertained, GWAP attracts people to play voluntarily, and also produce useful metadata as a by‐product. The games have shown promise in solving a variety of problems, which computer computation has been unable to resolve completely thus far. In this paper, we propose a metric, calledsystem gain, for evaluating the performance of GWAP systems, and also use analysis to study the properties of GWAP systems. We argue that it is important for GWAP systems to implement properpuzzle selection strategiesin order to collect human intelligence in a more efficient manner. Therefore, based on our analysis, we implement anOptimal Puzzle Selection Strategy(OPSA) to improve GWAP systems. Using a comprehensive set of simulations, we demonstrate that the proposed OPSA approach can effectively improve the system gain of GWAP systems, as long as the number of puzzles in the system is sufficiently large. Copyright © 2010 John Wiley & Sons, Ltd.

The “games with a purpose” (GWAP) genre is a type of “human computation” that outsources certain steps of the computational process to humans. Although most GWAP studies focus on the design and analysis of GWAP systems, a systematic and thorough evaluation of existing systems is lacking. We address the issue in this paper. Taking the ESP game as an example, we propose a metric, calledsystem utility, for evaluating the performance of GWAP systems, and use analysis to study the properties of the ESP game. We argue that GWAP systems should be designed and played with strategies. To this end, based on our analysis, we implement an optimal puzzle selection strategy (OPSA) to improve GWAP systems. Using a comprehensive set of simulations, we show that the proposed OPSA approach can improve the system utility of the ESP game significantly. In addition, we implement a quasi ESP game, called ESP Lite, which embeds three puzzle selection algorithms transparently and records the complete game trace for evaluation and further research. During a one-month experiment, we have investigated the inner properties of the three strategies in real-world GWAP systems, and verified that the OPSA scheme achieves the best system utility for the ESP game. The results of this study demonstrate that GWAP systems are more efficient if they are designed and played with strategies.

An opportunistic network is a type of challenged network that has attracted a great deal of attention in recent years. While a number of schemes have been proposed to facilitate data dissemination in opportunistic networks, there is an implicit assumption that each participating peer behaves collaboratively. Consequently, these schemes may be vulnerable if there are uncooperative or malicious peers in the network. In this study, we identify five types of non-collaborative behavior, namely free rider, black hole, supernova, hypernova, and wormhole behavior, in opportunistic networks. We also evaluate the impacts of the five types of behavior on the data transmission performance of three widely used routing schemes. Using simulations as well as real-world traces of network mobility, we show that the data forwarding performance degrades significantly as the number of non-collaborative peers, except wormholes, increases. Moreover, we find that the three compared routing schemes can benefit from wormhole behavior, especially when the network connectivity is poor and the buffer size is limited.

Geospatial tagging (geotagging) is an emerging and very promising application that can help users find a wide variety of location-specific information, and facilitate the development of future location-based services. Conventional geotagging systems share some limitations, such as the use of a two-p

The objective of YushanNet is to provide a reliable and robust system for hiker tracking in Yushan National Park, Taiwan. The aggregated information can help national parks to provide various services to tourists, and the collected hiking traces can provide more precise information to professional rescue teams if there are hikers lost in the mountains. YushanNet is a delay and disruption tolerant system. In the system, each hiker is required to carry a matchbox-size device, which consists of a ZigBee-based mote and a GPS receiver, and the device records its hiking trace, along with encounter information with other devices. Then, the recorded data is disseminated in the network in a store-carry-and-forward fashion, until it reaches one of the base stations along the trail. In this demo, we will present the design, implementation, and deployment of the YushanNet system, and we will demonstrate the system using a small-scale network scenario.

As the R&D experience accumulates, there is a rising interest of wireless sensor network (WSN) deployment in the urban environment. For mission critical applications such as healthcare and workplace safety, in particular, it is essential that the data dissemination mechanisms satisfy two important quality of service (QoS) requirements: (1) high delivery rate and (2) low transmission delay. Proposed in this work is a cross-layer designed data dissemination mechanism, referred to as Cross-Layer Diffusion (XD), in which notions in the path discovery (routing) component are exploited by data forwarding (MAC) component to improve the delivery rate and transmission delay. Using traces collected from a prototype WSN deployed in urban environment, we compare XD to the state-of-the-art mechanisms and find that XD is not only more efficient but also more practical.

With wireless technologies extending to every part of our daily lives, mobile networking applications are becoming increasingly popular for accessing the Internet. In this paper, we propose a peer-to-peer approach, called M-FTP, for mobile FTP services. Unlike traditional approaches, the proposed scheme implements a Collaborative Forwarding algorithm that takes advantage of opportunistic wireless connections, thereby improving network capacity by exploiting the diversity of network mobility. Using simulations as well as real- world network scenarios, we demonstrate that the proposed scheme provides a better mobile FTP service than traditional schemes, and thus facilitates more effective data transfer on the go.

With wireless technologies extending to every part of our daily lives, mobile networking applications are becoming increasingly popular for accessing the Internet. Among them, Web surfing is one of the most important applications because the World Wide Web has accelerated the dissemination of information and knowledge via the Internet unencumbered by geographic boundaries. In this paper, we propose a peer-to-peer approach, called PPWeb, for mobile Web surfing. Unlike traditional approaches, the proposed scheme implements a collaborative forwarding algorithm that takes advantage of opportunistic wireless connections, thereby improving network capacity by exploiting the diversity of network mobility. Using simulations as well as real-world network scenarios, we demonstrate that the proposed scheme provides a better Web surfing service than traditional schemes, and thus facilitates more effective Web surfing on the go.

We propose a Scalable and Collaborative Internet Access approach, called SCIA, for mobile web surfing. Unlike traditional approaches, the proposed scheme implements a Collaborative Forwarding algorithm that takes advantage of opportunistic wireless connections, and thus improves network capacity by exploiting the diversity of network mobility. Moreover, using the Layered Multiple Description Coding (LMDC) algorithm, the SCIA scheme allows the end user to 'preview' the web content, even before the data has been completely transferred. Using simulations as well as realistic opportunistic people network scenarios, we demonstrate that the SCIA scheme provides a better web surfing service than traditional schemes, and thus facilitates more effective web surfing on the go.

An opportunistic network is a type of challenged network that has attracted increasing attention recently. While a number of schemes have been proposed to facilitate data forwarding in opportunistic networks, an implicit assumption is made in common that each participating peer is collaborative in the network. Consequently, these schemes become vulnerable when there exist uncooperative peers in the network. In this study, we evaluate two widely used routing schemes in opportunistic networks with three types of uncooperative behaviors, namely free riders, black holes, and wormholes. Using simulation as well as realistic network scenarios, we show that the data forwarding performance degrades significantly as the number of free riders increases. Moreover, we show that the epidemic scheme is more resilient against black holes than the PROPHET scheme, while both schemes are robust against wormholes.

"Human computation" represents a new paradigm of applications that take advantage of people's desire to be entertained and produce useful metadata as a by-product. By creating games with a purpose, human computation has shown promise in solving a variety of problems that computer computation cannot currently resolve completely. Using the ESP game as an example, we propose a metric, called system gain, for evaluating the performance of human computation systems, and also use analysis to study the properties of the ESP game. We argue that human computation systems should be played with a strategy. To this end, we implement an optimal puzzle selection strategy (OPSA) based on our analysis to improve human computation. Using a comprehensive set of simulations, we demonstrate that the proposed OPSA approach can effectively improve the system gain of the ESP game, as long as the number of puzzles in the system is sufficiently large.

In this paper, we propose an analytical model for computing the utility of ESP games, i.e., the throughput rate of appropriate labels for given puzzles. The model targets generalized games, where the number of players, the consensus threshold, and the stopping condition are variable. Via extensive simulations, we show that our model can accurately predict the stopping condition that will yield the optimal utility of an ESP game under a specific setting. A service provider can therefore utilize the model to ensure that the hosted ESP games produce high-quality labels efficiently, given that the number of players willing to invest time and effort in the game is limited.

In this paper, we address the challenges of content transfer in opportunistic networks, and propose techniques to better facilitate data dissemination based on the characteristics of the content. To investigate this problem from its origins, we propose three message scheduling algorithms: Sequential Forwarding (SF), Full Interleaving (FI), and Block-based Interleaving (BI). Each algorithm is embedded in a specially tailored data dissemination technique to evaluate the benefits of applying it to different types of content and data dissemination methods. Three types of content (file, video and web) are considered and evaluated, and the dissemination methods considered are Layered Multiple Description Coding (LMDC) based and file-based. Using simulations as well as both synthetic and realistic network scenarios, we evaluate the proposed schemes in terms of latency and user perceived quality, and demonstrate how the schemes can achieve much better latency performance for file transfers. Furthermore, we show that using LMDC-based techniques leads to higher user perceived quality, since the end user is allowed to "preview" video file or web content, even before the data has been completely transferred. The effectiveness and robustness of our message scheduling algorithms and their corresponding content dissemination techniques make them ideal solutions that can go a long way toward effective data dissemination in opportunistic networks.

Opportunistic network is a type of delay tolerant networks (DTN) where network communication opportunities appear opportunistic. In this study, we investigate opportunistic network scenarios based on public network traces, and our contributions are the following: First, we identify the censorship issue in network traces that usually leads to strongly skewed distribution of the measurements. Based on this knowledge, we then apply the Kaplan-Meier Estimator to calculate the survivorship of network measurements, which is used in designing our proposed censorship removal algorithm (CRA) that is used to recover censored data. Second, we perform a rich set of analysis illustrating that UCSD and Dartmouth network traces show strong self-similarity, and can be modeled as such. Third, we pointed out the importance of these newly revealed characteristics in future development and evaluation of opportunistic networks.

We demonstrate a seamless vertical handoff solution, called Universal Seamless Handoff Architecture (USHA). USHA is simple and requires minimal modification to the current Internet infrastructure. Therefore, it is instantly ready for realworld deployment. In this demonstration, we present USHA in two scenarios. Using video streaming applications, we demonstrate that USHA is able to successfully maintain the application connectivity and achieve almost zero delay during a vertical handoff. Moreover, we present a vertical handoff detection technique based on the end-to-end link capacity monitoring. The ongoing work of this study is to improve the accuracy of the handoff detection and to enhance application QoS support for vertical handoffs given the accurate handoff detection is provided.

In this demonstration, we will showcase SenProbe, a lightweight capacity estimation technique specifically designed for the CSMA based wireless sensor networks. SenProbe is a packet train technique based on the concept pioneer by CapProbe, but designed specially for the wireless environment. We will use wireless devices with limited radio ranges to display the properties of multi-hop wireless networks (a table top experiment). More specifically, we will illustrate wireless properties with IEEE 802.15.4 enabled devices, and offer insights into how the capacity of a wireless path changes in real wireless environments can be deployed, measured, and used. This tool will be useful for network users as well as network designers to gain better understanding of their network, and plan their activities accordingly. This demonstration aims to provide results that can be of assistance in various facets of IEEE 802.15.4 enabled intelligent home/industrial networking environments.

With the increasingly deployed Wireless Personal Area Network (WPAN) devices, channel conflict has become very frequent and severe when one WPAN technology coexists with other WPAN technologies in the same interfering range. In this paper, we study the coexistence issue between various IEEE 802.15 based WPAN technologies. We present analytical models on the non-conflicting channel allocation probabilities, focusing on the coexistence scenarios of one WPAN technology coexisting with another. The results show that channel allocation conflicts occurs frequently in all cases, and is especially severe between IEEE 802.15.3 and IEEE 802.15.4 networks. On the other hand, the probability of non-conflict channel allocation is less dramatic between a single IEEE 802.15.1 and coexisting IEEE 802.15.4 networks. In addition, the proposed models in this paper are also applicable to other wireless technologies, as long as the channel allocation mechanisms are known.

In this paper we enhance the Digital Content Mediator (DCM) approach, a legitimate online service that uses financial incentives as an effective weapon to fight against online piracy. We provide needed network security support for the DCM service. The DCM mediator is a trusted notary to ensure fair and legitimate deals between digital content selling peers and buying peers. (1) In our design, the mediator sees no raw bits of digital contents. This saves storage and communication resource for the central mediator. (2) For the seller and buyer in a DCM-legitimized transaction, one wants payment and the other wants the content. The DCM protocol ensures that neither of them can stop the protocol in the middle to steal its service without serving the other party. (3) A digital content may have many legitimate copies from large amount of sellers. In a large-scale random network like the Internet, transaction fairness is defined as the condition that a buyer wants to buy the copy from the seller with shortest downloading delay (i.e., largest seller-to-buyer pairwise bandwidth) given the same amount of financial charge. DCM employs flow network security countermeasures to ensure that a seller keeps its bandwidth promises. Our experiments on the Internet confirm the effectiveness of our design.

Mobile ad-hoc networks (MANETs) are most useful in unprepared emergencies where critical applications must be launched quickly. However, they often operate in an adverse environment where end-to-end connectivity is highly susceptible to disruption. Adjusting the motion of existing nodes or deploying additional nodes can improve the connectivity under some circumstances, but for scenarios where connectivity cannot be immediately improved, disruption must be coped with properly. In this paper we propose the ad-hoc storage overlay system (ASOS). ASOS is a self-organized overlay of storage-abundant nodes to jointly provide distributed and reliable storage to data flows under disruption. ASOS is a delay-tolerant networking (DTN) approach that significantly improves the applicability of MANETs in practice

Based on the IEEE 802.15.4 LR-WPAN standard, the ZigBee standard has been proposed to interconnect simple, low rate, and battery powered wireless devices. The deployment of ZigBee networks is expected to facilitate numerous applications, such as home-appliance networks, home healthcare, medical monitoring, consumer electronics, and environmental sensors. An effective routing scheme in a ZigBee network is particularly important in that it is the key to achieve resource (e.g., bandwidth and energy) efficiency in ZigBee networks. Routing in a ZigBee network is not exactly the same as in a MANET. In particular, while full function devices (FFD) can serve as network coordinators or network routers, reduced function devices (RFD) can only associate and communicate with FFDs in a ZigBee network. Therefore, different from traditional MANET routing algorithms, which only take into account node mobility to figure out a best route to a given destination, node heterogeneity plays an important role in ZigBee network routing. In this paper, we perform extensive evaluation, using NS-2 simulator, to study the impact of node heterogeneity on ZigBee mesh network routing. The results show that the ZigBee mesh routing algorithm exhibits significant performance difference when the network is highly heterogenous. We also reveal that the node type and the role of the node plays a critical role in deciding routing performances.

IEEE 802.15.4 is an emerging wireless standard addressing the needs of low-rate wireless personal area networks with a focus on enabling various pervasive and ubiquitous applications that require interactions with our surrounding environments. In view of the application potential of IEEE 802.15.4, knowing the fundamental network properties soon becomes essential in fasten the interactivity between these devices. Among all, knowing effective capacity of a path in wireless networks is of particular importance in routing and traffic management. In this paper, we implement SenProbe, a recently proposed path capacity estimation tool specially designed for the multi-hop ad hoc wireless environment. We present an implementation of SenProbe in sensor operating system (SOS), and evaluate the behavior/effectiveness of SenProbe in various testbed setups; including an interfered setting that cannot be simulated. Experiment results validate the workings of SenProbe and offer insights into how the capacity of a wireless path changes in real wireless environments. Our efforts provide a basis for realistic results that can be of assistance in activities such as capacity planning, protocol design, performance analysis, and etc

In this article we study a smooth and efficient transport protocol for real-time video over wireless networks. The proposed scheme, named the video transport protocol (VTP), has a new and unique end-to-end rate control mechanism that aims to avoid drastic rate fluctuations while maintaining friendliness to legacy protocols. VTP is also equipped with an achieved rate estimation scheme and a loss discrimination algorithm, both end-to-end, to cope with random errors in wireless networks efficiently. We show by analysis that VTP preserves most of the convergence properties of AIMD and converges to its fair share fast. VTP is compared to two recent TCP friendly rate control (TFRC) extensions, namely TFRC Wireless and MULTFRC, in wired-cum-wireless scenarios in Ns-2. Results show that VTP excels in all tested scenarios in terms of smoothness, fairness, and opportunistic friendliness. VTP is also implemented to work with a video camera and an H.263 video codec as part of our hybrid testbed, where its good performance as a transport layer protocol is confirmed by measurement results.

In this work, we determine an analytical relationship between the average path length of traffic connections of a Bluetooth scatternet and the overall throughput and power consumption of the network. Results obtained implementing this analytical relationship to different scatternet topologies are presented and discussed. By reducing the hop count in a scatternet we can achieve better performance in terms of throughput and power consumption. Therefore, the issue of minimizing the hop count in the presence of mobility, changing traffic flows and varying interference receives an important role. In our analysis we also show the impact of the link quality on the overall throughput. The obtained results motivate the importance of heuristics aimed at reducing the communication path length in a scatternet.

Real-time streaming over wireless links is challenging. The streaming protocol must be efficient and robust to random wireless loss, fair to itself, and friendly to legacy TCP. Various solutions have been proposed in the literature, among which the video transport protocol (VTP), TFRC wireless, and MULTFRC are end-to-end representatives. In this paper we provide an in-depth comparison on the performance of VTP, TFRC wireless, and MULTFRC in various wireless scenarios. The results show that VTP and TFRC wireless both perform well and deliver similar performance, with VTP exhibiting greater efficiency and smoothness in presence of heavy errors. In contrast, MULTFRC performs less satisfactorily, as it experiences large rate fluctuation and slow convergence caused by the frequent changes in the number of simultaneous connections.

Vertical handoffs between different wireless technologies usually lead to dramatic changes in the link capacity. A successful QoS solution for vertical handoffs must be able to fast track the capacity changes and agilely adapt the delivery rates and qualities of the ongoing applications. Though traditional AIMD-based source adaptation schemes (as found in TCP, TFRC, etc.) have been well designed for mild, gradual rate adjustments required by load fluctuations and network congestion, their response time is inadequate when the rate must be adjusted to the drastic network capacity changes that are typical in vertical handoff scenarios. To expedite the response to such changes, we propose in this paper two adaptive algorithms, named the fast rate adaptation (FRA) and early rate reduction (ERR), that are launched when the handoff is from low to high capacity (LOW-to-HIGH) or from high to low capacity (HIGH-to-LOW), respectively. We also propose two vertical handoff notification mechanisms to work with FRA and/or ERR, i.e. the implicit handoff notification (IHN) and explicit handoff notification (EHN). We show by simulation that our proposed schemes are able to provide better QoS support than the traditional AIMD based schemes during vertical handoffs.

TCP throughput limitations over wireless links have received considerable attention in the last few years. One of the problems is that TCP congestion control interprets packet losses as an indication of congestion, whereas in wireless links, losses could be due to transient link quality degradations. In this paper, we propose and study a link layer solution and evaluate its effects on TCP in the context of Bluetooth. We enhance the Bluetooth link layer to make use of channel state information and accordingly adapt the Bluetooth packet type to enhance TCP throughput We propose a simple analytical method to determine the optimal packet type for a given channel state by adding FEC support or changing packet size. Since wireless interfaces, such as 802.11 or Bluetooth, can provide information regarding the channel state using relevant APIs, this simple enhancement can easily be added to the link layer. We implemented this functionality in the Bluetooth link layer. Our simulation experiments show that the proposed adaptive packet type solution significantly improves TCP throughput. The throughput enhancement increases with the error rate. For high error rates close to 0.1%, the link layer enhanced with the adaptive scheme is able to maintain good TCP throughput, whereas throughput is almost zero when the adaptive scheme is not used.

Wireless communication is inherently vulnerable to errors from the dynamic wireless environment. Link layer packets discarded due to these errors impose a serious limitation on the maximum achievable throughput in the wireless channel. To enhance the overall throughput of wireless communication, it is necessary to have a link layer transmission scheme that is robust to the errors intrinsic to the wireless channel. To this end, we present interleaved-forward error correction (I-FEC), a clever link layer coding scheme that protects link layer data against random and busty errors. We examine the level of data protection provided by I-FEC against other popular schemes. We also simulate I-FEC in Bluetooth, and compare the TCP throughput result with Bluetooth's integrated FEC coding feature. We show that I-FEC consistently and significantly outperforms other link layer coding schemes by providing an impressive amount of protection against heavy channel burst errors.

CapProbe is an inexpensive and accurate means to estimate capacity. CapProbe combines both dispersion and end-to-end delay to estimate the capacity of the narrowest link on a path. We evaluate in this paper the accuracy of CapProbe estimation, and its dependence on end systems speed, packet sizes, narrow link speeds, and other system parameters. We test kernel and user level implementations of CapProbe and find the kernel implementation to be much more accurate. We also evaluate through experiments the effect of probing packet size on the accuracy of CapProbe estimation. Finally, we explore the idea of a "passive CapProbe" within the context of a TCP flow. Passive here means that the dispersion and delay observed for the TCP flow data and ACK packets, without introducing any additional probing packets. We test active and passive versions of CapProbe with TCP. The active version is found to produce more accurate capacity estimates than the passive version.

Video streaming has become a popular form of transferring video over the Internet. With the emergence of mobile computing needs, a successful video streaming solution demands 1) uninterrupted services even with the presence of mobility and 2) adaptive video delivery according to current link properties. We study the need and evaluate the performance of adaptive video streaming in vertical handoff scenarios. We use universal seamless handoff architecture (USHA) to create a seamless handoff environment, and use the video transfer protocol (VTP) to adapt video streaming rates according to "eligible rate estimates". Using testbed measurements experiments, we verify the importance of service adaptation, as well as show the improvement of user-perceived video quality, via adapting video streaming in the vertical handoffs.

We present a new capacity estimation technique, called CapProbe. CapProbe combines delay as well as dispersion measurements of packet pairs to filter out samples distorted by cross-traffic. CapProbe algorithms include convergence tests and convergence speed-up techniques by varying probing parameters. Our study of CapProbe includes a probability analysis to determine the time it takes CapProbe to converge on the average. Through simulations and measurements, we found CapProbe to be quick and accurate across a wide range of traffic scenarios. We also compared CapProbe with two previous well-known techniques, pathchar and pathrate. We found CapProbe to be much more accurate than pathchar and similar in accuracy to pathrate, while providing faster estimation than both. Another advantage of CapProbe is its lower computation cost, since no statistical post processing of probing data is required.

The problem of estimating the capacity of an Internet path is one of fundamental importance. Due to the multitude of potential applications, a large number of solutions have been proposed and evaluated. The proposed solutions so far have been successful in partially addressing the problem, but have suffered from being slow, obtrusive or inaccurate. In this work, we evaluate CapProbe, a low-cost and accurate end-to-end capacity estimation scheme that relies on packet dispersion techniques as well as end-to-end delays. The key observation that enabled the development of CapProbe is that both compression and expansion of packet pair dispersion are the result of queuing due to cross-traffic. By filtering out queuing effects from packet pair samples, CapProbe is able to estimate capacity accurately in most environments, with minimal processing and probing traffic overhead. In fact, the storage and processing requirements of CapProbe are orders of magnitude smaller than most of the previously proposed schemes. We tested CapProbe through simulation, Internet, Internet2 and wireless experiments. We found that CapProbe error percentage in capacity estimation was within 10% in almost all cases, and within 5% in most cases.