Spectrum crowdsensing is a paradigm where participants upload their collected spectrum data to the cloud for extracting analytics. First movers like Microsoft Spectrum Observatory and Electrosense, though with support from leading industry, research, and government, still suffer from sustainability challenges. In this paper, we present Fiesta, a sustainable framework for spectrum crowdsensing. On the technology side, we use federated learning and blockchain to decentralize the data analysis computations. For individual participants, minimal invasion of privacy suppresses concerns regarding large-scale adoption. From organizations’ perspectives, using blockchain avoids single point of failure and enhances the robustness of the entire system against malicious attacks. On the policy side, we propose a reward quantification mechanism to motivate engagement. Potential funding sources to ensure ongoing sustainability are also discussed. We have demonstrated Fiesta through simulation testbeds and real-world deployments with two demo tasks. Results show that Fiesta, as a decentralized framework, can preserve user privacy, enhance system robustness, maintain data fidelity compared with traditional methods, and fairly reward participants. We believe Fiesta is a stepping stone for the future spectrum crowdsensing paradigm.

Weidi Luo, Shenghong Dai, Xiaogeng Liu, Suman Banerjee, Huan Sun, Muhao Chen, Chaowei Xiao. Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2025.

Machine learning models deployed locally on social media applications are used for features, such as face filters which read faces in-real time, and they expose sensitive attributes to the apps. However, the deployment of machine learning models, e.g., when, where, and how they are used, in social media applications is opaque to users. We aim to address this inconsistency and investigate how social media user perceptions and behaviors change once exposed to these models. We conducted user studies (N=21) and found that participants were unaware to both what the models output and when the models were used in Instagram and TikTok, two major social media platforms. In response to being exposed to the models' functionality, we observed long term behavior changes in 8 participants. Our analysis uncovers the challenges and opportunities in providing transparency for machine learning models that interact with local user data.

Deploying large language models (LLMs) locally on mobile devices is advantageous in scenarios where transmitting data to remote cloud servers is either undesirable due to privacy concerns or impractical due to network connection. Recent advancements have facilitated the local deployment of LLMs. However, local deployment also presents challenges, particularly in balancing quality (generative performance), latency, and throughput within the hardware constraints of mobile devices. In this paper, we introduce our lightweight, all-in-one automated benchmarking framework that allows users to evaluate LLMs on mobile devices. We provide a comprehensive benchmark of various popular LLMs with different quantization configurations (both weights and activations) across multiple mobile platforms with varying hardware capabilities. Unlike traditional benchmarks that assess full-scale models on high-end GPU clusters, we focus on evaluating resource efficiency (memory and power consumption) and harmful output for compressed models on mobile devices. Our key observations include: i) differences in energy efficiency and throughput across mobile platforms; ii) the impact of quantization on memory usage, GPU execution time, and power consumption; and iii) accuracy and performance degradation of quantized models compared to their non-quantized counterparts; and iv) the frequency of hallucinations and toxic content generated by compressed LLMs on mobile devices.

Radio frequency (RF) techniques have shown promise for continuous contactless healthcare applications. However, real-world indoor environments pose challenges for existing systems, which may struggle to detect subtle physiological signals. This paper proposes Medusa, a novel wireless vital-sign sensing system designed for multi-view setups. It enables users to deploy distributed Multiple Input Multiple Output (MIMO) arrays into their daily living environments, facilitating vital-sign sensing in real-world settings. Unlike most existing single Commercial Off-The-Shelf (COTS) radar-based systems that operate under controlled settings Medusa's primary novelty lies in the design of a first-of-its-kind flexible multi-view vital sign sensing system that is view-agnostic, pose-agnostic, contactless, and can sense basic human vitals with good accuracy. Through our well-engineered hardware and software co-design, Medusa enables real-time processing of large distributed MIMO arrays, while balancing the tradeoff between Signal-to-Noise Ratio (SNR) and spatial diversity gain across each of its four distributed 4 × 4 sub-arrays for increased robustness. This is achieved using our novel unsupervised learning model which effectively recovers vital sign waveforms by decomposing the received signals. Extensive evaluations with 21 participants demonstrate Medusa's spatial diversity gain for real-world vital-sign monitoring, enabling free movement and orientation of subjects in both familiar and unfamiliar indoor environments.

This paper presents Babel, the expandable modality alignment model, specially designed for multi-modal sensing. While there has been considerable work on multi-modality alignment, they all struggle to effectively incorporate multiple sensing modalities due to the data scarcity constraints. How to utilize multi-modal data with partial pairings in sensing remains an unresolved challenge.

Recent work highlights how gradient-level access can lead to successful inference and reconstruction attacks against federated learning (FL). In such settings, differentially private (DP) learning is known to provide resilience. However, approaches used in the status quo (i.e., central and local DP) introduce disparate utility vs. privacy trade-offs. In this work, we mitigate such trade-offs through hierarchical FL (HFL). For the first time, we demonstrate that by the introduction of a new intermediary level where calibrated noise can be added, better trade-offs can be obtained; we term this hierarchical DP (HDP). Our experiments with 3 different datasets (commonly used as benchmarks for FL in prior works) suggest that HDP produces models as accurate as those obtained using central DP, where noise is added at a central aggregator at a lower privacy budget.

First responders (FRs) navigate hazardous, unfamiliar environments in the field (e.g., mass-casualty incidents), making life-changing decisions in a split second. AR head-mounted displays (HMDs) have shown promise in supporting them due to its capability of recognizing and augmenting the challenging environments in a hands-free manner. However, the design space have not been thoroughly explored by involving various FRs who serve different roles (e.g., firefighters, law enforcement) but collaborate closely in the field. We interviewed 26 first responders in the field who experienced a state-of-the-art optical-see-through AR HMD, as well as its interaction techniques and four types of AR cues (i.e., overview cues, directional cues, highlighting cues, and labeling cues), soliciting their first-hand experiences, design ideas, and concerns. Our study revealed both generic and role-specific preferences and needs for AR hardware, interactions, and feedback, as well as identifying desired AR designs tailored to urgent, risky scenarios (e.g., affordance augmentation to facilitate fast and safe action). While acknowledging the value of AR HMDs, concerns were also raised around trust, privacy, and proper integration with other equipment. Finally, we derived comprehensive and actionable design guidelines to inform future AR systems for in-field FRs.

Mobile apps have embraced user privacy by moving their data processing to the user’s smartphone. Advanced machine learning (ML) models, such as vision models, can now locally analyze user images to extract insights that drive several functionalities. Capitalizing on this new processing model of locally analyzing user images, we analyze two popular social media apps, TikTok and Instagram, to reveal (1) what insights vision models in both apps infer about users from their image and video data and (2) whether these models exhibit performance disparities with respect to demographics. As vision models provide signals for sensitive technologies like age verification and facial recognition, understanding potential biases in these models is crucial for ensuring that users receive equitable and accurate services.We develop a novel method for capturing and evaluating ML tasks in mobile apps, overcoming challenges like code obfuscation, native code execution, and scalability. Our method comprises ML task detection, ML pipeline reconstruction, and ML performance assessment, specifically focusing on demographic disparities. We apply our methodology to TikTok and Instagram, revealing significant insights. For TikTok, we find issues in age and gender prediction accuracy, particularly for minors and Black individuals. In Instagram, our analysis uncovers demographic disparities in extracting over 500 visual concepts from images, with evidence of spurious correlations between demographic features and certain concepts.

Federated Learning (FL) enables collaborative training of machine learning models for edge devices (e.g., mobile phones) over a network without revealing raw data of the participants. The existing FL benchmarks mostly assume static distribution of local data over time failing to capture the behavior of real-world applications with space-time varying data (e.g., autonomous cars). Our framework addresses this limitation by leveraging popular open source physics simulator (CARLA) and FL framework (OpenFL) to allow collection of streaming data from the mobile agents and feeding them to a practically deployable FL engine for online collaborative training. It also provides the FL researchers with the ability to model data heterogeneity, annotate data with practically zero cost, and perform reproducible continual FL experiments. We believe that this is one of the first attempts to demonstrate online FL on realistic streaming datasets from a virtual world. The demo showcases this framework using a popular object detection use case.

Visual codes such as QR codes provide a low-cost and convenient communication channel between physical objects and mobile devices, but typically operate when the code and the device are in close physical proximity. We propose a system, called QfaR, which enables mobile devices to scan visual codes across long distances even where the image resolution of the visual codes is extremely low. QfaR is based on location-guided code scanning, where we utilize a crowd-sourced database of physical locations of codes. Our key observation is that if the approximate location of the codes and the user is known, the space of possible codes can be dramatically pruned down. Then, even if every "single bit" from the low-resolution code cannot be recovered, QfaR can still identify the visual code from the pruned list with high probability. By applying computer vision techniques, QfaR is also robust against challenging imaging conditions, such as tilt, motion blur, etc. Experimental results with common iOS and Android devices show that QfaR can significantly enhance distances at which codes can be scanned, e.g., 3.6cm-sized codes can be scanned at a distance of 7.5 meters, and 0.5m-sized codes at about 100 meters. QfaR has many potential applications, and beyond our diverse experiments, we also conduct a simple case study on its use for efficiently scanning QR code-based badges to estimate event attendance.

Understanding spectrum activity is challenging when attempted at scale. The wireless community has recently risen to this challenge in designing spectrum monitoring systems that utilize many low-cost spectrum sensors to gather large volumes of sampled data across space, time, and frequencies. These crowdsensing systems are limited by the uplink bandwidth available to backhaul the raw in-phase and quadrature (IQ) samples and power spectrum density (PSD) data needed to run various applications. This paper presents FlexSpec, a framework based on the Walsh-Hadamard transform to compress spectrum data collected from distributed and low-cost sensors for real-time applications. This transformation allows sensors to significantly save uplink bandwidth thanks to its inherent properties both when it is applied to IQ and PSD data. Additionally, by leveraging a feedback loop between the sensor and the edge device it connects to, FlexSpec carefully adapts the compression ratio over time to changes in the spectrum and different applications, jointly considering data size, application performance, and spectrum variations. We experimentally evaluate FlexSpec in several applications. Our results show that FlexSpec is particularly suitable for IoT transmissions and signals close to the noise floor. Compared with prior work, FlexSpec provides up to$7\times $more reduction of uplink data size for signal detection based on PSD data, and reduces up to$6\times $to$8\times $the number of undecodable messages for IQ sample decoding.

We describe a study conducted at a large public university campus in the United States which shows the efficacy of network log information for digital contact tracing and prediction of COVID-19 cases. Over the period of January 18, 2021 to May 7, 2021, more than 216 million client-access-point associations were logged across over 11,000 wireless access points (APs). The association information was used to find potential contacts for approximately 30,000 individuals. Contacts are determined using an AP colocation algorithm, which supposes contact when two individuals connect to the same WiFi AP at approximately the same time. The approach was validated with a truth set of 350 positive COVID-19 cases inferred from the log data by observing associations with APs in isolation residence halls reserved for individuals with a confirmed (clinical) positive COVID-19 test result. The network log data and AP-colocation have a predictive value of greater than 10%; more precisely, the contacts of an individual with a confirmed positive COVID-19 test have greater than a 10% chance of testing positive in the following 7 days (compared with a 0.79% chance if chosen at random, a relative risk ratio of 12.6). A cumulative exposure score is computed to account for exposure to multiple individuals that test positive. Over the duration of the study, the exposure score predicts positive cases with a true positive rate of 16.5% and missed detection rate of 79% at a specified operating point.

This paper revisits the opportunities of using WiFi offloading for Vehicle to Internet (V2I) communication, and how this has changed over the last decade. With the rollouts of provider-managed WiFi networks that are more structured and operate under authenticated regimes, WiFi offloading, or use of available (roadside) WiFi networks for V2I data communication, has different opportunities and challenges. To study the current landscape,we develop a system (X-Fi), which efficiently selects, associates to, authenticates with, and performs WiFi offloading for V2I communication with these networks, and a tool (X-Perf), which illustrates opportunities of WiFi offloading available today in these networks, with measurements and experiments across four metro areas across three continents over 22 months. Our results indicate the feasibility of achieving 1 GB/hour application goodput, an order of magnitude higher than the number provided by open WiFi networks in the past, which can take a significant load away from alternative communication paths for V2I systems. Moreover, we provide several implications on transport protocols and WiFi deployments to shed light on the use of such WiFi networks for V2I communication.

Detecting anomalous inputs, such as adversarial and out-of-distribution (OOD) inputs, is critical for classifiers (including deep neural networks or DNNs) deployed in real-world applications. While prior works have proposed various methods to detect such anomalous samples using information from the internal layer representations of a DNN, there is a lack of consensus on a principled approach for the different components of such a detection method. As a result, often heuristic and one-off methods are applied for different aspects of this problem. We propose an unsupervised anomaly detection framework based on the internal DNN layer representations in the form of a meta-algorithm with configurable components. We proceed to propose specific instantiations for each component of the meta-algorithm based on ideas grounded in statistical testing and anomaly detection. We evaluate the proposed methods on well-known image classification datasets with strong adversarial attacks and OOD inputs, including an adaptive attack that uses the internal layer representations of the DNN (often not considered in prior work). Comparisons with five recently-proposed competing detection methods demonstrates the effectiveness of our method in detecting adversarial and OOD inputs.

Context-based authentication is a method for transparently validating another device's legitimacy to join a network based on location. Devices can pair with one another by continuously harvesting environmental noise to generate a random key with no user involvement. However, there are gaps in our understanding of the theoretical limitations of environmental noise harvesting, making it difficult for researchers to build efficient algorithms for sampling environmental noise and distilling keys from that noise. This work explores the information-theoretic capacity of context-based authentication mechanisms to generate random bit strings from environmental noise sources with known properties. Using only mild assumptions about the source process's characteristics, we demonstrate that commonly-used bit extraction algorithms extract only about 10% of the available randomness from a source noise process. We present an efficient algorithm to improve the quality of keys generated by context-based methods and evaluate it on real key extraction hardware. MOONSHINE is a randomness distiller which is more efficient at extracting bits from an environmental entropy source than existing methods. Our techniques nearly double the quality of keys as measured by the NIST test suite, producing keys that can be used in real-world authentication scenarios.

Multipath TCP (MPTCP) is a promising protocol that aggregates the bandwidth of mobile client's multiple interfaces. However, currently it is still not widely adopted. A key reason for this slow adoption is that the layer-4 load balancers (LBs) used to scale TCP based services in data centers are not MPTCP-aware and forward the multiple TCP subflows of the same MPTCP connection independently to different backends (BEs). In this paper, we present RomanRoads (RR), an MPTCP-aware layer-4 LB which eliminates this hurdle to widespread MPTCP adoption. Compared with prior proposals, RR is easily deployable because it does not change the service provider's network configuration, makes no modification to ordinary TCP protocol and only minimal modification to the MPTCP connection setup process, and supports the case of multiple LBs. We implement RR in the form of a software LB and validate its correctness and high performance through extensive experiments. RR achieves 100% correctness at steady state, no connection disruption during LB churns, and line-rate throughput for packets from 5-tuple seen before. Moreover, we shed light on the desired properties of an LB-friendly multipath layer-4 protocol to provide guidance for future multipath protocol design.

LoRa has seen widespread adoption as a long range IoT technology. As the number of LoRa deployments grow, packet collisions undermine its overall network throughput. In this paper, we propose a novel interference cancellation technique -- Concurrent Interference Cancellation (CIC), that enables concurrent decoding of multiple collided LoRa packets. CIC fundamentally differs from existing approaches as it demodulates symbols by canceling out all other interfering symbols. It achieves this cancellation by carefully selecting a set of sub-symbols -- pieces of the original symbol such that no interfering symbol is common across all sub-symbols in this set. Thus, after demodulating each sub-symbol, an intersection across their spectra cancels out all the interfering symbols. Through LoRa deployments using COTS devices, we demonstrate that CIC can increase the network capacity of standard LoRa by up to 10x and up to 4x over the state-of-the-art research. While beneficial across all scenarios, CIC has even more significant benefits under low SNR conditions that are common to LoRa deployments, in which prior approaches appear to perform quite poorly.

Abstract Advances in deep learning have made face recognition technologies pervasive. While useful to social media platforms and users, this technology carries significant privacy threats. Coupled with the abundant information they have about users, service providers can associate users with social interactions, visited places, activities, and preferences–some of which the user may not want to share. Additionally, facial recognition models used by various agencies are trained by data scraped from social media platforms. Existing approaches to mitigate associated privacy risks result in an imbalanced trade-off between privacy and utility. In this paper, we address this trade-off by proposing Face-Off, a privacy-preserving framework that introduces strategic perturbations to images of the user’s face to prevent it from being correctly recognized. To realize Face-Off, we overcome a set of challenges related to the black-box nature of commercial face recognition services, and the scarcity of literature for adversarial attacks on metric networks. We implement and evaluate Face-Off to find that it deceives three commercial face recognition services from Microsoft, Amazon, and Face++. Our user study with 423 participants further shows that the perturbations come at an acceptable cost for the users.

The emergence of advanced in-vehicle infotainment (IVI) systems, such as Apple CarPlay and Android Auto, calls for fast and intuitive device pairing mechanisms to discover newly introduced devices and make or break a secure, high-bandwidth wireless connection. Current pairing schemes are tedious and lengthy as they typically require users to go through pairing and verification procedures by manually entering a predetermined or randomly generated pin on both devices. This inconvenience usually results in prolonged usage of old pins, significantly degrading the security of network connections.

Wireless video streaming has become an extremely popular application in recent years. Internet video streaming to mobile devices, however, faces several challenges, e.g., unstable wireless connections, long latency, high jitter and etc. Bitrate adaptive streaming and video transcoding solutions are widely used to address the above-mentioned issues, however, there are still several shortcomings of these approaches. Such challenges hinder providing satisfactory quality of video streaming service to the mobile users. We propose a hardware-assisted, real-time video transcoding solution implemented on a commercial off-the-shelf device, Raspberry Pi. We employ the software and hardware coupled architecture in order to improve the performance/quality of video streaming and enhance the user satisfactions in wireless network. Our video transcoding solution can be applied to both the downlink and uplink streaming: for downlink stream, it can provide agile bitrate adaptation to sudden network dynamics and enhance video quality by running our transcoding solution at the wireless edge. It can be used to uplink stream for broadcasting live streams in real-time. We present the design and implementation of our video transcoding system in both cases with practical scenarios. The evaluation results reveal that our transcoding solution enhances the performance of video streaming compared with other adaptive bitrate streamings and it provides higher video quality without causing rebuffering or video stall. We bridge the gap between the wireless channel capacity and the video quality while providing a better streaming experience to end user.

We present PANDA, a data acquisition technique for energy-constrained sensor nodes that reduces the energy per operation required to sample and preprocess analog sensor data. PANDA takes advantage of the energy consumption patterns of commodity microcontrollers by sampling input signals in short bursts followed by long periods of inactivity. This approach reduces the overhead of repetitively transitioning the CPU and analog components in and out of low-power sleep states. This nonuniformly-spaced input data is then fed to a nonuniform FFT algorithm that computes the frequency spectrum. We show that the spectrum computed with the nonuniform FFT is very close to the spectrum that would be computed from uniformly sampled data preprocessed with a conventional FFT. The output of the nonuniform FFT can be filtered or post-processed with conventional frequency domain analysis techniques, and a uniformly resampled output can be constructed with the conventional inverse FFT. We compare the energy consumption patterns of burst-mode sampling to those of conventional uniform sampling in several real sensor nodes. We demonstrate that for reasonably sized input datasets, burst mode sampling and postprocessing consumes more than 17 percent less energy than conventional uniform sampling, including the additional computations required to compute the nonuniform DFT.

Understanding spectrum characteristics with little prior knowledge requires fine-grained spectrum data in the frequency, spatial, and temporal domains; gathering such a diverse set of measurements results in a large data volume. Analysis of the resulting dataset poses unique challenges; methods in the status quo are tailored for specific spectrum-related applications (apps), and are ill equipped to process data of this magnitude. In this paper, we design BigSpec, a general-purpose framework that allows for fast processing of apps. The key idea is to reduce computation costs by performing computation extensively on compressed data that preserves signal features. Adhering to this guideline, we build solutions for three apps, i.e., energy detection, spatio-temporal spectrum estimation, and anomaly detection. These apps were chosen to highlight BigSpec's efficiency, scalability, and extensibility. To evaluate BigSpec's performance, we collect more than 1 terabyte of spectrum data spanning a year, across 300MHz-4GHz, covering 400 km2. Compared with baselines and prior works, we achieve 17× run time efficiency, sublinear rather than linear run time scalability, and extend the definition of anomaly to different domains (frequency & spatio-temporal). We also obtain high-level insights from the data to provide valuable advice on future spectrum measurement and data analysis.

Abstract Audio-based sensing enables fine-grained human activity detection, such as sensing hand gestures and contact-free estimation of the breathing rate. A passive adversary, equipped with microphones, can leverage the ongoing sensing to infer private information about individuals. Further, with multiple microphones, a beamforming-capable adversary can defeat the previously-proposed privacy protection obfuscation techniques. Such an adversary can isolate the obfuscation signal and cancel it, even when situated behind a wall. AudioSentry is the first to address the privacy problem in audio sensing by protecting the users against a multi-microphone adversary. It utilizes the commodity and audio-capable devices, already available in the user’s environment, to form a distributed obfuscator array. AudioSentry packs a novel technique to carefully generate obfuscation beams in different directions, preventing the multi-microphone adversary from canceling the obfuscation signal. AudioSentry follows by a dynamic channel estimation scheme to preserve authorized sensing under obfuscation. AudioSentry offers the advantages of being practical to deploy and effective against an adversary with a large number of microphones. Our extensive evaluations with commodity devices show that protects the user’s privacy against a 16-microphone adversary with only four commodity obfuscators, regardless of the adversary’s position. AudioSentry provides its privacy-preserving features with little overhead on the authorized sensor.

This half-day tutorial will explore edge computing through hands-on development activities using a physical edge computing platform. With the recent trends in the networking ecosystem-the rise of IoT devices, high bandwidth wireless, and powerful, energy-efficient, and inexpensive computation-edge computing is a promising technology for highly interactive and immersive environments. This tutorial aims to promote awareness of the possibilities of edge computing in general and introduce attendees to the tools that they can use to begin exploring this space. The tutorial's activities will give attendees hands-on experience with the ParaDrop edge computing platform developed at the University of Wisconsin-Madison. We hope to impart working knowledge about edge computing and our vision for its future. Recent developments in home wireless networks such as the proliferation of connected devices (e.g. Internet-of-Things) and gigabit wireless (e.g. 802.11ac) create an environment where edge computing can truly enhance applications. The availability of low latency and high bandwidth at the network edge but not necessarily end-to-end to the cloud, can enable interesting new applications involving video (e.g. augmented reality), low latency sensor-actuator coordination, and other public safety or educational applications. However, before such applications emerge, there needs to be a platform available for them. We argue that the availability of widespread platforms and standards for edge computing will unleash a new wave of innovation much like the creation of app markets for mobile devices. The hardware exists, e.g. in the form of millions of always-on Wi-Fi routers in homes and businesses, but it is not very programmable. With our tutorial, we aim to foster awareness of the possibilities that exist with edge computing and promote the advancement of research toward good standards for its realization. We introduce the ParaDrop platform as an open source demonstration of our vision for edge computing. We have been single-mindedly focused on researcher and developer needs in creating the ParaDrop platform. As such, applications that are already written to be run as cloud services can be easily modified to run on the ParaDrop platform in order to benefit from running at the network edge. Applications are not required to be written in a highly specialized language, and applications are able to leverage the rapidly growing software ecosystem surrounding Docker. Our tutorial will give researchers a hands-on experience with the platform and knowledge that they will be able to use beyond the workshop.

Mobile virtual reality (VR) headsets, such as Google Cardboard and Samsung GearVR, can reuse a smartphone as near-eye display to create immersive experience. But such devices barely support any user interaction, even for simple tasks such as menu selection and single-character input. In this paper, we design Inkput, a simple passive interface attached to the unexploited backside of the headset to enable touch sensing. Inkput is a piece of paper substrate with carbon ink patterns printed atop. It leverages the column of electrodes near the edge of the smartphone touchscreen to sense multi-touch on the 2D space, and is even able to locate finger hovering. Our experiments demonstrate that Inkput can precisely detect touch positions with mm-level precision. Our case studies in actual VR applications also verify that Inkput can support common VR interactions and can even outperform high-end handheld controllers in terms of efficiency.

Video streaming is one of the killer applications in recent years. Video transcoding plays an important role in the video streaming service to cope with the various purposes. Specifically, content owners and publishers heavily utilize video transcoders to reconfigure source video in a variety of formats, video qualities, and bitrate to provide end users with the best possible quality of service. In this paper, we present VideoCoreCluster, a low‐cost and energy‐efficient transcoder cluster that is suitable for live streaming services. We designed and implemented real‐time video transcoder cluster using cheap ($35), powerful, and energy‐efficient Raspberry Pi. The quality of transcoded video provided by VideoCoreCluster is similar to the best software‐based video transcoder while consuming significantly less energy (<3 W). We have proposed a scheduling algorithm based on priority of video stream and transcoding capacity. Our cluster manager provides reliable and scalable streaming services, because it uses the characteristics of adaptive bitrate scheme. We have deployed our transcoding cluster to provide IP‐based TV streaming services on our university campus.

This paper introduces GonioSense -- a wearable based system that can track and measure both active and passive joint motions. GonioSense leverages embedded inertial sensors to provide reliable and consistent measurement results. A key challenge for our system is how to efficiently detect body movement and accurately calculate range of joint motion. We built a preliminary version that can collect and analyze sensor data in real time on the Android platform. To test the system, we recruited five healthy volunteers and measured six kinds of joint motions. The same tools used by physical therapists, measurements gathered from a goniometer and an inclinometer is what we used as the ground truth data for calibrations and evaluation of our algorithm. In the initial version, GonioSense estimation errors are within ± 5% for more than 80% cases. By utilizing wearable technology, the benefits for both doctors and patients could bring forth an immense step forward in physical therapy treatment.

We present dBHound, a mobile sensing analytics app for tracking room occupancy and activities while preserving privacy for users. dBHound takes advantage of audio sensors available on mobile devices to collect data about the environment. dBHound is a layer in the mobile operating system API that gathers decibel (dB) data from the device's onboard audio sensors and strips it of sensitive information. Using mobile devices as sensors has the advantage that users do not have to purchase or manage additional sensing devices within the home. Among the different sensor modalities discussed in [4, 1, 2, 5, 3] many of which are limited to detecting a small subset of activities in-home, sound is a highly general and aligns well with everyday human activity. We demonstrate that having access to decibel-level data streams from multiple devices can give us more insight into activity and occupancy patterns than a single sensor, without compromising privacy.

Small cells form a critical component of next generation cellular networks, where spatial reuse is the key to higher spectral efficiencies. Interference management in the spatial domain through beamforming allows for increased reuse without having to sacrifice resources in the time or frequency domain. Existing beamforming techniques for spatial reuse, being coupled with client scheduling, face a key limitation in practical realization, especially with OFDMA small cells. In this context, we argue that for a practical spatial reuse system with beamforming, it is important to decouple beamforming from client scheduling. Further, we show that jointly addressing client association with beamforming is critical to maximizing the reuse potential of beamforming. Towards our goal, we propose ProBeam - a system for multi-cell beamforming and client association in OFDMA small cell networks. ProBeam incorporates two key components - a low complexity, highly accurate SINR estimation module that helps determine interference dependencies for beamforming between small cells; and an efficient, low complexity joint client association and beam selection algorithm for the small cells that accounts for scheduling at the small cells without being coupled with it. We have prototyped ProBeam on a WiMAX-based network of four small cells. Our evaluations reveal the accuracy of our SINR estimation module to be within 1 dB, and the reuse gains from joint client association and beamforming to be as high as 115 percent over baseline approaches.

In dense wireless deployments at homes, such as apartment buildings, neighboring home WLANs share the same unlicensed spectrum by deploying consumer-grade access points in their individual homes. In such environments, WiFi networks can suffer from intermittent performance issues such as wireless packet losses, interference from WiFi and non-WiFi sources due to the rapid growth and increasing diversity of devices that share the spectrum. In this paper, we propose a vendor-neutral cloud-based centralized framework called COAP to configure, coordinate and manage individual home APs using an open API implemented by these commodity APs. The framework, implemented using OpenFlow extensions, allows the APs to share various types of information with a centralized controller - interference and traffic phenomenon and various flow contexts, and in turn receive instructions - configuration parameters (e.g., channel) and transmission parameters (through coarse-grained schedules and throttling parameters). This paper describes the framework and associated techniques, applications to motivate its potential benefits, such as, upto 47% reduction in channel congestion and our experiences from having deployed it in actual home environments.

This paper introduces EcoDrive, a fuel consumption sensing and control system for modern vehicles, implemented in an embedded platform, to improve fuel efficiency and reduce carbon emissions. EcoDrive senses vehicle dynamics through the standard vehicle On-board diagnostics (OBD) port and models various vehicle forces, i.e., propulsion, drivetrain loss, wind resistance and grade resistance, as functions of instant fuel consumption. By sensing vehicular speed and controlling air/fuel injection rate in real time, EcoDrive can adjust speed carefully to improve fuel efficiency. We have collected more than 10,000 miles of driving traces from 12 different vehicles to build models of vehicle dynamics. Based on the models, a prototype of EcoDrive is implemented in an off-the-shelf embedded platform. The prototype is installed on a regular vehicle and evaluated through test drives of more than 100 miles across both urban and highway environments. In comparison with human drivers, EcoDrive achieves an average of 20% higher fuel efficiency in urban road segments and 30% higher fuel efficiency on highways.

Internet-enabled cameras pervade daily life, generating a huge amount of data, but most of the video they generate is transmitted over wires and analyzed offline with a human in the loop. The ubiquity of cameras limits the amount of video that can be sent to the cloud, especially on wireless networks where capacity is at a premium. In this paper, we present Vigil, a real-time distributed wireless surveillance system that leverages edge computing to support real-time tracking and surveillance in enterprise campuses, retail stores, and across smart cities. Vigil intelligently partitions video processing between edge computing nodes co-located with cameras and the cloud to save wireless capacity, which can then be dedicated to Wi-Fi hotspots, offsetting their cost. Novel video frame prioritization and traffic scheduling algorithms further optimize Vigil's bandwidth utilization. We have deployed Vigil across three sites in both whitespace and Wi-Fi networks. Depending on the level of activity in the scene, experimental results show that Vigil allows a video surveillance system to support a geographical area of coverage between five and 200 times greater than an approach that simply streams video over the wireless network. For a fixed region of coverage and bandwidth, Vigil outperforms the default equal throughput allocation strategy of Wi-Fi by delivering up to 25% more objects relevant to a user's query.

Next generation wireless networks (i.e., WiMAX, LTE) provide higher bandwidth and spectrum efficiency leveraging smaller (femto) cells with orthogonal frequency division multiple access (OFDMA). The uncoordinated, dense deployments of femtocells however, pose several unique challenges relating to interference and resource management in OFDMA femtocell networks. Towards addressing these challenges, we propose RADION, a distributed resource management framework that effectively manages interference across femtocells. RADION's core building blocks enable femtocells to opportunistically determine the available resources in a completely distributed and efficient manner. Further, RADION's modular nature paves the way for different resource management solutions to be incorporated in the framework. We implement RADION on a real WiMAX femtocell testbed deployed in a typical indoor setting. Two distributed solutions are enabled through RADION and their performance is studied to highlight their quick self-organization into efficient resource allocations.

We present WiScape, a framework for benchmarking and understanding the behavior of wide-area wireless networks, e.g., city-wide or nation-wide cellular data networks using active participation from clients. The goal of WiScape is to provide a coarse-grained view of a wide-area wireless landscape that allows operators and users to understand broad performance characteristics of the network. In WiScape, a centralized controller instructs clients to collect minimal measurement samples over time and space in an opportunistic manner. To limit the overheads of this measurement framework, WiScape partitions the world into zones, contiguous areas with relatively similar user experiences, and partitions time into zone-specific epochs over which network statistics are relatively stable. For each epoch in each zone, WiScape takes a minimalistic view-it attempts to collect a small number of measurement samples to characterize the client experience in a zone at a specific epoch, thereby limiting the bandwidth and energy overheads for collecting measurements at client devices. For this effort, we have collected ground truth measurements for three commercial cellular networks across a nation-wide area in USA for a period of more than one year. We justify our design choices of WiScape through collected data, demonstrate that WiScape can provide an accurate performance characterization of the networks over a wide-area (within 4 percent error for more than 70 percent of instances) with a low overhead on the clients, and illustrate multiple applications of this framework through a sustained and ongoing measurement study.

The work in this paper is a systematic research and engineering effort in exploring the design space of multi-interface wireless repeater systems. We present the design, implementation and evaluation of a wireless repeater system, Multifacet, which opportunistically utilizes multiple interfaces to enhance capacity. The system is designed to be transparent to application endpoints and does not need any end application modifications for adoption. Multifacet incorporates several techniques to achieve efficient bandwidth utilization across multiple interfaces -- (a) coordinated channel sharing, (b) a simplified backpressure based striping technique, (c) a single link abstraction, and (d) the ability to seamlessly migrate a client to the optimal channel. Multifacet is implemented on off-the-shelf dual band wireless repeater and demonstrates high speeds (337 Mbps) operation. On average, Multifacet performs 50% better than traditional AP/repeater setups, and in the best case, more than 2x better.

The present TV whitespace networks rely on spectrum occupancy databases to determine their operating channels. In this paper, we show that such databases cause non-negligible wastage of whitespace spectrum. We also report that whitespace channels can have very different quality due to interference from secondary devices and the leakage from TV broadcasts. Such disparity in channel quality is not captured by existing databases. We propose the use of spectrum measurements to overcome the above limitations of databases. In particular, we describe a system called V-Scope that leverages spectrum sensors on public vehicles to collect and report measurements from the road. These measurements are used as "anchor points" to construct various models to better determine whitespace spectrum, estimate its channel quality, and validate locations of primary and secondary devices.

We explore the use of TV whitespace communication systems for providing robust connectivity to vehicles. A key challenge in this setup is the asymmetry in transmit power limits -- the fixed base station is allowed to communicate at up to 4 W, while the mobile gateways in vehicles are limited to 100 mW. This paper presents a specific solution to deal with this asymmetry in which whitespace transceivers are used in the downlink direction while a more traditional cellular path is used in the uplink one. While heterogeneous communication systems have been considered before (e.g., in some satellite networks), our solution explores some unique opportunities that arise in vehicular systems. In particular, we describe a system called Scout that uses a front radio at the head of a vehicle to look ahead and identify the best channel parameters to be used when the rear radio eventually reaches the forward post. We use these channel estimates to adapt a number of transmission mechanisms for improving the performance of flows through this heterogeneous network. We have implemented and deployed this system on moving vehicles in an urban environment, and demonstrated 3 -- 8× performance improvement over simpler alternatives that do not use the scouting technique.

Although wireless broadband technologies have evolved significantly over the past decade, they are still insufficient to support the fast-growing mobile traffic, especially due to the increasing popularity of mobile video applications. Wireless multicast, aiming to exploit the wireless broadcast advantage, is a viable approach to bridge the gap between the limited wireless capacity and the ever-increasing mobile video traffic demand. In this paper, we propose MuVi, a Multicast Video delivery scheme through joint optimal resource allocation and adaptive modulation and coding scheme in OFDMA-based 4G cellular networks. MuVi differentiates video frames based on their importance in reconstructing the video and incorporates an efficient radio resource allocation algorithm to optimize the overall video quality across all users in the multicast group. MuVi is a lightweight solution with most of the implementation in the gateway, slight modification in the base station, and no modification at the clients. We implement MuVi on a WiMAX testbed and compare its performance to a Naive wireless multicast scheme that employs the most robust Modulation and Coding Scheme (MCS), and an Adaptive scheme that employs the highest MCS supportable by all clients. Experimental results show that MuVi improves the average video peak signal-to-noise ratio (PSNR) by up to 13 and 7 dB compared to the Naive and the Adaptive schemes, respectively. MuVi does not require modification to the video encoding scheme or the air interface. Thus, it allows speedy deployment in existing systems.

We present a long term and large scale study of the experience of mobile users through two popular but contrasting applications in the wild. To conduct this study, we implemented a measurement framework and library, called Insight, which has been deployed on these two applications that are available through Apple's App Store and Google's Android Market. One of them, Parallel Kingdom (PK), is a popular massively multiplayer online role-playing game (MMORPG) which has over a million unique users distributed more than 120 countries. The other application, StudyBlue (SB), is an educational application with over 160,000 unique users. Our study spans most of the life of the PK game (more than 3 years) while our deployment with SB has been running for over a year now. We use Insight to collect diverse information about network behavior, application usage and footprints, platform statistics, user actions, and various factors affecting application revenues.

Unlicensed, secondary users of TV whitespaces today rely on spectrum occupancy databases to determine what spectrum they can use for their communication needs. In this paper, we first show that such spectrum databases (that depend solely on propagation models as per guidelines of the FCC in the USA) can be quite inaccurate leading to under-utilization of spectrum. Next, we propose that these spectrum databases can be significantly augmented using opportunistic measurements when possible. Instead of incorporating primary detection functions in each secondary device, we propose to use vehicle-mounted spectrum sensors that collect and report measurements from the road, which can serve as useful "anchor points" to enhance existing propagation models.

Geographically distributed storage is an important method of ensuring high data availability in cloud computing and storage systems. With the increasing demand for moving file systems to the cloud, current methods of providing such enterprise-grade resiliency are very inefficient. For example, replication based methods incur large storage cost though they provide low access latencies. While erasure coded schemes reduce storage cost, they are associated with large access latencies and high bandwidth cost. In this paper, we propose a novel scheme named CAROM, an ensemble of replication and erasure codes, to provide resiliency in cloud file systems with high efficiency. While maintaining the same consistency semantics seen in today's cloud file systems, CAROM provides the benefit of low bandwidth cost, low storage cost, and low access latencies. We perform a large-scale evaluation using real-world file system traces and demonstrate that CAROM outperforms replication based schemes in storage cost by up to 60% and erasure coded schemes in bandwidth cost by up to 43%, while maintaining low access latencies close to those in replication based schemes.

We present a measurement study of wireless experience in a diverse set of home environments by deploying an infrastructure, we call WiSe. Our infrastructure consists of OpenWrt-based Access Points (APs) that have been given away to residents for free to be installed as their primary wireless access mechanism. These APs are configured with our specialized measurement and monitoring software that communicates with our measurement controller through an open API. We have collected wireless performance traces from 30 homes for a period in excess of 6 months. To analyze the characteristics of these home wireless environments, we have also developed a simple metric that estimates the likely TCP throughput different clients can expect based on current channel and environmental conditions. With this infrastructure, we provide multiple quantitative observations, some of which are anecdotally understood in our community. For example, while a majority of links performed well most of the time, we observed cases of poor client experience about 2.1% of the total time.

The recently released TV whitespaces offer a promising land for wireless communications. The secondary users of TV whitespaces today rely on spectrum occupancy databases to determine vacant TV channels for unlicensed communications. However, the accuracy of these databases (that depend solely on propagation models as per guidelines of the FCC) may be low. In this paper, we propose V-Scope - a vehicular sensing framework aimed to collect wide-area spectrum measurements for evaluating the accuracy of these databases. A key design feature of V-Scope is to leverage spectrum sensors mounted on public vehicles for collecting and reporting measurements from the road (opportunistic wardriving). We have currently deployed a version of our system on a single public transit bus traveling across a mid-sized city in the US. Based on measurements collected at over 1 million locations across a 100 square-km area, we find that databases tend to over-predict the coverage of certain TV broadcasts, unnecessarily blocking the usage of whitespace spectrum in a large area (up to 42% measured locations). We further propose ways of leveraging these measurements to enhance existing propagation models in databases.

Small cells form a critical component of next generation cellular networks, where spatial reuse is the key to higher spectral efficiencies. Interference management in the spatial domain through beamforming allows for increased reuse without having to sacrifice resources in the time or frequency domain. Existing beamforming techniques for spatial reuse, being coupled with client scheduling, face a key limitation in practical realization, especially with OFDMA small cells. In this context, we argue that for a practical spatial reuse system with beamforming, it is important to decouple beamforming from client scheduling. Further, we show that jointly addressing client association with beamforming is critical to maximizing the reuse potential of beamforming.

We present Scout, a communication system leveraging TV whitespaces to support robust and high-speed streaming services. Scout uses two key techniques to improve video performance. First, it extends network coverage through an asymmetric network architecture where whitespace transceivers are used for the downlink direction while a cellular path is used for the uplink. Scout further leverages some unique opportunities that arise in vehicular systems. In particular, it sends a front radio to lookahead and identify the best channel parameters when the rear radio eventually reaches the forward post. We demonstrate the performance of Scout by using a single base station to stream a high-quality video to a vehicle driving along a 1.3km road stretch.

To meet the capacity demands from ever-increasing mobile data usage, mobile network operators are moving toward smaller cell structures. These small cells, called femtocells, use sophisticated air interface technologies such as orthogonal frequency division multiple access (OFDMA). While femtocells are expected to provide numerous benefits such as energy efficiency and better throughput, the interference resulting from their dense deployments prevents such benefits from being harnessed in practice. Thus, there is an evident need for a resource management solution to mitigate the interference that occurs between collocated femtocells. In this paper, we design and implement one of the first resource management systems, FERMI, for OFDMA-based femtocell networks. As part of its design, FERMI: 1) provides resource isolation in the frequency domain (as opposed to time) to leverage power pooling across cells to improve capacity; 2) uses measurement-driven triggers to intelligently distinguish clients that require just link adaptation from those that require resource isolation; 3) incorporates mechanisms that enable the joint scheduling of both types of clients in the same frame; and 4) employs efficient, scalable algorithms to determine a fair resource allocation across the entire network with high utilization and low overhead. We implement FERMI on a prototype four-cell WiMAX femtocell testbed and show that it yields significant gains over conventional approaches.

All practical wireless communication systems are prone to errors. At the symbol level, such wireless errors have a well-defined structure: When a receiver decodes a symbol erroneously, it is more likely that the decoded symbol is a good “approximation” of the transmitted symbol than a randomly chosen symbol among all possible transmitted symbols. Based on this property, we define approximate communication, a method that exploits this error structure to natively provide unequal error protection to data bits. Unlike traditional [forward error correction (FEC)-based] mechanisms of unequal error protection that consume additional network and spectrum resources to encode redundant data, the approximate communication technique achieves this property at the PHY layer without consuming any additional network or spectrum resources (apart from a minimal signaling overhead). Approximate communication is particularly useful to media delivery applications that can benefit significantly from unequal error protection of data bits. We show the usefulness of this method to such applications by designing and implementing an end-to-end media delivery system, called Apex. Our Software Defined Radio (SDR)-based experiments reveal that Apex can improve video quality by 5–20 dB [peak signal-to-noise ratio (PSNR)] across a diverse set of wireless conditions when compared to traditional approaches. We believe that mechanisms such as Apex can be a cornerstone in designing future wireless media delivery systems under any error-prone channel condition.

The increase in mobile data usage is pushing broadband operators towards deploying smaller cells (femtocells) and sophisticated access technologies such as OFDMA. The expected high density of deployment and uncoordinated operations of femtocells however, make interference management both critical and extremely challenging. Femtocells have to use the same access technology as traditional macrocells. Given this, understanding the impact of the system design choices (originally tailored to well-planned macrocells) on interference management, forms an essential first step towards designing efficient solutions for next-generation femtocells. This in turn is the focus of our work. With extensive measurements from our WiMAX OFDMA femtocell testbed, we characterize the impact of various system design choices on interference. Based on the insights from our measurements, we discuss several implications on how to efficiently operate a femtocell network.

Localization techniques that allow inferring the location of wireless devices directly from received signals have exposed mobile users to new threats. Adversaries can easily collect required information (such as signal strength) from target users, however, techniques securing location information at the physical layer of the wireless communication systems have not received much attention. In this paper, we propose Phantom, a novel approach to allow mobile devices thwart unauthorized adversary's location tracking by creating forged locations. In particular, Phantom leverages cooperation among multiple mobile devices in close vicinity and utilizes synchronized transmissions among those nodes to obfuscate localization efforts of adversary systems. Through an implementation on software-defined radios (GNU Radios) and extensive simulation with real location traces, we see that Phantom can improve location privacy.

Although wireless broadband technologies have evolved significantly over the past decade, they are still insufficient to support the fast-growing mobile traffic, especially due to the increasing popularity of mobile video applications. Wireless multicast, aiming to exploit the wireless broadcast advantage, is a viable approach to bridge the gap between the limited wireless networking capacity and the ever-increasing mobile video traffic demand. In this work, we propose MuVi, a Multicast Video delivery scheme in OFDMA-based 4G wireless networks, to optimize multicast video traffic. MuVi differentiates video frames based on their importance in reconstructing the video and incorporates an efficient radio resource allocation algorithm to optimize the overall video quality across all users in the multicast group. MuVi is a lightweight solution with most of the implementation in the gateway, slight modification in the base-station, and no modification at the clients. We implement MuVi on a WiMAX testbed and compare its performance to a Naive wireless multicast scheme that employs the most robust MCS (Modulation and Coding Scheme), and an Adaptive scheme that employs the highest MCS supportable by all clients. Experimental results show that MuVi improves the average video PSNR (Peak Signal-to-Noise Ratio) by up to 13 and 7 dB compared to the Naive and the Adaptive schemes, respectively. MuVi does not require modification to the video encoding scheme or the air interface. Thus it allows speedy deployment in existing systems.

Next generation wireless networks (i.e., WiMAX, LTE) provide higher bandwidth and spectrum efficiency leveraging smaller (femto) cells with orthogonal frequency division multiple access (OFDMA). The uncoordinated, dense deployments of femtocells however, pose several unique challenges relating to interference and resource management in these networks. Towards addressing these challenges, we propose RADION, a distributed resource management framework that effectively manages interference across femtocells. RADION's core building blocks enable femtocells to opportunistically find the available resources in a completely distributed and efficient manner. Further, RADION's modular nature paves the way for different resource management solutions to be incorporated in the framework. We implement RADION on a real WiMAX femtocell testbed deployed in a typical indoor setting. We extensively evaluate two solutions integrated with RADION, both via prototype implementation and simulations and quantify their performance in terms of quick and efficient self-organization.

We describe our experiences with deploying a vehicular Internet access service on public transit buses. Our system, called WiRover, has been running on these buses since April 2010 for about 18 months till date providing a WiFi hotspot to which bus passengers can connect. As of Dec. 1, 2011 we have observed 17,567 unique client devices connect to the WiRover system. These devices have downloaded more than 337.53 GB and uploaded 48.19 GB of content through our system. Since the initial deployment, the buses have logged more than 9,331 hours of operation and have traveled over much of the northern Midwest of the United States. Through this paper we provide different insights acquired in deploying and running a continuous service with real users, present various user and usage characteristics in these systems, discuss various design and management strategies for these networks, and explore different network traffic optimizations possible.

Ternary Content-Addressable Memories (TCAMs) has become the industrial standard for high-throughput packet classification. However, one major drawback of TCAMs is their high power consumption, which is becoming critical with the boom of data centers, the growing classifiers and the deployment of IPv6. In this paper, we propose a practical and efficient solution which introduces a smart pre-classifier to reduce power consumption of TCAMs for multi-dimensional packet classification. We reduce the dimension of the problem through the pre-classifier which pre-classifies a packet on two header fields, source and destination IP addresses. We then return to the high dimension problem where only a small portion of a TCAM is activated and searched for a given packet. The smart pre-classifier is built in a way such that a given packet matches at most one entry in the pre-classifier, which make commodity TCAMs sufficient to implement the pre-classifier. Furthermore, each rule is stored only once in one of the TCAM blocks, which avoids rule replication. The presented solution uses commodity TCAMs, and the proposed algorithms are easy to implement. Our scheme achieves a median power reduction of 91% and an average power reduction of 88% on real and synthetic classifiers respectively.

This paper introduces models and a system for designing 802.11 wireless LANs (WLANs) using flexible channelization— the choice of an appropriate channel width and center frequency for each transmission. In contrast to current 802.11 systems that use fixed width channels, the proposed system, FLUID, configures all access points and their clients using flexible channels. We show that a key challenge in designing such a system stems from managing the effects of interference due to multiple transmitters employing variable channel widths, in a network-wide setting. We implemented FLUID in an enterprise-like setup using a 50 node testbed (with off-the shelf wireless cards) and we show that FLUID improves the average throughput by 59 percent across all PHY rates, compared to existing fixed-width approaches.

In this paper, we propose Airshark -- a system that detects multiple non-WiFi RF devices in real-time and using only commodity WiFi hardware. To motivate the need for systems like Airshark, we start with measurement study that characterizes the usage and prevalence of non-WiFi devices across many locations. We then present the design and implementation of Airshark. Airshark extracts unique features using the functionality provided by a WiFi card to detect multiple non-WiFi devices including fixed frequency devices (e.g., ZigBee, analog cordless phone), frequency hoppers (e.g., Bluetooth, game controllers like Xbox), and broadband interferers (e.g., microwave ovens). Airshark has an average detection accuracy of 91-96%, even in the presence of multiple simultaneously active RF devices operating at a wide range of signal strengths (-80 to -30 dBm), while maintaining a low false positive rate. Through a deployment in two production WLANs, we show that Airshark can be a useful tool to the WLAN administrators in understanding non-WiFi interference.

We present WiScape, a framework for measuring and understanding the behavior of wide-area wireless networks, e.g., city-wide or nation-wide cellular data networks using active participation from clients. The goal of WiScape is to provide a coarse-grained view of a wide-area wireless landscape that allows operators and users to understand broad performance characteristics of the network. In this approach a centralized controller instructs clients to collect measurement samples over time and space in an opportunistic manner. To limit the overheads of this measurement framework, WiScape partitions the world into zones, contiguous areas with relatively similar user experiences, and partitions time into zone-specific epochs over which network statistics are relatively stable. For each epoch in each zone, WiScape takes a minimalistic view --- it attempts to collect a small number of measurement samples to adequately characterize the client experience in that zone and epoch, thereby limiting the bandwidth and energy overheads at client devices. For this effort, we have collected ground truth measurements for up to three different commercial cellular wireless networks across (i) an area of more than 155 square kilometer in and around Madison, WI, in the USA, (ii) a road stretch of more than 240 kilometers between Madison and Chicago, and (iii) locations in New Brunswick and Princeton, New Jersey, USA, for a period of more than 1 year. We justify various design choices of WiScape through this data, demonstrate that WiScape can provide an accurate performance characterization of these networks over a wide area (within 4% error for more than 70% of instances) with a low overhead on the clients, and illustrate multiple applications of this framework through a sustained and ongoing measurement study.

The effectiveness of rate adaptation algorithms is an important determinant of 802.11 wireless network performance. The diversity of algorithms that has resulted from efforts to improve rate adaptation has introduced a new dimension of variability into 802.11 wireless networks, further complicating the already difficult task of understanding and debugging 802.11 performance. To assist with this task, in this paper we present and evaluate a methodology for accurately fingerprinting 802.11 rate adaptation algorithms. Our approach uses a Support Vector Machine (SVM)-based classifier that requires only simple passive measurements of 802.11 traffic. We demonstrate that careful conversion of raw packet traces into input features for SVM is necessary for achieving high classification accuracy. We tested our classifier on the four rate adaptation algorithms available in MadWifi, cards. The classifier performs with an accuracy of 95% – 100%. We also show that the classifier is robust over a variety of network conditions if the training data includes a sufficient sampling of the range of an algorithm's behavior.

While virtual machine (VM) migration is allowing data centers to rebalance workloads across physical machines, the promise of a maximally utilized infrastructure is yet to be realized. Part of the challenge is due to the inherent dependencies between VMs comprising a multi-tier application, which introduce complex load interactions between the underlying physical servers. For example, simply moving an overloaded VM to a (random) underloaded physical machine can inadvertently overload the network. We introduce AppAware-a novel, computationally efficient scheme for incorporating (1) inter-VM dependencies and (2) the underlying network topology into VM migration decisions. Using simulations, we show that our proposed method decreases network traffic by up to 81%compared to a well known alternative VM migration method that is not application-aware.

The demand for increased spectral efficiencies is driving the next generation broadband access networks towards deploying smaller cells (femtocells) with sophisticated air interface technologies (Orthogonal Frequency Division Multiple Access or OFDMA). The projected dense deployment of femtocells however, makes interference and hence resource management both critical and extremely challenging. In this paper, we design and implement one of the first resource management systems, FERMI, for OFDMA-based femtocell networks. As part of its design, FERMI (i) provides resource isolation in the frequency domain (as opposed to time) to leverage power pooling across cells to improve capacity; (ii) uses measurement-driven triggers to intelligently distinguish clients that require just link adaptation from those that require resource isolation; (iii) incorporates mechanisms that enable the joint scheduling of both types of clients in the same frame; and (iv) employs efficient, scalable algorithms to determine a fair resource allocation across the entire network with high utilization and low overhead. We implement FERMI on a prototype four-cell WiMAX femtocell testbed and show that it yields significant gains over conventional approaches.

This paper introduces models and a system for designing 802.11 wireless LANs (WLANs) using flexible channelization -- the choice of an appropriate channel width and center frequency for each transmission. In contrast to current 802.11 systems that use fixed width channels, the proposed system, FLUID, configures all access points and their clients using flexible channels. We show that a key challenge in designing such a system stems from managing the effects of interference due to multiple transmitters employing variable channel widths, in a network-wide setting. We implemented FLUID in an enterprise-like setup using a 50 node testbed (with off-the shelf wireless cards) and we show that FLUID improves the average throughput by 59% across all PHY rates, compared to existing fixed-width approaches.

Community wireless networks (CWNs) have been proposed to spread broadband network access to underprivileged, underprovisioned, and remote areas. Research has focused on optimizing network performance through intelligent routing and scheduling, borrowing solutions from mesh networks. Surprisingly, however, there has been no work on how to make efficient use of multiple channels in CWNs in the presence of multiple gateways and a single radio per device. In fact, today's deployments in underprivileged areas are primarily single-radio and do operate on a single channel. Frequency selection in such CWNs is very complex because it does not only determine the nodes' channel of operation, but also the gateway and the routing tree to the gateway-a rather computationally intensive task. In this paper, we propose, design, implement, and evaluate SWARM, a practical system that allows a CWN to make effective use of the available wireless channels in order to offer globally optimal performance. SWARM improves performance versus current single-channel protocols by up to 7.7 × in our experiments. Moreover, while we should be expecting performance gains due to channel diversity, we clearly demonstrate that up to 3.7 × improvement is attributed to the network organization into efficient traffic distribution structures.

All practical wireless communication systems are prone to errors. At the symbol level such wireless errors have a well-defined structure: when a receiver decodes a symbol erroneously, it is more likely that the decoded symbol is a good "approximation" of the transmitted symbol than a randomly chosen symbol among all possible transmitted symbols. Based on this property, we define approximate communication, a method that exploits this error structure to natively provide unequal error protection to data bits. Unlike traditional (FEC-based) mechanisms of unequal error protection that consumes additional network and spectrum resources to encode redundant data, the approximate communication technique achieves this property at the PHY layer without consuming any additional network or spectrum resources (apart from a minimal signaling overhead). Approximate communication is particularly useful to media delivery applications that can benefit significantly from unequal error protection of data bits. We show the usefulness of this method to such applications by designing and implementing an end-to-end media delivery system, called Apex. Our Software Defined Radio (SDR)-based experiments reveal that Apex can improve video quality by 5 to 20 dB (PSNR) across a diverse set of wireless conditions, when compared to traditional approaches. We believe that mechanisms such as Apex can be a cornerstone in designing future wireless media delivery systems under any error-prone channel condition.

We present a software-based solution to the multi-dimensional packet classification problem which can operate at high line speeds, e.g., in excess of 10 Gbps, using high-end multi-core desktop platforms available today. Our solution, called Storm, leverages a common notion that a subset of rules are likely to be popular over short durations of time. By iden-tifying a suitable set of popular rules one can significantly speed up existing software-based classification algorithms. A key aspect of our design is in partitioning processor resources into various relevant tasks, such as continuously computing the popular rules based on a sampled subset of traffic, fast classification for traffic that matches popular rules, dealing with packets that do not match the most popular rules, and traffic sampling. Our results show that by using a single 8-core Xeon processor desktop platform, it is possible to sustain classification rates of more than 15 Gbps for rep-resentative rule sets of size in excess of 5-dimensional 9000 rules, with no packet losses. This performance is signifi-cantly superior to a 8-way implementation of a state-of-the-art packet classification software system running on the same 8-core machine. Therefore, we believe that our design of packet classification functions can be a useful classification building block for RouteBricks-style designs, where a core router might be constructed as a mesh of regular desktop machines.

A recent approach-COPE, presented by Katti (Proc. ACM SIGCOMM 2006, pp. 243-254)-for improving the throughput of unicast traffic in wireless multihop networks exploits the broadcast nature of the wireless medium through opportunistic network coding. In this paper, we analyze throughput improvements obtained by COPE-type network coding in wireless networks from a theoretical perspective. We make two key contributions. First, we obtain a theoretical formulation for computing the throughput of network coding on any wireless network topology and any pattern of concurrent unicast traffic sessions. Second, we advocate that routing be made aware of network coding opportunities rather than, as in COPE, being oblivious to it. More importantly, our model considers the tradeoff between routing flows close to each other for utilizing coding opportunities and away from each other for avoiding wireless interference. Our theoretical formulation provides a method for computing source-destination routes and utilizing the best coding opportunities from available ones so as to maximize the throughput. We handle scheduling of broadcast transmissions subject to wireless transmit/receive diversity and link interference in our optimization framework. Using our formulations, we compare the performance of traditional unicast routing and network coding with coding-oblivious and coding-aware routing on a variety of mesh network topologies, including some derived from contemporary mesh network testbeds. Our evaluations show that a route selection strategy that is aware of network coding opportunities leads to higher end-to-end throughput when compared to coding-oblivious routing strategies.

Community wireless networks (CWNs) have been proposed to spread broadband network access to underprivileged, under-provisioned and remote areas. Research has focused on optimizing network performance through intelligent routing and scheduling, borrowing solutions from mesh networks. Surprisingly, however, there has been no work on how to make efficient use of multiple channels in CWNs in the presence of multiple gateways, and a single radio per device. In fact, today's deployments in under-privileged areas are primarily single radio and do operate on a single channel [20]. Frequency selection in such CWNs is very complex because it does not only determine the nodes' channel of operation but also the gateway and the routing tree to the gateway - a rather computationally intensive task. In this paper, we propose, design, implement, and evaluate SWARM, a practical system that allows a CWN to make effective use of the available wireless channels in order to offer globally optimal performance. SWARM improves performance versus current single channel protocols by up to 7.7× in our experiments. Moreover, while we should be expecting performance gains due to channel diversity, we clearly demonstrate that up to 3.7 x improvement is attributed to the network organization into efficient traffic distribution structures.

Enterprise WLANs have made a dramatic shift towards centralized architectures in the recent past. The reasons for such a change have been ease of management and better design of various control and security functions. The data path of WLANs, however, continues to use the distributed, random-access model, as defined by the popular DCF mechanism of the 802.11 standard. While theoretical results indicate that a centrally scheduled data path can achieve higher efficiency than its distributed counterpart, the likely complexity of such a solution has inhibited practical consideration. In this paper, we take a fresh, implementation and deployment oriented, view in understanding data path choices in enterprise WLANs. We perform extensive measurements to characterize the impact of various design choices, like scheduling granularity on the performance of a centralized scheduler, and identify regions where such a centralized scheduler can provide the best gains.Our detailed evaluation with scheduling prototypes deployed on two different wireless testbeds indicates that DCF is quite robust in many scenarios, but centralization can play a unique role in 1) mitigating hidden terminals - scenarios which may occur infrequently, but become pain points when they do and 2) exploiting exposed terminals - scenarios which occur more frequently, and limit the potential of successful concurrent transmissions. Motivated by these results, we design and implement CENTAUR - a hybrid data path for enterprise WLANs, that combines the simplicity and ease of DCF with a limited amount of centralized scheduling from a unique vantage point. Our mechanisms do not require client cooperation and can support legacy 802.11 clients.

The increasing density of WiFi access points (APs) in metropolitan areas is enabling an opportunistic model of wireless networking, whereby a "guest" user within range of one or more wireless APs can gain temporary Internet access through these APs. In this paper, we address the problem of TCP throughput prediction for opportunistic networks. Applications of opportunistic networking can benefit from such predictions by adapting to prevailing network conditions. Our approach is different from prior efforts to model wireless network throughput in that only the two communicating endpoints participate in the prediction, and no information about network topology or traffic loads generated by interfering sources is required. Our goal is to understand how accurate throughput predictions can be under the above assumptions. The physical environment considered in our study includes varying degrees of interference, indoor and outdoor networks, and nodes that are stationary or moving at walking or driving speeds. We use throughput predictors based on time series analysis and machine learning techniques, as they are well-suited to predicting phenomena with unknown variables. The prediction accuracy that our methods yield is cause for cautious optimism. We find that 80% to 100% of predictions are within a factor of two of actual throughput. This bound on accuracy means that predictions are useful for certain applications, because this bound (a) can be achieved by measurements lasting for as little as 0.3 seconds, and (b) holds even when nodes are driving at speeds of 15-25 mph.

We present a measurement study of a large-scale urban WiFi mesh network consisting of more than 250 Mesh Access Points (MAPs), with paying customers that use it for Internet access. Our study, involved collecting multi-modal data, e.g., through continuous gathering of SNMP logs, syslogs, passive traffic capture, and limited active measurements in different parts of the city. Our study is split into four components — planning and deployment of the mesh, success of mesh routing techniques, likely experience of users, and characterization of how the mesh is utilized. During our data collection process that spanned 8 months, the network changed many times due to hardware and software upgrades. Hence to present a consistent view of the network, the core dataset used in this paper comes from a two week excerpt of our dataset. This part of the dataset had more than 1.7 million SNMP log entries (from 224 MAPs) and more than 100 hours of active measurements. The scale of the study allowed us to make many important observations that are critical in planning and using WiFi meshes as an Internet access technology. For example, our study indicates that the last hop 2.4GHz wireless link between the mesh and the client is the major bottleneck in client performance. Further we observe that deploying the mesh access points on utility poles results in performance degradation for indoor clients that receive poor signal from the access points.

We present an experimental study of IEEE 802.11n (high throughput extension to the 802.11 standard) using commodity wireless hardware. 802.11n introduces a variety of new mechanisms including physical layer diversity techniques, channel bonding and frame aggregation mechanisms. Using measurements from our testbed, we analyze the fundamental characteristics of 802.11n links and quantify the gains of each mechanism under diverse scenarios. We show that the throughput of an 802.11n link can be severely degraded (up ≈85%) in presence of an 802.11g link. Our results also indicate that increased amount of interference due to wider channel bandwidths can lead to throughput degradation. To this end, we characterize the nature of interference due to variable channel widths in 802.11n and show that careful modeling of interference is imperative in such scenarios. Further, as a reappraisal of previous work, we evaluate the effectiveness of MAC level diversity in the presence of physical layer diversity mechanisms introduced by 802.11n.

It is well known that a packet loss in 802.11 can happen either due to collision or an insufficiently strong signal. However, discerning the exact cause of a packet loss, once it occurs, is known to be quite difficult. In this paper we take a fresh look at this problem of wireless packet loss diagnosis for 802.11-based communication and propose a promising technique called COLLIE. COLLIE performs loss diagnosis by using newly designed metrics that examine error patterns within a physical-layersymbolin order to expose statistical differences between collision and weak signal based losses. We implement COLLIE through custom driver-level modifications in Linux and evaluate its performance experimentally. Our results demonstrate that it has an accuracy ranging between 60-95% while allowing a false positive rate of up to 2%. We also demonstrate the use of COLLIE in subsequent link adaptations in both static and mobile wireless usage scenarios through measurements on regular laptops and the Netgear SPH101 Voice-over-WiFi phone. In these experiments, COLLIE led to throughput improvements of 20- 60% and reduced retransmission related costs by 40% depending upon the channel conditions.

We design, implement, and evaluate a technique to identify the source network interface card (NIC) of an IEEE 802.11 frame through passive radio-frequency analysis. This technique, called PARADIS, leverages minute imperfections of transmitter hardware that are acquired at manufacture and are present even in otherwise identical NICs. These imperfections are transmitter-specific and manifest themselves as artifacts of the emitted signals. In PARADIS, we measure differentiating artifacts of individual wireless frames in the modulation domain, apply suitable machine-learning classification tools to achieve significantly higher degrees of NIC identification accuracy than prior best known schemes.

Local network coding is growing in prominence as a technique to facilitate greater capacity utilization in multi-hop wireless networks. A specific objective of such local network coding techniques has been to explicitly minimize the total number of transmissions needed to carry packets across each wireless hop. While such a strategy is certainly useful, we argue that in lossy wireless environments, a better use of local network coding is to provide higher levels of redundancy even at the cost of increasing the number of transmissions required to communicate the same information. In this paper we show that the design space for effective redundancy in local network coding is quite large, which makes optimal formulations of the problem hard to realize in practice. We present a detailed exploration of this design space and propose a suite of algorithms, called CLONE, that can lead to further throughput gains in multi-hop wireless scenarios. Through careful analysis, simulations, and detailed implementation on a real testbed, we show that some of our simplest CLONE algorithms can be efficiently implemented in today's wireless hardware to provide a factor of two improvement in throughput for example scenarios, while other, more effective, CLONE algorithms require additional advances in hardware processing speeds to be deployable in practice.

Community wireless networks have been proposed as a powerful technique to spread broadband network access to underprivileged, under-provisioned and remote areas. These networks consist of a few Internet gateways which are reached by homes using multi-hop wireless links between wireless routers. The benefits of such networks include low costs for deployment due to reduced wiring needs, low maintenance and increased flexibility. Current practice in routing protocols for such networks (e.g. LQSR, OLSR and SrcRR) is for routing protocols to obtain information about the link quality (via some metric such as ETT, ETX) and select a gateway to whom a route minimizes the cost of the metric. All nodes operate on the same known frequency to maintain connectivity.

A wide range of transmit power control (TPC) algorithms have been proposed in recent literature to reduce interference and increase capacity in 802.11 wireless networks. However, few of them have made it to practice. In many cases this gap is attributed to lack of suitable hardware support in wireless cards to implement these algorithms. In particular, many research efforts have indicated that wireless card vendors need to support power control mechanisms in a fine-grained manner - both in the number of possible power levels and the time granularity at which the controls can be applied. In this paper we claim that even if fine-grained power control mechanisms were to be made available by wireless card vendors, algorithms would not be able to properly leverage such degrees of control in typical indoor environments. We prove this claim through rigorous empirical analysis and then build a tunable empirical model (Model-TPC) that can determine the granularity of power control that is actually useful. To illustrate the importance of our solution, we conclude by demonstrating the impact of choice of power control granularity on Internet applications where wireless clients interact with servers on the Internet. We observe that the number of feasible power was found to be between 2-4 for most indoor environments. We believe that the results from this study can serve as the right set of assumptions to build practically realizable TPC algorithms in the future.

A radio frequency identifier (RFID) system consists of inexpensive, uniquely-identifiable tags that are mounted on physical objects, and readers that track these tags (and hence these physical objects) through RF communication. In this paper we, therefore, address this load balancing problem for readers - given a set of tags that are within range of each reader, which of these tags should each reader be responsible for such that the cost for monitoring tags across the different readers is balanced, while guaranteeing that each tag is monitored by at least one reader. We show that a generalized variant of the load balancing problem is NP-hard and hence present a 2-approximation centralized algorithm. We next present an optimal centralized solution for a specialized variant. Subsequently, we present a localized distributed algorithm that is probabilistic in nature and closely matches the performance of the centralized algorithms. Our results demonstrate that our schemes achieve very good performance even in highly dynamic large-scale RFID systems.

We study the problem of supporting VoIP calls in a wireless mesh network. Specifically, we propose solutions for call admission control (CAC) and route selection for VoIP calls. Call admission decisions must evaluate how the capacity of the mesh network is utilized by the existing calls. We address this issue via a measurement-based modeling effort to model mutual interference between wireless links. The modeling approach evaluates whether capacity constraints (or, required QoS metrics) will be satisfied if a new call is admitted with a given route. Evaluations with a 6-node 802.11a testbed demonstrate excellent accuracy of the model and thus also the CAC performance. We address the issue of route selection by also using a modeling approach that considers models of transmission and interference ranges to develop a polynomial-time algorithm to search for feasible routes. This problem takes exponential time for wireless networks without such modeling. In addition to studying feasibility, we study several routing metrics such as shortest feasible path and maximum residual feasible path. Finally, we develop a new method for routing using call statistics that uses prior calling patterns to avoid potentially critical links. We evaluate the performance of these route selection techniques via extensive simulations and demonstrate the superiority of using max residual feasible path over simply shortest feasible path, and routing using call statistics over max residual feasible path.

A recent approach, COPE, for improving the throughput of unicast traffic in wireless multi-hop networks exploits the broadcast nature of the wireless medium through opportunistic network coding. In this paper, we analyze throughput improvements obtained by COPE-type network coding in wireless networks from a theoretical perspective. We make two key contributions. First, we obtain a theoretical formulation for computing the throughput of network coding on any wireless network topology and any pattern of concurrent unicast traffic sessions. Second, we advocate that routing be made aware of network coding opportunities rather than, as in COPE, being oblivious to it. More importantly, our work studies the tradeoff between routing flows "close to each other" for utilizing coding opportunities and "away from each other" for avoiding wireless interference. Our theoretical formulation provides a method for computing source-destination routes and utilizing the best coding opportunities from available ones so as to maximize the throughput. We handle scheduling of broadcast transmissions subject to wireless transmit/receive diversity and link interference in our optimization framework. Using our formulations, we compare the performance of traditional unicast routing and network coding with coding-oblivious and coding-aware routing on a variety of mesh network topologies, including some derived from contemporary mesh network testbeds. Our evaluations show that a route selection strategy that is aware of network coding opportunities leads to higher end-to-end throughput when compared to coding-oblivious routing strategies.

Wireless LANs are commonplace installations in enterprise environments. Their ease of use and deployment, however, are accompanied by a difficulty in their management and security. Proposed solutions to these problems are based on centralization; in the control plane through centralized authentication and allocation of channels and power levels, and in the data plane through time slotted medium access using centralized scheduling for interference mitigation. While centralization of some control plane tasks has been shown to be feasible, centralization on the data plane is significantly harder to realize. This is because it needs to take into account the inherent variability of the wireless medium while offering bounds on delay and jitter on the control paths. In this work, we present a study of the various problems that arise in centralization of the data plane in an enterprise WLAN. We believe that a pragmatic solution for data plane centralization is the key approachto provisioning an enterprise WLAN consisting of a dense deployment of APs.

Packet classification is the foundation of many Internet functions such as QoS and security. A long thread of research has proposed efficient software-based solutions to this problem. Such software solutions are attractive because they require cheap memory systems for implementation, thus bringing down the overall cost of the system. In contrast, hardware-based solutions use more expensive memory systems, e.g., TCAMs, but are often preferred by router vendors for their faster classification speeds. The goal of this paper is to find a "best-of-both-worlds" solution -- a solution that incurs the cost of a software-based system and has the speed of a hardware-based one. Our proposed solution, called smart rule cache achieves this goal by using minimal hardware -- a few additional registers -- to cache evolving rules which preserve classification semantics, and additional logic to match incoming packets to these rules. Using real traffic traces and real rule sets from a tier-1 ISP, we show such a setup is sufficient to achieve very high hit ratios for fast classification in hardware. Cache miss ratios are 2 ∼ 4 orders of magnitude lower than flow cache schemes. Given its low cost and good performance, we believe our solution may create significant impact on current industry practice.

Abstract — We introduce MobiStream — a video streaming system that exploits the perceptual value in the video content and characteristics of the link layer and physical layer channels to enable efficient error-resilient video streaming over wireless widearea networks (WWANs). The key building block in MobiStream is the use of link-layer based, but application-controlled, virtual channels (ViCs) abstraction. Each virtual channel in MobiStream offers a level of reliability and statistical loss gaurantee using ‘awareness ’ of the characteristics of link-layer and physical layer channels. Video applications can dynamically instantiate new virtual channels, control their loss behavior, and/or flexibly switch video transmission across channels. MobiStream achieves fine-grained errorresilience by partitioning the video frames into number of small, independently decodable, blocks of data (called ‘slices’) and assigns priority to each individual slice based on its perceptual (visual) usefulness. MobiStream augments a number of other enhancements for error-resilience: multiple description video coding, perceptual slice-structured coding, low-delay inter-frame and intra-frame slice interleaving, dynamic unequal error protection, and priority-based video-data scheduling to enable efficient and error-resilient video streaming over wireless wide-area links. MobiStream has been implemented and evaluated using loss distributions from tests conducted over a commercial wide-area wireless (CDMA2000 3G) network. Results show that, even in stationary conditions, MobiStream, on average, can improve video picture quality by at least 4 dB. We conclude that significant benefits to end-user experience can be obtained by deploying such a system. I.

Abstract — There is an inherent well-known conflict between fairness and throughput that arises in many networking scenarios. A number of researchers have studied this problem in the context of (single-hop) wireless local area networks (WLANs), where clients directly exchange traffic with access points (APs). More recently, researchers have proposed multi-hop extensions to WLANs where client traffic is forwarded via a series of client-client links. In this paper, we show that the objective of improving throughput without sacrificing fairness can be much better met in multi-hop WLANs. We decouple this objective into two separate but related problems. First, we need an algorithm to organize clients into a multi-hop structure such that fair bandwidth allocation within this structure leads to improved throughput. Second, we need algorithms for performing fair bandwidth allocation within the determined multi-hop structure. In this paper, we first design optimal fair bandwidth allocation algorithms for both max-min throughput fairness and max-min time fairness in multi-hop WLANs. Subsequently, design an efficient algorithm to find desirable multi-hop structures. With slight modification, our results in this paper can be generalized to other multi-hop wireless networks, such as the emerging wireless backhaul networks and wireless mesh networks. Our proposed solutions seamlessly integrate with legacy devices and hence are incrementally deployable. Simulation results demonstrate that our solutions can effectively improve throughput (by up to 114% or more) as well as network coverage while preserving fairness. I.

Abstract — We present Sieve – an infrastructure for fast content-based data distribution to interested users. The ability of Sieve to filter and forward high-bandwidth data streams stems from its distributed pipelined architecture. The complex message filtering task is broken-up into a sequence of light-weight filtering components resulting in high end-to-end throughput. Furthermore, since each component is assigned to a node based on its resource constraints, the queue buildup inside the nodes is minimal resulting in low end-to-end latency. Our experimental results based on real system implementation show that Sieve can sustain a throughput of more than 5000 messages per second for 100000 subscriptions with predicates of 10 attributes. Index Terms — Content-based Information Dissemination, Publish-Subscribe System, Event Stream Filtering

Abstract — We propose an efficient client-based approach for channel management (channel assignment and load balancing) in 802.11-based WLANs that lead to better usage of the wireless spectrum. This approach is based on a “conflict set coloring ” formulation that jointly performs load balancing along with channel assignment. Such a formulation has a number of advantages. First, it explicitly captures interference effects at clients. Next, it intrinsically exposes opportunities for better channel re-use. Finally, algorithms based on this formulation do not depend on specific physical RF models and hence can be applied efficiently to a wide-range of in-building as well as outdoor scenarios. We have performed extensive packet-level simulations and measurements on a deployed wireless testbed of 70 APs to validate the performance of our proposed algorithms. We show that in addition to single network scenarios, the conflict set coloring formulation is well suited for channel assignment where multiple wireless networks share and contend for spectrum in the same physical space. Our results over a wide range of both simulated topologies and in-building testbed experiments indicate that our approach improves application level performance at the clients by upto three times (and atleast 50%) in comparison to current best-known techniques. I.

Wireless 802.11 hotspots have grown in an uncoordinated fashion with highly variable deployment densities. Such uncoordinated deployments, coupled with the difficulty of implementing coordination protocols, has often led to conflicting configurations (e.g., in choice of transmission power and channel of operation) among the corresponding Access Points (APs). Overall, such conflicts cause both unpredictable network performance and unfairness among clients of neighboring hotspots. In this paper, we focus on the fairness problem for uncoordinated deployments. We study this problem from the channel assignment perspective. Our solution is based on the notion of channel-hopping, and meets all the important design considerations for control methods in uncoordinated deployments - distributed in nature, minimal to zero coordination among APs belonging to different hotspots, simple to implement, and interoperable with existing standards. In particular, we propose a specific algorithm called MAXchop, which works efficiently when using only non-overlapping wireless channels, but is particularly effective in exploiting partially-overlapped channels that have been proposed in recent literature. We also evaluate how our channel assignment approach complements previously proposed carrier sensing techniques in providing further performance improvements. Through extensive simulations on real hotspot topologies and evaluation of a full implementation of this technique, we demonstrate the efficacy of these techniques for not only fairness, but also the aggregate throughput, metrics.We believe that this is the first work that brings into focus the fairness properties of channel hopping techniques and we hope that the insights from this research will be applied to other domains where a fair division of a system's resources is an important consideration.

Probabilistic packet marking is a general technique which routers can use to reveal internal network information to end-hosts. Such information is probabilistically set by the routers in headers of regular IP packets on their way to destinations. A number of potential applications have been identified, such as IP traceback, congestion control, robust routing algorithms, dynamic network reconfiguration, and locating Internet bottlenecks, etc. In this paper, we define EPPM, an efficient general probabilistic packet marking scheme with a wide range of potential applications, of which locating Internet bottlenecks and IP traceback are investigated as two representative examples to demonstrate its effectiveness. Our proposed scheme imposes only a single-bit overhead in the IP packet headers. More importantly, it significantly reduces the number of IP packets required to convey the relevant information when compared to the prior best known scheme (almost by two orders of magnitude).

Wireless wide area networks (WWANs) are becoming ubiquitous across most geographic regions, enabling simultaneous coverage from multiple providers. WWAN channels exhibit both uncorrelated and correlated behaviour on a variety of levels. In this paper we examine the statistical properties of WWAN links, and illustrate the benefits in heterogeneity that can be exploited to improve statistical throughput and multimedia quality. Our results are based on real network measurements. We describe the design and implementation of a high quality multimedia streaming application that implements WWAN streaming optimisations utilising unequal error protection coding techniques, and we evaluate the performance over an operational WWAN network.

We present SPIDER-a system for fast replication or distribution of large content from a single source to multiple sites interconnected over Internet or via a private network. In order to exploit spatial diversity of the underlying network, SPIDER uses an overlay structure composed of dedicated transit nodes (TNs). The data transport mechanism in SPIDER leverages this overlay structure to provide a coordinated approach that minimizes the maximum time to replicate to all destination sites (the make span of content replication). In order to achieve this objective, SPIDER employs two orthogonal components: a) creation of multiple dynamic distribution trees using the transit nodes b) end-to-end reliable data transport with flow control on these trees by chaining point-to-point TCPs. We further present simulations based results to quantify benefits of tree construction algorithms in random topologies. We evaluate the real implementation of the SPIDER in Planet Lab and observe a 2-6 times speed up compared to different existing schemes.

We introduce MoB, an infrastructure for collaborative wide-area wireless data services. MoB proposes to change the current model of data services in the following fundamental ways: (1) it decouples infrastructure providers from services providers and enables fine-grained competition, (2) it allows service interactions on arbitrary timescales, and, (3) it promotes flexible composition of these fine-grained service interactions based on user and application needs.At the heart of MoB is an open market architecture in which mobile users can opportunistically trade various services with each other in a flexible manner. In this paper we first describe the overall architecture of MoB including various enablers like user reputation management, incentive management, and accounting services. We next present our experience from both simulations as well as our prototype implementation of MoB in enhancing application performance in multiple different scenarios --- file transfers, web browsing, media streaming, and location-enhanced services.

We address the problem of energy-efficient reliable wireless communication in the presence of unreliable or lossy wireless link layers in multi-hop wireless networks. Prior work [1] has provided an optimal energy efficient solution to this problem for the case where link layers implement perfect reliability. However, a more common scenario --- a link layer that is not perfectly reliable, was left as an open problem. In this paper we first present two centralized algorithms, BAMER and GAMER, that optimally solve the minimum energy reliable communication problem in presence of unreliable links. Subsequently we present a distributed algorithm, DAMER, that approximates the performance of the centralized algorithm and leads to significant performance improvement over existing single-path or multi-path based techniques.

Tiny, low-cost sensor devices are expected to be failure-prone and hence in many realistic deployment scenarios for sensor networks these nodes are deployed in higher than necessary densities to meet operational goals. In this paper we address the question of how nodes should be managed in such dense sensor deployments so that the network topology formed by the active sensors is able to provide connected-coverage to the entire area of interest and at the same time increase the lifetime of the network. In particular, we propose and study distributed, low-coordination node wakeup schemes to efficiently construct multiple independent (node-disjoint) sensor network topologies to achieve good fault tolerance. We propose and evaluate different distributed, random and pattern-based wakeup policies for sensor nodes to construct connected-covered topologies. Through analysis and simulations we demonstrate that in dense sensor deployment scenarios, these policies can construct near-optimal topologies (within 2.7% of the optimal) with zero coordination between nodes, as long as location information is available at the individual sensor nodes.Based on these observations, we develop and evaluate a few simple distributed, wakeup based topology construction algorithms that can realize similar performance bounds in realistic sensor deployments, with varying node densities. These algorithms differ in terms of the required level of coordination and the use of sensor location information, and generate connected-covered topologies efficiently, with very low message-exchange overhead.

We propose a new link metric called normalized advance (NADV) for geographic routing in multihop wireless networks. NADV selects neighbors with the optimal trade-off between proximity and link cost. Coupled with the local next hop decision in geographic routing, NADV enables an adaptive and efficient cost-aware routing strategy. Depending on the objective or message priority, applications can use the NADV framework to minimize various types of link cost.We present efficient methods for link cost estimation and perform detailed simulations in diverse scenarios. Our results show that NADV outperforms current schemes in many aspects: for example, in high noise environments with frequent packet losses, the use of NADV leads to 81% higher delivery ratio. When compared to centralized routing under certain settings, geographic routing using NADV finds paths whose cost is close to the optimum.

Success of peer-to-peer applications in many cases is attributed to user altruism, where a user contributes some of its own resources to facilitate performance of other users. This observation has been corroborated with some experimental evidence. In this paper we make a first attempt to demonstrate that there are many scenarios where peer-to-peer resource sharing is a natural behavior that selfish users can use to improve their own performance. In particular we examine such natural incentives that exist in a streaming media application which lead such greedy users to cooperate and share resources with each other in forming an efficient overlay multicast tree. We define a freestyle Bazaar environment in which streaming media receivers interact with each other and cooperatively construct an overlay tree for improving their perception of media streams from a single server. Through simulations we demonstrate the efficacy of our proposed environment.

Dynamic Host Configuration Protocol (DHCP) was defined to facilitate automatic configuration of IP addresses and other network parameters to hosts in a network. Efficiency of DHCP's address management is especially important today in part due to proliferation of mobile devices with transient network access patterns and the consequent increased demand on transient IP addresses in open-access networks. Unfortunately, DHCP's flexible design makes it susceptible to a variety of misconfigurations. The focus of this work is, therefore, to evaluate the performance and vulnerabilities of DHCP in operational networks today. To this end, we developed a tool called DHCP-Watch that facilitates DHCP-related network debugging and enables better capacity planning. We used this tool to perform a first-of-its-kind detailed measurement study of DHCP performance in operational university campus networks. Our measurements revealed various trends of IP address usage. Additionally, we discovered frequent anomalous operations due to network misconfigurations and presence of misbehaving hosts.

We propose a multi-hop wireless LAN architecture and demonstrate its benefits to wireless clients. For this architecture, we define implementation paths that allow interoperation with existing wireless LANs which can lead to an incremental deployment of this system. We quantify the performance benefits of the proposed schemes through measurements in realistic wireless LAN environments. We also examine the performance of such multi-hop wireless LANs through detailed simulation studies. Our results show that these multi-hop extensions can significantly improve the wireless access experience (in terms of data throughput, latency, etc.) for clients who enable such mechanisms. More interestingly, when multi-hop extensions are enabled by some of the clients, it also positively impacts the performance at other clients that are completely unaware of these extensions.

We present a comparative performance study of a wide selection of optimization techniques to enhance application performance in the context of wide-area wireless networks (WWANs). Unlike in traditional wired and wireless IP-based networks, applications running over WWAN cellular environments are significantly affected by the vagaries of the cellular wireless medium. Prior research has proposed and analyzed optimizations at individual layers of the protocol stack. In contrast, we introduce the first detailed experiment-based evaluation and comparison of all such optimization techniques in a commercial WWAN testbed. This paper, therefore, summarizes our experience in implementing and deploying an infrastructure to improve WWAN performance.The goals of this paper are: (1) to perform an accurate benchmark of application performance over such commercially deployed WWAN environments, (2) to implement and characterize the impact of various optimization techniques across different layers of the protocol stack, and (3) to quantify their interdependencies in realistic scenarios. Additionally, we also discuss measurement pitfalls that we experienced and provide guidelines that may be useful for future experimentation in WWAN environments.

We introduce MAR, a commuter mobile access router infrastructure that exploits wireless diversity (e.g. channel diversity, network diversity, and technology diversity) to provide improved data performance for wireless data users. Our system design stems from the observation that rather than choosing a single wireless service provider (e.g. Sprint, AT&T, BT, Vodafone), a single technology (e.g. GPRS, UMTS, CDMA, 802.11), or a single wireless channel, users can obtain significant benefits by using the multiplicity of choices available. MAR is a wireless multi-homed device that can be placed in moving vehicles (e.g. car, bus, train) to enable high-speed data access. MAR dynamically instantiates new channels based on traffic demand, aggregates the bandwidth and dynamically shifts load from poor quality to better quality channels. MAR, thus, provides a faster, more stable, and reliable communication channel to mobile users.We have implemented and tested the MAR system in our testbed which spans the networks of three different cellular providers. Through our experiments we have performed a detailed evaluation to quantify the benefits of MAR for different protocols and applications. For example, even in highly mobile environments, MAR, on average, improves the end-user experience of web-browsing and streaming applications by a factor of 2.8 and 4.4 respectively. Our results show that significant benefits can be obtained by exploiting the diversity in coverage offered by many cellular operators, different technology networks (e.g. GPRS, CDMA), and diverse wireless channels.

We present a low-overhead media streaming system, called SRMS (Scalable Resilient Media Streaming) that can be used to scalably deliver streaming data to a large group of receivers. SRMS uses overlay multicast for data distribution. to a large group of users. SRMS leverages a probabilistic loss recovery technique to provide high data delivery guarantees even under large network losses and overlay node failures. The clients in the SRMS system are able to interoperate with existing media streaming servers that use RTP for data transport. One of the interesting features of SRMS is that it can simultaneously support clients with disparate access bandwidths. It enables the necessary bandwidth adaptations using standard Real-time Transport Protocol (RTP) mechanisms, e.g. RTP translators. We have implemented and evaluated the SRMS system in detail on an emulated network as well as on a wide-area testbed with up to 128 clients. Our results show that clients using SRMS achieve high (97%) data delivery ratios with low overheads (<5%) even for a very dynamic network (up to five membership changes per minute).

This paper presents an overlay architecture where service providers deploy a set of service nodes (called MSNs) in the network to efficiently implement media-streaming applications. These MSNs are organized into an overlay and act as application-layer multicast forwarding entities for a set of clients. We present a decentralized scheme that organizes the MSNs into an appropriate overlay structure that is particularly beneficial for real-time applications. We formulate our optimization criterion as a "degree-constrained minimum average-latency problem" which is known to be NP-hard. A key feature of this formulation is that it gives a dynamic priority to different MSNs based on the size of its service set. Our proposed approach iteratively modifies the overlay tree using localized transformations to adapt with changing distribution of MSNs, clients, as well as network conditions. We show that a centralized greedy approach to this problem does not perform quite as well, while our distributed iterative scheme efficiently converges to near-optimal solutions.

We introduce PRM (Probabilistic Resilient Multicast): a multicast data recovery scheme that improves data delivery ratios while maintaining low end-to-end latencies. PRM has both a proactive and a reactive component; in this paper we describe how PRM can be used to improve the performance of application-layer multicast protocols, especially when there are high packet losses and host failures. Further, using analytic techniques, we show that PRM can guarantee arbitrarily high data delivery ratios and low latency bounds. As a detailed case study, we show how PRM can be applied to the NICE application-layer multicast protocol. We present detailed simulations of the PRM-enhanced NICE protocol for 10,000 node Internet-like topologies. Simulations show that PRM achieves a high delivery ratio (> 97%) with a low latency bound (600 ms) for environments with high end-to-end network losses (1-5%) and high topology change rates (5 changes per second) while incurring very low overheads (< 5%).

We define energy-efficient broadband and multicast schemes for reliable communication in multi-hop wireless networks. Unlike previous techniques, the choice of neighbors in the broadband and multicast trees in these schemes, are based not only on the link distance, but also on the error rates associated with the link. Our schemes can be implemented using both positive and negative acknowledgement based reliable broadcast techniques in the link layer. Through simulations, we show that our scheme achieves up to 45% improvement over previous schemes on realistic 100-node network topologies. A positive acknowledgment based implementation is preferred. Our simulations show that the additional benefits of a positive acknowledgement based implementation is marginal (1-2%). Therefore a negative acknowledgement based implementation of our schemes is equally applicable in constructing energy-efficient reliable and multicast data delivery paths.

We consider the problem of finding nearby application peers over the Internet. We define a new peer-finding scheme (called Tiers) that scales to large application peer groups. Tiers creates a hierarchy of the peers, which allows an efficient and scalable solution to this problem. The scheme can be implemented entirely in the application-layer and does not require the deployment of either any additional measurement services, or well-known reference landmarks in the network. We present detailed evaluation of Tiers and compare it to one previously proposed scheme called Beaconing. Through analysis and detailed simulations on 10,000 node Internet-like topologies we show that Tiers achieves comparable or better performance with a significant reduction in control overheads for groups of size 32 or more.

Current algorithms for minimum-energy routing in wireless networks typically select minimum-cost multi-hop paths. In scenarios where the transmission power is fixed, each link has the same cost and the minimum-hop path is selected. In situations where the transmission power can be varied with the distance of the link, the link cost is higher for longer hops; the energy-aware routing algorithms select a path with a large number of small-distance hops. In this paper, we argue that such a formulation based solely on the energy spent in a single transmission is misleading --- the proper metric should include the total energy (including that expended for any retransmissions necessary) spent in reliably delivering the packet to its final destination.We first study how link error rates affect this retransmission-aware metric, and how it leads to an efficient choice between a path with a large number of short-distance hops and another with a smaller number of large-distance hops. Such studies motivate the definition of a link cost that is a function of both the energy required for a single transmission attempt across the link and the link error rate. This cost function captures the cumulative energy expended in reliable data transfer, for both reliable and unreliable link layers. Finally, through detailed simulations, we show that our schemes can lead to upto 30-70% energy savings over best known current schemes, under realistic environments.

We describe a new scalable application-layer multicast protocol, specifically designed for low-bandwidth, data streaming applications with large receiver sets. Our scheme is based upon a hierarchical clustering of the application-layer multicast peers and can support a number of different data delivery trees with desirable properties.We present extensive simulations of both our protocol and the Narada application-layer multicast protocol over Internet-like topologies. Our results show that for groups of size 32 or more, our protocol has lower link stress (by about 25%), improved or similar end-to-end latencies and similar failure recovery properties. More importantly, it is able to achieve these results by using orders of magnitude lower control traffic.Finally, we present results from our wide-area testbed in which we experimented with 32-100 member groups distributed over 8 different sites. In our experiments, average group members established and maintained low-latency paths and incurred a maximum packet loss rate of less than 1% as members randomly joined and left the multicast group. The average control overhead during our experiments was less than 1 Kbps for groups of size 100.

We propose MRPC, a new power-aware routing algorithm for energy-efficient routing that increases the operational lifetime of multi-hop wireless networks. In contrast to conventional power-aware algorithms, MRPC identifies the capacity of a node not just by its residual battery energy, but also by the expected energy spent in reliably forwarding a packet over a specific link. Such a formulation better captures scenarios where link transmission costs also depend on physical distances between nodes and the link error rates. Using a max-min formulation, MRPC selects the path that has the largest packet capacity at the 'critical' node (the one with the smallest residual packet transmission capacity). We also present CMRPC, a conditional variant of MRPC that switches from minimum energy routing to MRPC only when the packet forwarding capacity of nodes falls below a threshold. Simulation based studies have been used to quantify the performance gains of our algorithms.

We introduce and analyze a scalable rekeying scheme for implementing secure group communications Internet protocol multicast. We show that our scheme incurs constant processing, message, and storage overhead for a rekey operation when a single member joins or leaves the group, and logarithmic overhead for bulk simultaneous changes to the group membership. These bounds hold even when group dynamics are not known a priori. Our rekeying algorithm requires a particular clustering of the members of the secure multicast group. We describe a protocol to achieve such clustering and show that it is feasible to efficiently cluster members over realistic Internet-like topologies. We evaluate the overhead of our own rekeying scheme and also of previously published schemes via simulation over an Internet topology map containing over 280 000 routers. Through analysis and detailed simulations, we show that this rekeying scheme performs better than previous schemes for a single change to group membership. Further, for bulk group changes, our algorithm outperforms all previously known schemes by several orders of magnitude in terms of actual bandwidth usage, processing costs, and storage requirements.

We introduce and analyze a scalable re-keying scheme for implementing secure group communications over IP multicast. We show that our scheme incurs constant processing, message, and storage overhead for a re-key operation when a single member joins or leaves the group, and logarithmic overhead for bulk simultaneous changes to the group membership. These bounds hold even when group dynamics are not known a priori. Our re-keying algorithm requires a particular clustering of the members of the secure multicast group. We describe a protocol to achieve such clustering and show that it is feasible to efficiently cluster members over realistic Internet-like topologies. We evaluate the overhead of our own re-keying scheme and also of previously published schemes via simulation over an Internet topology map containing over 280,000 routers. Through analysis and detailed simulations, we show that this re-keying scheme performs better than previous schemes for a single change to group membership. Further, for bulk changes, our algorithm outperforms all previously known schemes by several orders of magnitude in terms of actual bandwidth usage, processing costs and storage requirements.

In this paper we present a clustering scheme to create a hierarchical control structure for multi-hop wireless networks. A cluster is defined as a subset of vertices, whose induced graph is connected. In addition, a cluster is required to obey certain constraints that are useful for management and scalability of the hierarchy. All these constraints cannot be met simultaneously for general graphs, but we show how such a clustering can be obtained for wireless network topologies. Finally, we present an efficient distributed implementation of our clustering algorithm for a set of wireless nodes to create the set of desired clusters.