Distributed AI training often suffers from network faults. Network faults, especially at the last hop between a switch and a host, result in loss of connectivity, resulting in training job stalls and eventual failure. This is typically managed through a fail-stop mechanism, followed by a restart, incurring significant inefficiencies. We present ReCCL, the first network fault-tolerant collective communication library (CCL) that allows training progress to be preserved by seamlessly failing over to alternate paths when a network fault occurs. During failover, ReCCL keeps communication states synchronized while using dynamic channel load balancing and intra-host GPU routing to improve communication performance. Our evaluations demonstrate that ReCCL can perform failover seamlessly with minimal performance losses. Additionally, our simulations also demonstrate that failover can be effectively used to achieve significant savings in GPU hours for large-scale distributed AI training workloads.

Energy Harvesting (EH) technology has emerged to prolong the lifetime of Internet of Things (IoT) devices. However, in EH-IoTs, the reliance on external energy sources introduces challenges in maintaining up-to-date information. To quantify data freshness in such systems, researchers have introduced the Age-of-Information (AoI) metric, which measures the time elapsed since the generation of the most up-to-date information received by the user. Consequently, the problem of AoI minimization has been studied extensively in EH-IoTs to ensure timely data delivery. While data aggregation is a fundamental task for IoTs, existing works on AoI minimization in EH-IoTs have only considered scenarios where sensory data is updated by individual source nodes. The problem has not been investigated for data aggregation, in which the sensory data is aggregated from multiple source nodes. In this paper, we study the problem of AoI minimization for Data Aggregation in EHIoTs. To address this problem, we propose an energy-adaptive node scheduling algorithm consisting of both offline scheduling and online adjustment. Extensive simulations and testbed experiments verify the high performance of our algorithm in terms of AoI minimization and energy efficiency.

The ongoing virtualization of Radio Access Networks (vRANs) promises increased velocity for updating RAN software with new features and bug fixes. To support this, we developed SwapRAN, a live update system for both vRAN software components: the Distributed Unit (DU) and the Centralized Unit (CU). Unlike previous systems, SwapRAN operates in-place without requiring additional hardware like staging servers or a programmable switch. For DU updates, SwapRAN presents two techniques: (1) using OS thread priorities to safely initialize the new DU while overlapping with the old DU which is active, and (2) using the network interface card's embedded switch to redirect fronthaul traffic to the new DU. For CU updates, SwapRAN is the first working live update system, which we achieve by (1) decoupling the stateful CU-DU connection and transparently rerouting DU messages to the new CU, and (2) repurposing existing midhaul control plane messages to move users to the new CU. We evaluate SwapRAN on real 5G testbeds and demonstrate its practical deployability via integration with Kubernetes. Our evaluations show that SwapRAN completes DU or CU updates with just 1–2 seconds of user downtime.

We present SwapRAN, a live update system that brings Continuous Integration/Continuous Deployment (CI/CD) to virtualized RANs (vRANs), which includes both the Distributed Unit (DU) and the Centralized Unit (CU). In contrast to prior solutions, SwapRAN performs in-place software updates without relying on additional infrastructure such as staging servers or programmable switches. For DU updates, SwapRAN introduces two techniques: (1) leveraging OS thread priorities to safely bring up the new DU while the old DU remains active, and (2) redirecting fronthaul traffic to the new DU using the embedded switch found on modern network interface cards. For CU updates, SwapRAN is the first system to enable live updates, made possible by (1) decoupling the stateful connection between the CU and DU and transparently rerouting DU messages to the new CU, and (2) repurposing existing midhaul control plane messages to transfer users to the new CU. We demonstrate SwapRAN on our O-RAN testbed equipped with a commercial O-RU, showing that it can perform DU or CU updates with significantly reduced downtime, as low as 1–2 seconds, compared to existing update strategies in Kubernetes.

Photo sharing is increasingly popular, driven by social media platforms like Instagram and services such as Flickr and Google Photos. However, this growth has been accompanied by significant issues, particularly image theft. To address this issue, we introduce CAMPrints, a robust system for detecting image theft. CAMPrints verifies whether edited images found online contain camera fingerprints matching those of user-provided reference images. The system overcomes the challenges of identifying images altered by diverse image processing operations. We select a small yet representative set of operations by categorizing them based on their impact on pixel values and locations. A deep-learning model is trained to recognize and compare camera noise patterns pre- and post-editing. We conduct real-world evaluations involving 36 cameras across eight make-and-model combinations, along with over 40 image processing operations applied to more than 4,000 images. CAMPrints achieves an average AUC of 0.92, significantly outperforming the state-of-the-art methods by up to 1.8 times.

Drone delivery is envisioned to be the delivery mode of the future due to its capability to provide autonomous, end-to-end delivery. Such rapid growth of the drone market necessitates careful checks on drone flight delivery, as a failure in any of a drone’s parts can result in an overestimation of the drone’s battery life, an unexpected increase in delivery time, or even a drone crash. Prior works utilize onboard sensors to detect potential drone failures during flight, which is a reactive approach where the problem may have already occurred. In this work, we propose PADrone , a pre-flight and an automated drone abnormality detection system that leverages contactless radio frequency– (RF) based vibration sensing. PADrone utilizes an end-to-end deep learning pipeline to differentiate various abnormalities in motors, propellers, and other drone’s parts, by leveraging their unique vibration fingerprints . PADrone uses a frequency-modulated continuous wave radar-based RF system to capture these unique drone vibrations using an RF bandwidth of 150 MHz in the industrial, scientific, and medical band (5.8 GHz). Our real-world evaluations show that PADrone can classify various drone abnormalities with an average accuracy of 97.5%.

Multi-access networks are increasingly important for reliable end-to-end connectivity and enhanced throughput performance. A scalable multi-access solution is required to roll out multi-access networks at scale. However, existing CPU-based solutions can no longer scale sustainably, as network traffic has outgrown the CPU performance growth. Consequently, hardware accelerators offer a compelling alternative. This paper introduces JUNCTION, a scalable multi-access solution designed using programmable switches. JUNCTION features a multipath protocol tailored to the hardware constraints and optimized for efficient memory utilization, enabling it to handle a large number of multipath sessions. We validate JUNCTION on a 5G-WiFi multi-access testbed. Our analysis demonstrates that it can scale an order of magnitude better than existing solutions.

LoRa networks have emerged as a promising solution for long-range, low-power communication in IoT applications. However, a significant challenge in LoRa networks is the uneven battery depletion among nodes, due to the varying transmission configurations required to support different distances between the nodes and the gateway. This disparity in battery consumption poses a substantial challenge to the overall network lifetime. This paper introduces EGAL, a relay-based approach designed to address the battery imbalance issue by ensuring balanced energy consumption across all nodes. Unlike purely multi-hop networks, EGAL follows a hybrid approach that selectively uses relays when necessary while maintaining direct connections to the gateway. EGAL employs a reward-based algorithm that dynamically adjusts relay duties and integrates predictive analytics, thereby enhancing network lifespan with minimal overhead. The proposed solution is validated through NS3 simulation and real-world testbed, demonstrating significant improvements in network lifetime. Specifically, in simulations, EGAL exhibits up to a 457% increase in network lifetime over standard LoRaWAN, and in hardware tests, it shows up to a 70% decrease in total energy consumption. The key contributions of this work include the development of the EGAL algorithm, its integration with standard LoRaWan,and extensive validation of its effectiveness.

The proliferation of Internet of Things (IoT) applications relying on Low Power Wide Area Networks (LPWANs) demands robust and energy-efficient communication solutions. Among various LP-WAN technologies, LoRa emerges as a prominent choice due to its long-range capabilities and low energy consumption. However, the practical deployment of LoRa is hindered by significant signal degradation caused by channel and hardware noise, especially in urban environments. We introduce GLoRiPHY, a novel generative framework designed to enhance the reception quality of LoRaPHY signals through a channel-aware denoising mechanism. Utilizing a transformer-based architecture, GLoRiPHY leverages the known preamble of LoRaPHY signals to compensate for channel-induced distortions, thereby generating a clean signal suitable for direct demodulation. The system integrates Convolutional Neural Networks (CNNs) for efficient feature encoding and decoding, maintaining a compact model footprint even at higher Spreading Factors (SFs). Evaluations on real-world and simulated datasets show that in comparison to the current state-of-the-art solution, GLoRiPHY significantly lowers the Symbol Error Rate (SER) by up to 2.85x and demonstrates generalizability in unseen environments, while reducing inference times by up to 5.75x.

Automotive Keyless Entry (RKE) systems provide car owners with a degree of convenience, allowing them to lock and unlock their car without using a mechanical key. Today’s RKE systems implement disposable rolling codes, making every key fob button press unique, effectively preventing simple replay attacks. However, a prior attack called RollJam was proven to break all rolling code–based systems in general. By a careful sequence of signal jamming, capturing, and replaying, an attacker can become aware of the subsequent valid unlock signal that has not been used yet. RollJam, however, requires continuous deployment indefinitely until it is exploited. Otherwise, the captured signals become invalid if the key fob is used again without RollJam in place. We introduce RollBack, a new replay-and-resynchronize attack against most of today’s RKE systems. In particular, we show that even though the one-time code becomes invalid in rolling code systems, replaying a few previously captured signals consecutively can trigger a rollback-like mechanism in the RKE system. Put differently, the rolling codes become resynchronized back to a previous code used in the past from where all subsequent yet already used signals work again. Moreover, the victim can still use the key fob without noticing any difference before and after the attack. Unlike RollJam, RollBack does not necessitate jamming at all. In fact, it requires signal capturing only once and can be exploited at any time in the future as many times as desired. This time-agnostic property is particularly attractive to attackers, especially in car-sharing/renting scenarios in which accessing the key fob is straightforward. However, while RollJam defeats virtually any rolling code–based system, vehicles might have additional anti-theft measures against malfunctioning key fobs, hence against RollBack. Our ongoing analysis (with crowd-sourced data) against different vehicle makes and models has revealed that ∼ 50% of the examined vehicles in the Asian region are vulnerable to RollBack, whereas the impact tends to be smaller in other regions, such as Europe and North America.

The demand for masks and air filters with effective filtration capabilities is skyrocketing as there are many applications that require protecting users from inhaling air pollutants or hazardous particles. Unfortunately, we are witnessing a surge in the number of counterfeit and substandard filters attributed to malicious and inept manufacturers. Hence, users are left vulnerable in not knowing which products are reliable. Exacerbating the problem, there are diverse filter standards, each with a unique expression for filtration efficiencies, adding to user confusion. Moreover, the average user lacks the necessary tools, techniques, and knowledge to independently verify the filtration efficiency. Specifically, state-of-the-art solutions are lab-based machines that are extremely expensive and difficult to access for the general public. To solve this problem, we propose FilterOp, a novel smartphone-based mask and filter testing system. FilterOp is a practical solution that allows a user to estimate the filtration efficiency of a mask or a filter using only a pair of commodity smartphones. The novelty of FilterOp comes from its use of light absorption and scattering effects, observed when light propagates through the filter. We evaluate FilterOp in comprehensive real-world experiments using 256 filter instances across 27 different make-and-model products with varying filtration efficiencies. Comparing our results to those obtained with a state-of-the-art government-certified testing machine, we observe that FilterOp yields comparable results with a low mean absolute error of 2.7%, and detects substandard products with an overall accuracy of 96%.

Packet loss due to link corruption is a major problem in large warehouse-scale datacenters. The current state-of-the-art approach of disabling corrupting links is not adequate because, in practice, all the corrupting links cannot be disabled due to capacity constraints. In this paper, we show that, it is feasible to implement link-local retransmission at sub-RTT timescales to completely mask corruption packet losses from the transport endpoints. Our system, LinkGuardian, employs a range of techniques to (i) keep the packet buffer requirement low, (ii) recover from tail packet losses without employing timeouts, and (iii) preserve packet ordering. We implement LinkGuardian on the Intel Tofino switch and show that for a 100G link with a loss rate of 10−3, LinkGuardian can reduce the loss rate by up to 6 orders of magnitude while incurring only 8% reduction in effective link speed. By eliminating tail packet losses, LinkGuardian improves the 99.9th percentile flow completion time (FCT) for TCP and RDMA by 51x and 66x respectively. Finally, we also show that in the context of datacenter networks, simple out-of-order retransmission is often sufficient to significantly mitigate the impact of corruption packet loss for short TCP flows.

5G networks are envisioned to support various emerging use cases, such as telemedicine, remote construction, autonomous driving, industrial automation, drone control, and immersive entertainment. These applications demand low latency, high reliability, and in some cases require ultra-high-bandwidths. Specifically, these applications require 5G networks to provide 1ms end-to-end latency with 99.99% reliability [12] for ultra-reliable low-latency communications (URLLC). Various studies [11, 19, 26] have shown that the radio access network (RAN) remains the bottleneck in realizing low-latency communications.

Remote Direct Memory Access (RDMA) is widely used in high-performance computing (HPC) and data center networks. In this paper, we first show that RDMA does not work well with existing load balancing algorithms because of its traffic flow characteristics and assumption of in-order packet delivery. We then propose ConWeave, a load balancing framework designed for RDMA. The key idea of ConWeave is that with the right design, it is possible to perform fine granularity rerouting and mask the effect of out-of-order packet arrivals transparently in the network datapath using a programmable switch. We have implemented ConWeave on a Tofino2 switch. Evaluations show that ConWeave can achieve up to 42.3% and 66.8% improvement for average and 99-percentile FCT, respectively compared to the state-of-the-art load balancing algorithms.

Application offloads on modern high-speed programmable switches have been proposed in a variety of systems (e.g., key–value store systems and network middleboxes) so as to efficiently scale up the traditional server-oriented deployments. However, they largely achieve sub-optimal offloading efficiency due to the lack of (1) capability to perform control actions at sufficient rates, and (2) adaptability to workload changes. In this paper, we scrutinize the common stumbling blocks of existing frameworks with performance evaluations on real workloads. We present DySO (Dynamic State Offloading), a framework which enables expeditious on-demand control actions and self-tuning of management rules. DySO’s key insight is to perform control actions via a data-path instead of the switch control channel which is the bottleneck to read/write states into data plane. Our software simulations show up to 100% performance improvement compared to existing systems for various real world traces. On top of that, we implement and evaluate DySO on a commodity programmable switch, showing two orders of magnitude faster responsiveness to sudden workload changes compared to the existing systems.

Recently we have witnessed the rise of Artificial Intelligence of Things (AIoT) and the shift of sensing paradigm from cloud-centric to the edge-centric, which effectively improves the sensing capability of intelligence transportation systems. To improve the real-time sensing performance, in this work we propose an ensemble sensing based scheme to solve the time-constraint synchronized inference problem and achieve robust inference with heterogeneous IoT devices in intelligence transportation systems. We design and implement Ensen, which incorporates various novel techniques such as customized DNN model design, KD-based model training, and dynamic deep ensemble management, etc., to achieve improved accuracy and maximize the computational resource usage of the whole sensing group. Extensive evaluations on different types of common IoT devices have shown that Ensen achieves a robust performance and can be easily extended to different types of convolutional neural networks.

Packet corruption loss is a serious problem in datacenter networks. A large-scale study by Microsoft reported that the number of packets lost due to corruption is comparable to those lost due to congestion. Previous attempts to mitigate the impact of packet corruption loss seek to avoid the faulty links by routing around them, at the cost of reduced link capacities and disruption to the rest of the network.

As the use of machine learning (ML) models is becoming increasingly popular in many real-world applications, there are practical challenges that need to be addressed for model maintenance. One such challenge is to "undo" the effect of a specific subset of dataset used for training a model. This specific subset may contain malicious or adversarial data injected by an attacker, which affects the model performance. Another reason may be the need for a service provider to remove data pertaining to a specific user to respect the user's privacy. In both cases, the problem is to "unlearn" a specific subset of the training data from a trained model without incurring the costly procedure of retraining the whole model from scratch. Towards this goal, this paper presents a Markov chain Monte Carlo-based machine unlearning (MCU) algorithm. MCU helps to effectively and efficiently unlearn a trained model from subsets of training dataset. Furthermore, we show that with MCU, we are able to explain the effect of a subset of a training dataset on the model prediction. Thus, MCU is useful for examining subsets of data to identify the adversarial data to be removed. Similarly, MCU can be used to erase the lineage of a user's personal data from trained ML models, thus upholding a user's "right to be forgotten". We empirically evaluate the performance of our proposed MCU algorithm on real-world phishing and diabetes datasets. Results show that MCU can achieve a desirable performance by efficiently removing the effect of a subset of training dataset and outperform an existing algorithm that utilizes the remaining dataset.

We are witnessing a heightened surge in remote privacy attacks on laptop computers. These attacks often exploit malware to remotely gain access to webcams and microphones in order to spy on the victim users. While webcam attacks are somewhat defended with widely available commercial webcam privacy covers, unfortunately, there are no adequate solutions to thwart the attacks on mics despite recent industry efforts. As a first step towards defending against such attacks on laptop mics, we propose TickTock, a novel mic on/off status detection system. To achieve this, TickTock externally probes the electromagnetic (EM) emanations that stem from the connectors and cables of the laptop circuitry carrying mic clock signals. This is possible because the mic clock signals are only input during the mic recording state, causing resulting emanations. We design and implement a proof-of-concept system to demonstrate TickTock's feasibility. Furthermore, we comprehensively evaluate TickTock on a total of 30 popular laptops executing a variety of applications to successfully detect mic status in 27 laptops. Of these, TickTock consistently identifies mic recording with high true positive and negative rates.

Continuous monitoring is a major component of many applications in wireless sensor network (WSN). In these applications, to reduce the communication overhead, some form of data summary or aggregation can be performed. However, performing non-trivial in-network data processing such as finding frequent items, Top-K monitoring, and clustering efficiently are challenging in practice.In this paper, we present Low-Power Distinct Sum (LDS), a distributed in-network data aggregation primitive that performs the sum of unique items in WSN. LDS serves as the underlying primitive that can be used to implement many distributed data processing efficiently. To demonstrate LDS’s capabilities, we design and implement a distributed data streaming application with LDS running on Contiki OS. Compared to the baseline algorithm, LDS can reduce the completion time by up to 66%.

Networks regularly face various threats and attacks that manifest in their communication traffic. Recent works proposed unsupervised approaches, e.g., using a variational autoencoder, that are not only effective in detecting anomalies in network traffic, but also practical as they do not require ground truth or labeled data. However, the problem of characterizing anomalies into different attack behaviors is still less explored; in this work, we study this specific problem. We develop APEX, a framework that employs data mining approaches in a semisupervised way to extract the attack patterns from anomalous traffic and links them to specific attack types. APEX comprises two levels of mining: the first level extracts patterns in anomalous network flows, and the second level characterizes behaviors in the extracted patterns into different attack classes. We carry out extensive experiments on real network traces obtained from the MAWI traffic archive. The evaluations demonstrate that APEX is effective in extracting distinguishable behaviors of network attacks from anomalous traffic, and provide useful insights to security analysts investigating the anomalies.

We are witnessing a surge in the reported cases of counterfeit liquid products in the market including olive oil, honey, and alcohol. Counterfeiters often adulterate the liquid products by replacing a large portion of the authentic content with cheaper substitutes (e.g., mixing vodka with cheaper alcohol or potentially toxic methanol). Exacerbating the problem, the counterfeits are packaged and sealed to factory standards, rendering it extremely difficult for an average consumer to identify them. While solutions exist, they are often impractical for the general public as they require specialized and costly equipment. To overcome these limitations, we propose LiquidHash, a novel counterfeit liquid food product detection system. LiquidHash is a practical solution that only requires the use of a commodity smartphone to detect adulterated liquid products without opening the bottles. LiquidHash works by detecting and tracking the shape and movement of air bubbles that form inside the bottles. We implement LiquidHash and evaluate its feasibility with real-world experiments under varying conditions with a total of more than 500 minutes of video recording and observe an overall detection accuracy of up to 95%.

Counterfeit liquid food products, including olive oil, honey and alcohol, are continuing to pose severe threats to the general public as counterfeiters adulterate the authentic content with cheaper and potentially harmful substitutes, and package them in authentic bottles. Existing solutions are often impractical for the general public as they require specialized and costly equipment as well as taking liquid samples. We overcome these limitations by proposing LiquidHash, a novel detection system that only requires the use of a commodity smartphone to detect adulterated liquid products without opening the bottles. LiquidHash leverages computer vision and machine learning techniques to extract characteristics of air bubbles formed by flipping a bottle. We implement LiquidHash and evaluate its feasibility with real-world experiments and achieve an overall detection accuracy of up to 95%.

We present the design and prototyping of a framework to support multicast for remote direct memory accesses (RDMA), specifically the one-sided WRITE operation. We use P4 programmable hardware to augment fixed-function RDMA transport hardware found on commodity NICs to enable one-sided RDMA multicast with zero-CPU overhead. Finally, we outline the potential challenges and future directions in realizing the framework for large-scale data center deployments.

5G networks are gaining pace in development and deployment in recent years. One of 5G's key objective is to support a variety of use cases with different Service Level Objectives (SLOs). Slicing is a key part of 5G that allows operators to provide a tailored set of resources to different use cases in order to meet their SLOs. Existing works focus on slicing in the frontend or the C-RAN. However, slicing is missing in the fronthaul network that connects the frontend to the C-RAN. This leads to over-provisioning in the fronthaul and the C-RAN, and also limits the scalability of the network.

Slicing is a critical technology in 5G, as it allows operators to slice a physical network into multiple virtual networks, each dedicated to a different use case/Mobile Virtual Network Operator (MVNO) [2]. Network slicing enables network operators to deploy a tailored set of resources for specific use cases or MVNO. For example, high performance reliable hardware is required only for ultra-reliable low-latency (uRLLC) use cases such as autonomous vehicle networks. Such tailoring of services reduces costs for network operators. Further, 5G systems can now be deployed more quickly due to virtualization provided by slicing, thereby enabling faster time-to-market. To this end, there exists a large body of work that introduces slicing in different parts of the cellular network (see Fig. 1). PRAN [12] and FlexRAN [13] provide slicing in the Radio Access Network (RAN) while Orion [14] provides slicing for the frontend (wireless spectrum). The fronthaul connects the frontend base station to the RAN and carries digitized radio signals between the two parts of the cellular network. However, to the best of our knowledge, there exists no work on slicing in the fronthaul. This severely limits the benefits of slicing in the RAN and the frontend (see §1.1).

Sketches have successfully provided accurate and fine-grained measurements (e.g., flow size and heavy hitters) which are imperative for network management. In particular, Count-Min (CM) sketch is widely utilized in many applications due to its simple design and ease of implementation. There have been many efforts to build monitoring frameworks based on Count-Min sketch. However, these frameworks either support very specific measurement tasks or they cannot be implemented on high-speed programmable hardware (PISA).

Collaborative machine learning (ML) is an appealing paradigm to build high-quality ML models by training on the aggregated data from many parties. However, these parties are only willing to share their data when given enough incentives, such as a guaranteed fair reward based on their contributions. This motivates the need for measuring a party’s contribution and designing an incentive-aware reward scheme accordingly. This paper proposes to value a party’s reward based on Shapley value and information gain on model parameters given its data. Subsequently, we give each party a model as a reward. To formally incentivize the collaboration, we define some desirable properties (e.g., fairness and stability) which are inspired by cooperative game theory but adapted for our model reward that is uniquely freely replicable. Then, we propose a novel model reward scheme to satisfy fairness and trade off between the desirable properties via an adjustable parameter. The value of each party’s model reward determined by our scheme is attained by injecting Gaussian noise to the aggregated training data with an optimized noise variance. We empirically demonstrate interesting properties of our scheme and evaluate its performance using synthetic and real-world datasets.

Interest in fine-grained indoor localization remains high and various approaches including those based on Radio Frequency (RF), ultrasound, acoustic, magnetic field and light have been proposed. However, while the achieved accuracy may be high, many of these approaches do not work well in environments with lots of obstructions. In this paper, we present MagB, a decimeter-level localization scheme that uses the magnetometer available on many IoT devices. MagB estimates the bearing of magnetic beacons by detecting changes in the magnetic field strength. Localization is then performed based on Angle-of-Arrival (AoA) information. We have built a prototype of MagB using low cost, off-the-shelf components. Our evaluation shows that MagB is able to achieve a median accuracy of about 13cm and can localize devices even when they are placed in steel filing cabinet or inside the casing of a running PC.

Intra-group spreading is one of the common and frequent needs in many decentralized communication protocols. Such spreading is useful for quick and local dissemination of information among different clusters in large-scale decentralized systems like the Internet-of-Things (IoT). Complex decentralized protocols can carefully exploit such localized dissemination as a base unit for their efficient implementation. However, due to the inherent broadcast nature of wireless communication, efficient and simultaneous execution of multiple intra-group disseminations is difficult. In this work, we propose a novel and simple way to completely hide a wireless transmission without changing any channel or frequency. Next, we use this for supporting simultaneous intra-group disseminations. Rigorous evaluation of the proposed strategy over testbeds show significant improvement of upto 60% in reliability with similar average latency and radio-on time in comparison to the baseline where no additional mechanism is adopted for separation of intra-group communications.

In datacenter networks, flows need to complete as quickly as possible because the flow completion time (FCT) directly impacts user experience, and thus revenue. Link failures can have a significant impact on short latency-sensitive flows because they increase their FCTs by several fold. Existing link failure management techniques cannot keep the FCTs low under link failures because they cannot completely eliminate packet loss during such failures. We observe that to completely mask the effect of packet loss and the resulting long recovery latency, the network has to be responsible for packet loss recovery instead of relying on end-to-end recovery. To this end, we propose Shared Queue Ring (SQR), an on-switch mechanism that completely eliminates packet loss during link failures by diverting the affected flows seamlessly to alternative paths. We implemented SQR on a Barefoot Tofino switch using the P4 programming language. Our evaluation on a hardware testbed shows that SQR can completely mask link failures and reduce tail FCT by up to 4 orders of magnitude for latency-sensitive workloads.

Wireless sensor networks have emerged as an important information collection and monitoring tool. The data collected is typically uploaded to a central gateway for processing and analysis. However, the approach of forwarding all the sensed data to a sink for processing is not always practical due to the high communication cost. In this paper, we present InDP, a framework that is designed to support data dissemination and processing in the edge. InDP has a communication component and a computation component. The communication component supports a low duty cycle mode for an infrequent status update and a high throughput mode to support distributed computation. The computation component implements a distributed version of Principal Component Analysis (PCA). As an application, we have implemented an outlier detection component over InDP. InDP and the outlier detection application have been implemented on Contiki using a modified version of Codecast as the underlying many-to-many communication protocol. Our evaluations show that InDP can terminate as fast as 100ms and 1.3s on the average running on a testbed with more than 70 nodes. In terms of PCA computation, InDP's computation uses 91.6% less data than a centralized approach and is able to detect anomalies using a fraction of the sensed data.

Contemporary many-core architectures, such as Adapteva Epiphany and Sunway TaihuLight, employ per-core software-controlled Scratchpad Memory (SPM) rather than caches for better performance-per-watt and predictability. In these architectures, a core is allowed to access its own SPM as well as remote SPMs through the Network-On-Chip (NoC). However, the compiler/programmer is required to explicitly manage the movement of data between SPMs and off-chip memory. Utilizing SPMs for multi-threaded applications is even more challenging, as the shared variables across the threads need to be placed appropriately. Accessing variables from remote SPMs with higher access latency further complicates this problem as certain links in the NoC may be heavily contended by multiple threads. Therefore, certain variables may need to be replicated in multiple SPMs to reduce the contention delay and/or the overall access time. We present Coordinated Data Management (CDM), a compile-time framework that automatically identifies shared/private variables and places them with replication (if necessary) to suitable on-chip or off-chip memory, taking NoC contention into consideration. We develop both an exact Integer Linear Programming (ILP) formulation as well as an iterative, scalable algorithm for placing the data variables in multi-threaded applications on many-core SPMs. Experimental evaluation on the Parallella hardware platform confirms that our allocation strategy reduces the overall execution time and energy consumption by 1.84× and 1.83× , respectively, when compared to the existing approaches.

In order to provide greater network capacity, the use of small base stations such as Femtocells has increased to allow higher spectrum reuse. In these Femtocells, base station designers have started to explore the use of general purpose multi-core architectures to provide greater flexibility. Multi-core architectures allow power-performance trade-off possibilities through techniques such as Dynamic Voltage Frequency Scaling (DVFS) and Power Gating. In this work, we propose a power management framework based on reinforcement learning called PR3, which uses both DVFS and Power Gating. Our approach is unique as it introduces a feedback from the network scheduler and baseband processor to the Power Governor, so that information about both the network and computation workloads are included in the decision making. Evaluation on a hardware platform (Odroid XU3) running PHY LTE uplink baseband processing benchmark, shows that PR3performs well in terms of both power and latency. It is able to save upto 50% power while maintaining low processing latency. PR3is also adaptive, making it effective over a wide range of traffic loads.

This paper presents Codecast, a many-to-many communication protocol for low-power sensor networks that provide high throughput and reliable data sharing from multiple sources to multiple destinations of a network. Codecast uses physical layer capture on concurrent transmissions for high spatial reuse and a network-assisted network coding for high throughput as the core techniques. Our extensive evaluation in two large-scale testbed deployments (Indriya and Flocklab) shows that Codecast provides up to 4x the throughput of Chaos and 1.8x the throughput of LWB for many-to-many data communication. Finally, we demonstrate the utility of Codecast through a distributed channel selection mechanism and a link state based routing protocol.

Beacons, the backbone of the physical web and location-based services, are widely used to tag objects and places. However, as a beacon's wireless transmission is limited to a ranging technology, the localization information is only available when the beacon is nearby.

Many WiFi fingerprint-based indoor localization approaches have been proposed to ease deployment and minimize infrastructure requirement. While researchers have devoted extensive efforts to improving the accuracy of these approaches, the user experience of such deployments in practice is typically far below expectation. One reason that contributes to such discrepancy is that while researchers often evaluate their systems in stable and "benign" environments, the actual environments can be much more dynamic and noisy. In this paper, we address this issue in the following manner. First, we identify factors that can result in significant degradation of localization performance and explore how these factors can be modeled in the localization process. Next, we design a system, EvaLoc, that takes fingerprinting data collected as input and provides accuracy prediction on the localization performance under different conditions. Our evaluation in 15 different locations covering around 25000 m2 shows that EvaLoc is able to produce localization result that better matches the user experience.

While location is one of the most important context information in mobile and pervasive computing, large-scale deployment of indoor localization system remains elusive. In this work, we propose MPiLoc, a multi-floor indoor localization system that utilizes data contributed by smartphone users through participatory sensing for automatic floor plan and radio map construction. Our system does not require manual calibration, prior knowledge, or infrastructure support. The key novelty of MPiLoc is that it clusters and merges walking trajectories annotated with sensor and signal strengths to derive a map of walking paths annotated with radio signal strengths in multi-floor indoor environments. We evaluate MPiLoc over five different indoor areas. Evaluation shows that our system can derive indoor maps for various indoor environments in multi-floor settings and achieve an average localization error of 1.82 m.

Achieving fast and efficient many-to-many communication is one of the most complex communication problems, especially in wireless systems. A compact form of many-to-many communication in a distributed system has the potential to bring huge benefit to many distributed algorithms and protocols. Many-to-many communication can be implemented as a sequential instantiations of a network wide one-to-many communication. One limitation of such an approach is that each individual instance of a one-to-many communication has to be given enough time to propagate through the whole network before the next instance. In addition, there is large overhead in generating the schedule for the sequence of individual one-to-many communications. In this work, we show that many-to-many communication can be more efficiently implemented as many parallel many-to-one communications. In this direction, we first develop an efficient TDMA based many-to-one communication module, and then use it in many-to-many setting. Our approach achieves a minimum about 20% to 50% improvements on latency (radio-on time) over the state-of-the-art solutions in a 90-node wireless sensor network testbed.

In this paper, we address the fundamental problem of improving the performance of many-to-one and many-to-many communications. Our approach is Time Division Multiple Access (TDMA) based but addresses the limitations of existing TDMA implementations in a novel way. In a nutshell, we combine packets from many senders into a single large packet transmission by exploiting capture effect achieved through fine grained power control at the level of segments within a single packet. We applied our technique to the design of a one-hop, many-to-one communication protocol, SyncMerge, and a multi-hop, many-to-many communication protocol, ByteCast. Our evaluation shows that SyncMerge is able to achieve 2 to 15 times improvement over traditional many-to-one communication schemes. In addition, ByteCast is able to disseminate 1 byte of data from every node to all other nodes in about 600 ms with 99.5% reliability on a 90 node testbed. Compared to the state-of-the-art protocols such as LWB and Chaos, ByteCast is able to reduce the radio-on time by up to 90% while achieving similar reliability.

An effective neighbor discovery service on smartphones is required for many emerging applications—from proximity-based interactions to opportunistic phone-to-phone collaborations. For a smartphone neighbor discovery service to be usable and attractive, it needs to meet two conflicting goals: 1) phones should discover neighbors fast enough (in seconds), and 2) the service’s energy footprint should be negligible so it can be “always on” while incurring little impact on battery life. Researchers have developed an impressive collection of neighbor discovery protocols to meet these two goals. By putting these protocols into concrete smartphones settings, we identify different key factors that limit their performance. Guided by our analysis, we focus on locally synchronized protocols, where phones use time information from nearby Wi-Fi Access Points (APs) to help neighbor discovery. By overcoming the key challenges for such protocols, especially, the scalability problem under increasing number of APs and neighbors, we design a new protocol, R2, that achieves low discovery delay ($<30$seconds for at least 80 percent of all connections) with a low duty cycle (1 percent).

Passive indoor localization for smartphones requires no explicit cooperation of the smartphone and enables a new spectrum of applications such as passive user tracking, mobility monitoring, social pattern analysis, etc. However, existing passive localization methods either achieve coarse-grained localization accuracy or require expensive infrastructure support. In this paper, we present Pallas, a self-bootstrapping system for fine-grained passive indoor localization using non-intrusive WiFi monitors. Pallas uses off-the-shelf access point hardware to opportunistically capture WiFi packets to infer the location of smartphones in the indoor environment. The key novelty of Pallas lies in that the passive fingerprint database for localization is automatically constructed and updated without any active participation of WiFi devices or manual calibration. To achieve this, Pallas first identifies passive landmarks that are present in WiFi RSS traces. Given the knowledge of the indoor floor plan and the location of WiFi monitors, Pallas statistically maps the collected RSS traces to specific indoor pathways. With sufficient mapping opportunistically detected, Pallas is able to bootstrap a fine-grained passive fingerprint database and build Gaussian processes for localization automatically without requiring any additional calibration effort.

Synchronous transmission has been exploited recently for accelerating fundamental operations of wireless sensor networks like data dissemination, data collection, and network-wide agreement by an order of magnitude. Although these protocols (e.g., Glossy) have shown to be highly reliable for small packet sizes, the use of large packet sizes coupled with an excessive number of simultaneous transmissions can reduce reliability significantly. Our work is motivated by the observation that capture effect can improve the reliability of synchronous transmission. We experimentally study the effect of physical layer capture and identify guidelines in which it can be exploited to enhance reliability. Based on these observations, we propose Syncast, a data dissemination protocol that improves over Glossy by exploiting capture effect over multichannels. The evaluation shows that Syncast provides a robust, reliable, and scalable data dissemination of large packets with lower end-to-end delay and up to 92% reduced radio-on-time.

Mobile devices, in particular, smartphones are ubiquitous. Many of these devices are able to communicate using the Wi-Fi network. In order to support fast service discovery, Wi-Fi devices broadcast probe request frames to actively look for Wi-Fi Access Points (APs) nearby. These probe requests provide information that allows one to extract useful observation about user behavior. In this paper, we present our results obtained from a long-term passive scanning of Wi-Fi probe request packets collected in a popular outdoor food court in Singapore. The collection was done over a 32-month duration. We present measurements on the breakdown in device vendors, probe intervals, Service Set Identifier (SSID) leakage and a simple counting application.

Time synchronization is an enabling service that allows devices to share a consistent notion of time and thus makes it easier to build efficient and robust collaborative services. However, existing synchronization protocols based on wireless packet transmissions are not energy efficient because powering the radio often consumes a significant fraction of the energy budget. In this paper, we propose PSync, a visible light-based time synchronization protocol that relies on an LED light source and is highly energy efficient for the receivers. The key novelty in our protocol is the use of a De Bruijn sequence to provide a rough estimate of time using a minimum amount of information. Experiments show that our scheme achieves an average synchronization error of 1.3 timer ticks (32 μs per clock tick). In addition, the additional energy consumed for one round of synchronization based on PSync can be as low as 19% of the energy needed to receive a small packet (1 byte) using IEEE 802.15.4 radio.

Wireless multicast has been exploited to bridge the gap between the limited wireless bandwidth and the rapidly increasing mobile video traffic demand. Multicast of videos to a set of heterogeneous users over multiple wireless access points, however, is challenging because of the trade-offs between high transmission rate, load balancing, and multicast opportunities. In this paper, we present JurCast, a joint user and rate allocation scheme for video multicast over multiple APs. Our approach balances the trade-off between these factors by determining user to Access Points (APs) association, the video resolution version (quality) to be delivered for each session, and the transmission link rate for each video version. The aim of our solution is to maximize the overall received video quality over all users. We have implemented and evaluated our solution on a WiFi testbed as well as the simulation of a large scale deployment. The results indicate that our method considerably outperforms the baseline schemes and achieves up to 3dB and 55% improvements in terms of peak signal-to-noise ratio (PSNR) and goodput, respectively.

It is widely known that cross-technology interference (CTI) adversely affects the reliability of 802.15.4 (ZigBee) communications. With the 2.4 GHz ISM band getting increasingly congested because of multiple wireless technologies like IEEE 802.11 (WiFi), 802.16 (WiMax), and Bluetooth and common appliances like a microwave and a cordless phone sharing it, data collection becomes exceedingly challenging especially in unplanned CTI prevalent in urban environments. This demonstration presents Oppcast, a robust and energy-efficient data collection protocol which carefully exploits a combination of spatial and channel diversity to eliminate the need for performing expensive channel estimation in advance. By exploiting multiple routes over multiple channels, packets can be reliably communicated to the sink even in severely interfered environments.

In this paper, we present an approach to extract social behavior and interaction patterns of mobile users by passively monitoring WiFi probe requests and null data frames that are sent by smartphones for network control/management purposes. By analyzing the temporal and spatial correlations of the Receive Signal Strength Indicators (RSSI) of packets from these low rate transmissions, we are able to discover proximity relationships, occupancy patterns, and social interactions among users.

Mobile devices are increasingly becoming the dominant device for Internet access. The network throughput achieved by a mobile device directly affects the performance and user experience. Throughput measurement techniques thus play an important role in predicting expected performance. Measurement techniques that require the transfer of large amounts of data can provide higher accuracy but incur large overhead. Further, since most mobile cellular plans impose usage quota, the overhead of such measurements over cellular networks can become quite high. Smaller data transfers have also been used to measure the throughput. Due to the conservative TCP slow start behaviour, however, these measurements often underestimate the achievable throughput. Considering these weaknesses in existing throughput measurement techniques, we propose a throughput estimation technique for TCP Cubic that uses 1 MB of data transfer to predict the throughput for prevalent large transfer sizes in mobile traffic such as 5 MB, 10 MB and 20 MB. Our evaluation shows that our approach can achieve high accuracy with low overhead, in predicting the achievable throughput.

Experimental research in low-power wireless networking lacks a reference benchmark. While other communities such as databases or machine learning have standardized benchmarks, our community still uses ad-hoc setups for its experiments and struggles to provide a fair comparison between communication protocols. Reasons for this include the diversity of network scenarios and the stochastic nature of wireless experiments. Leveraging on the excellent testbeds and tools that have been built to support experimental validation, we make the case for a reference benchmark to promote a fair comparison and reproducibility of results. This abstract describes early design elements and a benchmarking methodology with the goal to gather feedback from the community rather than propose a definite solution.

Indoor environment inference is of great importance to mobile and pervasive computing. As high-level metadata of indoor environment, floor maps contain rich information and are widely required in many pervasive systems. However, despite significant research progress, automatic inference of indoor maps has been less studied. In this paper, we present iMap, a smartphone-based opportunistic sensing system that automatically constructs the indoor maps by merging crowdsourced walking trajectories from smart-phone users. Most importantly, indoor semantics, such as stairs, escalators, elevators and doors are also automatically detected and annotated to the constructed map in the same inference process. The evaluation result shows that iMap can accurately detect different indoor semantics and be applied to different indoor environments. With the capability of generating semantic-annotated indoor maps without requiring any prior knowledge of the indoor environment, iMap has the potential to be widely deployed in practice.

The ubiquity of the smartphones makes them ideal platforms for generating in-situ content. In well-attended events, photos captured by attendees have diverse views that could be subjected to occlusion and abnormal lighting effects that could obscure the view. Such unstructured photo collections also have significant redundancy. Thus, a scene that is partially occluded or has bad contrast in one photo may be captured in another photo, possibly with higher details. We propose an application called Autolink that automatically establishes content-based links between sensor-annotated photos in unstructured photo collections captured using smartphones, such that users could navigate between high-context and high-detail images. This hierarchically structured image collection facilitates the design of applications for navigation and discovery, analytics about user photography patterns, user taste, and content/event popularity. Autolink includes a framework that constructs this hierarchy efficiently and with little content-specific training data by combining photo content processing with associated sensor logs obtained from multiple participants. We evaluated the performance of Autolink on two real-world sensor tagged photo datasets. The result shows that Autolink is able to efficiently cluster photos at 20 times faster than candidate algorithms, into the appropriate hierarchy with at least 70% precision and 37% better recall than candidate algorithms.

Zoomable video streaming refers to a new class of interactive video applications, where users can zoom into a video stream to view a selected region of interest in higher resolutions and pan around to move the region of interest. The zoom and pan effects are typically achieved by breaking the source video into a grid of independently decodable tiles. Streaming the tiles to a set of heterogeneous users using broadcast is challenging, as users have different link rates and different regions of interest at different resolution levels. In this article, we consider the following problem: Given the subset of tiles that each user requested, the link rate of each user, and the available time slots, at which resolution should each tile be sent, to maximize the overall video quality received by all users. We design an efficient algorithm to solve this problem and evaluate the solution on a testbed using 10 mobile devices. Our method is able to achieve up to 12dB improvements over other heuristic methods.

The mismatch between increasingly large video resolution and constrained screen size of mobile devices has led to the proposal of zoomable video systems based on tiled video. In the current system, a tiled video frame is constructed from multiple tiles in a single resolution stream. In this paper, we explore the perceptual effect of mixed-resolution tiles in tiled video, in which tiles within a video frame could come from streams with different resolutions, with the aim to tradeoff bandwidth and perceptual video quality. To understand how users perceive the video quality of mixed-resolution tiled video, we conducted a psychophysical study with 50 participants on tiled videos where the tile resolutions are randomly chosen from two resolution levels with equal probability. The experiment results show that in many cases, we can mix tiles from HD (1920×1080p) stream and tiles from 1600×900p stream without being noticed by the viewers. Even when participants notice quality degradation in videos combined with tiles from HD stream and tiles from 960×540p stream, the majority of participants still accept the degradation when viewing videos with low and medium motion; and greater than 40% of participants accept the quality degradation when viewing video with dense motion.

Accelerometer is the predominant sensor used for low-power context detection on smartphones. Although low-power, accelerometer is orientation and position-dependent, requires a high sampling rate, and subsequently complex processing and training to achieve good accuracy. We present an alternative approach for context detection using only the smartphone's barometer, a relatively new sensor now present in an increasing number of devices. The barometer is independent of phone position and orientation. Using a low sampling rate of 1 Hz, and simple processing based on intuitive logic, we demonstrate that it is possible to use the barometer for detecting the basic user activities of IDLE, WALKING, and VEHICLE at extremely low-power. We evaluate our approach using 47 hours of real-world transportation traces from 3 countries and 13 individuals, as well as more than 900 km of elevation data pulled from Google Maps from 5 cities, comparing power and accuracy to Google's accelerometer-based Activity Recognition algorithm, and to Future Urban Mobility Survey's (FMS) GPS-accelerometer server-based application. Our barometer-based approach uses 32 mW lower power compared to Google, and has comparable accuracy to both Google and FMS. This is the first paper that uses only the barometer for context detection.

The metrics of information latency and reach ability are widely used for measuring information flow in opportunistic networks. We present an alternative measure that looks at the amount of information updates or changes, which is an important and yet relatively unexplored metric for characterizing opportunistic networks. In this paper, we propose a novel concept to measure information freshness - how much the latest information received differs from the most up-to-date information on the sender. Based on the information freshness, the awareness on how information propagated may have changed, or change awareness, can be computed. Change awareness can be exploited to design efficient algorithms, in particular, data aggregation algorithms. Evaluation on real world traces shows that change awareness based solutions can achieve similar results as data aggregation algorithm with connection oracles.

Proliferation of mobile devices with video recording capability has lead to a tremendous growth in the amount of user-generated mobile videos. Researchers have embarked on developing new interesting applications and enhancement algorithms for mobile video. There is, however, no standard dataset with videos that could represent characteristics of mobile videos captured in realistic scenarios. In this paper, we present our effort to create one such dataset, consisting of videos simultaneously recorded using mobile devices in an unconstrained manner by multiple users attending performance events. Each video is accompanied by concurrent readings from accelerometer and compass sensors. At the time of writing, the dataset contains 473 video clips, with a total length of 30 hours 41 minutes and total size of 122.8 GB. We believe this dataset is useful as a common benchmark dataset for a variety of different research topics on mobile videos, including video analytics, video quality enhancement, and automatic video mashups.

In this paper, we present an efficient routing algorithm, Plankton, for Delay/Disruptive Tolerant Network (DTN). Plankton utilizes replica control to reduce overhead and contact probability estimates to improve performance. Plankton has two major features. First, it uses a combination of both short-term bursty contacts and long-term association based statistics for contact prediction. Second, it dynamically adjusts replication quotas based on estimated contact probabilities and delivery probabilities. Our evaluation on extensive traces shows that Plankton achieves significantly better prediction accuracy than existing algorithms for contact probability prediction. In addition, we show that while Plankton incurs much lower communication overhead compared to Spray-and-Wait, MaxProp and RAPID with savings from 14% to 88%, it can also achieve similar if not better delivery ratios and latencies.

Understanding how people communicate with one another plays a very important role in many disciplines including social psychology, economics, marketing, and management science. This paper proposes and evaluates SocialWeaver, a sensing service running on smartphones that performs conversation clustering and builds conversation networks automatically. SocialWeaver uses a hybrid speaker classification scheme that exploits an adaptive histogram-based classifier to non-obtrusively bootstrap the in situ speaker model learning. The conversation clustering algorithm proposed is able to detect fine-grain conversation groups even if speakers are close together. Finally, to address energy constrain, a POMDP-based energy control scheme is incorporated.

As the current generation of mobile smartphones become more powerful, they are being used to perform more resource intensive tasks making battery lifetime a major bottle-neck. In this paper, we present a technique called dynamic AoV lookahead for reducing wireless interface power consumption upto 50% while playing a popular, yet resource intensive, mobile multiplayer games.

Large-scale failures are commonplace in commodity data centers, the major platforms for Distributed Stream Processing Systems (DSPSs). Yet, most DSPSs can only handle single-node failures. Here, we propose Meteor Shower, a new fault-tolerant DSPS that overcomes large-scale burst failures while improving overall performance. Meteor Shower is based on checkpoints. Unlike previous schemes, Meteor Shower orchestrates operators' check pointing activities through tokens. The tokens originate from source operators, trickle down the stream graph, triggering each operator that receives these tokens to checkpoint its own state. Meteor Shower is a suite of three new techniques: 1) source preservation, 2) parallel, asynchronous check pointing, and 3) application-aware check pointing. Source preservation allows Meteor Shower to avoid the overhead of redundant tuple saving in prior schemes, parallel, asynchronous check pointing enables Meter Shower operators to continue processing streams during a checkpoint, while application-aware check pointing lets Meteor Shower learn the changing pattern of operators' state size and initiate checkpoints only when the state size is minimal. All three techniques together enable Meteor Shower to improve throughput by 226% and lower latency by 57% vs prior state-of-the-art. Our results were measured on a prototype implementation running three real world applications in the Amazon EC2 Cloud.

Due to their thin size, vivid colors, high contrast and power efficiency, OLED (Organic Light-Emitting Diode) display and its variants such as AMOLED (Active Matrix OLED) displays are increasingly replacing traditional LCD (Liquid Crystal Display) screens in smart phones. However, the power efficiency of OLED screens greatly depends on the luminance and colors of the displayed contents on the screen. Web browsing is one of the most widely used applications in mobile devices. In this paper, we present our cloud service, which intelligently re-paints the web pages in real-time with power efficient colors and HVS (Human Visual System) based tone mapping techniques, without adversely affecting the identity (brand color) of the web pages as well as the user's browsing experience. El-pincel helps to save up to 60% of OLED energy with color combinations that ensure good legibility and pleasing affective response to human eyes.

With mobile devices becoming ubiquitous, collaborative applications have become increasingly pervasive. In these applications, there is a strong need to obtain a count of the number of mobile devices present in an area, as it closely approximates the size of the crowd. Ideally, a crowd counting solution should be easy to deploy, scalable, energy efficient, be minimally intrusive to the user and reasonably accurate. Existing solutions using data communication or RFID do not meet these criteria. In this paper, we propose a crowd counting solution based on audio tones, leveraging the microphones and speaker phones that are commonly available on most phones, tackling all the above criteria. We have implemented our solution on 25 Android phones and run several experiments at a bus stop, aboard a bus, within a cafeteria and a classroom. Experimental evaluations show that we are able to achieve up to 90% accuracy and consume 81% less energy than the WiFi interface in idle mode.

In this paper, we present a dual-radio framework for implementing MAC protocols in wireless sensor networks. The framework is based on the observation that MAC operations can be categorized into time-dependent and bandwidth-dependent. Unlike existing dual-radio systems, we do not propose a new MAC protocol. Instead, we show how a given MAC protocol can be re-implemented using a dual-radio framework resulting in substantial energy savings. Our approach is generic. Different categories of MAC protocols can be re-implemented using dual radios, including synchronous (e.g., SMAC), asynchronous (e.g., BMAC) and hybrid (e.g., SCP-MAC). The proposed framework is easy-to-implement. We have been able to re-implement SMAC, BMAC and SCP-MAC in TinyOS to dual-radio implementations using a combination of mica2 and micaZ motes. Extensive evaluation shows that a reduction of up to 5 times in energy consumption can be attained.

When low power mobile devices communicate over short range wireless networks using technologies such as WiFi, Bluetooth, ZigBee etc., the contacts are often opportunistic and intermittent due to range, mobility and energy constraint. While the DTN architecture provides a framework to support opportunistic and intermittent connectivities, existing DTN routing algorithms do not consider application requirements. In particular, if data block dependencies of applications are known to the network nodes, resource control mechanisms can utilize this knowledge to improve performance. The contributions of this paper are as follow. First, we propose a mechanism to quantify application data dependencies and show how this information can be exploited by resource control mechanisms. Next, we show how application data dependencies information can be incorporated into existing DTN routing algorithms to improve performance. Simulation results show that when application awareness is exploited, we can substantially improve application performance from 60% to 583% over the baseline DTN routing algorithms.

Due to their thin size, vivid colors, high contrast and power efficiency, OLED and its variants such as AMOLED screens are increasingly replacing traditional LCD screens in mobile phones (eg. Google Nexus One, Samsung Galaxy S phones). However, the power efficiency of OLED screens greatly depends on the luminance and colors of the displayed contents on the screen. Web browsing is one of the most widely used applications in mobile devices. We demonstrate our cloud service, which intelligently re-paints the web pages in real-time with power efficient colors and tone mapping techniques without adversely affecting the user experience in reading the page and the identity of the page. Transformation of images, flash contents and videos in client itself will incur significant computational and energy overhead. El-pincel is designed as a cloud service to avoid any additional overhead to the mobile device.

In this paper, we show how tone mapping techniques can be used to dynamically increase the image brightness, thus allowing the LCD backlight levels to be reduced. This saves significant power as the majority of the LCD's display power is consumed by its backlight. The Gamma function (or equivalent) can be efficiently implemented in smartphones with minimal resource cost. We describe how we overcame the Gamma function's non-linear nature by using adaptive thresholds to apply different Gamma values to images with differing brightness levels. These adaptive thresholds allow us to save significant amounts of power while preserving the image quality. We implemented our solution on a laptop and two Android smartphones. Finally, we present measured analytical results for two different games (Quake III and Planeshift), and user study results (using Quake III and 60 participants) that shows that we can save up to 68% of the display power without significantly affecting the perceived gameplay quality.

The current generation of mobile smartphones are not just devices for voice communication. Instead, they are frequently used as mobile PC replacements that are used to edit documents, browse the web, check email, and play games. However, this functionality comes at the cost of battery lifetimes. Indeed, to maintain the slim form factor required for these phones (which impacts the amount of battery that can be put into the phone), it is quite common for the battery lifetimes of smartphones to be significantly shorter compared to previous generation of "dumber" phones.

A large percentage of links in low-power wireless sensor networks are of intermediate quality. To the best of our knowledge, opportunistic exploitation is currently the only way to use these links. However, such exploitation requires overhearing which consumes a significant amount of energy. In this paper, we propose a new approach to exploit intermediate quality (IQ) links through channel diversity with a new protocol, called IQ Link Transformation Protocol (ILTP), that does not require overhearing. ILTP transforms IQ links into good links thus allowing us to exploit such links continuously rather than using them only opportunistically. Our key insight is that the packet reception ratios (PRR) across different channels on IQ links are not correlated and it is common on such links to find channels that change in quality on the time scale of a few minutes. Consequently, when the link quality of a channel is bad, it is highly likely that a good channel can be found and its quality will remain good for at least a few minutes. Our evaluations on three large-scale test beds demonstrate that ILTP is able to consistently transform the IQ links into good links. We observe that even a poor link with a PRR of 0.05 can be transformed into a good link with a PRR greater than 0.9. When ILTP is integrated with CTP, the default collection tree protocol for TinyOS, the average number of transmissions per end-to-end packet delivery is reduced by 24% to 58%.

Abstract The inherent temporal connectivity and existence of impairments in wireless channels pose challenges to network performance. Cooperative communication has been proposed as an effective technique to mitigate the imperfections of the wireless medium by exploiting channel diversity and availability of neighboring nodes that can act as relays. Although numerous cooperative communication techniques have been proposed in the literature, most of them do not consider Quality of Service (QoS) issues in a wireless sensor network. In this work, we study how cooperative communication can be applied to achieve differentiated QoS in a sensor network that uses Ultra‐Wideband (UWB) as its underlying PHY layer technology. SAUCeR is a slotted‐aloha based ultra‐wideband medium access control protocol with cooperative retransmissions that provides differentiated QoS in networks with varying traffic classes. Despite the high transmission rates provided by UWB, its impulse‐based nature renders many conventional carrier sensing MAC protocols incompatible. Consequently, SAUCeR utilizes slotted‐aloha to reduce packet collisions without the need for carrier sensing. Differentiated QoS is provided by allocating different resources (time slots) to varying traffic classes to segregate the contention between them. A QoS‐aware cooperative retransmission technique and two distributed relay selection schemes are also introduced to improve overall traffic throughput and reduce end‐to‐end delay, while preventing the starvation of any traffic class. Copyright © 2010 John Wiley & Sons, Ltd.

When Disruption Tolerant Network (DTN) is used in commercial environments, incentive mechanism should be employed to encourage cooperation among selfish mobile users. Key challenges in the design of an incentive scheme for DTN are that disconnections among nodes are the norm rather than exception and network topology is time varying. Thus, it is difficult to detect selfish actions that can be launched by mobile users or to pre-determine the routing path to be used. In this paper, we propose MobiCent, a credit-based incentive system for DTN. While MobiCent allows the underlying routing protocol to discover the most efficient paths, it is also incentive compatible. Therefore, using MobiCent, rational nodes will not purposely waste transfer opportunity or cheat by creating non-existing contacts to increase their rewards. MobiCent also provides different payment mechanisms to cater to client that wants to minimize either payment or data delivery delay.

Upon the occurrence of a phenomenon of interest in a wireless sensor network, multiple sensors may be activated, leading to data implosion and redundancy. Data aggregation and/or fusion techniques exploit spatio-temporal correlation among sensory data to reduce traffic load and mitigate congestion. However, this is often at the expense of loss in Information Quality (IQ) of data that is collected at the fusion center. In this work, we address the problem of finding the least-cost routing tree that satisfies a given IQ constraint. We note that the optimal least-cost routing solution is a variation of the classical NP-hard Steiner tree problem in graphs, which incurs high overheads as it requires knowledge of the entire network topology and individual IQ contributions of each activated sensor node. We tackle these issues by proposing: (i) a topology-aware histogram-based aggregation structure that encapsulates the cost of including the IQ contribution of each activated node in a compact and efficient way; and (ii) a greedy heuristic to approximate and prune a least-cost aggregation routing path. We show that the performance of our IQ-aware routing protocol is: (i) bounded by a distance-based aggregation tree that collects data from all the activated nodes; and (ii) comparable to another IQ-aware routing protocol that uses an exhaustive brute-force search to approximate and prune the least-cost aggregation tree.

Energy constraints in wireless sensor nodes necessitate the design and development of energy-efficient MAC protocols to arbitrate access to the shared communication medium. While there exists a plethora of sensor MAC protocols, these protocols do not individually vary the duty-cycle of each sensor according to local connectivity status, to maximize energy savings. In this paper, we propose A2-MAC - an Adaptive, Anycast MAC protocol for low-powered wireless sensor networks. It utilizes: (i) a random wakeup schedule, such that each node can independently and randomly wakeup in each cycle without coordination and time synchronization; (ii) adaptive duty-cycles based on network topology; and (iii) adaptive anycast forwarders selection, which allows each node to transmit to any member in its forwarding set. There are two key adaptive mechanisms in A2-MAC: (i) each node varies its duty-cycle and set of forwarding nodes such that energy consumption can be locally minimized for a given local delay performance objective; and (ii) nodes cooperatively reduce the duty-cycles required of their forwarding nodes, depending on local network connectivity. By allowing nodes to operate with different duty-cycles and forwarding sets, A2-MAC achieves better energy-latency tradeoffs and extends node lifetime substantially, while providing good end-to-end latency.

As bandwidth in the Internet continues to grow, there will be more and more long fat network pipes with abundant residual bandwidth. At the same time, there is also a gradual and steady increase in the deployment of Internet endpoints equipped with different variants of high speed TCP. In this work, we first illustrate drawbacks associated with two widely deployed high speed TCP variants, namely: cubic TCP and compound TCP. We show that even with common and reasonable settings, problems can arise. Next, we present synchronized TCP (Sync-TCP), a new delay-based high speed congestion control (HSCC) algorithm. The approach taken by Sync-TCP is novel in the following ways. First, Sync-TCP exploits synchronization. The key insight of Sync-TCP is that if competing flows could detect the same congestion signal through queue delay, these flows can coordinate their congestion control behaviors. Second, using only the basic mechanism, Sync-TCP will yield to legacy TCP when congestion is detected. Hence, Sync-TCP is designed to not hurt applications using legacy TCP or interactive applications. We performed extensive simulation and some testbed evaluations to show that Sync-TCP achieves its design goals and it performs better than existing HSCC approaches including Fast TCP, Compound TCP and Cubic TCP, especially in the trade-off between throughput and friendliness.

In this paper, we present MobTorrent, an on- demand, user-driven framework designed for vehicles which have intermittent high speed access to roadside WiFi access points (AP). Mobile nodes in MobTorrent use the WWAN network as a control channel. When a mobile client wants to initiate a download, instead of waiting for contact with the AP, it informs one (or multiple) selected AP(s) to prefetch the content. The scheduling algorithm in MobTorrent then replicates the prefetched data on the mobile helpers so that the total amount of data transferred and the average transfer rate to the mobile clients are maximized. Therefore, instead of limiting high speed data transfer to the short contact periods between APs and mobile clients, high speed transfers among vehicles are opportunistically exploited. Evaluation based on testbed measurement and trace-driven simulation shows that MobTorrent provides substantial improvement over existing architectures. For the case of a single AP, its performance approximates that of an off-line optimal scheduler. In case of multiple APs, our evaluation shows that MobTorrent's performance is robust in a variety of settings.

Asymmetric links commonly exist in low power wireless sensor networks. However, it is difficult to discover and exploit them efficiently. In this work, we propose DEAL, a link management scheme to Discover and Exploit Asymmetric Links efficiently in dense wireless sensor networks. Equipped with a novel feedback mechanism, DEAL dynamically adapts its link maintenance mechanism based on the estimated link quality, and manages the (small) neighbor table so as to retain the most useful information. We implement DEAL in TinyOS and evaluate its performance using both TOSSIM and testbed. The simulation results show that more than 80% of asymmetric links can be discovered and maintained with minimum overhead. Using a collection tree application and ETX as the routing metric, the average path ETX can be reduced by up to 20%. Testbed evaluation also shows that DEAL improves the network routing performance by identifying useful asymmetric links.

Connectivity-based routing protocols provide an attractive option for point to point communication in wireless networks due to its potential for low routing overhead. However, when the entire hop-count vector is used to address each node, the communication and storage overhead in the packets are often so high that it is not feasible to implement existing connectivity- based routing protocols infeasible on resource-constrained sensor networks. In this paper, we apply the technique of dimension reduction, in particular principle component analysis (PCA), to the hop-count vectors. Compared to the original hop-count vector, the embedding coordinates preserve the network geometry with much lower overhead, making their use much more practical on current sensor platform. Simulation results show that the coordinates computed by PCA can achieve higher packet delivery ratio, lower path stretch and shorter flooding range in local minimum cases. We have also implemented the PCA algorithm on MICAz motes and conducted experiments in a testbed containing 48 nodes deployed on two floors of an office building. With the use of 9 landmark nodes and only 3 dominant components, the PCA coordinates can achieve 95% of the delivery ratio obtained using full hop-count vector and maintain an low path stretch of 1.12.

A video streaming server needs to adapt its source/channel encoding parameters (or configurations) to changes in network conditions and to differences in users' connection profiles. The adaptation can be achieved by adjusting parameters such as frame rate, error protection ratio, and resolution. Ideally, the server should adapt the serving configurations with respect to the current network and user conditions to improve received video quality. However, adaptations that optimize playable frame rate require intensive computation, and storing all possible configurations requires a tremendous amount of storage. This brings forth the issues of how to obtain good video quality and reduce server resources usage at the same time. We address this issue in this paper. Our approach is based on the observation that transcoding between certain configurations can be performed very efficiently. We propose a framework to compute a set of configurations to store on the server by considering two opposing goals: (a) maximizing expected received quality of the video, and (b) minimizing server resource usage by lowering transcoding cost and expected number of switches between configurations. The second objective also reduces the number of configurations, and therefore reduces the total storage required. Our framework models the relationship among different configurations in a partial order, formulates the search of a good set of configurations as an energy minimization problem, and we use techniques in image segmentation to solve the problem. Experimental results show that our framework relieves the server load and increases the number of clients served, while only slightly reducing the expected frame rate.

10.1145/1368310.1368337

Connectivity-based routing protocols use the hop count vector to a group of anchors for packet forwarding. Due to the discrete nature of hop count based coordinates, without an effective recovery mechanism, these protocols will frequently encounter failures at network local minimum sites. In this paper, we propose a new connectivity-based routing protocol named Hop Distance Routing(HDR) with an efficient distance metric called hop distance. To ensure packet delivery, HDR complements greedy forwarding with a tree based recovery method, which allows the packet to traverse the branches and escape from local minimum locations. By labeling each tree node with an angle range, HDR can identify the subtree where the destination resides and forward the packet towards it until greedy forwarding can be resumed. Compared to other connectivity-based routing protocols such as NoGeo, BVR, LCR and VPCR, performance results show that the HDR protocol provides the highest packet delivery ratio and the lowest path stretch with the minimal communication overhead for each packet delivery on the node density range from 1.76 to 17.67.

In recent years, many high-speed congestion control (HSCC) algorithms have been proposed for utilizing network pipes with huge bandwidth-delay product (BDP), and some of them have also been implemented in popular operating systems. Considering the extremely long round trip propagation delay (RTPD) of satellite network, it is very likely that these algorithms are triggered when TCP flows pass through satellite network. But the existing algorithms are normally evaluated on network pipes with high bandwidth and moderate RTPD (< 400 ms). This paper is an attempt to study these algorithms on a simulated satellite network with moderate bandwidth and extremely long RTPD. Their effects on the existing applications, especially World Wide Web (WWW) and the emerging streaming applications, are emphasized. Different queue sizes are also used in simulation with the aim of investigating how to provision satellite link's queue for well accommodating flows driven by HSCC algorithms. Through this study, we find that Compound-TCP, Cubic-TCP, and H-TCP can not work well on satellite network. Currently, satellite link should use moderate queue size for accommodating flows driven by HSCC algorithms. As for end hosts, when they find that round trip time (RTT) is very long, HS-TCP should be adopted for high throughput while avoiding to hurt the existing applications as little as possible.

Topology discovery in sensor network is useful in any practical deployment of the network. Topology information helps the administrators monitor, debug and perform proactive control and management obtaining topology for dense wireless sensor network is expensive due to limited bandwidth and contention for the wireless channel. In this paper, we consider the problem of efficiently obtaining network topology where the locations and associated node identifiers are known by the central controller (sink). Although the locations of the nodes are known, the connectivity between any two nodes cannot be determined by distance information alone due to irregular radio coverage and obstacles. We propose a location-aided topology discovery algorithm (LAD) that requires O(log(|M|)) data units per node in common cases, and O(|M|) data units per node in the worst case, where M is the set of directly communicable neighbors of a particular node. Simulation results show that the overhead reduction ranges from 70% to over 95%.

Ultra-wideband (UWB) is a physical (PHY) layer technology that promises high transmission rates, as well as high resistance to noise and multipath effects. However, the impulse-based nature of UWB, coupled with its low transmission power, makes it difficult to enable efficient detection of the signals. Consequently, conventional carrier-sensing based MAC protocols cannot be used with a UWB PHY. In this paper, we propose SASW-CR - a Slotted Aloha MAC protocol for UWB networks with Sliding contention Window and Coperative Retransmissions, which provides QoS support without the use of carrier sensing. SASW-CR utilizes the slotted-Aloha technique to avoid carrier sensing and reduce packet collisions. In addition, it makes use of differentiated contention windows to provide varying classes of QoS for different traffic classes. A cooperative retransmission technique is also introduced to improve the overall traffic throughput and reduce end-to-end delay. The efficacy of our protocol is demonstrated through simulations.

While wireless networks are generally deployed in three dimensional environments, most geometric routing protocols are designed and evaluated with a focus on two dimensional space. In this paper, we present a connectivity based routing protocol in 3D space named Spherical Coordinate Routing(SCR), which exploits connectivity-based greedy forwarding for efficient routing paths and a spherical coordinate tree based recovery method to guarantee packet delivery. SCR can deploy multiple recovery trees simultaneously to enhance the routing efficiency and resilience against network dynamics. Simulation results show that the packet delivery ratio of SCR is 22% ~ 63% higher than that of BVR, LCR and NoGeo, at the critical node density range of 4 to 8. The amount of forwarding traffic required per packet delivery is considerably lower than protocols relying on scoped flooding based recovery. The SCR protocol can efficiently recover from topology changes due to clustered node failures, while maintaining excellent path stretch and traffic load performance in a 3D network topology.

Connectivity monitoring is useful in practical deployment of wireless sensor network. In order to understand the behavior and performance bottleneck, knowledge of the network connectivity is crucial. In this paper, we propose a flexible and efficient connectivity monitoring algorithm (H2CM) that has three components and operates in a divide and conquer manner. The components include hop vector distance based filtering, Bloom filters and signature hashing and are designed to work with different combinations of network and neighbor set sizes. In simulation, communication cost reduction of H2CM compare to maximal compression of neighborhood information varies from 65% to 85% for large networks (> 1000 nodes) and from 40% to 70% for a medium size network (a few hundred nodes). We have also implemented the algorithm in TinyOS and evaluated its performance on a testbed with 34 motes.

The inherent near-far effect in wireless networks causes nodes that are further away from the receiver to suffer from throughput degradation, as packets from nodes that are nearer are typically received with greater signal strengths. This unfair situation is traditionally overcome by power control. However, when power control is not feasible, for example in tiny sensor nodes with power-limited batteries, alternative solutions have to be utilized to achieve fairness in the network. In this paper, we propose U-LiBRA - an UWB Location Based Resource Allocation scheme to alleviate the contention between near and far nodes in a TH-UWB sensor network. U-LiBRA allocates different time slots to nodes that are at varying distances from the receiver, so that nodes that are further away from the receiver can achieve higher throughput than what they would typically obtain under the influence of the near-far effect. Simulation results show that U-LiBRA can effectively mitigate the near-far effect and improve fairness in the absence of power control.

In this work, we study the time-slot allocation problem in a multi-sink single-hop TH-UWB network scenario, where the traffic from a sensor node is anycasted via a single hop to any one of multiple sinks. The slot allocation problem is formulated as an optimization problem and shown to be NP-hard. We then present a heuristic to increase network throughput and fairness as compared to a random allocation. In the proposed heuristic, nodes that are of similar distances to any sinks are grouped together to utilize the same set of TH slots for transmissions. Simulations show that the proposed heuristic improves both throughput and fairness, scales with multiple sinks and can be used as a simple admission control mechanism.

In this work, we present, CaSMA, a packet scheduling mechanism for mobile ad hoc networks (MANETs) that takes into account both the congestion state and the end-to-end path duration. We show that CaSMA approximates an ideal scheduling mechanism in terms of maximizing the goodput and sharing the throughput (losses) fairly among the contending flows. Further, the simulation results show that both average delay for CBR flows and throughput for TCP can be improved substantially compared to FIFO.

As third-generation (3G) wireless networks with high data rate get widely deployed, optimizing the transmission control protocol (TCP) performance over these networks would have a broad and significant impact on data application performance. In this paper, we make two main contributions. First, one of the biggest challenges in optimizing the TCP performance over the 3G wireless networks is adapting to the significant delay and rate variations over the wireless channel. We present window regulator algorithms that use the receiver window field in the acknowledgment (ACK) packets to convey the instantaneous wireless channel conditions to the TCP source and an ACK buffer to absorb the channel variations, thereby maximizing long-lived TCP performance. It improves the performance of TCP selective ACK (SACK) by up to 100 percent over a simple drop-tail policy, with small buffer sizes at the congested router. Second, we present a wireless channel and TCP-aware scheduling and buffer sharing algorithm that reduces the latency of short flows while still exploiting user diversity for a wide range of user and traffic mix.

In recent years, many different kinds of wireless access networks have been deployed for the Internet and have become inseparable parts of the Internet. But TCP, the most widely used transport protocol of the Internet, was designed for stationery hosts. In particular, TCP faces severe challenges when user moves around in these networks and handoff occurs frequently. In this paper, TCP handoff (TCP-HO), a practical end-to-end mechanism, is proposed for improving TCP performance in heterogeneous mobile environments. TCP-HO assumes that a mobile host is able to detect the completion of handoff immediately and has a coarse estimation of new wireless link's bandwidth. When a mobile host detects handoff completion, it will immediately notify the server through two duplicate ACKs, whose TCP option also carries the bandwidth of new wireless link. After receiving this notification, the server begins to transmit immediately and keeps updating ssthresh according to the bandwidth from mobile host and its new RTT samples. This updating will be stopped after four RTT samples or after congestion is detected. TCP-HO has been implemented in FreeBSD 5.4. Experimental results show that TCP-HO does improve TCP performance without adversely affecting cross traffic in a heterogeneous mobile environment.

Geographic face routing provides an attractive way for packet delivery in wireless networks due to its high reliability and low overhead. A good face routing protocol should provide guaranteed packet delivery and efficient routing paths. In this paper, we present a new face routing method named GFRIS that has both features by actively probing each face for the face size and the unique face identification sequence -face ID. Face switch occurs only if the outgoing edge intersects with the local minimum-destination line at a progressing location and the edge is shared between two different faces. To avoid the huge performance penalty when selecting an inefficient face traversal direction on a large face, GFRIS uses the face size to trigger the bounded face traversal procedure as proposed earlier in GOAFR+. Simulation results show that, by using face ID to assist face switch and adaptively applying the normal and bounded face traversal rules according to the face size, GFRIS achieves lower path stretch factor compared to GFG, GPSR, GFG2 and GOAFR+. The worst case performance of GFRIS is even better than that of GOAFR+ in critical node densities from 4 to 7.

This paper presents AMCM, a traffic-adaptive multichannel MAC protocol that increases the capacity of wireless network by enabling multiple concurrent transmissions on orthogonal frequency channels using a single half- duplex transceiver. AMCM is based on the IEEE 802.11 MAC but provides fine-grain, asynchronous coordination among locally interfering nodes for channel negotiation. By incorporating load-awareness, channel availability awareness and batch transmissions, our window-based approach achieves high channel utilization under varying load, while avoiding the control-window saturation problem as the number of channels increases. For single-hop scenarios, we show that, at low load, AMCM is comparable to IEEE 802.11 MAC, while under high load, AMCM delivers almost Nx improvement gain over IEEE 802.11 MAC protocol, where N is the number of channels. AMCM also outperforms existing multi-channel MAC protocols by 100% and 150% respectively under high load at a lower hardware cost and complexity. In multi-hop scenarios, AMCM achieves performance improvement of 190% and 90% for both dense and sparse network over IEEE 802.11 MAC respectively. In both scenarios, AMCM achieves close to full utilization of all channels with good protocol efficiency.

In MANETs, majority of performance studies are carried out via simulations, where node mobility and radio propagation models play a crucial role. However, popular simulation tools, like NS-2 and GloMoSim, use simplistic random mobility patterns and free space radio propagation models. Such simplification ignores many crucial details in environment where movements are not random and obstacles are common. To have a better understanding of MANET protocols, there is a need for a tool that can generate more realistic mobility scenarios and provide better radio propagation model. This paper presents EGRESS, which is built on top of NS-2 and consists of two major components: SGT and ORPM. SGT is a tool that generates node movements in an urban environment with buildings and pathways. ORPM is an obstructive radio propagation model, which enhances the existing radio models in NS-2 by taking into account obstacles in a 3D environment. We believe EGRESS is the first open source tool that provides such integrated and more realistic features for simulating urban environment. Our simulation results show that using more realistic scenarios can have a significant impact on network topology and performance of routing protocols

A wireless sensor network is normally a multi-hop wireless ad hoc network whose purpose is to transmit events sensed by sensors to the sink. Since sensor nodes are normally powered by non-rechargeable and non-replaceable battery, an energy efficient event transport service is crucial for extending the life time of a wireless sensor network. Due to the possibility of crossing multiple lossy links between the source and the sink, many events may be lost before they arrive at the sink. In addition, the overhead of low layer protocols is very large due to the short event data. These two factors can result in significant event loss and waste of energy. Many mechanisms have been proposed on other networks to improve reliability and reduce overhead. Due to scarce memory and poor computing ability of sensor nodes in a sensor network, these mechanisms may not be feasible to implement and they may not work efficiently with respect to energy. In this paper, PMC (Pipelined Multiple Copies) is proposed for transmitting events in a wireless sensor network, with the goals of achieving low event loss rate and high energy efficiency. In PMC, multiple copies of an event are pipelined and transmitted in several consecutive frames according to Silent CSMA, a variant of CSMA/CA. Multiple copies of an event reduce end-to-end event loss rate and Silent CSMA reduces packet energy consumption significantly.

In this paper, we propose a call admission control mechanism for wireless ad hoc networks called interference-based call admission control (iCAC). iCAC is unique in that it does not treat interference uniformly instead classifies interference based on estimates of the position of the interfering nodes. iCAC relies on two novel techniques: (1) estimation of position of the interfering nodes (2) fair allocation using bandwidth acquisition and rate control. By incorporating these techniques, iCAC is able to increase the estimated available bandwidth substantially without overloading the network. We compare iCAC with perceptive admission control (PAC) and IEEE 802.11 without admission control. Simulation results show that iCAC is able to admit substantially more requests than PAC, achieves more than 80% of the throughput of IEEE 802.11 and at the same time maintains very low packet loss rate and average delay comparable to PAC

Environmental monitoring and surveillance is a popular application of wireless sensor network. In such an application, the data transmitted are tagged with geographic information. A network with better coverage provides better quality-of-service since it will be able to monitor its area of responsibility more effectively. In this work, we study the impact of congestion on coverage of the sensor network. Congestion can negatively impact the performance since it can result in reduced coverage and power wastage. In this paper, we present a buffer management scheme called most redundant drop (MRD) and a scheduling algorithm called coverage transmit (CT) that make use of spatial information in sensor data to improve network coverage. Compared to drop-tail and FIFO, MRD and CT improve coverage by up to 75% when exact sensor location is available. Furthermore, as exact locations may not be available in practice, MRD and CT are evaluated using a modified DV-hop scheme that provides approximate localization. Simulation results show that substantial improvement can also be obtained using only approximated locations

Coordination of cellular base stations (BS) with overlapping coverage enables joint optimization of radio resource allocation in a multiple cell environment. This paper extends existing Proportional Fairness model for wired networks and single wireless cell, to the context of multiple (probably heterogeneous) wireless cells with overlapping coverage. The proposed fair allocation achieves both global Pareto optimality and inter- cell fairness (load balance). However, the ideal allocation is not practical as it requires a mobile station (MS) be simultaneously associated with multiple BSs. Instead, we use a simple GLS (Greedy Logarithmic Sum) scheme, which associates each new arrival MS with only one BS, to approximate the optimal allocation. Simulation result shows that GLS performs close to optimal scheme in a wide range of network settings.

IP-based radio access networks (RAN) are expected to be the next generation access networks in UMTS and CDMA networks. The question of connectivity, i.e., how best to connect base stations to the radio network controllers (RNC) in an IP-based RAN, has not been addressed by researchers. Furthermore, given a connection configuration, an RNC selection algorithm that assigns an incoming call to an RNC is also necessary. This paper examines RAN connectivity and its impact on the performance and resiliency of the wireless network using different RNC selection algorithms. For homogeneous networks, we show that the proposed min-load-1 algorithm, which allows at most one hard handoff in order to accommodate each incoming call request, delivers performance close to the optimal algorithm. We also show that allowing a few base stations to connect to two RNCs (a 10 percent increase in the number of links in our network) results in resiliency to RNC failures that is comparable to the resiliency of RANs with full-mesh connectivity. Finally, for heterogeneous networks, we show that the min-load-k algorithm (with at most k hard handoffs per call) is effective in handling load imbalances. These results provide strong motivation for deploying IP-based RAN, as they suggest that enhancing current point-to-point RAN with few additional links and allowing a few hard handoffs to accommodate incoming calls can result in significant gains in performance and resiliency.

Rerouting has been used in traffic management to perform dynamic load balancing. The aim of rerouting is to reassign the path/bandwidth allocations of current traffic trunks in a network in order to minimize the probability of blocking future resource requests. We investigate how the effectiveness of rerouting can be affected by the characteristic of the underlying network topology. We established baseline measures through two resource allocation algorithms: a shortest distance path algorithm (SDP), that represents the best common practice without rerouting, and a global rerouting algorithm that is based on a provably /spl epsiv/-optimal algorithm for the multi-commodity flow problem. We propose two rerouting algorithms based on the basic SDP algorithm that selects for rerouting either from traffic trunks with the same source-destination pairs (local rerouting) or from all traffic trunks (global rerouting). The effectiveness of rerouting is highly related to the average node degree. As the connectivity of a graph increases, rerouting tends to be more effective. However, rerouting does not always perform better when connectivity is increased. Significant performance improvement only occurs within a relatively small range of connectivities when a rerouting algorithm is able to find alternative paths and SDP cannot. Furthermore, local rerouting is sufficient to exploit most of the benefits of rerouting and it is not necessary to utilize much more computationally intensive global rerouting algorithms. Finally, we investigate the rerouting frequency vs. blocking trade-off and show that for local rerouting, the best performance can be achieved by a rerouting frequency of only 30%.

In this paper, we present an approach for MANET routing based on stability and hop-count, where the stability metric considered is the residual lifetime of a link. We view stability based routing not as a separate routing protocol but as an enhancement to a hop-count based routing protocol (e.g. DSR or AODV), so that the expected residual lifetime as well as hop count of a route are taken into account. First, we investigate how residual link lifetime is affected by parameters such as speed and mobility pattern using simulation. The result shows that residual link lifetime is a function of current link age, mobility speed and mobility pattern and does not vary monotonically with age. Therefore, intuitive idea such as older links are more stable, which is used in existing stability-based routing algorithms like associativity based routing (ABR), does not hold across a large spectrum of mobility speeds and models. Instead, the reverse can be true. We propose stability and hop-count based routing algorithm (SHARC) using DSR as the basic routing protocol. The stability of a path is calculated using a simple histogram based estimator. Simulation results show that SHARC performs better than a hop-count only algorithm (DSR) and a stability only algorithm for both throughout of long-lived TCP and response time of short data transfer. SHARC also performs close to an algorithm with perfect knowledge of link residual lifetime in many cases.

In third generation (3G) cellular networks, base stations are connected to base station controllers by point-to-point (usually T1/E1) links. However, today's T1/E1 based buck haul network is not a good match for next generation wireless networks because symmetric T1s is not an efficient way to carry bursty and asymmetric data traffic. In this paper, we propose designing an IEEE 802.16-based wireless radio access network to carry the traffic from the base station to the radio network controller. 802.16 has several characteristics that make it a better match for 3G radio access networks including its support for time division duplex mode that supports asymmetry efficiently. In this paper, we tackle the following question; given a layout of base stations and base station controllers, how do we design the topology of the 802.16 radio access network connecting the base stations to the base station controller that minimizes the number of 802.16 links used while meeting the expected demands of traffic from/to the base stations? We make three contributions: we first show that finding the optimal solution to the problem is NP-hard. We then provide heuristics that perform close to the optimal solution. Finally, we address the reliability issue of failure of 802.16 links or nodes by designing algorithms to create topologies that can handle single failures effectively.

Distributed denial of service (DDoS) attacks pose a serious threat to service availability of the victim network by severely degrading its performance. Recently, there has been significant interest in the use of statistical-based filtering to defend against and mitigate the effect of DDoS attacks. Under this approach, packet statistics are monitored to classify normal and abnormal behaviour. Under attack, packets that are classified as abnormal are dropped by the filter that guards the victim network. We study the effectiveness of DDoS attacks on such statistical-based filtering in a general context where the attackers are "smart". We first give an optimal policy for the filter when the statistical behaviours of both the attackers and the filter are static. We next consider cases where both the attacker and the filter can dynamically change their behaviour, possibly depending on the perceived behaviour of the other party. We observe that while an adaptive filter can effectively defend against a static attacker, the filter can perform much worse if the attacker is more dynamic than perceived.

TCP, perhaps the most widely used transport protocol, was designed for wired links and stationary hosts. But more and more links with different characteristics are used to access the Internet. The variations in link characteristics present different problems to TCP and enhancement of TCP performance over these links remains an active area of research. Since the current core network of the Internet is composed mainly of optical fiber links and high speed routers, the characteristics of the access link (or the last link) usually dominate the characteristics of the end-to-end network path. As a result, if the client can acquire the characteristics of the access link (e.g. through measurement), the TCP server can utilize heuristics that exploit the access link's characteristics in order to improve the performance of TCP. In this paper, we propose and evaluate the performance of several heuristics that exploit knowledge of access link characteristics such as bandwidth and packet error rate. Simulation results show that these heuristics can indeed improve TCP performance significantly and demonstrate the potential benefits of using heuristics that adjust TCP congestion control behaviors to match the access link characteristics.

IP-based radio access networks (RAN) are expected to be the next generation access networks in UMTS and CDMA networks. There are several benefits of IP-based RAN including lower costs, flexibility of merging wired and wireless networks, and network scalability and reliability. While quality of service issues in IP-based RAN have been addressed by a number of researchers, the question of connectivity, i.e., how best to connect base stations to the radio network controllers (RNC) in an IP-based RAN, has not been addressed by any research literature. Furthermore, given a connection configuration, an RNC selection algorithm that assigns an incoming call to an RNC is also necessary. This paper examines radio access network (RAN) connectivity and its impact on the performance and resiliency of the wireless network using different RNC selection algorithms. The proposed min-load-1 algorithm, which allows at most one hard hand off in order to accommodate each incoming request, delivers performance close to the standard Min-Load algorithm using a RAN of much higher connectivity and is close to the optimal algorithm using the same RAN. We also find that using min-load-1 algorithm and allowing the base stations to connect to two RNCs result in resiliency to RNC failures that is similar to having full-mesh connectivity between base stations and RNCs

As third generation (3G) wireless networks with high data rate get widely deployed, optimizing TCP performance over these networks would have a broad and significant impact on data application performance. One of the biggest challenges in optimizing TCP performance over the 3G wireless networks is adapting to the significant delay and rate variations over the wireless channel. In this paper, we make two main contributions. First, we present a window regulator algorithm that uses the receiver window field in the acknowledgment packets to convey the instantaneous wireless channel conditions to the TCP source and an ack buffer to absorb the channel variations, thereby maximizing long-lived TCP performance. It improves the performance of TCP sack by up to 100% over a simple drop-tail algorithm for small buffer sizes at the congested router. Second, we present a wireless channel and TCP-aware scheduling and buffer sharing algorithm that reduces the latency of short TCP flows by up to 90% while still exploiting user diversity, thus allowing the wireless channel to be utilized efficiently.

This paper studies the effect of rerouting for augmenting aggregate based resource allocation in the trade-off between overhead and utilization. Aggregation is a common approach to address the scalability issue in resource allocation. However, resources committed in bulk may be under utilized while other resource requests are being turned down for lack of resources in some shared links. The aim of rerouting is to free up committed resources for better utilization by reusing resources vacated by terminated flows and by moving existing flows to alternative paths. Our results show that rerouting improves performance over a wide range of network loads on two different network topologies. In particular, we show that depending on the network load and topology, it is possible to reduce both blocking rate and routing cost.

Wireless link losses result in poor TCP throughput since losses are perceived as congestion by TCP, resulting in source throttling. In order to mitigate this effect, 3G wireless link designers have augmented their system with extensive local retransmission mechanisms. In addition, in order to increase throughput, intelligent channel state based scheduling have also been introduced. While these mechanisms have reduced the impact of losses on TCP throughput and improved the channel utilization, these gains have come at the expense of increased delay and rate variability. In this paper, we comprehensively evaluate the impact of variable rate and variable delay on long-lived TCP performance. We propose a model to explain and predict TCP's throughput over a link with variable rate and/or delay. We also propose a network-based solution called Ack Regulator that mitigates the effect of variable rate and/or delay without significantly increasing the round trip time, while improving TCP performance by up to 40%.

Network monitoring is an integral part of any network management system. In order to ensure end-to-end service quality stated in service level agreements (SLAs), managers of a service provider network need to gather quality-of-service (QoS) measurements from multiple nodes in the network. For a large network with over thousands of flows with end-to-end SLAs, the information exchanged between network nodes and a central network management system (NMS) could be substantial. We propose a mechanism called aggregation and refinement based monitoring (ARM) to reduce the amount of information exchange. ARM is a generic mechanism that can be configured to run with different objectives, including threshold-based, rank-based and percentile-based. The mechanism enables the NMS to collect data from network nodes using a dynamic QoS data aggregation/refinement technique, and to process these information differently depending on its measurement objective. Our simulation results show that for these various objectives, the selective refinement process is able to validate SLAs quickly, is an order of magnitude more efficient than a simple polling scheme, and performs well across a wide range of traffic loads.

In order to detect violations of end-to-end service level agreements (SLA) and to isolate trouble links and nodes based on a unified framework, managers of a service provider network need to gather quality of service (QoS) measurements from multiple nodes in the network. For a network carrying over thousands of flows with end-to-end SLAs, the information exchanged between network nodes and a central network management system (NMS) could be substantial. Moreover in situations where only, a small number of flows violate their respective SLAs, simple polling mechanisms can lead to huge unnecessary overhead in identifying these ill-behaved flows. We propose an algorithm called (ARM) (Aggregation and Refinement based Monitoring) to reduce the amount of information exchange. (ARM) uses a histogram-based dynamic QoS data aggregation/refinement technique at each network node and a reasoning engine at the NMS to minimized the amount of data exchange between network nodes and NMS. (ARM) not only reduces unnecessary reporting through selective refinement, it also performs well across a wide range of traffic loads. Our simulation results show that (ARM) is at least an order of magnitude more efficient than a simple polling scheme. It also outperforms two centralized highly optimized schemes that cannot be implemented in practice.

We propose and study a new technique, which we call cache-based compaction for reducing the latency of Web browsing over a slow link. The compaction technique trades computation for bandwidth. The key observation is that an object can be coded in a highly compact form for transfer if similar objects that have been transferred earlier can be used as references. The contributions of this paper are: (1) an efficient selection algorithm for selecting similar objects as references, and (2) an encoding/decoding algorithm that reduces the size of a Web object by exploiting its similarities with the reference objects. We verify the efficacy of our proposal through detailed experimental evaluations. This compaction technique significantly generalizes previous work on optimizing Web transfer using compression or differencing, and provides a systematic foundation that ties together caching, compression and prefetching.

In this paper, we study the application of a new technique, which we call cache-based compaction for reducing the latency of email transfer over a slow link. Our compaction technique trades computation for bandwidth. The key observation is that an object can be coded in a highly compact form for transfer if similar objects that have been transferred earlier can be used as references. The compaction algorithm has two components: (1) an efficient selection algorithm for selecting similar objects as references, and (2) an encoding/decoding algorithm that reduces the transfer size of an object by exploiting its similarities with a set of reference objects. Depending on the target applications, different instances of compaction algorithms can be derived. In this paper, an instance of the compaction algorithm for optimizing email transfer is presented. Our compaction technique significantly generalizes previous framework on optimizing data transfer using caching, differencing and compression.

ATM switching platforms are well suited for transporting multimedia streams with quality-of-service (QOS) requirements. This paper describes the system design of a high performance connection management system for xbind, a flexible open programmable signaling system for ATM switching platforms. The latency and throughput of call processing is improved by caching, message aggregation, and processing of requests in parallel. Using a set of general purpose UNIX work stations, we are able to attain a maximum throughput of close to 600000 call operations/h (setup and delete) with an average call setup time of 85 ms. With a low traffic load of 3600 call operations/h, an average call setup latency of 11 ms can be obtained. The system is adaptive. By adjusting various control parameters, the connection manager(s) can be dynamically configured to trade off between throughput and call setup time.

Outlines a methodology for developing network control systems which allows for an evaluation of the dynamic behavior and overall performance at an early stage of the development process. Our approach is to build a software prototype which is designed according to the architecture under consideration and runs the intended control algorithms. The functional and dynamic properties of this prototype are tested and evaluated on an emulation platform that we built for this purpose. By providing support for real-time visualization and interactive emulation, this platform can be used to study multimedia networks in various scenarios, such as different load patterns, network sizes and management operations. The current implementation runs on a KSR-1 and an SP2 parallel processor, which are connected to a graphics workstation via ATM links. We use the platform in several projects, one of which aims at developing an architecture for managing multimedia network services.

Emerging ATM-based virtual private network (VPN) services offer customers a flexible way to interconnect customer premises networks (CPNs) via high-speed links. Compared with traditional leased lines, these services allow for rapid provisioning of VPN bandwidth through cooperative control between customer and provider. Customers can dynamically renegotiate the VPN bandwidth according to their current needs, paying only for the resources they actually use. In order to meet the various requirements and demands of different classes of VPN customers, a VPN provider must provide customers with the flexibility to choose their own control schemes and objectives. The focus of this paper is on enhancing the customer's capability of controlling a VPN. First, we propose a new scheme for a broadband VPN service, which is based on the virtual path group (VPG) concept. In our scheme, the customer performs VP control operations without interacting with the VPN provider, thus enabling the following merits: (1) the customer can share bandwidth among VPs that traverse the same physical network link in the provider's domain, thus using the VPN bandwidth more efficiently; (2) customers can perform VP control operations according to their own requirements and control objectives. Second, we outline an architecture for a customer-operated control system, which utilizes a VPG-based VPN service. The system is structured into three layers of control, which execute on different time scales. The functionalities of these layers are call processing, VP control, and VPN control, respectively. Finally, we evaluate the effectiveness of the control system, with respect to VP control.

No abstract available.