LoRa networks enable battery-powered IoT devices to deliver data over kilometre distances, yet their chirp-spread-spectrum links collapse once the gateway’s signal-to-noise ratio (SNR) falls below the dechirp threshold. We present LoRaFlow—a gateway-side, rectified-flow diffusion model that reconstructs the full complex LoRa waveform from heavily corrupted IQ samples, allowing reliable demodulation at SNR several decibels lower than the conventional limit. LoRaFlow combines a lightweight DiT backbone with straight-line rectified trajectories and is trained mainly on procedurally generated chirps plus a single real sample per class. Across –40 to –10 dB, it reduces symbol-error rate by up to 40% relative to standard dechirp and achieves roughly one-third fewer errors than the recent NELoRa decoder, while using one model for all spreading factors and demanding no firmware changes on end devices. By recovering packets that would otherwise be lost, LoRaFlow extends communication range, curtails retransmissions, and lengthens sensor lifetime, offering a practical drop-in upgrade for existing LoRaWAN gateways and illustrating how diffusion models can bolster other low-power wide-area networks.

Early-exit models improve efficiency by allowing inputs to terminate inference at shallow layers, while class-aware variants further prioritize critical classes for faster processing. Yet existing solutions assume centralized training and cannot accommodate distributed settings where devices may have distinct class priorities. We introduce CAFED (Class-Aware Federated Early-Exit Network), the first framework that integrates class-aware early exits into federated learning. CAFED combines clustered aggregation with layer freezing: shared lower layers capture general features, while personalized upper layers host exits aligned with local priorities. Experiments show that CAFED accelerates inference for prioritized classes by up to 40% and improves accuracy by up to 30% over conventional federated and early-exit baselines, establishing a new path for privacy-preserving collaborative learning with class prioritization.

Reliable data transmission in noisy environments poses a significant challenge, where retransmissions are costly in terms of bandwidth, time, and energy. Traditional error detection methods, such as Cyclic Redundancy Check (CRC) and Low-Density Parity Check (LDPC) codes, focus on bit-level accuracy and retransmit all corrupted packets regardless of task relevance. In this paper, we focus on semantic communication, where not all bits contribute equally to the end task. To make this concrete, we study image transmission, where certain regions carry higher semantic importance than others. We leverage DI-NOv2 attention maps and Vision Transformers to identify high-importance regions and employ reinforcement learning to optimize retransmission policies accordingly. This enables selective retransmissions guided by semantic value rather than uniform error detection. Experiments on CIFAR-100 with simulated span corruption show that our framework reduces retransmissions from over 90% with CRC to below 25%, while preserving high classification accuracy above 89%. These findings highlight the potential of Bit Error Tolerance communication systems for efficient and task-aware transmission in noisy environments.

As the integration of edge and IoT devices continues to surge, the need for streamlined machine learning solutions, notably Deep Neural Networks (DNNs), becomes paramount. However, the inherent computational demands and significant memory requirements of DNNs present hurdles for seamless edge integration. Early-Exit Networks (EENs) have emerged as a solution by adding early exits to DNNs, allowing for early inference and facilitating dynamic deployment on devices with varying capabilities. Despite their advantages, current EEN models uniformly handle samples from all classes, which is suboptimal in many edge scenarios where different classes require varying processing speeds and response times. Our research introduces RACENet, a novel architecture with adaptive early-exit capabilities that adjusts class prioritization in real-time. This flexibility enables RACENet to respond to evolving class priorities during runtime. Extensive evaluations on vision tasks, and network intrusion detection demonstrate that RACENet maintains accuracy while dynamically managing class priorities and accelerating inference for high-priority classes, all without adding additional computational burdens on constrained edge devices.

This paper presents a study on applying on-device machine learning (ML) algorithms to enhance MAC layer protocols in wireless communications. It focuses on the MU-MIMO Grouping algorithm and explores the benefits of executing ML models directly on devices such as computers, smartphones, and IoT devices. This approach promises improved speed, privacy, security, and adaptability in dynamic networks. The paper evaluates the effectiveness of this strategy in Wi-Fi and Mas-sive MIMO scenarios, demonstrating significant system capacity enhancement, latency reduction, and improved user experience. Additionally, it examines the interaction between on-device ML and changing network environments, underscoring the method's adaptability and robustness. This research represents a significant advancement in MAC layer protocols using on-device ML and may inspire future innovations in wireless networks.

Edge-based and IoT devices have seen phenomenal growth in recent years, driven by the surge in demand for emerging applications that leverage machine learning models, such as Deep Neural Networks (DNNs). However, a primary drawback of DNNs is their substantial storage/memory needs and high computational overhead, making their adoption in edge devices challenging. This limitation prompted the development of early-exit models like BranchyNet, which enable decisions to be made at earlier stages by incorporating dedicated exits within the architecture’s inner layers. Nonetheless, these existing early-exit models lack control over the specific class that should exit and when. The necessity for such class-aware models is evident in numerous edge applications, where particular high-priority classes must be detected earlier due to their time-sensitive nature. In this paper, we introduce ClassyNet, the first early-exit architecture designed to return only selected classes at each exit. This feature facilitates faster inference times for critical classes, allowing the initial layers to operate on edge devices. This strategy conserves considerable computational time and resources on the edge without compromising accuracy. Through extensive experiments, we show the effectiveness of ClassyNet compared to other models under various scenarios.

The exponential growth of IoT and edge devices has led to their widespread use across various applications. However, the security of these devices remains a significant concern due to their vulnerability to a broad spectrum of cyber-attacks. Network Intrusion Detection Systems (NIDS) are crucial for identifying and mitigating such threats. Traditional NIDS approaches, while effective, struggle to detect sophisticated modern attacks and often require substantial computational power and memory, which may not be feasible for edge devices. Machine learning and neural network-based methods have demonstrated promising improvements in NIDS detection accuracy. Yet, their deployment on resource-constrained edge devices presents a challenge. This has led to the development of Dynamic Neural Networks, an approach that allows models to adapt according to the input, making them more efficient and lightweight. However, these networks are class-agnostic, rendering them unsuitable for handling cases with uneven classification priorities. In this paper, we introduce ClassyNet, a platform designed for efficient, classaware NIDS on edge devices. ClassyNet leverages class-specific feature extraction and a class-specific neural network architecture to enhance intrusion detection efficiency. Experimental results indicate that our proposed approach matches the detection accuracy of traditional machine learning and neural network-based methods while significantly improving resource efficiency.

Edge-based IoT devices have experienced phenomenal growth in recent years due to rapidly increasing demand in various emerging applications which typically utilize machine learning (ML) models such as Deep Neural Network (DNN) and demand low latency and low power consumption. To support the edge requirements, ML models have to support faster inference and less computation. Dynamic DNNs (D2NN) have been proposed to support low-latency and power-saving on edge devices by enabling conditional computations and context dependant activation of the network model for inference; saving computational time and edge resources, hence they are becoming popular for edge applications. In this paper, we show that D2NN are vulnerable to our novel adversarial attack, Dynamic DNN Adversarial attacks (DDAS). Unlike conventional adversarial attacks that target classification accuracy, DDAS targets the IoT device resources such as the battery, latency, and so on. We show that our attack is effective under various attack scenarios with a high attack success rate. We also provide a retraining scheme as a countermeasure to DDAS and show its effectiveness.

With the emergence of machine learning technology that supports the development of synthetic models, many new use cases and challenges are emerging in the fields of computer vision and security. The main model behind this technology is Generative Adversarial Networks (GANs), with their ability to model unknown distributions accurately and perform well in generating synthetic data such as images and videos. However, the application of this technology by the networking community has been lacking. Given this motivation, we introduce MirageNet; our vision for a GAN-based synthetic network traffic generation framework, which can automatically create synthetic network models of protocols, applications, and devices. With the potential to build many applications for privacy, security, and network optimization. In this paper, we present MiragePkt; the first component of MirageNet. It is a GAN-based model to synthetically generate network packets. We describe the different challenges, limitations, and solutions for generating synthetic network packets. Finally, we validate and evaluate the performance of our framework with the synthesizing DNS packets.

In this paper, we propose a novel communication framework, Acoustic-WiFi, that utilizes the acoustic channel (i.e., microphone/speaker) as a control channel to develop more efficient Wi-Fi networks. We used the Acoustic-WiFi framework to develop a Power Saving Mode (PSM) scheme;A2PSM, and a smart contention resolution scheme;Harmony. InA2PSM, we propose an acoustic channel assisted power saving scheme for the Wi-Fi interface, which addresses the inefficiency of the existing power saving schemes in smart devices. In this scheme, we leverage the low power consumption of the acoustic interfaces to reduce the wakeup events of the Wi-Fi interface when it is in power saving mode. We develop a small-scale prototype testbed on real smartphones to evaluate the proposedA2PSMscheme. Experiments show thatA2PSMcould save up to more than 25% more power than the existing schemes. InHarmony, we address the overhead of the Wi-Fi backoff scheme (i.e., contention window countdown, DIFS) and reduce the overall collisions among the devices.Harmonyuses the acoustic channel for contention resolution in Wi-Fi networks. We also evaluateHarmonyusing real testbed and simulation. Experiments using the testbed show a 40% gain in throughput over the 802.11 Wi-Fi backoff scheme, while simulation results show a gain of more than 27% for dense networks.

MU-MIMO is a high-speed technique in IEEE 802.11ac and upcoming 802.11ax technologies that improves spectral efficiency by allowing concurrent communication between one Access Point and multiple users. In this paper, we present MuVIS, a novel framework that proposes MU-MIMO-aware optimization for multi-user multimedia applications over IEEE 802.11ac/ax. Taking a cross-layer approach, MuVIS first optimizes the MU-MIMO user group selection for the users with the same characteristics in the PHY/MAC layer. It then optimizes the video bitrate for each group accordingly. We present our design and its evaluation on smartphones and laptops over 802.11ac WiFi.

In this paper, we present LATTE, a novel framework that proposes MU-MIMO group selection optimization for multi-user video streaming over IEEE 802.11ac. Taking a cross-layer approach, LATTE first optimizes the MU-MIMO user group selection for the users with the same characteristics in the PHY/MAC layer. It then optimizes the video bitrate for each group accordingly. We present our design and its evaluation on smartphones over 802.11ac WiFi.

Analysis of the human gait is used in many applications such as medicine, sports, indoor localization, and person identification. Several research studies focus on the use of MEMS inertial sensors for gait analysis and have shown promising results. Step detection and step length estimation are two basic and important gait analysis tasks. Although researchers have proposed many methods for step detection, all of them rely on experimental thresholds selected based on a limited number of subjects and walking conditions. Also, most of these methods do not distinguish walking from other activities; they can only recognize motion state from an idle state. On the other hand, step length estimation methods used in the literature either need constant calibration for each user, rely on impractical sensor placement, or both. In this paper, the human walking bipedal nature is employed for gait analysis using two waist mounted MEMS gyroscopes. This setup allows the step detection and discrimination from other non-bipedal activities without the need for magnitude thresholds. The hip rotation angle in the sagittal plane is also calculated that allowed us to estimate the step length without needing for constants calibration. By mounting an accelerometer on the center of the back of the waist, an auto-calibration method for the Weinberg equation's constant is developed. This method also improves the accuracy of the Weinberg equation for step length estimation.

With the current significant penetration of mobile devices (i.e. smartphones and tablets) and the tremendous increase in the number of the corresponding mobile applications, they have become an indispensable part of our lives. Nowadays, there is a significant growth in the number of sensitive applications such as personal health applications, personal financial applications, home monitoring applications, etc. In addition, with the significant growth of Internet-of-Things (IoT) devices, smartphones and the corresponding applications are widely considered as the Internet gateways for these devices. Mobile devices mostly use wireless LANs (WLANs) (i.e., WiFi networks) as the prominent network interface to the Internet. However, due to the broadcast nature of WiFi links, wireless traffics are exposed to any eavesdropping adversary within the WLAN. Despite WiFi encryption, studies show that application usage information could be inferred from the encrypted wireless traffic. The leakage of this sensitive information is very serious issue that will significantly impact users' privacy and security. In addressing this privacy concern, we design and develop a lightweight programmable privacy framework, called PrivacyGuard. PrivacyGuard is inspired by the vision of pushing the Software Defined Network (SDN)-like paradigm all the way to wireless network edge, is designed to support of adopting privacy preserving policies to protect the wireless communication of the sensitive applications. In this paper, we demonstrate and evaluate a prototype of PrivacyGuard framework on Android devices showing the flexibility and efficiency of the framework.

Nowadays, transportation network management and efficiency enhancement have become a necessity due to the rapidly increasing demand for transportation. The availability of vehicle trajectories is essential to draw a complete picture of traffic flow and investigate traffic dynamics in order to improve transportation network performance. Obtaining this information using the conventional Intelligent Transportation Systems (ITS) is expensive due to their complexity and rigidity. Over the last decade, the widely spread mobile devices, as well as vehicles, become heavily equipped with wireless communication technologies such as Bluetooth motivate researchers to utilize these technologies in addressing several transportation challenges and developing novel services. In this paper, we present BlueMap - a cost- effective Bluetooth based vehicle trajectory reconstruction system, in which we utilize the advantages of Bluetooth and the mobility of compact computing hardware to provide mobile data gathering units carried on public vehicles to collect Bluetooth data from in-vicinity vehicles in order to identify traveled vehicles trajectories. The evaluation shows the ability of BlueMap to accurately reconstruct the vehicle trajectories.

This paper introduces inLaneCom - a framework to facilitate message exchange between vehicles on the same lane. Given the GPS infeasibility for a lane-level accuracy, inLaneCom utilizes smartphone's inertial sensors to generate a unique lane signature used as lane ID in the exchanged messages. Lane signature consists of combined features extracted from phone's inertial sensors. Two main features types were exploited: physical features and traffic features. Physical features capture the characteristics of the lane surface such as bumps, potholes, and energy. Traffic features capture traffic characteristics such as speed, number of stops, waiting time per stop, acceleration, and deceleration. inLaneCom was evaluated under various road conditions, on highways and urban roads, using different types of vehicles. To the best of our knowledge, inLaneCom is the first framework to enable communication between vehicles on the same lane. This framework would enable several safety and navigation applications that require data exchange between vehicles on the same lane.

We present FlexStream, a programmable framework realized by implementing Software-Defined Networking (SDN) functionality on end devices. FlexStream exploits the benefits of both centralized and distributed components to achieve dynamic management of end devices, as required and in accordance with specified policies. We evaluate FlexStream on one example use case -- the adaptive video streaming, where bandwidth control is employed to drive selection of video bitrates, improve stability and increase robustness against background traffic. When applied to competing streaming clients, FlexStream reduces bitrate switching by 81%, stall duration by 92%, and startup delay by 44%, while improving fairness among players. In addition, we report the first implementation of SDN-based control in Android devices running in real Wi-Fi and live cellular networks.

Edge computing is an increasingly popular paradigm, wherein computation comes closer to the sources of data. A key challenge for edge computing is discovering and utilizing the heterogeneous resources of the vast number of mobile devices at the network edge. Mobile edge devices hide behind private networks, they are mobile and their owners hesitate to share them due to privacy considerations. We propose Kinaara a framework for discovering and allocating collective resources for edge computing applications. First, Kinaara uses a multi-tier architecture to organize geographically proximal edge devices in clusters and provide their resources via trusted mediator entities. Second, it uses a novel resource representation to name and encode devices in terms of the resources advertised by their users. Third, it uses a novel distributed indexing scheme to organize devices in a proximal cluster on a ring logical structure based on their resource similarity. Our extensive experimental evaluation on a large Wi-Fi mobility dataset of 150K users and 345 APs show that compared to existing approaches Kinaara reduces resource discovery overhead by up to 70% and is robust to device mobility.

In heavily congested urban areas, the rapid growth of population is becoming more and more of an issue. Affected cities quickly demand solutions to areas such as quality of life, waste management, public transportation, and accessibility to main resources. However, since the number of impacted areas of population growth is endless, we focus on public parking. Studies show that drivers spend a large portion of their travel time locating vacant parking spots. For this reason, we present MagnoPark, a crowdsourced approach to identifying unoccupied spots accessible to the general public, who are typically free. MagnoPark is a smartphone based sensing solution that detects empty parking spots using internal sensors of cellphones. While a pedestrian is walking on the sidewalk, we exploit magnetometer changes near metal objects in identifying where cars are located. The amplitude and rate of change shift dramatically when approaching or passing cars that are parked beside the street, giving us a great platform towards solving the defined problem. With empirical evaluation, we show that not only is our solution a notable step towards economical parking management but also significantly efficient and as accurate as traditional sensor- based parking solutions.

Using smartphones in intelligent transportation systems (ITS) is gaining an increasing interest among researchers and developers. Typically, the set of sensors that come with smartphones are utilized to develop tools and services in order to enhance safety and driving experience. GPS, cameras, Bluetooth and inertial sensors are used to detect and analyze drivers' behavior and vehicles' motion. Given that inertial sensors reading are in reference to the device local coordinate system, one of the main challenges in processing the inertial sensors in vehicles is to translate the sensors reading values to the vehicle movement context (i.e., vehicle coordinate system). In this work we develop and evaluate a framework to map the inertial sensors readings from the device coordinate to the vehicle coordinate. The proposed framework in this paper performs better than the existing methods and it enhances the accuracy of tracking and analyzing various vehicle dynamics such as vehicle's stops, lane changes and accurate vehicle speed calculation that, in turn, it will enable development of new ITS applications and services.

In wireless network edges, knowing the network flow types and applications can enable various policy-driven network managements (i.e. traffic offloading, BYOD, E2E QoS etc.). However, applying network policies at wireless links between the mobile devices and access points (APs) requires greater visibility and control on generated traffic generated from mobile devices. The recent advent of Software-Defined Networking (SDN) could enable fine-grained network management at the edge. However, in existing solutions, SDN uses external Deep-Packet Inspection (DPI) engine that requires additional and potentially heavy loaded computational resources to perform the packet analysis. Moreover, mobile applications become more dynamic (rapid install/update), diverse and complex (individual applications generate multiple traffic types) in which scalability and granularity requirements challenging current DPI solutions. In addition, DPI is unreliable in classifying application's encrypted packets. Therefore, in this paper we present the design and the development of TrafficVision that extends the SDN's layer architecture to have fine-grained and real-time policy making at wireless network edges. More specifically, we carefully extends the SDN framework to develop tools that allow to have scalable, efficient and flexible way to classify the network traffic flows at fine-grained fashion using Machine-Learning (ML) based technique. We evaluate our system using the performance of CPU utilization, network overhead and network throughput metrics. Finally, as a proof of concept, we develop a simple case study of traffic management application that exploits TrafficVision.

Insufficient intersection management is one of the top congestion causes. Intersection performance enhancement requires its usage data and statistics, such as vehicle queue lengths, waiting times, etc. Intelligent Transportation Systems (ITS) provide transportation researchers and engineers with access to such data. However, the majority of the state of the art ITS are expensive, hard to maintain or inadequate. Over the last decade, the widely spread mobile devices and vehicles have become heavily equipped with wireless communication technologies like Bluetooth, which is broadly adopted to address variety of challenges due to its popularity and simplicity. Yet, Bluetooth utilization in ITS is limited. In this paper, we present itsBlue, a Bluetooth-based framework that utilizes Bluetooth advantages to collect Bluetooth data from off-the-shelf road users devices and extract their features to provide ITS applications with required statistics and information. itsBlue is a cost-effective framework for intersection management ITS applications that overcomes recent practices drawbacks such as installation complications and third-party application requirement. The evaluation showed itsBlue ability to provide intersection management ITS applications with reliable information.

We are experiencing huge growth of video streaming traffic, which is creating big challenges for video providers in guaranteeing a satisfactory level of viewing experience to end users. Furthermore, the increase in video streaming demands on mobile devices over dynamic wireless links is creating another obstacle toward providing a high quality video service. In order to overcome most of these challenges, HTTPs adaptive video streaming technology was introduced, along with other great features for streaming videos. However, we found that HTTPs adaptive protocol can still suffer under certain situations and conditions. Mostly, these issues are likely experienced when multiple concurrent players compete over the same bottleneck. Several studies have proposed a network side solution at the home gateway or at the cloud aiming to assist the video players to maximize the viewing experience to all users sharing the same bottleneck. Although these proposed systems could provide some enhancements to the video streaming, they are unable to provide fine-grained monitoring and understanding of the video traffic to apply desire level of dynamic resource management. Considering the above issues, in this paper we conduct extensive analysis of the video traffic of YouTube; the most popular HTTPs adaptive video player. In our study, we argue that through a deep understanding and careful analysis of the HTTPs video traffic, valuable information about the competing streams can be obtained and could be utilized in developing a network based solution that can significantly improve the video QoE and assist the video players to perform much better.

In this paper, we utilize a novel communication framework Acoustic-WiFi to develop a smart contention resolution scheme Harmony among the contending devices to address the overhead of the traditional Wi-Fi backoff scheme (i.e. contention window countdown, DIFS) and reduce the overall collisions among the devices. Harmony uses the acoustic channel for contention resolution in Wi-Fi networks. To the best of our knowledge, Harmony is the first to leverage the acoustic interface on commodity smart devices as an addition control channel in parallel with the Wi-Fi interface. We evaluate our scheme using real testbed and simulation. Testbed experiments show more than 40% throughput gain over traditional Wi-Fi networks, while simulation results show more than 27% gain for dense networks.

In response to the rapid growth of mobile video traffic, major video providers such as YouTube and Netflix have recently adopted HTTP adaptive streaming protocol for video delivery. However, the possible degradation and instability in the perceived video quality are still the major issues with this protocol. Moreover, this protocol can cause unfairness among competing players in addition to bandwidth inefficiency. In fact, these issues are more common on mobile devices due to the instability nature of wireless links. In this paper, we present a dynamic SDN-based traffic shaping technique (DASH-SDN) that aims to address all the aforementioned issues. DASH-SDN is different from the existing shaping techniques in that it utilizes the long OFF periods of the most commercial adaptive video players to increase the throughputs for all traversing streams. Consequently, the video playback buffer can be proactively filled and the possible fluctuation in wireless link throughput can be effectively countered. Furthermore, DASH-SDN takes into consideration, in managing the bandwidth, the heterogeneity of the clients in terms of wireless channel qualities, round trip delays, and battery levels. Our initial results have shown great improvement in video QoE, bandwidth utilization, and power consumption.

In this paper, we propose a simple and practical scheme for optimization the inquiry process in Bluetooth standards in order to achieve fast, reliable and controlled responses. We show how the inquiry process is related to clock setting in Bluetooth devices and how clock optimization could be used to shorten the inquiry process time. Theoretical analysis and simulation results shows the efficiency of the proposed optimized scheme. We also develop a traffic transportation application, i.e. speed measurement, as an example of many applications where this scheme could benefit. Results show how the proposed optimized scheme outperforms in comparison with others.

In this paper, we describe the in-lane communication framework that enables the communication between vehicles on the same lane. The system creates a unique signature for each lane using smartphone inertial sensors. The calculated lane signature will be attached with each broadcasted message, where other vehicles on the street compare it with their own lane signature to decide if the received message came from a vehicle on the same lane or not. The paper concentrates on finding the right features that can be used to identify a lane in a road. To identify a lane with a unique signature, the paper proposes a combination of driving features and road features. The paper uses the inertial sensors because the current GPS systems are not accurate enough to locate vehicles with a lane level accuracy. The ability to establish a communication channel between vehicles on the same lane can enhance the road safety and enable several interesting navigation applications.

Smartphones have created a strong tie with human beings. It is likely to have plenty of smartphones in the same proximity. However, users are limited with the capabilities of a single device. In this work we propose ColPhone a framework that facilitates collaboration between smartdevices in the vicinity to share their resources providing users hardware upgrade for their devices. Further, we implemented an application (3D Story Teller) on top of ColPhone, and measured the impact of collaboration on smartphone's power consumption.

No abstract available.

In this paper, we design, implement and evaluate the SpyLoc localization system. The design goal of SpyLoc is to develop a light weight and high accuracy localization system for off-the-shelf smartphones. SpyLoc leverages both the acoustic interface (microphone/speaker) and the Wi-Fi interface at the kernel-level of smartphones as well as the inertial sensors in smartphones to achieve high localization accuracy. SpyLoc does not require any central controller unit nor any collaboration with nearby devices. Furthermore, in SpyLoc, each user's smartphone works autonomously to estimate its location. We implement and evaluated the complete SpyLoc using commercial off-the-shelf smartphones. Our result shows that SpyLoc can achieve less then 1 meter accuracy for more than 90% of the time for both indoor and outdoor environments.

Mobile devices such as smartphones have a number of sensors that can be exploited to solve a number of problems related to health care delivery. In this paper we use accelerometer, gyroscope, and compass sensors on smart-phones to solve a location tracking problem common to many emergency departments. An emergency department typically has unique characteristics including not being friendly to be visually surveyed, since the layout consists of many isolated islands and the workstation layout is not standardized. The spaghetti diagram is a tool to help to identify areas where time can be saved by visualizing unnecessary movement of products, staff or patients. We report the development of an automated tool to create spaghetti diagrams of movements of personnel in a non-intrusive way. A preliminary prototype to produce paths showed very encouraging results. We also identify challenges and our approach to meet them.

In this paper, we propose and addresses the challenge of designing a light-weight high-accuracy indoor/outdoor localization system (SpyLoc) for off-the-shelf smartphones. In SpyLoc, we want to leverages both the acoustic interface (microphone/speaker) and the Wi-Fi interface at the kernel-level of the smartphones as well as the inertial sensors in the smartphones to achieve high localization accuracy. In the ranging-based approach, we utilizes the RF-Beep ranging scheme[?], and in dead reckoning-based approach, we fuses the inertial sensors of the smartphone to estimate the direction and the distance of user's movements. In RF-Beep [?], we develope a ranging scheme that utilizes the Time-Difference-of-Arrival (TDoA) between the acoustic and the radio-frequency (RF) signal. The RF-Beep basically leverages the slow propagation speed of the acoustic signal with respect to the RF signal to estimate the relative range. The well known acoustic range based localization scheme, Cricket [?] also utilizes the same concept of using the difference in arrival times of concurrent transmissions of radio and ultrasound signals at the target device to infer the distance. Unlike the Cricket, which was designed with special hardware, our localization scheme is applicable to the smartphones. In RF-Beep, we address the different challenges of implementing such ranging scheme in smartphones by leveraging the existing functionalities of the audio driver and the WiFi driver. Further details on RF-Beep scheme could be found in~\cite{Uddin:RF}. The ranging based localization scheme typically requires at least three reference points (e.g., anchor points), in order to calculate the location. However all these three reference points must be in Line of Sight(LoS) to the target device. In \textit{SpyLoc} localization system, we use the combination of both ranging-based and dead reckoning approaches to reduce the constraint of having three LoS anchor/reference points all the time. The basic idea of the SpyLoc is to leverage the benefits of both the dead-reckoning and the ranging scheme to build a practical localization system. Given the high errors of the inertial sensors, SpyLoc uses a novel ranging scheme based on both the acoustic and WiFi interfaces to mitigate this error in order to improve the localization accuracy. Unlike the ranging-based or RF-based localization schemes that require multiple reference points (e.g., achor points), using the dead reckoning in SpyLoc reduces the required number of reference points to only one reference to locate and track users accurately. This low dependency on ranging scheme make SpyLoc practically applicable to high mobility environment. Finally, In the SpyLoc system, user's device (i.e. smartphone) works autonomously to determine its location. This system does not require any coordination from the nearby smartphones or from a central controller. Furthermore, this localization system does not require user's device to transmit any acoustic signal or RF messages to nearby smartphones. Therefore increasing the number of user's devices have no impact on the complexity of the SpyLoc system, which make SpyLoc a light-weight localization system. In addition, such characteristics make the localization system privacy preserving and energy efficient for the user's smartphone.

In this paper, we design, implement and evaluate RF-Beep - a high-accuracy, one-way sensing, energy efficient and light-weight ranging scheme for smart devices. RF-Beep is based on the well known Time-Difference-of-Arrival (TDoA) scheme that utilizes the different propagation speeds of both the acoustic and the radio-frequency (RF) signals. Unlike the previous works, RF-Beep utilizes both the audio interface (i.e., microphone, speaker and sound driver) and the RF interface (i.e., WiFi) at the kernel-level of commercial-off-the-shelf smart devices. Implementing the scheme at lower levels enables us to understand and address the challenges related to the timing uncertainties in transmitting and receiving the acoustic signal. Moreover, RF-Beep does not require any special hardware or infrastructure support. In this paper, we describe the complete implementation of RF-Beep at the kernel space of Linux OS. We evaluate RF-Beep under different indoor and outdoor real scenarios. Results show that the error in the estimated range is less than 50cm for more than 93% of the time.

Instead of traditional content adaptation at servers or clients, we propose context-aware content adaptation at the Access Point (AP). We show that because of the special characteristics of the AP such as: being the last node at the edge of the network, no power constraint, and powerful computing capabilities, it can be the best candidate for context-aware content adaptation in high dynamic wireless mobile networks. In fact, it can adapt content more accurately and faster in a high diverse and dynamic context. Content adaptation at the application and MAC layers of the AP is introduced in this paper.

In this paper we present the design and the evaluation of a framework MagnoTricorder, a system that utilizes the magnetic sensor in smartphones to detect the running devices at home thru a singlepoint sensing. MagnoTricorder leverages the effect of Electro Magnetic Interference (EMI) generated by the AC current in the main power-line at home. This EMI induces a magnetic field that highly fluctuates the reading of the magnetic sensor in smartphones. In this paper, we utilize this characteristic for detecting and identifying the running devices at home thru the Circuit Breaker Panel. Experimental evaluation demonstrates the feasibility of the developed framework. Results show that MangoTricorder can detect and identify individual devices with 93%-98% accuracy.

Tracking energy consumption for individual operating machines (e.g., home appliance) is a prerequisite for making energy conservation and management efficient. In order to meet the requirement for monitoring home energy consumption, several industries and researchers came up with different solutions. Unfortunately, all these solutions require invasive and expensive installation of sensor devices. Furthermore, many of these solutions can't measure the energy consumption of individual machines. In this paper, we propose and evaluate the feasibility of using smart phones in machines's energy monitoring system. We call our system EnergySniffer in which it exploits various sensors, such as magnetic sensor, light, microphone, temperature, camera, WiFi, in smart phones to build a multi sensing framework. This framework is used to build a unique fingerprint profile for each individual machine. As a proof of concept, we develop a simple sensing framework prototype that utilizes only the microphone sensor on the phone. We call this framework a sound sensing framework. Experimental evaluation on sound sensing framework demonstrate the feasibility of continuously identifying and monitoring individual machine in real-time.

Wireless networks which support multiple modulation techniques and data rates suffer from the so-called multi-rate anomaly problem resulting in unfair environments; in which all the users, regardless of their data rate choices, observe the same, low throughput values. This situation violates the baseline fairness principle which states that the desired throughput of a node competing against any nodes running at different speeds is equal to the throughput that the node would achieve in an existing single-rate network in which all the competing nodes were running at its own rate. In our work presented here, we propose a novel algorithm which utilizes the retry limit parameter of the CSMA/CA algorithm in order to differentiate different data rate user groups. Our algorithm, MORAL (Multiuser data Rate Adjustment Algorithm), dynamically adjusts the retry limits of the users according to the channel characteristics observed and the neighbor information gathered from previous communications. MORAL chooses the most appropriate retry limit to use in order to establish a more baseline-fair environment.

In recent years, the enforcement of safety features like seatbelt and reactive automotive technologies like lane assistance system etc. have enabled drivers to drive more safely and effectively. Better safety can be achieved by using proactive technologies to predict driver's intentions ahead of time and inform surrounding drivers of the course of the current action. This work introduces a novel system to predict driver's intention based on Electroencephalography (EEG).

Reputation management systems have been proposed as a cooperation enforcement solution in ad hoc networks. Typically, the functions of reputation management (evaluation, detection, and reaction) are carried out homogeneously across time and space. However, the dynamic nature of ad hoc networks causes node behavior to vary both spatially and temporally due to changes in local and network-wide conditions. When reputation management functions do not adapt to such changes, their effectiveness, measured in terms of accuracy (correct identification of node behavior) and promptness (timely identification of node misbehavior), may be compromised. We propose an adaptive reputation management system that realizes that changes in node behavior may be driven by changes in network conditions and that accommodates such changes by adapting its operating parameters. We introduce a time-slotted approach to allow the evaluation function to quickly and accurately capture changes in node behavior. We show how the duration of an evaluation slot can adapt according to the network's activity to enhance the system accuracy and promptness. We then show how the detection function can utilize a Sequential Probability Ratio Test (SPRT) to distinguish between cooperative and misbehaving neighbors. The SPRT adapts to changes in neighbors' behavior that are a by-product of changing network conditions, by using the node's own behavior as a benchmark. We compare our proposed solution to a nonadaptive system, showing the ability of our system to achieve high accuracy and promptness in dynamic environments. To the best of our knowledge, this is the first work to explore the adaptation of the reputation management functions to changes in network conditions.

Directional antennas have been introduced to improve the performance of 802.11 based wireless networks. Station equipped with directional antennas can beamform data in a specific direction. However, IEEE 802.11 is developed with omni-directional antennas in mind. Thus, deploying IEEE 802.11 in a directional antenna environment leads stations to be conservative in blocking their own transmissions in favor of the ongoing transmissions. In this paper we design and evaluate two opportunistic enhancements for the IEEE 802.11 networks using directional antennas. The first enhancement is to augment the 802.11 MAC protocol with additional information (i.e., location of the stations) that gives a node the flexibility to transmit data while there are ongoing transmissions in its vicinity. The second enhancement, using the augmented MAC protocol, alters the way the 802.11 accessing its MAC data queue. Simulation shows improvement in network throughput up to 40% in case of applying the first enhancement, and up to 60% in case of applying the second enhancement.

Directional antennas have been introduced to improve the performance of 802.11 based wireless networks. Station equipped with directional antennas can beamform data in a specific direction. However, IEEE 802.11 is developed with omni-directional antennas in mind. Thus, deploying IEEE 802.11 in a directional antenna environment leads stations to be conservative in blocking their own transmissions in favor of the ongoing transmissions. In this paper we propose two opportunistic enhancements for the IEEE 802.11 networks using directional antennas. The first enhancement is to augment the 802.11 MAC protocol with additional information (i.e., location of the stations) that gives a node the flexibility to transmit data while there are ongoing transmissions in its vicinity. The second enhancement, using the augmented MAC protocol, alters the way the 802.11 accessing its MAC data queue.

Directional antennas have been introduced to improve the performance of 802.11 based wireless networks. Station equipped with directional antennas can beamform data in a specific direction. However, IEEE 802.11 is developed with omni-directional antennas in mind. Thus, deploying IEEE 802.11 in a directional antenna environment leads stations to be conservative in blocking their own transmissions in favor of the ongoing transmissions. In this paper we study and analyze the limitation of IEEE 802.11 networks with directional antennas. We show analytically that a station with directional antenna and using 802.11 protocol is conservative in terms of assesses channel availability, with as much as 60% of unnecessary blocking assessments. Moreover, by altering the way the 802.11 accessing its MAC data queue, we show that the unnecessary blocking assessments of a node could be increased up to 90%.

Traffic querying, road sensing and mobile content delivery are emerging application domains for vehicular networks whose performance depends on the throughput these networks can sustain. Rate adaptation is one of the key mechanisms at the link layer that determine this performance. Rate adaptation in vehicular networks faces the following key challenges: (1) due to the rapid variations of the link quality caused by fading and mobility at vehicular speeds, the transmission rate must adapt fast in order to be effective, (2) during infrequent and bursty transmission, the rate adaptation scheme must be able to estimate the link quality with few or no packets transmitted in the estimation window, (3) the rate adaptation scheme must distinguish losses due to environment from those due to hidden-station induced collision. Our extensive outdoor experiments show that the existing rate adaptation schemes for 802.11 wireless networks under utilize the link capacity in vehicular environments. In this paper, we design, implement and evaluate CARS, a novel context-aware rate selection algorithm that makes use of context information (e.g. vehicle speed and distance from neighbor) to systematically address the above challenges, while maximizing the link throughput. Our experimental evaluation in real outdoor vehicular environments with different mobility scenarios shows that CARS adapts to changing link conditions at high vehicular speeds faster than existing rate-adaptation algorithms. Our scheme achieves significantly higher throughput, up to 79%, in all the tested scenarios, and is robust to packet loss due to collisions, improving the throughput by up to 256% in the presence of hidden stations.

We propose to use two link adaptation strategies together in order to find energy-efficient routing paths in an ad-hoc network: exploiting different modulation techniques supported by the physical layer and the fragmentation size of the packets. Our proposed mechanism finds the best (modulation technique, fragmentation size) pair for a given link and uses this information for finding the minimum energy routing paths. We provide a modified version of the well known AODV protocol and show that our proposed mechanism performs better than the two mechanisms deployed individually and much better than the original protocol.

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

Recent advances in wireless inter-vehicle communication systems enable the establishment of vehicular ad-hoc networks (VANET) and create significant opportunities for the deployment of a wide variety of applications and services to vehicles. In this work, we investigate the problem of developing services that can provide car drivers with time-sensitive information about traffic conditions and roadside facilities. We introduce the vehicular information transfer protocol (VITP), a location- aware, application-layer, communication protocol designed to support a distributed service infrastructure over vehicular ad- hoc networks. We describe the key design concepts of the VITP protocol and infrastructure. We provide an extensive simulation study of VITP performance on large-scale vehicular networks under realistic highway and city traffic conditions. Our results demonstrate the viability and effectiveness of VITP in providing location-aware services over VANETs.

In this paper, we propose an enhancement to the IEEE 802.11 distributed coordination function (DCF). The enhancement improves the level of channel spatial reuse; thus, it improves overall network data throughput in dense deployments. Our modification, named the location-enhanced DCF (LED), incorporates location information in DCF frame exchange sequences so that stations sharing the communication channel are able to make better interference predictions and blocking assessments. Hence, more concurrent transmissions can be conducted in densely deployed wireless LANs. The potential performance enhancement of LED is studied both analytically and via ns-2 simulations. The results show that the LED method achieves significant throughput improvements over the original DCF.

Ad hoc networks can be used not only as data carriers for mobile devices but also as providers of a new class of services specific to ubiquitous computing environments. Building services in ad hoc networks, however, is challenging due to the rapidly changing operating contexts, which often lead to situations where a node hosting a certain service becomes unsuitable for hosting the service execution any longer. We propose a novel model of service provisioning in ad hoc networks based on the concept of context- aware migratory services. Unlike a regular service that executes always on the same node, a migratory service can migrate to different nodes in the network in order to accomplish its task. The migration is triggered by changes of the operating context, and it occurs transparently to the client application. We designed and implemented a framework for developing migratory services. We built TJam, a proof-of-concept migratory service that predicts traffic jams in a given region of a highway by using only car-to-car short-range wireless communication. The experimental results obtained over an ad hoc network of personal digital assistants (PDAs) show the effectiveness of our approach in the presence of frequent disconnections. We also present simulation results that demonstrate the benefits of migratory services in large-scale networks compared to a statically centralized approach.

VANETs (vehicular ad hoc networks) are emerging as a new network environment for intelligent transportation systems. Many of the applications built for VANETs will depend on the data push communication model, where information is disseminated to a group of vehicles. In this paper, we present a formal model of data dissemination in VANETs and study how VANET characteristics, specifically the bidirectional mobility on well defined paths, affects the performance of data dissemination. We study the data push model in the context of TrafficView, a system we have implemented to disseminate information about the vehicles on the road. Traffic data could be disseminated using vehicles moving on the same direction, vehicles moving in the opposite direction, or vehicles moving in both directions. Our analysis as well as simulation results show that dissemination using only vehicles in the opposite direction increases the data dissemination performance significantly

Abstract Physical topology of an ad hoc wireless network imposes fundamental limits on its throughput capacity. In this work, we present a network design algorithm for configuring node locations in an 802.11 ad hoc network with the goal of improving the throughput capacity of the network. Given a network configuration (i.e., a mapping of network nodes to physical locations), we establish an inverse proportional relationship between the interference degree of the network and the guaranteed link‐throughput. Motivated by this observation, we present an algorithm which progressively guides the network towards better configurations with lower interference degrees, which in turn increases the network's throughput capacity. Packet level simulations using the ns‐2 simulator shows that the reconfigured topologies obtained by our algorithm consistently outperform the original network topologies for throughput and delay‐related metrics. In particular, for many cases, we observed over 100% increase in the total network throughput and the minimum guaranteed throughput, over 60% increase in throughput fairness and over 50% reduction in the mean‐service delay of packets. Our algorithm admits a simple distributed implementation and can be viewed as a distributed mobility control primitive for improving the throughput performance of mobile ad hoc networks. Alternately, our techniques can also be employed by a centralized designer during network creation time to obtain a network configuration with a high throughput capacity. To the best of our knowledge, ours is the first work which explores the use of guided network configuration strategies for improving the throughput capacity in ad hoc wireless networks. Copyright © 2006 John Wiley & Sons, Ltd.

In this paper, we propose an enhancement to the existing IEEE 802.11 distributed coordination function (DCF) MAC to improve channel spatial reuse efficiency, and thus improve overall network data throughput. Our modification, named the location enhanced DCF (LED) for IEEE 802.11, incorporates location information in DCF frame exchange sequences so that stations sharing the communication channel are able to make better interference predictions and blocking assessments. Utilizing an underlying physical layer design that supports frame capture, the LED enhanced interference estimation can increase overall network data throughput by permitting more concurrent transmissions. In this paper we also analytically study the potential performance enhancement of the LED over the original IEEE 802.11 DCF. The results are verified using the ns-2 simulator, which shows that up to 35% of DCF blocking decisions are unnecessary and our LED method can achieve up to 22% more throughput than the original DCF.

Wireless sensor networks have recently gained a lot of attention from the research community. Wireless sensor networks are often partitioned into clusters, each managed by a cluster head (gateway). In this paper, we study the performance of the IEEE 802.11 based wireless sensor networks in noisy environments. We show that using the standard binary exponential backoff (BEB) mechanism in noisy clustered sensor networks results in a poor throughput performance due to its inability of differentiating between the causes of unsuccessful packet transmissions and verify the analytical model using the ns-2 simulator. We show the enhanced BEB mechanism in [T. Nadeem et al. (2004)] enhances clustered sensor network performance up to order of magnitudes with respect to the network error rates (noise level). We also study the noise effect on the fairness of IEEE 802.11 and show that our proposed mechanism maintains the channel fairness between the competing sensors.

EZCab is a proof-of-concept ubiquitous computing application that allows people to book nearby cabs using their cell phones or PDAs equipped with short-range wireless network interfaces. EZCab discovers and books free cabs using mobile ad hoc networks of vehicles. We have implemented an EZCab prototype on top of smart messages, a middleware architecture based on execution migration, which we had developed to provide a common execution environment for outdoor ubiquitous computing applications. The experimental and simulation results have demonstrated the feasibility of EZCab

We define techniques to compute energy-efficient paths, using the IEEE 802.11 fragmentation mechanism, within the framework of on-demand routing protocols. We focus on one specific on-demand routing protocol, namely the ad-hoc on-demand vector routing protocol (AODV), and show how it should be adapted to compute energy-efficient paths. The choice of energy-efficient paths depends on link error rates on different wireless links, which in turn depend on channel noise. We show how our scheme accounts for such channel characteristics in computing such paths and how it exploits the IEEE 802.11 fragmentation mechanism to generate optimum energy-efficient paths. We perform a detailed study of the AODV protocol and our energy-efficient variants, under various noise and node mobility conditions. Our results show that our proposed variants of on-demand routing protocols can achieve orders of magnitude improvement in energy-efficiency of reliable data paths.

Vehicles are part of people's life in modern society, into which more and more high-tech devices are integrated, and a common platform for inter-vehicle communication is necessary to realize an intelligent transportation system supporting safe driving, dynamic route scheduling, emergency message dissemination, and traffic condition monitoring. TrafficView, which is a part of the e-Road project, defines a framework to disseminate and gather information about the vehicles on the road. Using such a system will provide a vehicle driver with road traffic information, which helps driving in situations such as foggy weather, or finding an optimal route in a trip several miles long. This paper describes the basic design of TrafficView and different algorithms used in the system.

We study analytically the performance of the IEEE 802.11 DCF (distributed coordination function) for infrastructure networks in noisy environments. We show that using the standard binary exponential backoff (BEB) mechanism in noisy environments results in a poor throughput performance due to its inability to differentiate between the causes of unsuccessful packet transmissions and verify the analytical model using the ns-2 simulator. We propose an enhanced BEB mechanism that enhances IEEE 802.11 with a capability of differentiating between different types of unsuccessful transmissions. We study the proposed mechanism analytically, verify it using ns-2 and show that the new mechanism enhances the network performance by up to orders of magnitude with respect to the network error rates.

Finding a previously visited page during Web navigation is a very common and important process. Although most commercial browsers incorporate a history mechanism, when accessing pages that were recently visited, users still rely mainly on the "Back" button. In this paper, we studied the effectiveness of visually enhanced history browser mechanisms on Web navigation. We used three different history mechanisms as the experiment treatments. 21 college students were the subjects of the experimental trials. With a between-subject design, three experimental groups were equally divided according to subject profiles. Users read and completed the given scenario. A quiz was given to measure user performance on the tested mechanism. At the end of the quiz, a subjective questionnaire was given to measure user satisfaction. The results showed that there is a significant statistical difference among the three mechanisms. The more visually enhanced history mechanism provided to be more effective in Web browsing.