With the rapid growth of the Internet of Things (IoT), integrating artificial intelligence (AI) on extremely weak embedded devices has garnered significant attention, enabling improved real-time performance and enhanced data privacy. However, the resource limitations of such devices and unreliable network conditions necessitate error-resilient device-edge collaboration systems. Traditional approaches focus on bit-level transmission correctness, which can be inefficient under dynamic channel conditions. In contrast, we propose SemanticNN, a semantic codec that tolerates bit-level errors in pursuit of semantic-level correctness, enabling compressive and resilient collaborative inference offloading under strict computational and communication constraints. It incorporates a Bit Error Rate (BER)-aware decoder that adapts to dynamic channel conditions and a Soft Quantization (SQ)-based encoder to learn compact representations. Building on this architecture, we introduce Feature-augmentation Learning, a novel training strategy that enhances offloading efficiency. To address encoder-decoder capability mismatches from asymmetric resources, we propose XAI-based Asymmetry Compensation to enhance decoding semantic fidelity. We conduct extensive experiments on STM32 using three models and six datasets across image classification and object detection tasks. Experimental results demonstrate that, under varying transmission error rates, SemanticNN significantly reduces feature transmission volume by 56.82–344.83× while maintaining superior inference accuracy.

The rapid evolution of the Web of Things (WoT) has created new opportunities for connectivity and standardization across heterogeneous devices, enabling the development of increasingly complex systems. However, edge devices deployed in resource-constrained scenarios face significant challenges. These devices require lightweight and efficient models to achieve high accuracy while operating within strict memory constraints. Typical approaches to model generation, Neural Architecture Search (NAS), have proven effective in automating the search for optimal architectures. However, existing NAS methods suffer from two critical limitations: (1) they fail to incorporate quantization into the search space, which can result in overlooking larger models that might perform better after quantization; and (2) current model evaluation methods struggle to provide accurate assessments within a short time period. To address these challenges, we propose EdgeGen, a novel NAS framework that integrates multiple quantization methods into the search process, enabling the discovery of larger models that need quantization to satisfy the constraint. EdgeGen employs a multi-beam Monte Carlo Tree Search (MCTS) algorithm and a constraint validator to explore the expanded search space efficiently, searching vast original and quantized models. Furthermore, EdgeGen evaluates the model performance following a GNN-based performance predictor, which provides a rapid and precise prediction. Across multiple benchmarks, EdgeGen consistently outperforms state-of-the-art NAS methods. Code available at: https://doi.org/10.5281/zenodo.18323232.

High-density low-power and lossy networks (LLNs) for IoT applications (e.g., smart cities) urgently require low latency in emergencies like fires. However, existing hash-based or traffic-aware Time Slotted Channel Hopping (TSCH) scheduling methods fall short in such emergency scenarios. These methods lack both dedicated scheduling mechanisms for emergency packets and effective conflict resolution strategies, thereby resulting in frequent collisions and urgent traffic latency that fails to meet response demands. While reinforcement learning (RL)-based schemes can mitigate these issues through multi-objective optimization, their high computational and energy overheads place stringent demands on the capabilities of edge nodes. To address these challenges, We propose ATSE, an adaptive TSCH scheduling for emergency packets in high-density LLNs. TSE employs subtree-segregated time-slot allocation to assign regular and emergency packets to independent subtree resources for precise scheduling, combined with temporal emergency time-slot adjustment to dynamically reallocate slots for non-real-time emergency packets, ensuring low-latency delivery. For conflict avoidance, we design an Nhash-based conflict avoidance method via selective time-slot segment partitioning, allowing conflicting packets to request resources from sinks in dedicated segments, thereby reducing collisions and control overhead. Additionally, ATSE integrates multi-sink load balancing to evenly distribute dynamic network traffic, improving overall transmission performance and resource utilization. We implement ATSE on Contiki OS and evaluate it in large-scale, high-density scenarios. Experimental results show that ATSE outperforms baseline methods, reducing latency by 18.6%, improving reliability by 10.5%, and lowering energy consumption by 10.6%.

Device-cloud collaborative inference is often necessary for resource-constrained IoT devices that cannot support full on-device models. To minimize bandwidth and support concurrency, existing methods typically compress images before transmission. However, these approaches often ignore the significant overhead of decoding native JPEG camera output, especially for high-resolution frames. Our measurements show that the on-device (Raspberry Pi 4B) decoding overhead for 700KB JPEG format is$\sim$14.4x the latency of on-cloud (GeForce RTX 3090) meter recognition inference. To reduce on-device decoding overhead, we design DC Camera, which is built upon a JPEG camera and leverages partial JPEG decoding to efficiently extract DC features from high-resolution images, significantly mitigating on-device image processing overhead. These DC features can preserve structural information better than conventional downsampled images. We utilize DC Camera to implement fast meter recognition system and deploy the system in material science laboratory to monitor multiple meters. Our evaluation demonstrates that compared to state-of-the-art (SOTA) methods, DC Camera can reduce on-device computation overhead by$\sim$5.8x and decrease transmission volume by$\sim$90.9x, without inference accuracy degradation.

Popular apps in the mainstream app markets are installed on millions of devices, attracting malicious actors to steal their income or distributing malware with their repackaged versions. To evade detection, their creators often obfuscate these apps and posing challenge to an effective repackaging detection. Many approaches have been proposed to address the issue. However, their resilience to obfuscation is limited and applies only to a narrow range of techniques. What is worse, a more advanced obfuscation technique have emerged, referred to ascall manipulation. These techniques disrupt the original call structure by introducing new invocations or concealing existing one, significantly undermining the effectiveness of all the existing repackaging detection tools. In this paper, we present RepkHunter, a repackaging detection tool that is obfuscation-resilient against a wide range of obfuscation techniques, including thecall manipulation. Specifically, we employ inter-class method invoking to construct theinter-class method invoking graphas the robust feature of an app, ensuring that it is: (i) independent of identifiers and package structures, and (ii) unaffected by instruction-level changes, such as randomized control flow and removed or inserted instructions. Then, we proposecontext-based call filtering. Specifically, we classify the observed call manipulation methods into two categories and propose two corresponding contexts to differentiate these manipulated calls from original invocations. Leveraging these contexts, we filter out calls introduced by obfuscators and restore the original call structure. Our experiments demonstrate that: (i) RepkHunter is resilient to all obfuscation techniques employed by five widely used obfuscators, outperforming the state-of-art tools. (ii) RepkHunter is scalable to large-scale, real-world datasets and helps to to identify more repackaged apps in the wild. We deploy RepkHunter for six months and identify 205 previously unreported repackaged applications in Google Play.

The rapid advancement of large language models (LLMs) in recent years has made it feasible to establish domain-specific LLMs for specialized fields. However, in practical development, acquiring domain-specific knowledge often requires a significant amount of professional expert manpower. Moreover, even when domain-specific data is available, the lack of a unified methodology for benchmark dataset establishment often results in uneven data distribution. This imbalance can lead to an inaccurate assessment of the true model capabilities during the evaluation of domain-specific LLMs. To address these challenges, we introduce SDBench, a generic framework for generating evaluation datasets for domain-specific LLMs. This method is also applicable for establishing the LLM instruction datasets. It significantly reduces the reliance on expert manpower while ensuring that the collected data is uniformly distributed. To validate the effectiveness of this framework, we also present the BridgeBench, a novel benchmark for bridge engineering knowledge, and the BridgeGPT, the first LLM specialized in bridge engineering, which can solve bridge engineering tasks.

To safeguard user data privacy, on-device inference has emerged as a prominent paradigm on mobile and Internet of Things (IoT) devices. This paradigm involves deploying a model provided by a third party on local devices to perform inference tasks. However, it exposes the private model to two primary security threats: model stealing (MS) and membership inference attacks (MIA). To mitigate these risks, existing wisdom deploys models within Trusted Execution Environments (TEEs), which is a secure isolated execution space. Nonetheless, the constrained secure memory capacity in TEEs makes it challenging to achieve full model security with low inference latency.

Monitoring the structural health of bridges, especially sea-crossing bridges, is vital for their longevity and safety. Among the potential risks during the operation of sea-crossing bridges, vortex-induced vibration (VIV) has drawn significant attention due to its potential to cause collapse. To accurately detect VIV, introducing new sensors such as GNSS has become a trend. However, current methods often suffer from high latency and low reliability. Additionally, the integration of GNSS poses significant challenges to power supply. To address these challenges, this paper proposes a novel deep learning-based adaptive GNSSIMU fusion method for low-latency, high-accuracy, and energyefficient detection of bridge VIV. We have employed a lightweight Temporal Convolutional Network (TCN) to replace traditional VIV detection methods, significantly reducing the latency of VIV detection. Additionally, Our method is the first to leverage the unique reflection mechanisms of the ocean surface to model GNSS multipath effects, significantly enhancing the accuracy of adaptive data fusion. Finally, the adaptive fusion mechanism has also substantially decreased the power consumption associated with integrating GNSS. The method is validated using both synthetic data and real vibration testbed data. Experimental results demonstrate that, compared to traditional VIV monitoring methods, our approach reduces detection latency from minutes to seconds while maintaining high detection accuracy. Under the same power supply conditions, the deployment duration is extended by 41.8%. We have also designed an IoT-based bridge Structural Health Monitoring (SHM) station to execute our algorithm.

Bluetooth Low Energy (BLE) and LoRa are two dominant wireless protocols for the Internet of Things (IoT), each built with specific design goals, rendering them non-interoperable. Cross-Technology Communication (CTC) enables heterogeneous communication between BLE and LoRa. Despite these efforts, existing works suffer from low throughput and short communication range. In this paper, we present a novel system named BoRa, which enables bi-directional communication between the COTS BLE and COTS LoRa chips. The key idea of BoRa to emulate the linear frequency changes of the LoRa chirp is continuously tuning its Modulation Offset (MO), which is originally designed to compensate for the Carrier Frequency Offset (CFO) in BLE chips. BoRa can be readily run on COTS chips without any hardware modifications.

Internet of Things (IoT) systems fundamentally rely on sensing, planning, and execution capabilities. Although existing systems excel at sensing, they often lack advanced abilities of reasoning and planning to generalize across user's diverse and open-ended requests. To address this, we propose IoTAgent, a hierarchical multi-agent system empowered by large language models (LLMs) to enhance intelligence in IoT environments. IoTAgent interprets user's natural language requests to perform heterogeneous tasks such as complex question answering and real-time event triggering, by coordinating high-level task planning and low-level execution and memory management. It integrates sensor data and domain-specific expert knowledge to achieve deeper understanding of the IoT environment. Evaluations on real-world datasets demonstrate that IoTAgent enhances the quality of question answering, and reduces the latency of rule execution by up to 84.3% while balancing energy consumption.

In recent years, the swift expansion of Internet of Things (IoT) applications has been notable. However, developing a comprehensive IoT application is highly challenging for non-expert developers due to the highly diverse characteristics of embedded operating systems. LLM-based methods provide a paradigm for code generation through natural language, which can greatly simplify and accelerate the development of IoT applications. While promising, existing works have failed to account for the specific characteristics of embedded operating system, resulting in the lower quality of generated IoT code. In this paper, we present IoTPilot, an LLM-driven embedded IoT programming tool. We have observed that the conflicts between LLM internal APIs/headers and external OS-specific APIs/headers are key factors leading to the low quality of generated embedded IoT applications. Thus, we introduce two effective self-thinking chains to integrate internal LLM knowledge with external documentation, addressing conflicts in APIs and headers. We provide embedded IoT benchmarks (IoTEval), which are built on RIOT, Zephyr, Contiki and FreeRTOS. Results show that IoTPilot can improve the performance of IoT code generation on all the three embedded OSes compared with existing state-of-the-art (SOTA) methods.

Blockchain-based distributed storage systems offer enhanced security, transparency, and lower costs compared to traditional centralized storage, making them ideal for peer-to-peer collaboration. However, with the trend towards the Web of Things (WoT), lower transaction speeds and higher computational requirements limit their access to high-density data such as IoT. To address this, we propose TimeChain, an efficient off-chain blockchain storage system for IoT time series data. TimeChain batches discrete time series data, storing only the hash value of each batch on-chain while keeping the complete data off-chain. This significantly reduces storage overhead on the blockchain and storage latency by 37.4 times. TimeChain adopts an adaptive packaging mechanism to reduce the additional latency in range queries by converting the batch processing problem into a graph partitioning problem. To reduce the overhead of node selection, TimeChain integrates a node selection mechanism based on consensus protocol, combining node selection and consensus processes together. TimeChain also proposes a Locality-Sensitive Hashing tree-based data integrity verification mechanism to reduce transmission size. Our evaluation shows a reduction in query latency by 64.6% and storage latency by 35.3% compared to existing systems.

Web of Things (WoT) is an emerging concept to connect IoT devices to the web using standard interfaces. This provides interoperability between different IoT platforms and enables seamless integration with web and cloud services. However, running sophisticated web services directly on resource-constrained IoT devices is challenging due to limitations in memory, computation, and energy. This paper proposesWaWoT, aWasm-based framework for flexible and efficientWebofThings services.WaWoTallows flexible WoT service development using annotations and automatic partitioning. It also enables dynamic service migration using WebAssembly modules to adapt placement between IoT devices and web clients. We also introduce an ahead-of-time compiler optimized for low memory usage through techniques like streamed compilation and trimming. For energy efficiency, we use optimizations like bulk instruction writing and direct I/O accessing. Safety is ensured through compile-time and run-time analyses to guarantee sandboxed execution. Evaluations demonstrateWaWoTexhibits better flexibility than existing WoT development approaches. Furthermore,WaWoTcan also reduce RAM usage by 84.9x and energy consumption by 1.9x-4.9x over existing WebAssembly runtimes. Overall, it enables efficient, safe, and flexible WoT services on constrained IoT devices.

Bluetooth Low Energy (BLE) technology supports various Internet-of-Things (IoT) applications. However, because of their limited transmission power and channel interference, their performance is deficient over weak links. Extending physical layer symbols or using error correction code to the link layer is effective somehow. Introducing excessive BLE bits to both respectively can also decrease the network throughput. To optimize the BLE technology performance, we proposeCPL, a combining PHY and link layer optimization technology that adaptively allocates BLE bits to both the physical layer and link layer. Then we propose theCross-Layer BLE Bits Dynamic Allocation Modelthat unifies the gain of BLE bits in different layers. Finally, we propose aInterference-Aware Controlled CFO Fine-Tuning Methodthat calibrates the model according to different interference patterns. We implementCPLon Commercial-Off-The-Shelf (COTS) BLE chips and SDR. The experiment results show that under various interference conditions,CPLachieves 50× and 32.16% throughput improvement than RSBLE and Symphony.CPLreduces energy consumption by 60.42% to 97.95% compared to RSBLE, and 11.04% to 25.15% compared to Symphony.

WebAssembly has shown promising potential on various IoT devices to achieve the desired features such as multi-language support and seamless device-cloud integration. The execution performance of WebAssembly bytecode is directly influenced by compilation sequences. While existing research has explored the optimization of compilation sequences for native code, these approaches are not suitable to WebAssembly bytecode due to its unique instruction format and control flow graph structure. In this work, we propose WasmRL, a novel efficient deep reinforcement learning (DRL)-based compiler optimization framework tailored for WebAssembly bytecode. We conduct a fine-grained analysis of the characteristics of WebAssembly instructions and associated compilation flags. We observe that the same compilation sequence may yield contrasting performance outcomes in WebAssembly and native code. Motivated by our observation, we introduce a WebAssembly-specific DRL state representation that simultaneously captures the impact of various compilation sequences on the WebAssembly bytecode and its runtime performance. To enhance the training efficiency of the DRL model, we propose a tree-based action space refinement method. Furthermore, we develop a pluggable cross-platform training strategy to optimize WebAssembly bytecode across different IoT devices. We evaluate the performance of WasmRL extensively on PolybenchC, MiBench, Shootout public datasets and real-world IoT applications. Experimental results show: (1) The DRL model trained on a specific device achieves 1.4x/1.1x speedups over -O3 for seen/unseen programs; (2) The DRL model trained on different devices simultaneously achieves 1.21x/1.06x improvements respectively. The code has been available at https://github.com/CarrollAdmin/WasmRL .

The fifth-generation (5G) has empowerd various applications. Effective bandwidth and delay prediction in 5G cellular networks are essential for many applications, such as virtual reality and holographic video streaming. However, accurate bandwidth and delay prediction in 5G networks remains a challenging task due to the short-distance coverage and frequent handover properties of 5G base stations. In this paper, we propose HYPER, a hybrid bandwidth and delay prediction approach that uses an Auto Regressive Moving Average (ARMA) time series predictive model for intra-cell prediction and a Random Forest (RF) regression model for cross-cell prediction. Our ARMA model takes prior information as its input, while the RF model further uses related network and physical features to predict future performance. We conduct a measurement study in commercial 5G networks to analyze the relationship between these features and bandwidth/delay. Moreover, we also propose a handover window adaptation algorithm to automatically adjust the handover window size and determine which model to use during handover for accurate bandwidth and delay prediction. We use commercial 5G smartphones for data collection and conducted extensive experiments in diverse urban environments. Experimental results show that HYPER can reduce the prediction error by more than 13% compared to state-of-the-art prediction approaches.

Mobile application updates occur frequently, and they continue to add considerable traffic over the Internet. Differencing algorithms, which compute a small delta between the new version and the old version, are often employed to reduce the update overhead. Transforming the old and new files into the decoded similarity spaces can drastically reduce the delta size. However, this transformation is often hindered by two practical reasons: (1) insufficient decoding (2) long recompression time. To address this challenge, we have proposed two general approaches to transforming the compressed files (more specifically, deflate stream) into the full decoded similarity space and partial decoded similarity space, with low recompression time. The first approach uses recompression-aware searching mechanism, based on a general full decoding tool to transform deflate stream to the full decoded similarity space with a configurable searching complexity, even when it cannot be recompressed identically. The second approach uses a novel solution to transform a deflate stream into the partial decoded similarity space with differencing-friendly LZ77 token reencoding. We have also proposed an algorithm called MDiffPatch to exploit the full and partial decoded similarity spaces. The algorithm can well balance compression ratio and recompression time by exposing a tunable parameter. Extensive evaluation results show that MDiffPatch achieves lower compression ratio than state-of-the-art algorithms and its tunable parameter allows us to achieve a good tradeoff between compression ratio and recompression time.

In recent years, Bluetooth Low Energy (BLE) has become one of the most wildly used wireless protocols and it is common that users carry one or more BLE devices. With the extensive deployment of BLE devices, there is a significant privacy risk if these BLE devices can be tracked. However, the common wisdom suggests that the risk of BLE location tracking is negligible. The reason is that researchers believe there are no stable BLE fingerprints that are stable across different scenarios (e.g., temperatures) for different BLE devices with the same model. In this paper, we introduce a novel physical-layer fingerprint named Transient Dynamic Fingerprint (TDF), which originated from the negative feedback control process of the frequency synthesizer. Because of the hardware imperfection, the dynamic features of the frequency synthesizer are different, making TDF unique among different devices, even with the same model. Furthermore, TDF keeps stable under different thermal conditions. Based on TDF, we propose BTrack, a practical BLE device tracking system and evaluate its tracking performance in different environments. The results show BTrack works well once BLE beacons are effectively received. The identification accuracy is 35.38%-57.41% higher than the existing method, and stable over temperatures, distances, and locations.

Virtualized radio access networks (vRAN) enable network operators to run RAN functions on commodity servers instead of proprietary hardware. It has garnered significant interest due to its ability to reduce costs, provide deployment flexibility, and offer other benefits, particularly for operators of 5G private networks. However, the non-deterministic computing platforms pose difficulties to effective quality of service (QoS) provision, especially in the case of hybrid deployment of time-critical and throughput-demanding applications. Existing approaches including network slicing and other resource management schemes fail to provide fine-grained and effective QoS support at the User Equipments level. In this paper, we propose UQ-vRAN, a UE-level QoS provision framework. UQ-vRAN presents the first comprehensive analysis of the complicated impacts among key network parameters, e.g., network function splitting, resource block allocation, and modulation/coding scheme selection and builds an accurate and comprehensive network model. UQ-vRAN also provides a fast network configurator which gives feasible configurations in seconds, making it possible to be practical in actual 5G vRAN. We implement UQ-vRAN on OpenAirInterface and use simulation and testbed-base experiments to evaluate it. Results show that compared with existing works, UQ-vRAN reduces the delay violation rate by 12%–41% under various network settings, while minimizing the total energy consumption.

Internet of Things (IoT) applications have recently been widely used in safety-critical scenarios. To prevent sensitive information leaks, IoT device vendors provide hardware-assisted protections, called Trusted Execution Environments (TEEs), like ARM Trust-Zone. Programming a TEE-based application requires separate code for two components, significantly slowing down the development process. Existing solutions tackle this issue by automatic code partition while not successfully applying it in two complicated scenarios: adding trusted logic and interactions with secure peripherals.We propose dTEE, a declarative approach to secure IoT applications based on TrustZone. dTEE proposes a rapid approach that enables developers to declare tiered-sensitive variables and functions of existing applications. Besides, dTEE automatically transforms device drivers into trusted ones. We evaluate dTEE on four real-world IoT applications and seven micro-benchmarks. Results show that dTEE achieves high expressiveness for supporting 50% more applications than existing approaches and reduces 90% of the lines of code against handcrafted development.

In recent years, the swift expansion of Internet of Things (IoT) applications has been notable. However, developing a comprehensive IoT application is highly challenging for non-expert developers due to the highly diverse characteristics of embedded operating systems. The LLM-based approach shows promise in generating code from natural language, but its performance in IoT code generation is poor. This stems from the LLM's insufficient understanding of the embedded IoT code context, leading to missed and conflicting OS-specific APIs. In this paper, we present IoTPilot, a LLM-driven multi-agent IoT programming framework. We develop a clustering-based progressive RAG strategy and auto-calibrating self-debug mechanism to enhance the quality of generated IoT applications.

Bluetooth Low Energy (BLE) is one of the most popular wireless protocols for building IoT applications because of its low energy, low cost, and wide compatibility nature. However, BLE communication performance can be easily affected by interference and blockages because of its low transmission power. This paper presents BLEW, a technique to improve the BLE communication performance over weak links by exploiting adaptive symbol extension and DNN-based demodulator to combat channel interference and maximize network throughput. First, we propose a phase peak clustering-based preamble detection method that coherently adds up the phase difference of preambles to combat the interference. We then propose a multi-domain DNN-based demodulator to fully extracts the temporal and spectrum features of the signal and enhance the demodulation performance. Finally, we model the throughput of Commercial Off-The-Shelf (COTS) BLE chips transmitting extended packets, which can be used to optimize the symbol length in an adaptive manner. We implement BLEW with USRP B210 and COTS nRF52840 platform. Experiments show that BLEW can increase throughput by up to 157.39 Kb/s compared with native BLE over typical weak links. Compared with existing approaches, BLEW has up to 25.70% higher preamble detection rate and up to 3.37 dB demodulation gain.

With the increasing demand for Web of Things (WoT) and edge computing, the efficient utilization of limited computing power on edge devices is becoming a crucial challenge. Traditional neural networks (NNs) as web services rely on deterministic computational resources. However, they may fail to output the results on non-deterministic computing power which could be preempted at any time, degrading the task performance significantly. Multi-exit NNs with multiple branches have been proposed as a solution, but the accuracy of intermediate results may be unsatisfactory. In this paper, we propose MEEdge, a system that automatically transforms classic single-exit models into heterogeneous and dynamic multi-exit models which enables Memory-Elastic inference at the Edge with non-deterministic computing power. To build heterogeneous multi-exit models, MEEdge uses efficient convolutions to form a branch zoo and High Priority First (HPF)-based branch placement method for branch growth. To adapt models to dynamically varying computational resources, we employ a novel on-device scheduler for collaboration. Further, to reduce the memory overhead caused by dynamic branches, we propose neuron-level weight sharing and few-shot knowledge distillation(KD) retraining. Our experimental results show that models generated by MEEdge can achieve up to 27.31% better performance than existing multi-exit NNs.

The high energy consumption of 5G New Radio (NR) poses a major challenge to user experience. A major source of energy consumption in User Equipments (UE) is the radio tail, during which the UE remains in a high-power state to release the radio sources. Existing energy optimization approaches cut radio tails by forcing the UE to enter a low-power state. However, these approaches introduce extra promotion delays and energy consumption with soon-coming data transmissions. In this paper, we first conduct an empirical study to reveal that the 5G radio tail introduces significant energy waste on UEs. Then we propose 5GSaver, a two-phase energy-saving approach that utilizes the inactive state of New Radio to better eliminate the tail in 5G cellular networks. 5GSaver identifies the end of App communication events in the first phase and predicts the next packet arrival time in the second phase. With the learning results, 5GSaver can automatically help the UE determine which radio resource control state to enter for saving energy. We evaluate 5GSaver using 15 mobile Apps on commercial smartphones. Evaluation results show that 5GSaver can reduce radio energy consumption by 9.5% and communication delay by 12.4% on average compared to the state-of-the-art approach.

Multi-exit neural networks have gained popularity in edge computing to leverage the computing power of diverse devices. However, real-time tasks in edge applications often face frequent unpredictable exits caused by power outages or high-priority preemptions, which have been largely overlooked by multi-exit models. To address this challenge, it is crucial to determine the appropriate exit point in the multi-exit model to ensure desirable results during unpredictable exits. In this paper, we propose EINet, a sample-wise planner for real-time multi-exit deep neural networks. EINet enables efficient Elastic Inference with unpredictable exits while ensuring best-effort accuracy on various edge platforms. Our approach involves partitioning a trained deep neural network into multiple blocks, each with its exit. Furthermore, EINet utilizes block-wise model profiles, which include accuracy and inference time information for each block. By leveraging these profiles, EINet dynamically determines the optimal exit plan for each sample during the inference process. We introduce Confidence Score Predictors to adapt to the unique characteristics of input samples and employ the Search Engine to efficiently find near-optimal plans for elastic inference. Extensive evaluations of EINet using multiple deep neural networks and datasets with unpredictable exits demonstrate its superior performance. EINet exhibits significant accuracy improvements: 0.13%–16.5% compared to static plans, 0.79%–4.1% compared to other dynamic plans, and over 50% compared to predictable inference in typical scenarios.

Mobile application updates occur frequently, and they continue to add considerable traffic over the Internet. Differencing algorithms, which compute a small delta between the new version and the old version, are often employed to reduce the update overhead. Researchers have proposed many differencing algorithms over the years. Unfortunately, it is currently unknown how these algorithms quantitatively perform for different categories of applications. It is also challenging to know the impacts of different techniques and whether a technique in one algorithm can be integrated into another algorithm for further performance improvement. This paper conducts the first systematic study to understand the performance of four widely used differencing algorithms for mobile application updates, including xdelta3, bsdiff, archive-patcher, and HDiffPatch with respect to five key metrics, including compression ratio, differencing time/memory overhead, and reconstruction time/memory overhead. We perform measurements for 200 mobile applications, and analyze key techniques (such as decompressing-before-differencing, sliding window, and copy instructions merging) that influence the performance of these algorithms. We have provided four important findings which give insights to further optimize for performance improvement. Guided by these insights, we have also proposed a novel algorithm,sdiff, which achieves the smallest compression ratio to state-of-the-art algorithms by combining an appropriately chosen set of key techniques.

IoT rule engines are important middlewares that allow users to easily create custom trigger-action programs (TAPs) and interact with the physical world. Users expect their TAPs to give a timely response within a certain deadline. Existing works provide this support by boosting the process of trigger event identification. Many IoT rule engines now run in containerized environments, bringing about new challenges and opportunities. Prior solutions can no longer satisfy the need of mitigating the end-to-end latency of containerized TAPs. In this work, we propose EdgeRuler, which couples the IoT rule engine and the container runtime to assure the performance of latency-critical TAPs. To enable such capability, EdgeRuler precisely models the end-to-end latency by exploiting information from both the physical and the cyber world. EdgeRuler then enforces a deadline-aware life-cycle control and resource provision for meeting the TAP constraints in a lightweight and efficient way. We prototype and evaluate EdgeRuler on top of production-ready open-source components, which shows that EdgeRuler reduces the end-to-end latency by 28.6%-96.2% compared to existing scheduling algorithms and 68.4%-89.1% to that of the state-of-the-art IoT rule engines, incurring negligible runtime overhead.

Trigger-Action Program (TAP) is a popular and significant form of Internet of Things (IoT) applications, commonly utilized in smart homes. Existing works either just perform actions based on commands or require human intervention to generate TAPs. With the emergence of Large Language Models (LLMs), it becomes possible for users to create IoT TAPs in zero-code manner using natural language. Thus, we propose ChatIoT, which employs LLMs to process natural language in chats and realizes the zero-code generation of TAPs for existing devices.

Multi-exit neural networks have recently boomed in edge computing to maximize the computing power of different devices. However, many real-time tasks running on edge computing applications have encountered unpredictable exiting frequently due to system power outages, high-priority preemption, etc., which have been overlooked by multi-exit models until now. To tackle this issue, it is critical to decide at which branch the multiexit model exits so that the unpredictable exit will always come with desirable results. In this paper, we propose EINet, a samplewise planner of real-time multi-exit deep neural networks, which achieves efficient Elastic Inference with unpredictable exit while guaranteeing best-effort accuracy on different edge platforms. Therefore, a given trained deep neural network is first partitioned into multiple blocks with one exit each by EINet. Then EINet obtains the block-wise model profiles, including the block-wise accuracy and inference time. Using the model profiles, EINet is able to dynamically determine which exits to take during the inference task for each sample. We introduce Confidence Score Predictors to dynamically adapt the uniqueness of the input samples, and the Search Engine to efficiently find the near-optimal plan during the elastic inference. EINet is evaluated extensively using multiple DNNs and datasets with unpredictable exits. Results show that EINet can achieve the highest average accuracy compared with multiple baselines.

The popularity of the wireless network and the embedded system has promoted the widespread application of the Internet of Things (IoT) monitoring system which detects anomalies for discovering emergency events to avoid losses by deploying various sensors. In the face of anomaly detection algorithms for multi-dimensional data in the IoT monitoring system, the currently popular method is to implement a model based on the graph neural network (GNN). Graph attention network (GAT), a mutant of GNN, is widely used because of the attention mechanism in capturing multi-dimensional nonlinear features between nodes. However, the GAT’s weakness in processing complex temporal features leads to the fact that anomalies with distinctive temporal features are hard to detect. In this paper, we propose Time-Series Graph Attention Network (TSGAT), an innovative GAT-based model for anomaly detection. Given the simplified attention mechanism of the GAT and its weakness in extracting temporal pattern features, we embed the time-series layer to enhance the ability to process time-series data. In addition, we designed the loss function and anomaly scoring mechanism based on data reconstruction features, multidimensional features, and temporal features. TSGAT achieves the improvement of 7%-8% F1 score on two classical public datasets for IoT monitoring: SWaT and WADI, compared with the state-of-the-art GNN-based anomaly detection model.

Audio sensing applications on embedded and mobile devices have recently enjoyed increasing popularity. Their performance can be significantly improved by edge computing which offloads computation-intensive tasks to edge servers through wireless links. The quality of wireless links is essential to offloading performance. However, existing edge computing solutions can hardly predict the link quality accurately and efficiently in a dynamic wireless environment, resulting in less optimal offloading decisions and unsatisfied user-perceived Quality of Experience (QoE). In this article, we present WiEdge, a distributed edge computing framework for audio sensing applications with accurate wireless link prediction. By combining cross-layer information extracted from recently received WiFi beacons, TCP-level statistics, and the past throughput observations, WiEdge can predict the throughput of wireless links accurately and efficiently in the near future. Based on the prediction, WiEdge makes optimal offloading decisions for QoE maximization. We formulate the offloading decision problem as a stochastic optimal control problem and propose an efficient solution based on model predictive control from the control-theoretic perspective. We implement WiEdge and evaluate its performance extensively in three representative real-world scenarios. Results show that WiEdge achieves high prediction accuracy and improves average normalized QoE by 2%, 11%, and 40% in three different scenarios, compared with state-of-the-art approaches.

Recent years, the rapid development of Internet of Things (IoT) technologies and applications have been witnessed. Three important features are characterized in modern IoT applications: 1) device heterogeneity; 2) long-range communication; and 3) cloud/edge-device integration. Difficulties are raised by the above features toward IoT application developers, e.g., predicting and evaluating the performance of the entire IoT application system. To deal with the above difficulties, we design and implement an IoT simulator, TinySim, which satisfies the requirements of high fidelity, high scalability, and seamless transplantation. TinySim takes advantage of the hardware-independent features of TinyLink programming language. Hence, a similar code can be used for both simulation and execution on real hardware platforms. Many virtual IoT devices can be simulated by TinySim at the PC end. These IoT devices can send or receive messages from the cloud or smartphones, making it possible for the developers to evaluate the entire system without the actual IoT hardware. We connect TinySim with Unity 3-D to provide high interactivity. To reduce the event synchronization overhead between TinySim and Unity 3-D, a dependence graph-based approach is proposed. We design an approximation-based approach to reduce the number of simulation events, greatly speeding up the simulation process. We carefully evaluate TinySim using benchmarks and two concrete case studies. TinySim can simulate representative IoT applications, such as smart flowerspot and shared bikes. We conduct extensive experiments to evaluate the performance of TinySim. Results show that TinySim can achieve high accuracy with an error ratio lower than 9.52% in terms of energy and latency. Further, TinySim can simulate 4000 devices within 11.2 physical-minutes for ten simulation-minutes, which is about$3\times $faster than the state-of-art approach.

Text entry for smartwatches is a useful service for many applications like sending text messages and replying emails. Traditional touchscreen-based approaches are two-handed text entry methods, that could be cumbersome when the user is performing other tasks with one hand. Therefore, we propose AirText, the first one-handed text entry method which achieves accurate and practical handwriting in the air for commercial smartwatches. By analyzing the inertial readings from the smartwatch worn on the wrist, AirText is able to accurately recognize the in-air handwritten characters. However, the wrist movements, which produce the inertial readings, are harmful to the user to focus on the screen. In order to address this challenge, AirText uses a novel cross-modal supervision design to achieve accurate character recognition from small wrist movements. AirText further includes a novel word recommendation method to speed up the text entry. We implement AirText on five smartwatches and evaluate its performance extensively with eight volunteers and more than 25,000 in-air handwritten characters. Results show that AirText outperforms two baseline methods and achieves comparable text entry speed as two-handed approaches.

Containers have emerged as a popular technology for edge computing platforms. Although there are varieties of container orchestration frameworks, e.g., Kubernetes to provide high-reliable services for cloud infrastructure, providing real-time support at the containerized edge systems (CESs) remains a challenge. In this paper, we propose EdgeMan , a holistic edge service management framework for CESs, which consists of (1) a model-assisted event-driven lightweight online scheduling algorithm to provide request-level execution plans; (2) a bottleneck-metric-aware progressive resource allocation mechanism to improve resource efficiency. We then build a testbed that installed three containerized services with different latency sensitivities for concrete evaluation. Additionally, we adopt real-world data traces from Alibaba and Twitter for large-scale emulations. Extensive experiments demonstrate that the deadline miss ratio of time-sensitive services run with EdgeMan is reduced by 85.9% on average compared with that of existing methods in both industry and academia.

Customizing and deploying an edge system are time-consuming and complex tasks because of hardware heterogeneity, third-party software compatibility, diverse performance requirements, and so on. In this article, we present TinyEdge, a holistic framework for the low-code development of edge systems. The key idea of TinyEdge is to use a top-down approach for designing edge systems. Developers select and configure TinyEdge modules to specify their interaction logic without dealing with the specific hardware or software. Taking the configuration as input, TinyEdge automatically generates the deployment package and estimates the performance with sufficient profiling. TinyEdge provides a unified development toolkit to specify module dependencies, functionalities, interactions, and configurations. We implement TinyEdge and evaluate its performance using real-world edge systems. Results show that: (1) TinyEdge achieves rapid customization of edge systems, reducing 44.15% of development time and 67.79% of lines of code on average compared with the state-of-the-art edge computing platforms; (2) TinyEdge builds compact modules and optimizes the latent circular dependency detection and message routing efficiency; (3) TinyEdge performance estimation has low absolute errors in various settings.

In the “network-as-a-service” paradigm, network operators have a strong need to know the performance of critical paths running services to the users. Network tomography is an attractive methodology for inferring internal network characteristics from end-to-end measurements between monitors. Motivated by previous results that uniquely identifying the path metrics can require a large number of monitors, we focus on calculating the performance bounds of a set of interesting paths, i.e., bound-based network tomography for interesting paths. We present an efficient solution to obtain the tightest upper and lower bounds of all interesting paths in an arbitrary network with a given set of end-to-end measurements. Based on this solution, we further develop an algorithm to place new monitors over existing ones such that the bounds of interesting paths can be maximally tightened. We formally prove the effectiveness of the proposed algorithms. We implement the algorithms and conduct extensive experiments on real ISP topologies. Compared with state-of-the-art approaches, our algorithms achieve up to$1.2\sim 1.9$times more reduction on the bound interval lengths of all interesting paths and use up to 50.4%~62.5% fewer monitors in various network settings.

The broadcast nature of wireless media makes WLANs easily attacked by rogue Access Points (APs) . Rogue AP attacks can potentially cause severe privacy leakage and financial loss. Hardware fingerprinting is the state-of-the-art technology to detect rogue APs since an attacker would find it difficult to set up a rogue AP with specific hardware fingerprints. However, existing hardware fingerprints not only depend on the AP but also depend on the client, significantly limiting their application scenarios. In this work, we investigate two novel client-agnostic fingerprints, which can be extracted using commercial off-the-shelf WiFi devices, to detect rogue APs. One is the power amplifier non-linearity fingerprint and the other is the frame interval distribution fingerprint . These two fingerprints remain consistent over time and space for the same AP but vary across different APs even with the same brand, model, and firmware. We use the fingerprint similarity between the candidate AP and the authorized AP for device authentication in typical indoor environments. We have also proposed a threshold-improved authentication scheme to improve the robustness of our system in dynamic environments. Our schemes can be implemented without modifying the infrastructural APs and can work well with new clients without rebuilding the fingerprint database. We evaluate our scheme in both in-lab and field scenarios, by analyzing 18 million WiFi packets. Results show that our scheme achieves an overall 96.55% positive detection rate and a 4.31% false alarm rate. Moreover, the threshold-improved authentication scheme can further reduce the false alarm rate by 13.0%-44.8% for dynamic environments.

Bluetooth Low Energy (BLE) has gained large popularity as an important infrastructure of Internet of Things (IoT). Recently, researchers have integrated the TCP/IP stack with the BLE stack for interoperability, supporting more upper-layer protocols and applications. However, these works gain extremely low TCP goodput due to connection event inefficiency when TCP cooperates with BLE. How to improve the performance is an urgent problem. This paper proposes TCPle, a performant TCP-over-BLE stack that presents a novel adaption layer design bridging the gap between TCP and BLE without violating their specifications. TCPle improves the efficiency of connection events significantly by two fundamental mechanisms: connection event length adaption and connection event maintenance which correspond to the two root causes of low goodput. The connection event length adaption mechanism predicts the data size to send based on an online learning method and updates the connection event capacity adaptively. This mechanism avoids the long waiting time of ACK to back to the TCP sender. The connection event maintenance mechanism prefetches data packets to maintain the connection event. This mechanism avoids the long waiting time of data packets when out of the sender after the ACK is back. We implemented TCPle on nRF52840 DK with RIOT OS based on lwIP stack and NimBLE stack and conducted extensive experiments to evaluate its performance. Results show that TCPle 1) is lightweight and well-suited for resource-constrained IoT devices; 2) improves TCP goodput by up to 101.6% compared with other existing TCP-over-BLE stacks.

Effective bandwidth prediction in the fifth-generation (5G) cellular networks is essential for bandwidth-consuming applications, such as virtual reality and holographic video streaming. However, accurate bandwidth prediction in 5G networks remains a challenging task due to the short-distance coverage and frequent handover properties of 5G base stations. In this paper, we propose HYPER, a HYbrid bandwidth PrEdiction appRoach using commercial smartphones. Hyper uses an AutoRegressive Moving Average (ARMA) time series predictive model for intra-cell bandwidth prediction and a Random Forest (RF) regression model for cross-cell bandwidth prediction. Our ARMA model takes prior bandwidth usage as its input, while the RF model further uses related network and physical features to predict future bandwidth. We conduct a measurement study in commercial 5G networks to analyze the relationship between these features and bandwidth. Moreover, we also propose a handover window adaptation algorithm to automatically adjust the handover window size and determine which model to use during handover. We use commercial 5G smartphones for data collection and conduct extensive experiments in diverse urban environments. Experimental results based on one TB of cellular data show that HYPER can reduce the bandwidth prediction error by more than 13% compared to state-of-the-art bandwidth prediction approaches.

Interoperability among a vast number of heterogeneous IoT nodes is a key issue. However, the communication among IoT nodes does not fully interoperate to date. The underlying reason is the lack of a lightweight and unified network architecture for IoT nodes having different radio technologies. In this paper, we design and implement TinyNet, a lightweight, modular, and unified network architecture for representative low-power radio technologies including 802.15.4, BLE, and LoRa. The modular architecture of TinyNet allows us to simplify the creation of new protocols by selecting specific modules in TinyNet. We implement TinyNet on realistic IoT nodes including TI CC2650 and Heltec IoT LoRa nodes. We perform extensive evaluations. Results show that TinyNet (1) allows interoperability at or above the network layer; (2) allows code reuse for multi-protocol co-existence and simplifies new protocols design by module composition; (3) has a small code size and memory footprint.

Recent years have witnessed the progressive integration between IoT (Internet of Things) devices and the cloud server, which promotes the efficiency and interoperability of IoT applications. WebAssembly, known for its performance and portability, is considered a promising technology to bridge the heterogeneity between devices and the server. Nevertheless, resource-constrained devices, which are commonly deployed in the wild, have difficulty participating in this device-cloud integration because they can hardly run WebAssembly efficiently.

Containers have emerged as a popular technology for edge computing platforms. Although there are varieties of container orchestration frameworks, e.g., Kubernetes to provide high-reliable services for cloud infrastructure, ensuring realtime service at the containerized edge systems (CESs) remains a challenge. In this paper, we propose Edgeman,a holistic edge service management framework for CESs, which consists of (1) a model-assisted event-driven lightweight online scheduling algorithm to provide request-level execution plans; (2) a bottleneck-metric-aware progressive resource allocation mechanism to improve resource efficiency. We then build a testbed that installed three containerized services with different latency sensitivities for concrete evaluation. Besides, we adopt real-world data traces from Alibaba and Twitter for large-scale emulations. Extensive experiments demonstrate that the deadline miss ratio of Edgemanis reduced 85.9% on average compared with existing methods in both industry and academia.

With the fast growth of the Internet of Things, people now are surrounded by plenty of devices. To achieve efficient interaction with these devices, human-device interaction technologies are evolving. Because existing methods (mobile App) require users to remember the mapping between the real-world device and the digital one, an important point is to break such a gap. In this paper, we propose VSLink, which offers human-device interaction in an Augmented-Reality-like manner. VSLink achieves fast object identification and pervasive interaction for fusing the physical and cyberspace. To improve processing speed and accuracy, VSLink adopts a two-step object identification method to locate the interaction targets. In VSLink, visual SLAM and object detection neural networks detect stable/-movable objects separately, and detection prior from SLAM is sent to neural networks which enables sparse-convolution-based inference acceleration. VSLink offers a platform where the user could customize the interaction target, function, and interface. We evaluated VSLink in an environment containing multiple objects to interact with. The results showed that it achieves a 33% network inference acceleration on state-of-the-art networks, and enables object identification with 30FPS video input.

Gesture-related human-computer interaction systems have been developed with different purposes. Wearable-based gesture recognition is well studied and considered to be effective. However, unexpected body movement of a user, such as walking and turning, is one issue that affects the robustness of classification. Collecting a sufficient dataset for every unexpected movement would be time-consuming and labor-intensive. To reduce the burden on users and address the lack of training data, we proposed RGC, which is a framework for IMU data augmentation. RGC adopts Generative-Adversarial-Networks-based and rotation-based augmentation to enhance the diversity of the dataset, which helps the classifier learn more effective representations for gesture recognition. We collected a dataset of 21120 arm gesture samples under different kinds of unexpected movements to evaluate RGC. The experiments showed that our method provides an 8.8% to 1.1% accuracy improvement for different SOTA classifiers with different original dataset sizes.

Programming a complete IoT application usually requires separated programming for device, edge and/or cloud sides, which slows down the development process and makes the project hardly portable. Existing solutions tackle this problem by proposing a single coherent language while leaving two issues unsolved: efficient migration among the three sides and the platform dependency of the binaries. We propose WiProg, an integrated approach to IoT application programming based on WebAssembly. WiProg proposes an edge-centric programming approach that enables developers to write the IoT application as if it runs on the edge. This is achieved by the peripheral-accessing SDKs and annotations specifying the computation placement. WiProg automatically processes the program to insert auxiliary code and then compile it to WebAssembly. At runtime, WiProg leverages dynamic code offloading with compact memory snapshotting to achieve efficient execution. WiProg also provides interfaces for the customization of offloading policies. Results on real-world applications and computation benchmarks show that WiProg achieves an average reduction by 18.7%~54.3% and 20.1%~57.6% in terms of energy consumption and execution time.

We advocate ThingSpire OS, a new IoT operating system based on WebAssembly for cloud-edge integration. By design, WebAssembly is considered as the first-class citizen in ThingSpire OS to achieve coherent execution among IoT device, edge and cloud. Furthermore, ThingSpire OS enables efficient execution of WebAssembly on resource-constrained devices by implementing a WebAssembly runtime based on Ahead-of-Time (AoT) compilation with a small footprint, achieves seamless inter-module communication wherever the modules locate, and leverages several optimizations such as lightweight preemptible invocation for memory isolation and control-flow integrity. We implement a prototype of ThingSpire OS and conduct preliminary evaluations on its inter-module communication performance.

Mobile advertising plays a vital role in the mobile app ecosystem. A major threat to the sustainability of this ecosystem is click fraud, i.e., ad clicks performed by malicious code or automatic bot problems. Existing click fraud detection approaches focus on analyzing the ad requests at the server side. However, such approaches may suffer from high false negatives since the detection can be easily circumvented, e.g., when the clicks are behind proxies or globally distributed. In this paper, we present AdSherlock, an efficient and deployable click fraud detection approach at the client side (inside the application) for mobile apps. AdSherlock splits the computation-intensive operations of click request identification into an offline procedure and an online procedure. In the offline procedure, AdSherlock generates both exact patterns and probabilistic patterns based on URL (Uniform Resource Locator) tokenization. These patterns are used in the online procedure for click request identification and further used for click fraud detection together with an ad request tree model. We implement a prototype of AdSherlock and evaluate its performance using real apps. The online detector is injected into the app executable archive through binary instrumentation. Results show that AdSherlock achieves higher click fraud detection accuracy compared with state of the art, with negligible runtime overhead.

Many IoT applications have the requirements of conducting complex IoT events processing (e.g., speech recognition) that are hardly supported by low-end IoT devices due to limited resources. Most existing approaches enable complex IoT event processing on low-end IoT devices by statically allocating tasks to the edge or the cloud. In this article, we present Queec, a QoE-aware edge computing system for complex IoT event processing under dynamic workloads. With Queec, the complex IoT event processing tasks that are relatively computation-intensive for low-end IoT devices can be transparently offloaded to nearby edge nodes at runtime. We formulate the problem of scheduling multi-user tasks to multiple edge nodes as an optimization problem, which minimizes the overall offloading latency of all tasks while avoiding the overloading problem. We implement Queec on low-end IoT devices, edge nodes, and the cloud. We conduct extensive evaluations, and the results show that Queec reduces 56.98% of the offloading latency on average compared with the state-of-the-art under dynamic workloads, while incurring acceptable overhead.

With the increasing relevance of the Internet of Things and large-scale location-based services, LoRa localization has been attractive due to its low-cost, low-power, and long-range properties. However, existing localization approaches based on received signal strength indicators are either easily affected by signal fading of different land-cover types or labor intensive. In this work, we propose SateLoc, a LoRa localization system that utilizes satellite images to generate virtual fingerprints. Specifically, SateLoc first uses high-resolution satellite images to identify land-cover types. With the path loss parameters of each land-cover type, SateLoc can automatically generate a virtual fingerprinting map for each gateway. We then propose a novel multi-gateway combination strategy, which is weighted by the environmental interference of each gateway, to produce a joint likelihood distribution for localization and tracking. We implement SateLoc with commercial LoRa devices without any hardware modification, and evaluate its performance in a 227,500-m urban area. Experimental results show that SateLoc achieves a median localization error of 43.5 m, improving more than 50% compared to state-of-the-art model-based approaches. Moreover, SateLoc can achieve a median tracking error of 37.9 m with the distance constraint of adjacent estimated locations. More importantly, compared to fingerprinting-based approaches, SateLoc does not require the labor-intensive fingerprint acquisition process.

Customizing and deploying an edge system is a time-consuming and complex task, considering the hardware heterogeneity, third-party software compatibility, diverse performance requirements, etc. In this paper, we present TinyEdge, a holistic system for the rapid customization of edge systems. The key idea of TinyEdge is to use a top-down approach for designing the software and estimating the performance of the customized edge systems under different hardware specifications. Developers select and conFigure modules to specify the critical logic of their interactions, without dealing with the specific hardware or software. Taking the configuration as input, TinyEdge automatically generates the deployment package and estimate the performance after sufficient profiling. TinyEdge provides a unified customization framework for modules to specify their dependencies, functionalities, interactions, and configurations. We implement TinyEdge and evaluate its performance using real-world edge systems. Results show that: 1) TinyEdge achieves rapid customization of edge systems, reducing 44.15% of customization time and 67.79% lines of code on average compared with the state-of-the-art edge platforms; 2) TinyEdge builds compact modules and optimizes the latent circular dependency detection and message queuing efficiency; 3) TinyEdge performance estimation has low average absolute error in various settings.

Network tomography is an attractive methodology for inferring internal network states from accumulated path measurements between pairs of monitors. Motivated by previous results that identifying all link metrics can require a large number of monitors, we focus on calculating the performance bounds of a set of interesting links, i.e., bound-based network tomography. We develop an efficient solution to obtain the tightest upper bounds and lower bounds of all interesting links in an arbitrary network with a given set of end-to-end path measurements. Based on this solution, we further propose an algorithm to place new monitors over existing ones such that the bounds of interesting links can be maximally tightened. We theoretically prove the effectiveness of the proposed algorithms. We implement the algorithms and conduct extensive experiments based on real network topologies. Compared with state-of-the- art approaches, our algorithms can achieve 2.2~3.1 times more reduction on the bound interval lengths of all interesting links and reduce the number of placed monitors significantly in various network settings.

With the increasing relevance of the Internet of Things (IoT) and large-scale Location-Based Services (LBS), LoRa localization has been attractive due to its low cost, low power and long range properties. However, existing localization approaches based on Received Signal Strength Indicator (RSSI) are either easily affected by signal fading of different land-cover types or labor-intensive. In this work, we propose SateLoc, a LoRa localization system that utilizes satellite images to generate virtual fingerprints. Specifically, SateLoc first uses high-resolution satellite images to identify land- cover types. With the path loss parameters of each land-cover type, SateLoc can automatically generate a virtual fingerprinting map for each gateway. We then propose a novel multi-gateway combination strategy, which is weighted by the environment interference of each gateway, to produce a joint likelihood distribution for localization. We implement SateLoc with commercial LoRa devices without any hardware modification, and evaluate its performance in a 227,500m2urban area. Experimental results show that SateLoc achieves a median localization error of 47.1m, improving more than 40% compared to the state-of-the-art model-based approaches. More importantly, compared to the fingerprinting-based approach, SateLoc does not require the labor-intensive fingerprint acquisition process.

With the continued proliferation of smart mobile devices, Quick Response (QR) code has played an important role in daily life. They may be distorted and partially invisible due to bright spots, folding and stains When they are printed on soft materials such as plastic bags. Existing scanners may fail in detecting and decoding QR codes due to distortion. In this paper, we propose a simple but effective approach to decoding distorted and partial QR codes. First, we improve an existing QR code detection algorithm to extract QR codes. Then based on the structural features of QR codes that white and black modules are staggered, we propose a novel distortion correction mechanism that uses an adaptive window to match each module. In order to tackle the problem of invisibility, we print multiple QR codes and capture them in an image. Considering confidence of each module in separate, we reconstruct a relatively complete QR code. Extensive experiments have been conducted to evaluate the performance of our approach. The results show that our approach improves the decoding rate by 50% – 60% compared to the other two baselines.

Channel State Information (CSI)-based wireless sensing has recently attracted extensive attention from both academia and industry. However, it is still challenging and time-consuming to develop a CSI-based sensing application due to the use of complex signal processing algorithms and the requirements of accuracy and responsiveness. In this paper, we present TinyCSI, a rapid development framework for CSI-based sensing applications. With TinyCSI, developers only need to write a main script to determine the CSI collection settings and a callback function to process the collected CSI signals using the well-abstracted Matlab/C-based library, without dealing with the connection/transmission details of the sensing nodes. To achieve fast performance tuning, TinyCSI also provides three working modes for different deployment requirements: a remote mode for fast iteration of the sensing algorithms and their parameters, an efficient mode for making full use of computing resources and improving sensing responsiveness, and a standalone mode for offline running sensing systems on individual nodes. We implement three representative demos and conduct real-world user studies to show the workflows and benefits of TinyCSI. Experimental results show that TinyCSI helps reduce the lines of code significantly compared to the original implementation. More importantly, the efficient mode can generate an optimal computing resource allocation solution and significantly improve the sensing responsiveness.

Fitness training is becoming an increasingly popular way of maintaining overall health and preventing illness. However, in some cases the training could be risky and fitness-related injuries have increased by 48% in the USA. The training itself will not cause hurt, but if it is performed in a crucial-improper form (using the wrong technique), it will injure the exerciser. Research has explored the potential of using wearables to monitor fitness training, but the consideration of proper/improper form is not included. In this paper, we propose WearCoach, a wearable based fitness training assistant, which acquires the user's fitness form information and generates real-time feedback during training. WearCoach differs from previous work of training assistant in 1) it employs a two-phase tracking algorithm to achieve accurate and real-time tracking of body motion, 2) it analyzes the captured arm posture and generates training guidance based on the user's form, 3) it uses joint orientation as an exercise classification feature to improve recognition accuracy. We conducted experiments with eight participants and nine exercises. Three kinds of feedback are generated, including injury alert, movement correction and symmetry analysis.

The recent years have witnessed the rapid growth of IoT (Internet of Things) applications. A typical IoT application usually consists of three essential parts: the device side, the cloud side, and the client side. The development of a complete IoT application is very difficult for non-expert developers because it involves drastically different technologies and complex interactions between different sides. Unlike traditional IoT development platforms which use separate approaches for these three sides, we present TinyLink 2.0, an integrated IoT development approach with a single coherent language. It achieves high expressiveness for diverse IoT applications by an enhanced IFTTT rule design and a virtual sensor mechanism which helps developers express application logic with machine learning. Moreover, TinyLink 2.0 optimizes the IoT application performance by using both static and dynamic optimizers, especially for resource-constrained IoT devices. We implement TinyLink 2.0 and evaluate it with eight case studies, a user study, and a detailed evaluation of the proposed programming language as well as the performance optimizers. Results show that TinyLink 2.0 can speed up IoT development significantly compared with existing approaches from both industry and academia, while still achieving high expressiveness.

The broadcast nature of wireless medium makes WLANs easily be attacked by rogue Access Points (APs). Rogue AP attacks can potentially cause severe privacy leakage and financial lost. Hardware fingerprinting is the state-of-the-art technology to detect rogue APs, since an attacker would find it difficult to set up a rogue AP with specific hardware fingerprints. However, existing hardware fingerprints not only depend on the AP, but also depend on the client, significantly limiting their applicable scenarios. In this work, we investigate two novel client-agnostic fingerprints, which can be extracted using commercial off-the-shelf WiFi devices, to detect rogue APs. One is the power amplifier non-linearity fingerprint and the other is the frame interval distribution fingerprint. These two fingerprints remain consistent over time and space for the same AP but vary across different APs even with the same brand, model and firmware. We use the fingerprint similarity between the candidate AP and the authorized AP for device authentication. Our scheme can be implemented without modifying the infrastructural APs and can work well with new clients without rebuilding the fingerprint database. We evaluate our scheme in both in-lab and field scenarios, by analyzing 12 million WiFi packets. Results shows that our scheme achieves an overall 96.55% positive detection rate and a 4.31% false alarm rate.

Multihop low-power wireless communication is a key feature of wireless sensor networks (WSNs). Although with a lot of efforts, the performance of multihop WSNs in practical deployment could still be unpredictable and unsatisfactory, such as high packet transmission latency and short network lifetime. Narrowband Internet of Things (NB-IoT) is an emerging cellular technology for providing wide-area coverage. In this article, we investigate the possibilities of deploying 802.15.4 and NB-IoT equipped sink nodes (i.e., gateways) into existing WSNs to achieve the predictable and improved performance in terms of end-to-end reliability, end-to-end latency, and network lifetime. To calculate these key network-wide metrics, we propose a hybrid analytical model for 802.15.4 and NB-IoT traffic, taking the new features of NB-IoT into account. Based on the established model, we formally consider two types of fundamental problems, i.e., the minimum gateway deployment and the optimal κ-gateway deployment. We implement our framework and extensively evaluate it in the realistic as well as synthetic WSNs. Results demonstrate that our approach can provide significant performance gains, e.g., the network lifetime can be extended by 37.4%-80.3% by deploying only one sink node. Compared with a state-of-the-art method, our approach improves the network lifetime by 12.6%-31.8%.

Indoor intrusion detection (IID) is an essential technology to enable various important applications. Recently, an extensive amount of research has been carried out to develop device-free intrusion detection systems based on the WiFi signal due to its ubiquitous existence and minimum privacy disclosure. However, existing WiFi-based intrusion detection systems typically suffer from false alarms caused by pets, limiting their usage in practice. In this article, we propose PetFree to revisit the IID problem with careful consideration of pet interference. PetFree uses fine-grained channel state information (CSI) of WiFi signals to detect whether there is a human or a pet in the monitoring area. The basic idea of PetFree is to use the effective interference height (EIH) differences between humans and pets. We propose a novel CSI-EIH model to characterize the relationship between CSI measurements and the EIH of the target. Based on this model, PetFree manages to achieve accurate pet identification using only a single WiFi link. We implement and evaluate PetFree extensively with commodity WiFi devices in three different indoor scenarios. Results show that PetFree achieves an overall 93.7% intrusion detection rate and decreases the false alarm rate caused by home pets to 6.5%, significantly outperforming the state-of-the-art approach.

Most sensor networks employ dynamic routing protocols so that the routing topology can be dynamically optimized with environmental changes. The routing behaviors can be quite complex with increasing network scale and environmental dynamics. Knowledge on the routing path of each packet is certainly a great help in understanding the complex routing behaviors, allowing effective performance diagnosis and efficient network management. We propose PAT, a universal sensornet path tracing approach. PAT includes an intelligent path encoding scheme that allows efficient decoding at the PC side. To make PAT more scalable, we propose techniques to accurately estimate the degree information by exploiting timing information, allowing more compact path encoding. Moreover, we employ subpath concatenation to infer excessively long paths with a high recovery probability. We propose an analytical model to quantify the benefits of PAT with varying network scale, network density, routing dynamics and packet delivery performance. We evaluate PAT's performance using testbed experiments, trace-driven study, and extensive simulations. Results show that PAT significantly outperforms existing approaches.

Rapid development is essential for IoT (Internet of Things) application developers to obtain first-mover advantages and reduce the development cost. In this article, we present TinyLink, a holistic system for rapid development of IoT applications. The key idea of TinyLink is to use a top-down approach for designing both the hardware and the software of IoT applications. Developers write the application code in a C-like language to specify the key logic of their applications, without dealing with the details of the specific hardware components. Taking the application code as input, TinyLink automatically generates the hardware configuration as well as the binary program executable on the target hardware platform. TinyLink provides unified APIs for applications to interact with the underlying hardware components. We implement TinyLink and evaluate its performance using real-world IoT applications. Results show that (1) TinyLink achieves rapid development of IoT applications, reducing 52.58% of lines of code on average compared with traditional approaches; (2) TinyLink searches a much larger design space and thus can generate a superior solution for the hardware configuration, compared with the state-of-the-art approach; (3) TinyLink incurs acceptable overhead in terms of execution time and program memory.

Most sensor networks employ distributed and dynamic routing protocols. The flexibility that each node can choose the best forwarder from a diverse candidate set could offer excellent routing performance when the network is highly dynamic. However, it sacrifices routing predictability since it is possible that routing loops are frequently formed. Can we increase the network predictability by controlling the network? As a step towards solving this problem, we introduce FlexCut, a flexible approach for cutting off wireless links, which essentially limits the candidate forwarder set of each node. Unlike existing SDN solutions, FlexCut introduces flexible control over existing distributed and dynamic routing protocols. FlexCut can trade arbitrary amounts of routing diversity for better network performance by exposing to network operators a parameter which quantifies the aggressiveness. We propose novel algorithms, both centralized and distributed, to cut off user-defined number of links so that loops can be alleviated while routing flexibility can be preserved to the largest extent. We evaluate FlexCut extensively by both testbed experiments and simulations. Results show that FlexCut improves the performance by 40%~90% compared with a baseline algorithm in terms of our optimization goal. Results also show that FlexCut can improve the network performance of a sensor network by 20%~35%, 30%~50%, 25% respectively, in terms of packet delivery ratio, transmission delay, and radio duty cycle.

Low-power and multihop wireless networking is envisioned as a promising technology to achieve both energy efficiency and easy deployment for many Internet of Things (IoT) applications. Measuring packet-level path is crucial for managing large-scale multihop wireless networks. Packet-level path information encodes the routing path, a packet that takes through a network. The availability of packet-level path information can greatly facilitate many network management tasks. It is challenging to reconstruct packet-level paths using a small overhead, especially for large-scale networks. While there is a long list of existing path reconstruction algorithms, these algorithms focus on specific network scenarios, e.g., periodic monitoring networks or event detection networks. There lacks a unified model for systematically understanding and comparing the performance of these algorithms in different network scenarios. In this paper, we fill this gap by proposing an abstract model. Using this model, it is possible to derive a decision space for selecting the best algorithm for different networks. Furthermore, this model also guides us to devise better path reconstruction algorithms (cPathτ,cPaths,and cPathsT) with respect to path reconstruction ratio. Extensive experiments demonstrate the prediction power of our model as well as the advantages of our proposed algorithms. The results show that our algorithm (cPathsT) improves a path reconstruction ratio from 94.4%, 34.3%, and 30.8% to 98.9%, 99.9%, and 60.1% on average in three network scenarios, respectively, compared with the best state-of-the-art algorithms.

Inferring fine-grained link metrics by using aggregated path measurements, known as network tomography, is essential for various network operations, such as network monitoring, load balancing, and failure diagnosis. Given a set of interesting links and the changing topologies of a dynamic network, we study the problem of calculating the metrics of these interesting links by end-to-end cycle-free path measurements among selected monitors, i.e., preferential link tomography. We propose MAPLink, an algorithm that assigns a number of nodes as monitors to solve this tomography problem. As the first algorithm to solve the preferential link tomography problem in dynamic networks, MAPLink guarantees that the assigned monitors can calculate the metrics of all interesting links in each possible topology of a dynamic network. We formally prove the above property of MAPLink based on graph theory. We implement MAPLink and evaluate its performance using two real-world dynamic networks, including a vehicular network and a sensor network, both with constantly changing topologies due to node mobility or wireless dynamics. Results show that MAPLink achieves significant better performance compared with four baseline solutions in both of these two dynamic networks.

With the development of LPWA (Low Power Wide Area)technology, the emerging NB-IoT (Narrowband Internet of Things)has attracted much attention and enabled a wide range of applications. An uplink of NB-IoT is a link from a user equipment (UE)to a base station (BS). Uplink transmission is a key component of NB-IoT, accomplishing the sensor data collection task for many applications. However, the performance of uplink transmission has not been rigorously analyzed in the current literature, while uplink performance degradation like long latency could be harmful to many applications with strict uplink performance requirements. In this work, we show a way of mathematically analyzing the performance of uplink transmission for NB-IoT systems, concerning the transmission latency and transmission reliability. Our model is accurate with consideration of the protocol details and the new features of NB-IoT, including link quality, packet size, channel access contention, and etc. We validate the analytical results through detailed simulations. Results show that our analytical model can achieve 83% accuracy for latency calculation and 96% accuracy for reliability calculation. Moreover, we demonstrate that the analytical results can be used to aid protocol design for performance optimization, e.g., repetition number tuning for reducing the transmission latency.

For network operators, knowing the problems in their networks before users complain can help improve their Quality-of-Experience (QoE)and bring great commercial value. However, accurate measurement of the internal performance of a network generally requires the deployment of additional dedicated instrumentation in the network. Another possible approach is to use network tomography technique to infer these internal states from measurements obtained from the edge of the network. However, network tomography usually requires that the network topology and the path of each probe are known. In this paper, using a sparse crowdsourced measurement dataset from mobile users to varies types of network resources, we build a BRAS (broadband remote access server)-level QoE model to help the network operator locate network problems quickly. By reasonably mining the intrinsic correlation among the measurement data, the proposed scheme is able to model the user QoE accurately and achieve an accuracy of 97 %.

LimbMotion is a limb tracking system which enables accurate and real-time tracking with one wearable device on the wrist/ankle of a user. By integrating inertial sensing and acoustic sensing, LimbMotion significantly reduces the search space of a moving limb, and provides accurate limb tracking for further human computer interaction (HCI). Objectives of this demo are to show how LimbMotion works and two HCI applications supported by LimbMotion.

Knowing the trajectory of each packet in a network enables a large range of network debugging and management tasks. Existing packet trajectory tracing approaches for software-defined networking (SDN) either require high message/computational overhead or only focus on one kind of network topology. In this paper, we propose UniROPE, a robust and lightweight packet trajectory tracing approach that supports various network topologies. Using the flow information, UniROPE dynamically selects one of the two proposed packet trajectory tracing algorithms to achieve a better tradeoff between accuracy and efficiency. We implement UniROPE using P4, a high-level language for programming SDN switch operations, and evaluate its performance in networks with different topologies, scales, and link failure probabilities. Results show that UniROPE achieves a high successful ratio of packet trajectory tracing with small message/computational overheads in various networks. We also use three case studies to show the effectiveness of the traced packet trajectory information for network debugging and management.

Dedicated Short Range Communications (DSRC), a promising vehicle-to-vehicle communication technology, has been under active research and large scale DSRC deployment is expected to start shortly. However, before all vehicles are deployed with DSRC, there will be a relatively long partial DSRC deployment period where DSRC-equipped vehicles and non-DSRC-equipped vehicles both exist on roads. More importantly, it is reported that the probability a DSRC-equipped vehicle will benefit from a safety application is only of 1% during the initial DSRC deployment. Therefore, we propose MVS, a Multiple Vehicle Sensing approach to improve the collision avoidance effectiveness under partial DSRC deployment. The design of MVS is based on the observation that vehicles are able to sense the kinematic states of its adjacent vehicles by using existing computer vision technologies and/or on-board radar technologies. Therefore, we focus on improving the efficiency of sharing these sensed kinematic states among DSRC-equipped vehicles. By using the sensed data from multiple adjacent vehicles, the kinematic states of a non-DSRC-equipped vehicle can be accurately estimated. MVS is implemented and evaluated through a trace-driven study based on two realistic vehicle mobility traces. Results show that MVS reduces the collision probability by 61.5% and 60.1% in the two traces.

Inferring fine-grained link metrics by using aggregated path measurements, known as network tomography, is essential for various network operations, such as network monitoring, load balancing, and failure diagnosis. Given a set of interesting links and the changing topologies of a dynamic network, we study the problem of calculating the link metrics of these links by end-to-end cycle-free path measurements among selected monitors, i.e., preferential link tomography. We propose MAPLink, an algorithm that assigns a number of nodes as monitors to solve this tomography problem. As the first algorithm to solve the preferential link tomography problem in dynamic networks, MAPLink guarantees that the assigned monitors can calculate the link metrics of all interesting links for all topologies of the dynamic network. We formally prove the above property of MAPLink based on graph theory. We implement MAPLink and evaluate its performance using two real-world dynamic networks, including a vehicular network and a sensor network, both with changing topologies due to node mobility or wireless dynamics. Results show that MAPLink achieves significant better performance compared with three baseline methods in both of the two dynamic networks.

Most sensor networks employ dynamic routing protocols so that the routing topology can be dynamically optimized with environmental changes. The routing behaviors can be quite complex with increasing network scale and environmental dynamics. Knowledge on the routing path of each packet is certainly a great help in understanding the complex routing behaviors, allowing effective performance diagnosis and efficient network management. We propose PAT, a universal sensornet path tracing approach. PAT includes an intelligent path encoding scheme that allows efficient decoding at the PC side. To make PAT more scalable, we propose techniques to accurately estimate the degree information by exploiting timing information, allowing more compact path encoding. Moreover, we employ subpath concatenation to infer excessively long paths with a high recovery probability. We carefully evaluate PAT's performance using testbed experiments and and extensive simulations with up to 4,000 nodes. Results show that PAT significantly outperforms existing approaches.

For mobile apps, user-perceived delays are critical for user satisfaction. According to our measurement, long delays are commonly caused by network and storage I/O operations while short delays are mainly caused by UI rendering. Short delays are not uncommon, which account for 55.3% in our measurement cases. Previous app performance studies have largely focused on I/O operations but the understanding of UI rendering impact is limited. In this work, we propose DRAW, a system that performs two UI rendering analyses to help app developers pinpoint rendering problems and resolve short delays. The first analysis outlines the wasted rendering time on invisible or covered UI components, namely the overdraw problem. The second analysis is to identify the responsible UI components and rendering operations that cause overall low rendering efficiency. We implement DRAW on Android and apply it to study 1,158 real-world Android apps. Results show that DRAW is helpful as it can pinpoint the responsible UI components and specific rendering operations. Four concrete case studies of real-world apps are further presented to show how DRAW can help developers improve the UI rendering performance of their apps.

Rapid development is essential for IoT (Internet of Things) application developers to obtain first-mover advantages and reduce the development cost. In this paper, we present TinyLink, a holistic system for rapid development of IoT applications. The key idea of TinyLink is to use a top-down approach for designing both the hardware and the software of IoT applications. Developers write the application code in a C-like language to specify the key logic of their applications, without dealing with the details of the specific hardware components. Taking the application code as input, TinyLink automatically generates the hardware configuration as well as the binary program executable on the target hardware platform. TinyLink provides unified APIs for applications to interact with the underlying hardware components. We implement TinyLink and evaluate its performance using real-world IoT applications. Results show that: (1) TinyLink achieves rapid development of IoT applications, reducing 52.58% of lines of code in average compared with traditional approaches; (2) TinyLink searches a much larger design space and thus can generate a superior solution for the hardware configuration, compared with the state-of-the-art approach; (3) TinyLink incurs acceptable overhead in terms of execution time and program memory.

The past decade has witnessed a tremendous growth in the variety and complexity of mobile applications (apps). Although considerable amount of efforts have been spent to improve app performance, smartphones nowadays still face many performance challenges. We discover that the resource contention of multiple running apps, caused by resource bottleneck(s), is a key factor that affects the smartphone performance. In this paper, we present APB, an Automatic tool that detects Performance issues caused by resource Bottleneck(s) on commodity Android smartphones. APB employs an innovative bottleneck-hypersurface model to quantify performance issues given a specific system state. Then, based on the model, APB identifies a list of apps that contribute most to the resource contention, which can well inform the end user to take action such as killing background apps to resolve the performance issue. We implement APB on commodity Android platforms and widely evaluate its effectiveness with real user studies. Results show that APB outperforms three baseline approaches and helps users to improve smartphone performance by 10 to 67 percent, with less than 1 percent runtime overhead.

Inferring fine-grained link metrics by using aggregated path measurements, known as network tomography, is an effective and efficient way to facilitate various network operations, such as network monitoring, load balancing, and failure diagnosis. Given the network topology and a set of interesting links, we study the problem of calculating the link metrics of these links by end-to-end cycle-free path measurements among selected monitors, i.e., preferential link tomography. Since assigning nodes as monitors usually requires non-negligible operational cost, we focus on assigning a minimum number of monitors to identify these interesting links. We propose an optimal monitor assignment (OMA) algorithm for preferential link tomography in communication networks. OMA first partitions the graph representing the network topology into multiple graph components. Then, OMA carefully assigns monitors inside each graph component and at the boundaries of multiple graph components. We theoretically prove the optimality of OMA by proving: 1) the monitors assigned by OMA are able to identify all interesting links and 2) the number of monitors assigned by OMA is minimal. We also implement OMA and evaluate it through extensive simulations based on both real topologies and synthetic topologies. Compared with two baseline approaches, OMA reduces the number of monitors assigned significantly in various network settings.

In this paper, we study the problem of decomposing the end-to-end delay into the per-hop delay for each packet, in multi-hop wireless ad hoc networks. Knowledge on the per-hop per-packet delay can greatly improve the network visibility and facilitate network measurement and management. We propose Domo, a passive, lightweight, and accurate delay tomography approach to decomposing the packet end-to-end delay into each hop. We first formulate the per packet delay tomography problem into a set of optimization problems by carefully considering the constraints among various timing quantities. At the network side, Domo attaches a small overhead to each packet for constructing constraints of the optimization problems. By solving these optimization problems by semi-definite relaxation at the PC side, Domo is able to estimate the per-hop delays with high accuracy as well as give a upper bound and lower bound for each unknown per-hop delay. We implement Domo and evaluate its performance extensively using both trace-driven studies and large-scale simulations. Results show that Domo significantly outperforms two existing methods, nearly tripling the accuracy of the state-of-the-art.

Rapid prototyping of IoT platforms is essential for developers to obtain first-mover advantage, validate feasibility of innovative ideas and the key technologies. In this paper, we present TinyLink, an approach for rapid and cost-effective prototyping of IoT platforms. With TinyLink, developers can specify the key platform functionalities and let TinyLink deal with the details of hardware components. Then TinyLink creates user constraints from the functionalities, as well as the inherent hardware constraints. The high level goal is to automatically select the hardware components so that they can satisfy the user requirements with the lowest cost. TinyLink solves the optimization problem and outputs a list of hardware components. We implement TinyLink and evaluate it using real-world IoT platform requirements. Results show that TinyLink achieves a lower cost compared with an existing IoT application, without affecting the functionalities of the platform.

Air quality monitoring has attracted a lot of attention from governments, academia and industry, especially for PM2.5due to its significant impact on our respiratory systems. In this paper, we present the design, implementation, and evaluation of Mosaic, a low cost urban PM2.5 monitoring system based on mobile sensing. In Mosaic, a small number of air quality monitoring nodes are deployed on city buses to measure air quality. Current low-cost particle sensors based on light-scattering, however, are vulnerable to airflow disturbance on moving vehicles. In order to address this problem, we build our air quality monitoring nodes, Mosaic-Nodes, with a novel constructive airflow-disturbance design based on a carefully tuned airflow structure and a GPS-assisted filtering method. Further, the buses used for system deployment are selected by a novel algorithm which achieves both high coverage and low computation overhead. The collected sensor data is also used to calculate the PM2.5of locations without direct measurements by an existing inference model. We apply the Mosaic system in a testing urban area which includes more than 70 point-of-interests. Results show that the Mosaic system can accurately obtain the urban air quality with high coverage and low cost.

Network measurement, which provides detailed information about the behaviors of operational networks, is essential for network management in wireless sensor networks. In the literature, there have been many approaches focusing on measuring individual aspect of the network, e.g., per-packet routing path and per-hop delay. However, there lacks a general support for conducting different measurement tasks. When managing an operational network, a network operator often needs to switch the current measurement task to a different one, in order to diagnose the observed symptoms. In this paper, we propose TinySDM, a software-defined measurement architecture for WSNs. TinySDM provides a general support for conducting different measurement tasks. TinySDM defines a set of carefully selected hooks that allow the users to easily execute their own measurement tasks. In addition, TinySDM provides a C- like language called TinyCode Language (TCL) to enable easy customization of measurement tasks. By only transmitting the binary code of the measurement task, TinySDM significantly reduces the size of the disseminated data compared with existing reprogramming approaches. We implement TinySDM on the TinyOS/TelosB platform and evaluate its performance extensively in a testbed with 60 nodes. We also use TCL to implement four specific measurement tasks. Results show that TinySDM is flexible, efficient and easily programmable.

In wireless sensor networks, sensor nodes are usually self-organized, delivering data to a central sink in a multi-hop manner. Reconstructing the per-packet routing path enables fine-grained diagnostic analysis and performance optimizations of the network. The performances of existing path reconstruction approaches, however, degrade rapidly in large scale networks with lossy links. This paper presents Pathfinder, a robust path reconstruction method against packet losses as well as routing dynamics. At the node side, Pathfinder exploits temporal correlation between a set of packet paths and efficiently compresses the path information using path difference. At the sink side, Pathfinder infers packet paths from the compressed information and employs intelligent path speculation to reconstruct the packet paths with high reconstruction ratio. We propose a novel analytical model to analyze the performance of Pathfinder. We further evaluate Pathfinder compared with two most related approaches using traces from a large scale deployment and extensive simulations. Results show that Pathfinder outperforms existing approaches, achieving both high reconstruction ratio and low transmission cost.

The notion of global time is of great importance for many sensor network applications. Time reconstruction methods aim to reconstruct the global time with respect to a reference clock. To achieve microsecond accuracy, MAC-layer timestamping is required for recording packet transmission and reception times. The timestamps, however, can be invalid due to multiple reasons, such as imperfect system designs, wireless corruptions, or timing attacks, etc. In this paper, we propose ART, an accurate and robust time reconstruction approach to detecting invalid timestamps and recovering the needed information. ART is much more accurate and robust than threshold-based approach, especially in dynamic networks with inherently varying propagation delays. We evaluate our approach in both testbed and a real-world deployment. Results show that: 1) ART achieves a high detection accuracy with low false-positive rate and low false-negative rate; 2) ART achieves a high recovery accuracy of less than 2 ms on average, much more accurate than previously reported results.

Recent wireless sensor networks (WSNs) are becoming increasingly complex with the growing network scale and the dynamic nature of wireless communications. Many measurement and diagnostic approaches depend on per-packet routing paths for accurate and fine-grained analysis of the complex network behaviors. In this paper, we propose iPath, a novel path inference approach to reconstructing the per-packet routing paths in dynamic and large-scale networks. The basic idea of iPath is to exploit high path similarity to iteratively infer long paths from short ones. iPath starts with an initial known set of paths and performs path inference iteratively. iPath includes a novel design of a lightweight hash function for verification of the inferred paths. In order to further improve the inference capability as well as the execution efficiency, iPath includes a fast bootstrapping algorithm to reconstruct the initial set of paths. We also implement iPath and evaluate its performance using traces from large-scale WSN deployments as well as extensive simulations. Results show that iPath achieves much higher reconstruction ratios under different network settings compared to other state-of-the-art approaches.

Inferring per-link metrics through aggregated path measurements, known as network tomography, is an effective way to facilitate various network operations, such as network monitoring, load balancing, and fault diagnosis. We study the problem of identifying additive link metrics of a set of interesting links from end-to-end cycle-free path measurements among selected monitors, i.e., preferential link tomography. Since assigning a node as a monitor usually requires non-negligible operational cost, we focus on assigning the minimum number of monitors (i.e., optimal monitor assignment) to identify all interesting links. By modeling the network as a connected graph, we propose Scalpel, a scalable preferential link tomography approach. Scalpel trims the original graph by a two-stage graph trimming algorithm and reuses an existing method to assign monitors in the trimmed graph. We theoretically prove Scalpel has several key properties: 1) the graph trimming algorithm in Scalpel is minimal in the sense that further trimming the graph does not reduce the number of monitors; 2) the obtained assignment is able to identify all interesting links in the original graph; and 3) an optimal monitor assignment in the graph after trimming is also an optimal monitor assignment in the original graph. We implement Scalpel and evaluate it based on both synthetic topologies and real network topologies. Compared with state-of-the-art, Scalpel reduces the number of monitors by 39.0% to 98.6% when 50% to 1% of all links are interesting links.

We introduce Mosaic, a mobile sensor network system for city scale sensing. Currently, we focus on accurate PM2.5 measurements. We design and implement low-cost smart sensors for this purpose. Those sensors are specifically designed for moving vehicles such as buses and taxies. The mobile sensing capability of our system enables a wide coverage of a city with a modest number of sensors. We incorporate a three-layer architecture in Mosaic. Mosaic faces many practical challenges which need to be properly addressed, e.g., reliable node hardware and software design, and accurate PM2.5 measurements. We have built three versions of Mosaic nodes. We present a novel data calibration approach combining traditional Artificial Neural Network (ANN) and Support Vector Machines (SVMs). We launch real-world deployments of Mosaic in two main cities in China -- Hangzhou and Ningbo. Our initial efforts show that Mosaic is a feasible approach for city scale sensing and our system can produce highly accurate PM2.5 measurements.

Wireless Sensor Networks (WSNs) have been successfully applied in many application areas. Understanding the wireless link performance is very helpful for both protocol designers and network managers. Loss tomography is a popular approach to inferring the per-link loss ratios from end-to-end delivery ratios. Previous studies, however, are usually targeted for networks with static or slowly changing routing paths. In this work, we propose Dophy, a Dynamic loss tomography approach specifically designed for dynamic WSNs where each node dynamically selects the forwarding nodes towards the sink. The key idea of Dophy is based on an observation that most existing protocols use retransmissions to achieve high data delivery ratio. Dophy employs arithmetic encoding to compactly encode the number of retransmissions along the paths. Dophy incorporates two mechanisms to optimize its performance. First, Dophy intelligently reduces the size of symbol set by aggregating the number of retransmissions, reducing the encoding overhead significantly. Second, Dophy periodically updates the probability model to minimize the overall transmission overhead. We implement Dophy on the Tiny OS platform and evaluate its performance extensively using large-scale simulations. Results show that Dophy achieves both high encoding efficiency and high estimation accuracy. Comparative studies show that Dophy significantly outperforms traditional loss tomography approaches in terms of accuracy.

Energy efficiency is one of the most important factors that affect the applicability of wireless sensor networks (WSNs) in many practical scenarios. Many low-power media access control (MAC) protocols have been proposed in the past decade to improve the energy efficiency of sensor nodes. In these low-power MAC protocols, preambles are widely used to wake up the receivers asynchronizely. However, the data delivery potential of these preambles has not been exploited. In this paper, we propose a coded preamble (COPE), which exploits the data delivery potential of preambles by encoding the preambles by network coding. COPE has two salient features. First, a passive receiver set selection scheme enables nodes to decide whether to receive the overheard preamble packets, without introducing extra communication overhead. Second, COPE supports multiple routing primitives, such as unicast and broadcast, making it be a versatile 2.5 layer between the low-power link layer (layer 2) and the network layer (layer 3). We analyze COPE by a novel analytical model. Results show that COPE is able to improve the energy efficiency of both unicast and broadcast significantly. We also implement COPE in TinyOS/TelosB platform and evaluate its energy efficiency. Results show that COPE significantly reduces the radio-on-time in practical network settings.

In multi-hop wireless ad-hoc networks, packet delivery delay is one of the most important performance metrics. While a lot of research efforts have been spent on measuring and optimizing the end-to-end delay performance, there usually lack accurate and lightweight methods for decomposing the end-to-end delay into the per-hop delay for each packet. Knowledge on the per-hop per-packet delay can greatly improve the network visibility and facilitate network measurement and management. In this paper, we propose Domo, a passive, lightweight and accurate delay tomography approach to decomposing the packet end-to-end delay into each hop. The basic idea is to formulate the problem into a set of optimization problems by carefully considering the constraints among various timing quantities. At the network side, Domo attaches a small overhead to each packet for constructing constraints of the optimization problems. At the PC side, Domo employs semi-definite relaxation and several other methods to efficiently solve the optimization problems. We implement Domo and evaluate its performance extensively using large-scale simulations. Results show that Domo significantly outperforms two existing methods, nearly tripling the accuracy of the state-of-the-art.

We study the problem of identifying additive and static link metrics of a set of interesting links in a communication network, by using end-to-end cycle-free path measurements among selected monitors. To uniquely identify the metrics of these interesting links, three questions should be addressed: monitor assignment (which nodes should serve as monitors), paths selection (which cycle-free paths connecting each pair of monitors will be used), and link metric calculation. Since assigning a node as a monitor usually requires non-negligible operational cost, we focus on assigning the minimum number of monitors (i.e., Optimal monitor assignment) to identify all interesting links. By modeling the network as a connected graph, we propose Scalpel, an efficient preferential link tomography approach. Scalpel trims the original graph by a two-stage graph trimming algorithm and reuses existing method to assign monitors in the trimmed graph. We theoretically prove Scalpel has several key properties: 1) the graph trimming algorithm in Scalpel is minimal in the sense that further trimming the graph cannot reduce the number of monitors, 2) the obtained assignment is able to identify all interesting links in the original graph, and 3) an optimal monitor assignment in the graph after trimming is also an optimal monitor assignment in the original graph. Extensive simulations based on both synthetic topologies and real network topologies show the effectiveness of Scalpel. Compared with state-of-the-art, our approach reduces the number of monitors by 39.0%~98.6% when 50%~1% of all links are interesting links.

The notion of global time is of great importance for many sensor network applications. To achieve microsecond accuracy, MAC-level timestamping is required for recording packet transmission and reception times. The MAC-level timestamps, however, are known to be error-prone, especially with low power listening techniques. In this paper, we propose ART, an accurate and robust time reconstruction approach to detecting invalid timestamps and recovering the needed information. We evaluate our approach in both testbed and a real-world deployment. Results show ART is accurate and robust for deployed sensor networks.

In wireless sensor networks, sensor nodes are usually self-organized, delivering data to a central sink in a multi-hop manner. Reconstructing the per-packet routing path enables fine-grained diagnostic analysis and performance optimizations of the network. The performances of existing path reconstruction approaches, however, degrade rapidly in large scale networks with lossy links. In this paper, we propose Pathfinder, a robust path reconstruction method against packet losses as well as routing dynamics. At the node side, Pathfinder exploits temporal correlation between a set of packet paths and efficiently compresses the path information using path difference. At the PC side, Pathfinder infers packet paths from the compressed information and employs intelligent path speculation to reconstruct the packet paths with high reconstruction ratio. We evaluate several variations of Pathfinder as well as two most related approaches using traces from a large scale deployment and extensive simulations. Results show that Pathfinder outperforms existing approaches, achieving both high reconstruction ratio and low transmission overhead.

Reprogramming over the air is important for maintaining a wireless sensor network. Traditional reprogramming approaches assume always-on link layers. Given the energy limitation of sensor nodes, always-on link layers are often not desired for most sensor network applications. In this paper, we propose ROLP, a novel reprogramming protocol built on a widely used low power link layer in wireless sensor networks. ROLP employs an efficient control packets self-suppression scheme for reliable data transmission. ROLP also employs an adaptively falling asleep scheme based on the neighbor information to reduce the energy consumption. We implement ROLP based on TinyOS and evaluate its performance in two different indoor networks. Compared with the standard reprogramming protocol in TinyOS, ROLP is able to reduce the radio-on time by 57.6% and 39.0% in the two networks. Since radio operations cost most of the energy, these reductions save significant amount of energy and prolong the lifetime of a wireless sensor network.

Wireless sensor networks (WSNs) can be successfully applied in a wide range of applications. Efficient data dissemination is a fundamental service which enables many useful high-level functions such as parameter reconfiguration, network reprogramming, etc. Many current data dissemination protocols employ network coding techniques to deal with packet losses. The coding overhead, however, becomes a bottleneck in terms of dissemination delay. We exploit the concurrency potential of sensor nodes and propose MT-Deluge, a multithreaded design of a coding-based data dissemination protocol. By separating the coding and radio operations into two threads and carefully scheduling their executions, MT-Deluge shortens the dissemination delay effectively. An incremental decoding algorithm is employed to further improve MT-Deluge's performance. Experiments with 24 TelosB motes on four representative topologies show that MT-Deluge shortens the dissemination delay by 25.5-48.6 percent compared to a typical data dissemination protocol while keeping the merits of loss resilience.

A distributed access control module in wireless sensor networks (WSNs) allows the network to authorize and grant user access privileges for in-network data access. Prior research mainly focuses on designing such access control modules for WSNs, but little attention has been paid to protect user's identity privacy when a user is verified by the network for data accesses. Often, a user does not want the WSN to associate his identity to the data he requests, particularly in a single-owner multi-user WSN. In this paper, we present the design, implementation, and evaluation of a novel approach, Priccess, to ensure privacy-preserving access control. In addition to the theoretical analysis that demonstrates the security properties of Priccess, this paper also reports the experimental results of Priccess in a network of Imote2 motes, which show the efficiency of Priccess in practice.

Anomaly detection is an essential functionality of Wireless Sensor Networks (WSNs) due to their complex behaviors and the wireless dynamics. In dynamic WSNs, many characteristics such as network topology, locations of sensor nodes, change frequently over time. We observe that indirect correlations among multiple attributes of a sensor node can be utilized to capture and model the historical behaviors. Prior studies overlooked indirect correlations while in this study we exploit it for detecting anomaly efficiently and accurately. Therefore, we propose ICAD, an indirect correlation based anomaly detection approach. By applying the Markov chain, the state transition probability matrix is calculated and it is subsequently used to detect anomalies. Compared to prior approaches, ICAD can detect different types of anomalies simultaneously. Furthermore, ICAD is implemented based on TinyOS and evaluated in a test-bed with 17 TelosB motes. Evaluation results show that ICAD has high detection accuracy with acceptable overhead.

We propose ReXOR, a lightweight and density-aware reprogramming protocol for wireless sensor networks using XOR. It employs XOR encoding in the retransmission phase to reduce the communication cost. In sparse and lossy networks, it delivers much better performance than Deluge, a typical reprogramming protocol for sensor networks. Compared to prior coding-based reprogramming protocols, it has two salient features. First, it is computationally much more lightweight than previous coding-based reprogramming protocols using Random Linear Codes or Fountain Codes. Second, it is density-aware by adapting its interpage waiting time. Hence, it achieves good performance in both dense and sparse networks. We have implemented ReXOR based on TinyOS and evaluate its performance extensively. Results show that ReXOR is indeed lightweight compared with previous coding-based reprogramming protocols in terms of computation overhead. The results also show that ReXOR achieves good network-level performance in both dense and sparse networks, compared with Deluge and a typical coding-based reprogramming protocol, Rateless Deluge.