Real-world applications such as on-orbit disaster assessment, precision agriculture, and maritime surveillance demand immediate semantic understanding of satellite imagery. Unfortunately, the lack of efficient, high-quality annotation for large-scale satellite imagery limits the responsiveness of low Earth orbit (LEO) constellations in dynamic observation scenarios. We propose SatCLA, a lightweight framework for collaborative LLM-driven annotation of LEO satellite imagery under resource-constrained edge environments. SatCLA enables lightweight large language models (LLMs) to perform zero-shot semantic labeling onboard by retrieving relevant visual-text features and generating structured prompts to guide annotation. However, challenges such as label inconsistency, LLM hallucination, and limited semantic generalization arise under constrained compute and data settings. To address these, SatCLA introduces three key designs: (1) a multimodal retrieval pipeline for context-aware prompting, (2) a structured prompt template to guide task-specific semantic constraints, and (3) an inter-satellite re-annotation mechanism that allows peer satellites to collaboratively verify and refine low-confidence outputs. We conduct experiments within an emulated heterogeneous LEO satellite-cluster testbed and on multiple remote sensing benchmarks, demonstrating that SatCLA achieves significant gains in annotation accuracy and consistency compared to baseline methods, while maintaining low latency and high throughput.

Efficient and adaptive traffic signal control is essential for alleviating urban congestion, improving traffic efficiency, and reducing emissions. However, existing methods often rely on centralized or single-point optimization strategies, lacking the capacity to model spatial dependencies and enable coordination among intersections, especially for rapidly changing traffic flow and emergencies. To address these limitations, we design an IoT-enhanced end-edge-cloud collaborative computing architecture, and propose a traffic signal control algorithm, GTALight. GTALight integrates graph neural networks and Transformer attention mechanisms within a multi-agent reinforcement learning framework. It models intersection topology and dynamic traffic states, enabling adaptive control through graph-based message passing and coordinated decision-making, and uses a decoupled reward mechanism encourages global cooperation. Experiments on real-world datasets show that, compared with the best-performing baseline CoLight, GTALight reduces average travel time by 6.5%, queue length by 40.7%, and fuel consumption by 9.8%, while improving average speed by 13.4% and throughput by 4.6%. These results highlight the efficacy of GTALight in optimizing traffic flow while concurrently enhancing environmental sustainability.

Mobile users can offload their tasks to adjacent edge servers to enhance service quality. These servers require suitable reimbursements to cover the operational and energy consumption costs incurred while assisting with offloaded tasks. Although previous studies have examined market mechanisms for multiple users offloading tasks to multiple servers, most of them have not investigated the market mechanism for splittable task offloading, where tasks can be divided into multiple subtasks and offloaded to multiple servers. In this work, we propose a novel edge computing marketplace that focuses on splittable task offloading in multi-user multi-server scenarios with the aim of maximizing social welfare. Designing such a marketplace presents several challenges. First, the problem of task and computing resource division introduced in this context results in a complex solution space, and the division decisions are interdependent. Second, the users and edge servers have conflicting objectives and hidden utility/cost information. To overcome these challenges and achieve socially optimal market operation, we devise an Iterative DoublE Auction (IDEA) mechanism.IDEAemploys a broker to facilitate the interactions between users and edge servers and induces truthful reporting of hidden information through iterative updates to the allocation and pricing rules. Rigorous theoretical analysis and extensive simulations demonstrate the effectiveness of the proposedIDEAmechanism in achieving optimal social performance.

Detecting leakages in a water distribution network (WDN) is a challenging task due to the complexity of data patterns caused by the pipeline leakages and the volatility of the daily demands. Usually, the data under normal operations are collected and different machine learning algorithms are developed to predict anomalies due to the leaks. However, these methods are overwhelmingly rely on the time domain modeling and ignore the information in the frequency domain, and lack a comprehensive modeling of the data patterns such as shapelet, trend, seasonality and point outliers. In this paper, we propose a joint time-frequency attention (JTFA) approach to detect the WDN leakages. In essence, the received signals are decomposed into trend and residual components to represent the incipient and abrupt leaks separately. Attention models are then applied on both time and frequency domain signals to learn the corresponding patterns. In particular, the spectrum is divided into different frequency bands to better attend the detailed information in the higher frequency bands. The desired signals are subsequently reconstructed and compared to the input signal to generate an anomaly score. Experiments from simulated water supply networks are organized and the results demonstrate that the proposed approach performs better than existing leak detection methods and time-frequency analysis methods.

In the context of Low Earth Orbit Satellite Edge Computing (LEO-SEC), the rapid expansion of satellite constellations and increasing concurrent demands in hotspot areas high-light the limitations of individual satellites. A single satellite often lacks the resources needed to handle task processing with both low latency and high reliability. As a result, the importance of solving challenges related to multi-satellite collaborative computing and task offloading has grown significantly. To address these challenges, this paper proposes the Decentralized Multi-Agent MAB Task Offloading (DMAMTO) algorithm, which integrates the incremental learning mechanism of the Multi-Armed Bandit (MAB) into multi-satellite collaborative computing. Based on the framework of a multi-agent repeated stochastic game, it achieves joint optimization of distributed collaborative satellite selection and discrete task allocation. Simulation results demonstrate that the DMAMTO algorithm enables autonomous collaborative computing in LEO-SEC scenarios, effectively reduces task offloading costs, and consistently outperforms benchmark algorithms in simulation scenarios involving the Starlink, Kuiper, and OneWeb constellations.

Flood mapping using satellite remote sensing images plays an important role in disaster monitoring and emergency response. However, traditional change detection methods encounter dual challenges in complex environments: ambiguous flood delineation and inadequate fusion of multi-source remote sensing data. To address these limitations, we propose the improved cross-modal change detection CNN (CCDCNet), specifically designed for cross-modal flood change detection tasks in synthetic aperture radar (SAR) and multispectral images. This network adopts a dual-stream encoder-decoder structure. We designed RS_DBlock module to expand the receptive field, enabling the network to capture more flood region information at once. The RDC module is designed to achieve multi-scale feature extraction, enhancing the model's ability to understand complex scenes. Additionally, the CBAM residual structure is introduced in the decoding part, which implements channel-spatial attention mechanisms for adaptive feature selection. Experimental results demonstrate that the proposed method achieves performance enhancement compared to baseline methods, with notable improvements in key metrics such as mIoU, Precision and F1 score, providing an effective solution for cross-modal high-precision flood mapping. The source codes are available at https://github.com/llya6/CCDCNet.git

Forest environment monitoring is crucial for detecting and predicting natural disasters, such as forest fires. Uncrewed aerial vehicle (UAV) is frequently used to acquire environmental data in forest ecosystems for monitoring purposes. However, the complex geographical terrain of forests, limited UAV battery capacity, and power constraints of ground IoT devices pose significant challenges to data collection. In this article, we design a 3-D trajectory optimization framework for a UAV to collect forest environmental data, considering the altitude limitations of the UAV in complex forest environments and the need to recharge IoT devices using wireless power transfer (WPT) technology, aiming to extend the operational lifetime of the network. Specifically, we aim to achieve a desirable balance between the amount of data collected by the UAV and its energy consumption incurred by data collection and sensor recharging. To this end, we first formulate the trajectory design of the UAV for data collection in forest areas as a nonlinear optimization problem, aiming to maximize the amount of sensor data collected while minimizing the energy consumption of the UAV. To solve this problem, we propose a converging-trajectory design and data collection (C-TDDC) method, which includes two subalgorithms. The first is an ant colony optimization-based traveling salesman problem (ACO-TSP) algorithm to generate optimal UAV trajectories. The second is a proximal policy optimization-based reinforcement learning algorithm, which balances data collection and energy consumption for the UAV. The simulation results demonstrate that the proposed C-TDDC algorithm exhibits more stable convergence and performs better in completing data collection tasks, outperforming both the state-of-the-art algorithm and baseline schemes.

In satellite Internet of Things (IoT), the remote sensing satellites capture images and then transmit them to a ground station through low Earth orbit (LEO) communication satellites for model inference. However, this process results in significant transmission latency and communication overhead. In response, researchers have proposed various satellite on-orbit model inference methods. Nonetheless, the limited computation capacity and memory space of a single remote sensing satellite impose processing delays when dealing with large quantities of high-resolution images, thereby making it difficult to ensure real-time service. To tackle this issue, we propose an on-orbit deep neural network (DNN) distributed inference framework for remote sensing images in satellite IoT, leveraging the availability of numerous LEO computing satellites. Designing such a framework involves two crucial questions: first, determining which LEO satellites should participate in distributed DNN inference, and second, how to partition the images among the selected LEO satellites. To address these questions, we formulate the distributed inference process as a mixed integer nonlinear optimization problem, which is known to be NP-hard. The objective is to minimize overall energy consumption while ensuring that the distributed inference is accomplished when the satellite dynamic network remains unchanged. We initially propose a dynamic optimization algorithm that derives the optimal solution with rigorous theoretical guarantees. Subsequently, to reduce computational complexity, we introduce an approximate solution based on an improved simulated annealing algorithm. We demonstrate that the approximate algorithm performs within a limited range of the optimal algorithm. Finally, we build a heterogeneous testbed based on Kubernetes and conduct extensive experiments to validate that our proposed algorithms reduce energy consumption by an average of 24.63% and 25.98% on the Faster-RCNN inference model, 47.09% and 47.51% on the RetinaNet inference model, and 53.36% and 48.08% on the Yolov5 inference model on the two datasets compared to the baselines.

With the advancement of satellite computing capabilities, the diversity of satellite communication services imposes varied quality of service (QoS) requirements. Limited satellite resources necessitate remote deployment and updates of running services for QoS testing, increasing testing difficulty. Existing testing tools are limited in functionality or reliant on specific infrastructures, failing to meet the QoS testing needs of edge cloud services in mobile satellite scenarios. In this paper, we present SECTest, an integrated testing platform for QoS in satellite edge clouds. More precisely, SECTest can integrate changes in satellite network topology, create and manage satellite edge cloud cluster testing environments on heterogeneous edge devices, customize experiments for users, support deployment and scaling of various integrated testing tools, provide test data persistence function to manage data life cycle and store data hierarchically, and publish and visualize test results. We have built a real satellite edge cloud cluster based on Kubernetes, integrating both physical and virtual machines, and deploying a variety of integrated testing tools using containerization technology. Currently, we have evaluated the quality of service in terms of processing latency, packet drop rate, throughput, and average response time for object detection microservice applications, web microservice applications, and data transfer tasks. To demonstrate SECTest's scalability in testing network communication protocols, we evaluated the performance of HTTP and gRPC in microservice communication within the cluster. Our experimental results validate SECTest's ability to test key service quality metrics in a real satellite edge cloud cluster.

Ultra-dense networks are a key technology for 5G, characterized by having more cells than active users. However, as the density of small cell base stations (SBS) increases, ultra-dense networks introduce challenges in the exploration of resource allocation methods, particularly with respect to computational complexity and network scalability. To address these issues, this paper proposes a computation offloading and scheduling model for each SBS in ultra-dense networks. Specifically, in the proposed model, each SBS receives computing tasks from mobile devices and stores them in a queue buffer. Subsequently, each SBS first determines how many tasks need to be processed locally and whether offloading them is necessary. To tackle the complexity of this decision-making process, we employ game theory abstraction to reduce the large-scale state space and design a fast, adaptive computation offloading algorithm (FACOA). This algorithm enables each SBS to learn efficiently and quickly adapt to strategy changes made by other SBSs. Finally, simulation and experiments show that the proposed algorithm shorten the convergence time by 6 times.

Federated learning (FL) is a privacy-preserving distributed machine learning framework that allows collaborative model training among multiple clients without disclosing their raw training data. Despite data staying localized in FL, clients are still susceptible to privacy threats as their private or sensitive information can be inferred from their shared model updates or gradients. To address this issue, differential privacy (DP) techniques have been widely employed in FL by introducing random noises to obfuscate clients’ shared model parameters, thus alleviating privacy leakage. However, the introduction of DP to FL often significantly compromises the accuracy of model training. In this work, we propose a novel differentially private federated learning method to strike a desirable balance between model training accuracy and client privacy. The core idea of our method is that in each round, each client first clips the model difference between their updated model and the received global model and introduces random noises to the clipped model difference, which is then uploaded to the server. In particular, we design an adaptive clipping strategy where the clipping bound is dynamically adjusted. We conduct extensive experimental evaluations of our proposed method, and the results show that it achieves superior model accuracy under the same level of privacy protection.

The diversification of satellite communication services imposes varied requirements on network service quality, making quality of service (QoS) testing for microservices running on satellites more complex. Existing testing tools have limitations, potentially offering only single-functionality testing, thus failing to meet the requirements of QoS testing for edge cloud services in mobile satellite scenarios. In this paper, we propose a framework for integrating quality of service testing in satellite edge clouds. More precisely, the framework can integrate changes in satellite network topology, create and manage satellite edge cloud cluster testing environments on heterogeneous edge devices, customize experiments for users, support deployment and scaling of various integrated testing tools, and publish and visualize test results. Our experimental results validate the framework’s ability to test key service quality metrics in a satellite edge cloud cluster.

The proliferation of machine learning (ML) applications has given rise to a new and popular data marketplace paradigm. These marketplaces facilitate ML model requesters in obtaining data from data owners to train their desired models. To mitigate the privacy concerns of data owners, federated learning (FL) has been introduced, enabling collaborative model training without raw data trading. Furthermore, researchers have incorporated differential privacy (DP) techniques into FL, resulting in differentially private federated learning (DPFL) to enhance privacy preservation. However, existing designs of DPFL-based data marketplaces consider a simplified but unrealistic scenario where the model requester holds dominant market power, and data owners cannot set their own prices. In this work, we propose a novel DPFL-based data marketplace that accommodates both price-taking and price-setting data owners. We model the interactions among the model requester and these two types of data owners as a three-stage Stackelberg game, focusing on maximizing the model requester's profit. We rigorously establish that the formulated game is a convex game with a unique subgame perfect equilibrium. Moreover, we devise iterative algorithms to determine the equilibrium strategies for the model requester and price-setting data owners. Notably, our algorithms allow data owners to operate without requiring complete information about the model requester or other data owners. Numerical experiments demonstrate the superiority of our proposed three-stage framework in terms of the model requester's profitability compared to scenarios where only price-taking data owners are involved. Furthermore, we reveal that price competition among price-setting data owners reduces equilibrium market prices.

With the rapid development of vehicle-to-everything communication technologies, many emerging compute-intensive in-vehicle applications have emerged. Vehicle edge computing (VEC) leverages the computational resources available at edge nodes to alleviate the strain on public network transmission and reduce task processing latency. However, the dynamic nature of the vehicle environment, the challenge of incentivizing vehicles to share idle resources, and the uncertainty surrounding the number of resources shared by vehicles present significant obstacles in designing task offloading and resource allocation methods for VEC systems. In this paper, we propose a hybrid offloading model wherein task vehicles can offload tasks to roadside units (RSUs) or other vehicles sharing resources. To maximize the benefits derived from task vehicles, RSUs, and shared resource vehicles, we first introduce an adaptive type selection algorithm (ALTS) for shared resource vehicles based on the multi-armed bandit (MAB) theory. Furthermore, we model the three-party interaction as a multi-stage Stackelberg game involving a computational resource lease contract. Experimental results demonstrate the superiority of the proposed ALTS algorithm over existing learning algorithms, thereby showcasing the effectiveness of the lease contract and the three-party transaction mechanism. Comparative experiments also reveal that integrating RSUs and idle vehicle resources offers better services compared to mechanisms relying solely on edge servers or shared resource vehicles.

To recognize the target classes with only a few samples, few-shot learning (FSL) uses prior knowledge learned from the source classes and is usually expressed as a special domain adaptation problem. However, Existing few-shot learning methods make the implicit assumption that the few target class samples are from the same domain or different domain as the source class samples, which greatly limits their application in the wild. This paper introduces a few-shot learning method in multiple scenarios which requires no prior knowledge on the label set. For a given target domain labels set, it may overlap with the set of source domain labels to varying degrees, thereby bringing up an additional class gap based on the domain gap. In order to solve this problem in a unified framework, we propose a novel domain adaptation network which is designed to address a specific challenge: How to achieve domain adaptation whilst maintaining source/target per-class discriminativeness when the target domain label set is unseen. Our solution is to design corresponding soft label transfer network and minimax entropy network for different target domain label set, then we quantify the transferability of the source domain label set during the training process to discover the relationship between the source domain label set and the target domain label set. Further, we broaden the model's understanding of the data by learning from self-supervised signals how to solve a jigsaw puzzle on the same sample. Extensive experiments show that our model outperforms the state-of-the-art models.

Edge computing provides efficient and low-latency computing services for Internet of Things applications. Container virtualization technology is widely used as an indispensable key technology in edge computing. However, the creation of the container requires reading the corresponding image file. If the image file is not stored locally, it will take a lot of time to download, which increases the user’s extremely high service delay. Aiming at decreasing the download time of image files, we develop a two-stage optimization storage strategy of image files to decrease its download time based on edge computing. This strategy optimizes the image file placement in the initialization stage and the runtime stage, respectively. In the initialization stage, we propose a pseudo-polynomial time algorithm to filter all image files and select the image file combination, which best meets the capacity of the edge node for placement. In the runtime stage, we continue to optimize the local image repository based on the historical access records of the edge node. This operation can reduce the number of downloads of image files, thereby further reducing the user’s service delay. In addition, we created a real data set according to the service requirements and the structure of image files on the smart factory and the access records on the DockerHub. A large number of experiments are carried out based on the data set. Experimental results show that the two-stage optimization storage strategy can greatly reduce the download time of image files, thus reducing the service delay of edge nodes and improving the service quality of edge nodes.

How to efficiently coordinate multiple readers to work together is critical for high throughput in RFID systems. Existing researchs focus on designing efficient reader scheduling strategies that arrange adjacent readers to work in different time to avoid signal collisions. However, the impact of unbalanced tag number of readers on tag read throughput is still challenging. In RFID systems, the distribution of tags is usually variable and uneven, making the number of tags covered by each reader (i.e., the load) imbalanced. This imbalance leads to different execution time for readers: the heavily loaded readers take longer time to collect all tags, while the other readers whose finish execution earlier have to wait in vain. To avoid this useless waiting and improve the system throughput, this paper focuses on the load balancing problem of multiple readers, which is an NP-hard problem. In this paper, we design heuristic algorithms to adjust readers interrogation regions and efficiently balance their loads. The amazing advantage of our algorithm is that it can be adopted by almost all existing protocols in multi-reader systems, including the reader scheduling protocol, to improve system throughput. Extensive experiments demonstrate that our algorithm can significantly improve the throughput in various scenarios.

As technology continuously shrinks, radiation-induced soft errors have become a great threat to the circuit reliability. Among all the causes, the Single-Event Transient (SET) effect is the dominating one for the radiation-induced soft errors. SET-induced soft errors can be mitigated by multiple methods. In terms of area and power overhead, blocking SET propagation is considered to be the most efficient way for soft error reduction. It is found that the SET pulse width can be shrunk by a pulse quenching effect, which can be utilized to mitigate soft errors without introducing any area and power overhead. In this article, we present an effective detailed placer to exploit the pulse quenching effect for soft error reduction in combinational circuits. In our method, the quenching effect enhancement is globally optimized while the cell displacement is minimized. The experimental results demonstrate that our method reduces the soft error vulnerability of the circuits by 29.53% versus 18.38% of the state-of-the-art solution. Meanwhile, our method has a minimal effect on the displacement and half-perimeter wire length (HPWL) compared with the previous solutions, which means a minimum timing influence to the original design.

Contactless respiration detection methods have a wide range of applications in medical services and personnel testing because they eliminate the need for contact between the device and the person being tested. However, existing respiration detection algorithms usually face the problem that the subject is in an unstable breathing state in a non-stationary situation, making it difficult to detect accurately. Therefore, in order to reduce the negative interference of different respiration states on the accuracy of respiration detection, a Bayesian classifier-based respiration state classification method, Res-Classifier, is proposed in this paper as a pre-processing step before respiration detection. Res-Classifier enables respiration detection to select a more targeted method based on respiration state to improve the accuracy of respiration detection. First, Res-Classifier extracts the energy distribution in the LoRa reflected signal spectrum from the frequency domain and the periodicity of the signal from the embedding space, while selects the Power Spectrum Density-Ratio, linear regression variance of the embedding space as features. The detected respiratory signals are then classified by these characteristics into three cases: normal breathing, unstable breathing, and stopped breathing, as a pre-processing step for respiratory state classification prior to respiratory detection. We use LoRa transmitting and receiving nodes to collect real signal data and evaluate Res-Classifier’s performance in experiments. The experiment results show that the classification accuracy of the proposed Res-Classifier reaches 97%, which can demonstrate that the Res-Classifier improves the quality of the respiration signal and the robustness of the respiration frequency estimation.

Satellite networks can provide network coverage in remote areas without terrestrial infrastructure and offer ground users an offload option. However, using satellite networks to provide computation offload services requires consideration not only of the dynamics of the satellite system, but also of how ground users offload tasks and how the limited resources of the satellite are allocated. Therefore, in this paper, we propose a computation offloading algorithm based on the optimal allocation of satellite resources (CO-SROA) and formulate an objective function to minimize the delay and energy consumption for ground users to process the computation tasks. The algorithm decomposes the optimization problem into two subproblems. One is the optimal allocation of satellite resources with determinate offloading decisions in a single time slot, which is solved based on the Lagrange multiplier method. The other is the long-term user offloading decision problem, which is solved by formulating it as a Markovian decision process and using a deep reinforcement learning (DRL) algorithm. Simulation results show that the CO-SROA can achieve better long-term returns in terms of delay and energy consumption.

Accurate and reliable indoor location estimate is crucial for many Internet-of-Things (IoT) applications in the era of smart buildings. However, the positioning accuracy and security of the existing positioning works cannot meet the demands in the large-scale smart buildings scenarios covering multiple multifloor buildings. Therefore, in this article, we focus on the reliable and accurate localization under multibuilding and multifloor environments. We propose two novel designs, including a two-step reliable feature selector and a multitask collaborative positioning model. First, we design a two-step reliable feature selector based on an access point (AP) confidence model and manifold learning, to help select the most representative and reliable fingerprint features. Second, we propose a multitask cooperative positioning model, which consists of a multiscale feature fusion module to adaptively fuse multiscale features and a multitask joint learning module to effectively constrain the cumulative error of multiscale position. Finally, based on the above two, we propose a reliable multibuilding and multifloor localization method (RMBMFL), which can achieve accurate and reliable location estimates with low computational complexity in a smart building complex. We did real-world experiments in a 20 000${m^{2}}$site that covers three multistory buildings to evaluate the performance of the proposed RMBMFL. The experimental results show that RMBMFL achieves a building identification accuracy and a floor identification accuracy of 99%, and a room-level indoor localization with an average positioning error within 2 m, and outperforms state-of-the-art solutions.

Radio-frequency identification (RFID) has been utilized in many applications, such as supply chain and stock management in supermarkets. RFID systems in such practical applications are inherently dynamic because tags may move in and out frequently. One important but challenging problem in such systems is how to estimate the number of dynamic tags fast and reliably. This article proposes effective solutions to this problem, which guarantee the accuracy of estimation and time efficiency. Especially, we want to simultaneously estimate the number of tags that moved out of the system (missing tags) and the number of tags that entered the system (unknown tags) in a specified time interval. We design a novel method called time slot reuse (TSR) that generates two logical frames corresponding to the two types of tags from only one physical frame. Based on TSR, we propose a protocol called SSR that can accurately estimate the number of dynamic tags by using the generated logic frames. However, the performance of SSR degrades significantly when the disparity between the number of the two types of tags is remarkable. We further propose an enhanced version of SSR (ESSR), which overcomes this drawback by partitioning the frame into ranges and mapping different types of tags into different ranges. Rigorous theoretical analysis is performed to tune parameters in SSR and ESSR to minimize the execution time. The simulation results demonstrate up to 80% improvement in time efficiency when compared with state-of-the-art solutions to the same problem.

Multi-agent reinforcement learning system is used to solve the problem that agents achieve specific goals in the interaction with the environment through learning policies. Almost all existing multi-agent reinforcement learning methods assume that the observation of the agents is accurate during the training process. It does not take into account that the observation may be wrong due to the complexity of the actual environment or the existence of dishonest agents, which will make the agent training difficult to succeed. In this paper, considering the limitations of the traditional multi-agent algorithm framework in noisy environments, we propose a multi-agent fault-tolerant reinforcement learning (MAFTRL) algorithm. Our main idea is to establish the agent's own error detection mechanism and design the information communication medium between agents. The error detection mechanism is based on the autoencoder, which calculates the credibility of each agent's observation and effectively reduces the environmental noise. The communication medium based on the attention mechanism can significantly improve the ability of agents to extract effective information. Experimental results show that our approach accurately detects the error observation of the agent, which has good performance and strong robustness in both the traditional reliable environment and the noisy environment. Moreover, MAFTRL significantly outperforms the traditional methods in the noisy environment.

The complicated interior structure of the high-rise buildings brings great difficulties for fire escape routes planning. Existing two-dimensional (2D) emergency evacuation models are utilized to solve the problem of guidance and rescue for fire responders. However, these models are faced with a bottleneck of low security due to limited environmental information and no consideration of trapped personnel behavior features. In this paper, we propose DERP, a dynamic escape route planning method that achieves accurate disaster site avoidance and safety route planning considering fire situation awareness in a smart building. DERP is enabled by two novel designs. First, a three-dimensional (3D) fire information model is constructed by cellular automata considering the overall situation of indoor 3D topological structure, fire situation and crowd distribution. Second, a multiple constraints 3D indoor emergency escape route planning algorithm is designed based on a 3D path safety function. The experimental results show that DERP can plan and adjust the escape route timely and dynamically, thus increasing the escape probability of the trapped people.

We describes four scheduling schemes on the predefined path in large-scale wireless networks supported by full-duplex radios. The end-to-end throughput on multi-hop path with these schemes is discussed, indicating the effect of scheduling schemes and the influence of full-duplex radios on multi-hop data transmission. We make a simulator and compare the throughput results in these schemes. Results show that the new scheduling methods may improve end-to-end throughput on multi-hop path in wireless networks moderately.

The rapidly growing traffic flow exceeds the capacity of the existing infrastructure. It will cause traffic congestion and increase travel time and carbon emissions. Intelligent traffic signal control is a significant element in intelligent transportation system. In order to improve the efficiency of intelligent traffic signal control, the traffic information needs to be collected and processed in real-time. In this paper, we propose a deep reinforcement learning model for traffic signal control. In this model, intersections are divided into several grids of different sizes, which represents the complex traffic state. The switching of traffic signals are defined as actions, and the weighted sum of various indicators reflecting traffic conditions is defined as rewards. The whole process is modeled as Markov Decision Process (MDP), and Convolutional Neural Network (CNN) is used to map the states to rewards. We evaluated the efficiency of the model through Simulation of Urban Mobility (SUMO), and the simulation results proved the efficiency of the model.

Edge computing is envisioned as a prominent technology that provides high computing demand services by offloading computation-intensive and delay-sensitive task from mobile or Internet of Things(IoT) devices to nearby edge servers. However, more and more edge servers are deployed near the terminal devices. The uneven distribution of devices will lead to insufficient resource utilization of edge servers, so the social benefits of the edge computing system decrease. In this paper, we propose an effective task offloading strategy in the scenario of multi-users and multi-edge servers. Terminal devices broadcast task offloading request to edge servers, and the edge servers compete for tasks to improve their resource utilization. We use a non-cooperative game to describe the competition of edge servers, and model an optimization problem as the multi-edge servers resource allocation problem. We then design an iterative algorithm to solve the optimization problem and prove that the problem has an unique Nash equilibrium. Simulation results show that the proposed offloading strategy not only improves the resource utilization of edge servers, but also guarantees the demand of terminal devices.

Achievable capacity in multi-hop wireless networks is seriously lower than single-hop communication. Two full-duplex nodes potentially have 2× capacity in wireless communications, compared to two half-duplex nodes. Organize the two nodes as one unit and reorganize nodes to be the units in multi-hop paths, the capacity can achieve 1× to 2× under space division simultaneous transmission mode, which is verified by analysis and simulation results.

With the maturity and popularity of the Internet of Things (IoT), wireless communication techniques have been vastly applied in daily lives. However, indoor localization has been remained as a challenge due to the insufficient accuracy. In this paper, a cooperative localization method called "Reliable And Cooperative Indoor Localization (RACIL)" framework is proposed to determine a target location under the coverage of multiple WSN schemes like WiFi, Blutooth/BLE, Zigbee, and so on. The calculated "intermediate result" of target location from each WSN scheme are further evaluated by a confidence degree mechanism on the edge computing platforms for a "weighted center" as the final target location. In such a way, both the accuracy and reliability of localization are improved. In order to evaluate the proposed RACIL framework, Matlab simulation and a real tunnel environment emulating coal mining scenario are set up separately for the analysis of location accuracy and capability. The experimental results show that RACIL improves not only the location accuracy but also the location rate in the coverage area with the presence of unreliable anchor nodes in the network.

Internet of Things (IoT) have been widely used in many fields including smart city, industry Internet and automatic driving. Because IoT end devices usually have only limited capability in computation and power supply, they are not suitable to execute energy-consuming computational tasks. In many cases, we need to offload computational tasks from IoT end devices to edge servers in order to save energy consumption on the end devices. This process is usually termed as computing offloading. In this paper, we study computing offloading in radio frequency identification (RFID) systems built with mobile readers. We analyze the energy consumption characteristics of different components in mobile RFID systems, based on which we propose a framework to perform energy-aware offloading for such systems. By using tag searching as an example, we illustrate how our framework can help offload computational intensive tasks to edge servers to save energy consumption on mobile readers while satisfying the constraint on total execution time. Simulation results shown that the energy consumption of mobile readers can be greatly reduced by using our offloading framework.

Energy harvesting (EH) technology is an effective way to resolve the energy supply problem for green Internet of Things application. However, it is extremely vulnerable to the unpredictable environmental changes, and as a result, sensor nodes are charged in an uncontrollable way. In this paper, we focus on EH cooperative wireless sensor networks (EHC-WSNs), a new type of WSNs that integrates EH and wireless energy transfer technologies to provide the continuous and controllable energy supply. We propose the graphene-grid deployment strategy to guarantee energy coverage and network connectivity, whereby a Graphene-based Energy Cooperation Management (GECM) mechanism is designed under the energy-neutral operation. Furthermore, we divide GECM into two different phases, i.e., graphene-based energy cooperative charging strategy and graphene-based opportunistic cooperative routing algorithm, which are optimized according to the graphene-grid structure. Extensive simulations show that the proposed GECM can maximize the harvested energy utilization and prolong the network lifetime.

Indoor localization has elicited increasing attention because it has been widely used in indoor location-based services. At present, many complex scenarios for indoor localization require position estimation not only in single-floor environments but also in multifloor ones. However, existing works exhibit certain limitations in solving the problems that involve high computational complexity and floor localization accuracy. In this paper, a multifloor identification system based on WiFi fingerprint database is designed to address these issues. This floor identification system is divided into offline and online phases. In the offline phase, a localization fingerprint database is built based on WiFi nodes and a multifloor identification model is proposed based on linear discriminant analysis (LDA), called MA_LDA. In the online phase, the final floor number is determined, and the trained model is combined with the majority voting mechanism. After determining the floor number, an algorithm based on the k-nearest neighbor (KNN), called LL_KNN, is proposed to obtain the location information of a target on the floor. Real experiment results show that our system can identify the floor number by using only a little WiFi node fingerprint information rather than all the nodes to reduce the computational complexity. It works efficiently and achieves high fault-tolerance performance compared with existing approaches in locating targets in a multifloor environment.

Aiming at the information overload and driving safety problems existing in VANET, this paper proposes a multidimension context-based service recommendation algorithm in VANET based on the recommended middleware architecture of VANET service. The middleware architecture not only shields the heterogeneity of the underlying devices, but also quickly captures the vehicle's rich real-time contextual information. The algorithm belongs to the content-based recommendation category. Firstly, the service station is filtered according to the context information, and the optional service station is selected. Secondly, the user preference model is calculated according to the user history service record. Then, the similarity between the service provided by the service station and the user preference model is calculated. Finally, the recommendation coefficient is calculated and sorted according to the recommendation coefficient, and the service that meets the personalized requirement is recommended for the user. In this paper, the Yelp real data set is used to simulate the algorithm. The simulation results show that the recommended results of the algorithm are more in line with the user's individual needs, and the accuracy of the recommendation results is improved, and the bypass probability caused by the service is reduced.

Fog computing has been proposed as an extension of cloud computing to provide computation, storage, and network services in network edge. For smart manufacturing, fog computing can provide a wealth of computational and storage services, such as fault detection and state analysis of devices in assembly lines, if the middle layer between the industrial cloud and the terminal device is considered. However, limited resources and low-delay services hinder the application of new virtualization technologies in the task scheduling and resource management of fog computing. Thus, we build a new task-scheduling model by considering the role of containers. Then, we construct a task-scheduling algorithm to ensure that the tasks are completed on time and the number of concurrent tasks for the fog node is optimized. Finally, we propose a reallocation mechanism to reduce task delays in accordance with the characteristics of the containers. The results showed that our proposed task-scheduling algorithm and reallocation scheme can effectively reduce task delays and improve the concurrency number of the tasks in fog nodes.

The safety prediction of rail transit system is a fundamental problem in rail transit modeling and management. In this paper, we propose a safety prediction model based on deep learning for rail transit safety, which has been implemented as a deep belief network (DBN). It can learn effective features for rail transit prediction in an unsupervised fashion, which has been examined and found to be effective for many areas such as image and audio classification. To increase the accuracy of prediction, we introduce user satisfaction and rare-event probability, the new input prediction factors, into safety prediction. The former takes account of human and the latter is computed by statistic model checking. To show proof of the model, a real-world subway data sets based on the Beijing Metro in China is presented to demonstrate the feasibility of the model. Experiments on data sets show good performance of our prediction. These positive results demonstrate that deep learning and new factors are promising in rail transit research.

Cayley data centers are well known patterns of completely wireless data centers (WDCs). However, low link reliability and link interference will affect the construction of virtual networks. This paper proposes a virtual resource mapping algorithm on the basis of Cayley structures. First, we analyze the characteristics of Cayley WDCs and model networks in WDCs, where a virtual network is modeled as a traditional undirected graph, while a physical topology is modeled as a directed graph. Second, we propose a virtual resource mapping and coloring algorithm based on link interference called VRMCA-LI. We build a connection interference matrix for each node and use a coloring method to avoid interference. VRMCA-LI uses the same color for the nodes that are within the transmitting angle of a sending node and whose signal-to-noise ratio is less than a threshold. These nodes with the same color cannot be allocated to virtual nodes at the same time. The allocation of nodes and links are concurrent, which performs dynamic adjustment to save mapping time. Third, our experimental results show that VRMCA-LI outperforms WVNEA-LR and PG-VNE in terms of mapping time of virtual nodes, acceptance rate of virtual networks, and average node utilization rate.

Energy savings optimization becomes one of the major concerns in the wireless sensor network (WSN) routing protocol design, due to the fact that most sensor nodes are equipped with the limited nonrechargeable battery power. In this paper, we focus on minimizing energy consumption and maximizing network lifetime for data relay in one-dimensional (1-D) queue network. Following the principle of opportunistic routing theory, multihop relay decision to optimize the network energy efficiency is made based on the differences among sensor nodes, in terms of both their distance to sink and the residual energy of each other. Specifically, an Energy Saving via Opportunistic Routing (ENS_OR) algorithm is designed to ensure minimum power cost during data relay and protect the nodes with relatively low residual energy. Extensive simulations and real testbed results show that the proposed solution ENS_OR can significantly improve the network performance on energy saving and wireless connectivity in comparison with other existing WSN routing schemes.

WiFi access points, mesh routers, wireless sensors and any other wireless routers along the road can serve as roadside unit (RSU), and these RSUs can provide infrastructural supports for wireless access and data dissemination in cyber-transportation systems. We present a hybrid routing scheme in vehicular networks for inter-vehicle, vehicle-to-roadside and inter-roadside data dissemination in urban hybrid networks. First, a location-based crowdsensing framework, including online sensing and offline crowdsourcing, is proposed to retrieve the number and location of available RSU resources. Then, we combine RSU resources and ad hoc solutions to design a routing switch mechanism, which can guarantee quality of data dissemination under various network connectivity and deployment configurations. The performance of our hybrid data dissemination scheme is evaluated using both simulation and real testbed experiments.

Core Java is a framework which extends the programming language Java with built-in regression analysis, i.e., the capability to do parameter estimation for a function. Core Java is unique in that functional forms for regression analysis are expressed as first-class citizens, i.e., as Java programs, in which some parameters are not a priori known, but need to be learned from training sets provided as input. Typical applications of Core Java include calibration of parameters of computational processes, described as OO programs. If-then-else statements of Java language are naturally adopted to create piecewise functional forms of regression. Thus, minimization of the sum of least squared errors involves an optimization problem with a search space that is exponential to the size of learning set. In this paper, we propose a combinatorial restructuring algorithm which guarantees learning optimality and furthermore reduces the search space to be polynomial in the size of learning set, but exponential to the number of piece-wise bounds.

Proposed and implemented is the language CoReJava (constraint optimization regression in Java), which extends the programming language Java with regression analysis, i.e., the capability to do parameter estimation for a function. CoReJava is unique in that functional forms for regression analysis are expressed as first-class citizens, i.e., as Java programs, in which some parameters are not a priori known, but need to be learned from training sets provided as input. Typical applications of CoReJava include calibration of parameters of computational processes, described as OO programs. To implement regression learning, the CoReJava compiler (1) analyses the structure of the parameterized Java program that represent a functional form, (2) automatically generates a constraint optimization problem, in which constraint variables are the unknown parameters, and the objective function to be minimized is the sum of squares of errors w.r.t. the training set, and (3) solves the optimization problem using an external non-linear optimization solver. CoReJava then executes as a regular Java program, in which the initially unknown parameters are replaced with the found optimal values. CoReJava syntax and semantics are formally defined and exemplified using a simple supply chain example.

Key management is a fundamental security service in distributed sensor networks. In this paper, the focus is to design a lightweight group key management scheme to safeguard the data packet passing on the sensor networks under different types of attacks. We propose an energy efficient level-based hierarchical system and build the secure route from the source node to sink node. We compromise between the energy consumption and the shortest path by utilizing the number of neighbors (NBR) of a sensor when renewing a cluster head or choosing the next hop in the hierarchical clusters. In addition, our protocol contains group communication policies, group membership requirements and an algorithm for generating a distributed group key for secure communication

The eXtensible Markup Language (XML) is widely used to transmit data across the Internet. XML schemas are used to defile the syntax of XML messages. XML-based applications can receive messages from arbitrary applications, as long as they follow the protocol defined by the schema. A receiving application must either validate XML messages, process the data in the XML message without validation, or modify the XML message to ensure that it conforms to the XML schema. A problem for developers is how well the application performs the validation, data processing, and, when necessary, transformation. This paper describes and gives examples of a method to generate tests for XML-based communication by modifying and then instantiating XML schemas. The modified schemas are based on precisely defined schema primitive perturbation operators