Actual Deployment Research Articles

Semisupervised Federated-Learning-Based Intrusion Detection Method for Internet of Things

Federated learning (FL) has become an increasingly popular solution for intrusion detection to avoid data privacy leakage in Internet of Things (IoT) edge devices. Existing FL-based intrusion detection methods, however, suffer from three limitations: (1) model parameters transmitted in each round may be used to recover private data, which leads to security risks, (2) not independent and identically distributed (non-IID) private data seriously adversely affects the training of FL (especially distillation-based FL), and (3) high communication overhead caused by the large model size greatly hinders the actual deployment of the solution. To address these problems, this paper develops an intrusion detection method based on semi-supervised FL scheme via knowledge distillation. First, our proposed method leverages unlabeled data via distillation method to enhance the classifier performance. Second, we build a model based on convolutional neural networks (CNN) for extracting deep features of the traffic packets, and take this model as both the classifier network and discriminator network. Third, the discriminator is designed to improve the quality of each client’s predicted labels, to avoid the failure of distillation training caused by a large number of incorrect predictions under private non-IID data. Moreover, the combination of hard-label strategy and voting mechanism further reduces communication overhead. Experiments on the real-world traffic dataset with three non-IID scenarios show that our proposed method can achieve better detection performance as well as lower communication overhead than state-of-the-art methods.

Instance Segmentation of Irregular Deformable Objects for Power Operation Monitoring Based on Multi-Instance Relation Weighting Module

Electric power operation is necessary for the development of power grid companies, where the safety monitoring of electric power operation is difficult. Irregular deformable objects commonly used in electrical construction, such as safety belts and seines, have a dynamic geometric appearance which leads to the poor performance of traditional detection methods. This paper proposes an end-to-end instance segmentation method using the multi-instance relation weighting module for irregular deformable objects. To solve the problem of introducing redundant background information when using the horizontal rectangular box detector, the Mask Scoring R-CNN is used to perform pixel-level instance segmentation so that the bounding box can accurately surround the irregular objects. Considering that deformable objects in power operation workplaces often appear with construction personnel and the objects have an apparent correlation, a multi-instance relation weighting module is proposed to fuse the appearance features and geometric features of objects so that the relation features between objects are learned end-to-end to improve the segmentation effect of irregular objects. The segmentation mAP on the self-built dataset of irregular deformable objects for electric power operation workplaces reached up to 44.8%. With the same 100,000 training rounds, the bounding box mAP and segmentation mAP improved by 1.2% and 0.2%, respectively, compared with the MS R-CNN. Finally, in order to further verify the generalization performance and practicability of the proposed method, an intelligent monitoring system for the power operation scenes is designed to realize the actual deployment and application of the proposed method. Various tests show that the proposed method can segment irregular deformable objects well.

Streaming From the Air : Enabling Drone-Sourced Video Streaming Applications on 5G Open-RAN Architectures

Enabling high data-rate uplink cellular connectivity for drones is a challenging problem, since a flying drone has a higher likelihood of having line-of-sight propagation to base stations that terrestrial UEs normally do not have line-of-sight to. This may result in uplink inter-cell interference and uplink performance degradation for the neighboring ground UEs when drones transmit at high data-rates (e.g., video streaming). We address this problem from a cellular operator's standpoint to support drone-sourced video streaming of a point of interest. We propose a low-complexity, closed-loop control system for Open-RAN architectures that jointly optimizes the drone's location in space and its transmission directionality to support video streaming and minimize its uplink interference impact on the network. We prototype and experimentally evaluate the proposed control system on a dedicated outdoor multi-cell RAN testbed, which is the first measurement campaign of its kind. Furthermore, we perform a large-scale simulation assessment of the proposed control system using the actual cell deployment topologies and cell load profiles of a major US cellular carrier. The proposed Open-RAN control scheme achieves an average 19% network capacity gain over traditional BS-constrained control solutions and satisfies the application data-rate requirements of the drone (e.g., to stream an HD video).

Authentication Technology in Internet of Things and Privacy Security Issues in Typical Application Scenarios

This paper mainly summarizes three aspects of information security: Internet of Things (IoT) authentication technology, Internet of Vehicles (IoV) trust management, and IoV privacy protection. Firstly, in an industrial IoT environment, when a user wants to securely access data from IoT sensors in real-time, they may face network attacks due to the data being transmitted through an open channel. In order to solve this problem, we innovatively propose a user and device authentication model integrated with cloud computing, introduce an algorithm related to protocol design, and summarize the research direction of developing a more lightweight algorithm when designing security protocols in the future. Secondly, for mobile IoT applications, such as IoV, information collection and distribution is realized by establishing a network between vehicles and infrastructure. IoV will face security threats such as information insecurity and privacy disclosure. We introduce a typical trust management model for the IoV, which solves the problem of information unreliability by storing vehicle trust values. In the future, we are committed to making the process of computing node credibility using a trust model more robust. Finally, aiming at the privacy protection of the IoV, we propose a cross-domain anonymous authentication system model based on blockchain. The user’s auxiliary authentication information is stored in the blockchain, and the auxiliary authentication information of any registered user can be obtained from the blockchain. The privacy protection of cross-domain authentication can be realized through anonymous authentication, which greatly saves the communication cost of cross-domain authentication. In the future, we will try to use deep learning or federated learning to integrate with blockchain for actual deployment.

Identifying the effect of coherent precipitates on the deformation mechanisms by in situ neutron diffraction in an extruded magnesium alloy under low-cycle fatigue conditions

As a new class of heat-treatable magnesium alloys, the low-alloyed Mg-Al-Ca-Mn alloy has great engineering potential because of its excellent extrudability and high strength by the dispersion of Guinier-Preston (G.P.) zones. The complex deformation mechanisms associated with hexagonal crystals, the interactions between defects with such G.P. zones, as well as their impacts on fatigue lifetime, are all critical issues before the actual deployment of such materials. In this study, this type of Mg alloy with and without G.P. zone dispersion during cyclic deformation was investigated by in situ neutron diffraction measurements. The relationship between the macroscopic cyclic deformation behavior and the microscopic response (particularly twinning and detwinning) at the grain level was established. The general deformation mechanism evolution in samples under the solution-treated (S.T.) state was generally similar to that in samples under the peak-aged (P.A.) state. Both samples plastically deformed by extension twinning during compression, and by a sequential process of detwinning and dislocation motion under reverse tension. Results suggest that the precipitates provided the greatest strengthening against the prismatic slip (∼ 45% increase), the moderate to the {101¯2} extension twinning (∼ 27 %), and the least to the basal slip (∼ 11% increase). The P.A. samples have shorter fatigue lifetime primarily due to the excessive dislocation pileups and the higher backstress that promote crack initiation.

Evaluation of the effects of a complex intervention

Appraising the effects of a complex intervention is one step in a more broadly based research process, from the construction or modelling of the intervention to its actual deployment. It consists in measuring the effectiveness or impact of the intervention, i.e. analyzing its capacity to produce change. The aim of this article is to obtain unbiased measurement of the average effects of an intervention, based on a panel of predetermined parameters and on the assumption of a causal link between the intervention and the measured result. This article is consequently devoted to evaluation of the effects of a complex intervention and focuses on the methodological challenges of its three key stages:1) modelling of the intervention and feasibility analysis, essential prerequisites;2) the choice of study design and of the effects to be measured, that is to say the methodological premises;and 3) process analysis, carried out in parallel with the evaluation of effects, leading to an indispensable appraisal of the intervention implementation and of the context into which it is integrated. The article is illustrated by five intervention impact assessment projects. A specific objective when evaluating the effects of a complex intervention consists in (a) moving away from a simple search for causality involving the intervention and its effects and (b) toward understanding of the effectiveness mechanisms, once again taking into account the context and the actual conditions of implementation. The challenge is to embrace rather than limit the complexity of the intervention, this being an essential prerequisite for its successful deployment and eventual generalization.

Utilizing Continuous Time Markov Chain for analyzing video-on-demand streaming in multimedia systems

In multimedia systems, Tree and Mesh-based topologies are the popular ways to organize the peers in an overlay and provide video streaming services. Unlike Content Delivery Networks (CDN), providing streaming services through a Peer-to-Peer (P2P) overlay poses the challenge of unpredictable churn for both provider and consumer peers. The dynamically arriving and departing peers, with Video Cassette Recorder (VCR) actions like jump, pause, forward, rewind, etc., controlled by end users, makes it challenging to reason about such systems. Though techniques like prototype test beds, simulation, etc., are successful to some extent in analyzing the behavior of such complex systems before the actual deployment, they either require lots of effort to set up or fail to capture the finer anomalies. Therefore, there is a need to have a formal framework that can specify the system at the desired abstraction level and capture the stochastic behavior exhibited by the actors of such systems. This paper addresses this issue and provides a platform for analyzing the behavior of real-time Video-on-Demand (VoD) streaming behavior with VCR actions over P2P overlays. To capture the stochastic behavior of actors, the Continuous Time Markov Chain (CTMC) is used to design a model of both the tree and mesh-based systems. This CTMC model is written in the Probabilistic Symbolic Model Checker (PRISM) language, which assists in containing the state space explosion and specifies the QoUX properties using Continuous Stochastic Logic (CSL). The proposed methodology uses the scenarios described in various simulation-based studies and industry standards of VoD systems available in the proposed formal model and finds a striking resemblance in the results. Eventually, given a joining-delay threshold of 170 ms, a P2P-Tree-based architecture resulted in an optimal tree height of 4-5 levels; after that, the startup delay of playback gets affected. Similarly, for a given startup delay threshold of 700 ms after departure and seek-jump actions, the maximum P2P-Tree height should be 5. We have also compared the performance properties of P2P tree and mesh-based VoD streaming systems with industry standards of centralized VoD systems. It is observed that the performance of P2P-tree and P2P-mesh-based VoD streaming systems also meet the industry standards. The results we got from the proposed formal framework are conclusive and establish the applicability of the approach.

A Smart Cloud and IoVT-Based Kernel Adaptive Filtering Framework for Parking Prediction

Smart vehicle parking is a collaborative effort of technology and human innovation where the efforts are to be minimized to save time and efforts. In smart cities it is one of the common challenges to introduce smart parking to increase parking efficiency and combat numerous issues like identification of free parking slot and real-time dynamic updation on traffic to save fuel and energy. In this work, a new cloud-based smart parking architecture is proposed that can help in predicting the available free parking slots in smart cities. Initially, the methodology collects the car count at any near by parking using Internet of Things (IoT) and Cloud-based approach. Later, the approach uses the Kernel Least Mean Square algorithm to make heuristic predictions about future vacancy using auto-regression. The proposed approach thus utilizes the online learning or model training. To validate the efficacy of the proposed work, the testing is done on the real-time dataset. The extensive numerical investigation is performed on parking lots of four international airports of a smart city in actual deployment scenarios. The experimentation has revealed superior performance of the method in terms of vacancy prediction.

SPAP: Simultaneous Demand Prediction and Planning for Electric Vehicle Chargers in a New City

For a new city that is committed to promoting Electric Vehicles (EVs), it is significant to plan the public charging infrastructure where charging demands are high. However, it is difficult to predict charging demands before the actual deployment of EV chargers for lack of operational data, resulting in a deadlock. A direct idea is to leverage the urban transfer learning paradigm to learn the knowledge from a source city, then exploit it to predict charging demands, and meanwhile determine locations and amounts of slow/fast chargers for charging stations in the target city. However, the demand prediction and charger planning depend on each other, and it is required to re-train the prediction model to eliminate the negative transfer between cities for each varied charger plan, leading to the unacceptable time complexity. To this end, we design an effective solution of S imultaneous Demand P rediction A nd P lanning ( SPAP ): discriminative features are extracted from multi-source data, and fed into an Attention-based Spatial-Temporal City Domain Adaptation Network ( AST-CDAN ) for cross-city demand prediction; a novel Transfer Iterative Optimization ( TIO ) algorithm is designed for charger planning by iteratively utilizing AST-CDAN and a charger plan fine-tuning algorithm. Extensive experiments on real-world datasets collected from three cities in China validate the effectiveness and efficiency of SPAP . Specially, SPAP improves at most 72.5% revenue compared with the real-world charger deployment.

An efficient Supervised Learning Models for Multimodal MR Brain Image classification

The practicality of distinct vehicular communication tissue classifiers is based on lighting training records that replicate a place or purchase circumstance. The use of transfer learning technologies to address sampling mistakes caused by sparse annotations during supervised learning on automated tumour segmentation is recommended. The comprehensive record of a recognised event might be rather extensive. The suggested method is based on a simple and sparse description, and it effectively corrects systematic sampling mistakes for diverse tissue types using domain correction methodologies. A retrospective examination of the 2013 challenge data sets and a multimodal MR image from 19 malignant gliomas patients verified the present strategy. When compared to training on entirely marked outcomes, the time to mark and train is reduced by more than 70 and 180 seconds respectively. This considerably facilitates the creation and ongoing extension of annotated large datasets in a variety of circumstances and imaging environments; this is an important step in the actual deployment of tissue categorization learning algorithms.

Prognostic value of mitral valve haemodynamic parameters obtained by intraprocedural echocardiography in transcatheter edge-to-edge repair.

To assess whether intraprocedural transesophageal echocardiographic (TEE)-derived haemodynamic parameters predict outcomes in patients undergoing transcatheter edge-to-edge repair (TEER) for mitral regurgitation (MR). This is a single-centre, retrospective analysis encompassing 458 (IQR, 104-1035) days of follow-up after 926 consecutive patients [481 (52%) with functional MR] referred to an isolated mitral TEER between 2013 and 2020. Cases without actual clip deployment, or in whom prior mitral procedures had taken place, were excluded. The primary outcome was the combined rate of all-cause mortality or heart failure (HF) hospitalizations. Secondary endpoints included single components of the primary outcome, as well as MR severity at one month and one year following the procedure. A multivariable analysis identified two intraprocedural echocardiographic observations made after clip deployment as independent predictors of the primary outcome: an above mild MR (HR for whole study period 1.49, 95% CI 1.05-2.13, P = 0.026) and a 100% or more increase from baseline in the transmitral mean pressure gradient (TMPG) (HR for whole study period 1.32, 95% CI 1.01-1.72, P = 0.039). Also, MR grade of above mild and the absence of a normal pulmonary venous flow pattern (PVFP) bilaterally were associated with an increased risk for HF hospitalizations and greater-than-mild 1-month MR. No prognostic role was demonstrated for the change in MR severity, the absolute TMPG, or the mere improvement in PVFP. Immediate post-TEER MR severity and the relative change in TMPG are predictive of clinical and echocardiographic outcomes following the procedure.

Gebotskostenförderung in Windenergie-Auktionen

Abstract Germany faces the challenge of rapidly breaking free from its dependence on fossil fuels. This requires a comprehensive transformation of the energy system. Onshore wind energy is one of the most important generation sources that must be developed on a large scale in order to decarbonise the energy sector. While the expansion targets were raised significantly with the Renewable Energy Law (Erneuerbare-Energien-Gesetz) 2023, the actual deployment of onshore wind energy remains in crisis. To accelerate the expansion, much has been discussed about providing land and simplifying the approval process. However, this article emphasises that reimbursing bid preparation costs can also contribute to driving the expansion of onshore wind energy.

Towards Deploying DNN Models on Edge for Predictive Maintenance Applications

Almost all rotating machinery in the industry has bearings as their key building block and most of these machines run 24 × 7. This makes bearing health prediction an active research area for predictive maintenance solutions. Many state of the art Deep Neural Network (DNN) models have been proposed to solve this. However, most of these high performance models are computationally expensive and have high memory requirements. This limits their use to very specific industrial applications with powerful hardwares deployed close the the machinery. In order to bring DNN-based solutions to a potential use in the industry, we need to deploy these models on Microcontroller Units (MCUs) which are cost effective and energy efficient. However, this step is typically neglected in literature as it poses new challenges. The primary concern when inferencing the DNN models on MCUs is the on chip memory of the MCU that has to fit the model, the data and additional code to run the system. Almost all the state of the art models fail this litmus test since they feature too many parameters. In this paper, we show the challenges related to the deployment, review possible solutions and evaluate one of them showing how the deployment can be realized and what steps are needed. The focus is on the steps required for the actual deployment rather than finding the optimal solution. This paper is among the first to show the deployment on MCUs for a predictive maintenance use case. We first analyze the gap between State Of The Art benchmark DNN models for bearing defect classification and the memory constraint of two MCU variants. Additionally, we review options to reduce the model size such as pruning and quantization. Afterwards, we evaluate a solution to deploy the DNN models by pruning them in order to fit them into microcontrollers. Our results show that most models under test can be reduced to fit MCU memory for a maximum loss of 3% in average accuracy of the pruned models in comparison to the original models. Based on the results, we also discuss which methods are promising and which combination of model and feature work best for the given classification problem.

Beyond Peace and Security: The UN Transition Assistance Group in Namibia and its Importance for Contemporary Constitution-Making

In her article “Constitution-Making as a Technique of International Law: Reconsidering the Post-war Inheritance,” Anna Saunders focuses on constitution-making as an international practice of the past three decades and suggests that its “epistemic boundaries”—namely, the separation between the formal and material dimensions of constitution-making and the latter's exclusion from contemporary constitution-making assistance—were primarily established by scholarly work on constitution-making in the post-war era. Saunders explicitly acknowledges that her account is not the only possible history of constitution-making assistance. In this essay, I add a different layer to that history, focusing on the UN Transition Assistance Group (UNTAG) in Namibia. UNTAG is often considered the first instance of international constitution-making assistance, a practice that is generally understood to have emerged after the end of the Cold War. However, UNTAG's mandate, including its constitution-making assistance component, was in fact conceived many years before its actual deployment, dating back to the 1960s and 1970s. The essay shows that UN constitution-making assistance pre-dates the end of the Cold War and is linked to UN efforts to forge modern nation-states in the context of decolonization. I argue that this early case of constitution-making practice was an important blueprint for further iterations of international constitution-making assistance, not least because of the continuous involvement of individual international civil servants. Lastly, the case of Namibia is significantly different from the cases that inspired scholarly work in the post-war era, and we might ask to what extent the post-war inheritance affected this early international practice. I end with a brief reflection on Saunders's call to address the material dimension of constitution-making and caution against overemphasizing substantive questions in constitution-making assistance.

Efficient and reliable encapsulation for perovskite/silicon tandem solar modules.

Perovskite/silicon tandem solar cells have a tremendous potential to boost renewable electricity production thanks to their very high performance combined with promising cost structure. However, for actual field deployment, any solar cell technology needs to be assembled into modules, where the associated processes involve several challenges that may affect both the performance and stability of the devices. For instance, due to its hygroscopic nature, ethylene vinyl acetate (EVA) is incompatible with perovskite-based photovoltaics. To circumvent this issue, we investigate here two alternative encapsulant polymers for the packaging of perovskite/silicon tandems into minimodules: a thermoplastic polyurethane (TPU) and a thermoplastic polyolefin (TPO) elastomer. To gauge their impact on tandem-module performance and stability, we performed two internationally established accelerated module stability tests (IEC 61215): damp heat exposure and thermal cycling. Finally, to better understand the thermomechanical properties of the two encapsulants and gain insight into their relation to the thermal cycling of encapsulated tandems, we performed a dynamic mechanical thermal analysis. Our understanding of the packaging process of the tandem module provides useful insights for the development of commercially viable perovskite photovoltaics.

The revamped European Alternative Fuel Observatory (EAFO): progress so far and new developments ahead

The European Alternative Fuel Observatory (EAFO) is the main, central reference portal for publicly available information on alternative fuels (AF) vehicles and AF infrastructure in the EU, supporting policy-decision making. EAFO oversees the continuous improvement of open data availability, by collecting, verifying and analysis of relevant alternative fuels, vehicles, and infrastructure data to enable the assessment of the European Commission of the actual deployment of infrastructure by the Member States. The move of EAFO to the EC Europa domain in 2022 shows a clear signal to the Member States and to public authorities, about the growing relevance of alternative fuels. Besides the current process of AF vehicle fleet and sales and AF-infrastructure data for road transport, the scope of the portal extends to the maritime section and the development of an AF aviation section and rail.

ENHANCED DESIGN AND DEVELOPMENT SIMULATION AND TESTING

Enhanced design and development simulation and testing using digital twins represent a significant advancement in the field of renewable energy systems. Digital twins, which are virtual replicas of physical assets, allow for comprehensive simulation and testing of various scenarios and new technologies within a virtual environment. This approach offers numerous benefits, including the ability to optimize designs, predict performance, and troubleshoot potential issues before actual deployment. By leveraging high-fidelity models and real-time data, digital twins provide an invaluable tool for improving the efficiency, reliability, and sustainability of renewable energy systems. The use of digital twins in renewable energy extends across different stages of the project lifecycle. During the design phase, engineers can test multiple configurations and materials, ensuring that the final design meets all performance and durability requirements. This reduces the need for physical prototypes, saving both time and costs. In the development phase, digital twins enable the simulation of operational conditions, helping to identify and mitigate risks early. For instance, they can model the impact of environmental factors such as wind, temperature, and load variations on wind turbines or solar panels. Once the systems are operational, digital twins continue to play a crucial role in maintenance and optimization. They provide continuous monitoring and diagnostics, predicting failures before they occur and suggesting maintenance schedules that minimize downtime. This proactive maintenance approach enhances the longevity and efficiency of renewable energy assets. Moreover, digital twins facilitate the integration of renewable energy systems with existing power grids, ensuring smooth and efficient operation. The implementation of digital twins also supports the testing of innovative technologies and strategies, such as energy storage solutions and smart grid applications, in a controlled, virtual environment. This capability is essential for advancing renewable energy technologies and achieving higher levels of sustainability and energy efficiency. As the renewable energy sector continues to evolve, digital twins will undoubtedly become an integral part of the design, development, and operational processes, driving innovation and improving the overall effectiveness of renewable energy systems. Digital twins offer a transformative approach to enhancing the design and development of renewable energy systems. By enabling detailed simulation and testing in a virtual environment, they provide significant advantages in terms of optimization, risk mitigation, and operational efficiency, ultimately contributing to the advancement of sustainable energy solutions. Keywords: Design Optimization, Stimulation, Digital Twin, Virtual Environment.

RESEARCH OF FUNCTIONAL PARAMETERS OF INFOCOMMUNICATION NETWORKS 6G

The work presents the results of research into 6G technology, which will allow integrating all types of access in one mobile phone, ensuring smooth switching between services. New options for the use of mobile networks, such as sensing and artificial intelligence, are considered, which form new indicators of the quality of the information service – the resolution of sensing and the probability of output. For new applications, the prospects for further research into related indicators, such as flexibility and scalability to support own artificial intelligence services and network reliability level, are considered. Prospects for the development of electronic health care based on mobile communication systems with increased requirements for reliability, availability, security and confidentiality are considered. A platform for aerial mobility of various high-resolution sensors and cameras for various industries based on unmanned aerial vehicles has been investigated. The fundamental energy limitations of calculations related to information processing, which are an important step on the way to the successful deployment of 6G wireless networks, are considered. In the design of next-generation wireless communication networks, it has been determined that traditional AI optimization algorithms (such as federated learning) usually consider the bandwidth or delay of wireless connections as a weight for distributed multiprocessor data exchange, without considering the power energy bounds between different devices in different regions. This ambiguous consideration of AI energy limitations or power costs may lead to a large divergence between the design of a wireless network and the actual deployment of AI in the future. For this reason, equal importance must be attached to green AI and green communications. Therefore, when designing the architecture of the 6G information communication system, it is necessary to fully consider the impact of artificial intelligence models, algorithms and equipment on energy consumption to provide economic benefits to customers with corresponding operating costs of the system. The peculiarities of providing users with access to Internet services at any time through the same device regardless of their location, which is one of the ultimate goals of creating effective wireless networks, are discussed. The proposed concept of green AI called Oncefor-All (once and forever), according to which it is proposed to train an information communication network with further specialization during deployment will allow for efficient logical implementation on many devices, taking into account the given resource constraints. At the same time, terrestrial and non-terrestrial 6G networks will be fully integrated at the system level, ensuring the convergence of services, radio interfaces, networks and user devices. By organically combining these two access environments into one converged multi-layer heterogeneous network covering the entire globe, 6G technology will provide users with the same service. Ensuring the global delivery of mobile services will be an important aspect of the development of the 6G network.

Situation-Aware IoT Data Generation towards Performance Evaluation of IoT Middleware Platforms.

With the increasing growth of IoT applications in various sectors (e.g., manufacturing, healthcare, etc.), we are witnessing a rising demand of IoT middleware platform that host such IoT applications. Hence, there arises a need for new methods to assess the performance of IoT middleware platforms hosting IoT applications. While there are well established methods for performance analysis and testing of databases, and some for the Big data domain, such methods are still lacking support for IoT due to the complexity, heterogeneity of IoT application and their data. To overcome these limitations, in this paper, we present a novel situation-aware IoT data generation framework, namely, SA-IoTDG. Given a majority of IoT applications are event or situation driven, we leverage a situation-based approach in SA-IoTDG for generating situation-specific data relevant to the requirements of the IoT applications. SA-IoTDG includes a situation description system, a SySML model to capture IoT application requirements and a novel Markov chain-based approach that supports transition of IoT data generation based on the corresponding situations. The proposed framework will be beneficial for both researchers and IoT application developers to generate IoT data for their application and enable them to perform initial testing before the actual deployment. We demonstrate the proposed framework using a real-world example from IoT traffic monitoring. We conduct experimental evaluations to validate the ability of SA-IoTDG to generate IoT data similar to real-world data as well as enable conducting performance evaluations of IoT applications deployed on different IoT middleware platforms using the generated data. Experimental results present some promising outcomes that validate the efficacy of SA-IoTDG. Learning and lessons learnt from the results of experiments conclude the paper.

Demand Prediction and Optimal Allocation of Shared Bikes Around Urban Rail Transit Stations

The imbalance between the supply and demand of shared bikes is prominent in many urban rail transit stations, which urgently requires an efficient vehicle deployment strategy. In this paper, we propose an integrated model to optimize the deployment of shared bikes around urban rail transit stations, incorporating a seasonal autoregressive integrated moving average with long short-term memory (SARIMA-LSTM) hybrid model that is used to predict the heterogeneous demand for shared bikes in space and time. The shared bike deployment strategy was formulated based on the actual deployment process and under the principle of cost minimization involving labor and transportation. The model is applied using the big data of shared bikes in Xicheng District, Beijing. Results show that the SARIMA-LSTM hybrid model has great advantages in predicting the demand for shared bikes. The proposed allocation strategy provides a new way to solve the imbalance challenge between the supply and demand of shared bikes and contributes to the development of a sustainable transportation system.