Engineering School of Information and Digital Technologies

Automatic emotion recognition constitutes one of the great challenges providing new tools for more objective and quicker diagnosis, communication and research. Quick and accurate emotion recognition may increase possibilities of computers, robots, and integrated environments to recognize human emotions, and response accordingly to them a social rules. The purpose of this paper is to investigate the possibility of automated emotion representation, recognition and prediction its state-of-the-art and main directions for further research. We focus on the impact of emotion analysis and state of the arts of multimodal emotion detection. We present existing works, possibilities and existing methods to analyze emotion in text, sound, image, video and physiological signals. We also emphasize the most important features for all available emotion recognition modes. Finally, we present the available platform and outlines the existing projects, which deal with multimodal emotion analysis.

Data transparency is essential in the modern supply chain to improve trust and boost collaboration among partners. In this context, Blockchain is a promising technology to provide full transparency across the entire supply chain. However, Blockchain was originally designed to provide full transparency and uncontrolled data access. This leads many market actors to avoid Blockchain as they fear for their confidentiality. In this paper, we highlight the requirements and challenges of supply chain transparency. We then investigate a set of supply chain projects that tackle data transparency issues by utilizing Blockchain in their core platform in different manners. Furthermore, we analyze the projects’ techniques and the tools utilized to customize transparency. As a result of the projects’ analyses, we identified that further enhancements are needed to set a balance between the data transparency and process opacity required by different partners, to ensure the confidentiality of their processes and to control access to sensitive data.

Edge Computing (EC) is a promising concept to overcome some obstacles of traditional cloud data centers to support Internet of Things (IoT) applications, especially time-sensitive applications. However, EC faces some challenges, including the resource allocation for heterogeneous applications at a network edge composed of distributed and resource-restricted nodes. A relevant issue that needs to be addressed by a resource manager is the service placement problem, which is the decision-making process of determining where to place different services (or applications). A related issue of service placement is how to distribute workloads of an application placed on multiple locations. Hence, we jointly investigate the load distribution and placement of IoT applications to minimize Service Level Agreement (SLA) violations due to the limitations of EC resources and other conflicting objectives. In order to handle the computational complexity of the formulated problem, we propose a multi-objective genetic algorithm with the initial population based on random and heuristic solutions to obtain near-optimal solutions. Evaluation results show that our proposal outperforms other benchmark algorithms in terms of response deadline violation, as well as terms of other conflicting objectives, such as operational cost and service availability.

Data aggregation is one of the key features used in databases, especially for Business Intelligence (e.g., ETL, OLAP) and analytics/data mining. When considering SQL databases, aggregation is used to prepare and visualize data for deeper analyses. However, these operations are often impossible on very large volumes of data regarding memory-and-time-consumption. In this paper, we show how NoSQL databases such as MongoDB and its key-value stores, thanks to the native MapReduce algorithm, can provide an efficient framework to aggregate large volumes of data. We provide basic material about the MapReduce algorithm, the different NoSQL databases (read intensive vs. write intensive). We investigate how to efficiently modelize the data framework for BI and analytics. For this purpose, we focus on read intensive NoSQL databases using MongoDB and we show how NoSQL and MapReduce can help handling large volumes of data.

International audience

A novel domain for Index Modulation (IM) named “Filter Domain” is proposed. This new domain generalizes many existing modulations and IM domains. In addition, a novel scheme “Filter Shape Index Modulation” (FSIM) is proposed. This FSIM scheme allows a higher Spectral Efficiency (SE) gain than the time and frequency IM dimensions in Single-Input Single-Output (SISO) systems. In the FSIM system, the bit-stream is mapped using an Amplitude Phase Modulation (APM) as QAM or PSK, and an index of a filter-shape changing at the symbol rate. This filter shape, being changed at each symbol, enables a SE gain in SISO system without sacrificing any time or frequency resources. Compared to an equivalent 8QAM and 16QAM schemes and at the same SE, the FSIM with QPSK using 2 and 4 non-optimal filter shapes achieves a gain of 3.8 dB and 1.7 dB respectively at BER= 10 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">-4</sup> , and this superiority is maintained in frequency selective fading channel compared to equivalent SISO-IM schemes. A low complexity detection scheme, approaching the maximum likelihood detector performance, is proposed along with a full performance characterization in terms of theoretical probability of filter index error and BER lower bound. Finally, FSIM can achieve better spectral and energy efficiencies when a filter bank and an ISI cancellation technique are optimally designed.

Many of today's companies use Smart Contracts to represent and execute their business processes. Smart contracts are self-executed programs running over blockchain. In this context, composite smart contracts are used to represent collaborative business processes. A composite smart contract is a smart contract that needs to execute other contracts using external calls to achieve its tasks. Composite smart contracts, through the use of external calls and the execution of other smart contracts that might belong to other owners or companies, bring many challenges with regard to security requirements. As a result, special efforts must be done to ensure composite smart contracts security verification. In this paper, we propose a novel approach to verify the security and the correctness of the composite smart contracts written in solidity in Ethereum blockchain. This approach is based on the finite state machine models and model checking method for modeling and verifying the composite smart contracts respectively. We consider seven security properties as well as the security issues that depend on the contract context to be checked in the composite smart contract. For this, we provide two different yet complementary types of verification. The first type of verification is applied to all smart contracts with properties called in our approach “standard properties” that represent the generic ones, while the second type considers the context-dependent properties that we called “specific properties” varying from one smart contract to another. Finally, we express all properties using computation tree logic formulae and we use the nuXmv symbolic model checker to verify the model against all properties. This approach is validated using a different set of solidity smart contracts.

During the last years we have seen a dramatic increase of new Cloud providers, applications, services, management platforms, data, etc. reaching a level of complexity that implies the necessity of new solutions to deal with such vast, shared and heterogeneous services and resources. Consequently, challenges often related to interoperability, portability, security, discovery, selection, negotiation and description of cloud service and resource may take place. In this sense, Semantic Web Technologies, holding a great potential to cloud computing, have been proven as an efficient means to relive these challenges. This paper examines and explores the role of Semantic Web Technologies in the cloud from a wide variety of literatures. Various approaches, architectures, and frameworks are screened and evaluated based on eight prime research questions. At the end of the review, research opportunities in the form of a roadmap are discussed.

This paper presents Internet of Things in a wider context. Main enabling factor of this concept is the integration of various technologies. In this paper, we describe the key technologies involved in the implementation of Internet of Things and the major application domain where the Internet of Things will play a vital role. Later we will discuss about the open issues which are to be addressed before the worldwide acceptance of these technologies. There are lots of open issues to address. Here we address the most relevant among them in detail.

Reviewing the current state of knowledge on sustainable production, this paper opens the Special Section entitled “Sustainability in production in the context of Industry 4.0”. The fourth industrial revolution (Industry 4.0), which embodies a vision for the future system of manufacturing (production), focuses on how to use contemporary methods (i.e. computerization, robotization, automation, new business models, etc.) to integrate all manufacturing industry systems to achieve sustainability. The idea was introduced in 2011 by the German government to promote automation in manufacturing. This paper shows the state of the art in the application of modern methods in sustainable manufacturing in the context of Industry 4.0. The authors review the past and current state of knowledge in this regard and describe the known limitations, directions for further research, and industrial applications of the most promising ideas and technologies.

Plant pathogens are a serious threat to agriculture for long-term viability and cost loss of billions of dollars yearly. Many pathogens have been documented in the literature that infect a variety of crops. Nevertheless, new pathogens emerge and often result in disease outbreaks, leading to million-dollar losses. Currently, several diagnostic approaches with enhanced sensitivity and specificity for the identification of widespread and/or unknown plant pathogens are constantly being developed. Whereas the extensively used approaches for plant pathogen diagnostics are mostly serological and nucleic acid-based assays, many different nucleic acid-based approaches for amplifying target DNA/RNA have also emerged over time. However, these approaches lack precision, specificity, and rapidity, making them unsuitable for on-field analysis. As a result, there is a lot of interest arising in field-deployable point of care (POC) devices and artificial intelligence (AI)-assisted pathogens’ detection accurately at an early stage within a minute. Similarly, development of a cell-lysis and purification-free DNA/RNA extraction process is also crucial for quick sample preparation for molecular diagnosis of plant pathogens at field level. In this review, we have discussed advanced tools that are trending not only to extract nucleic acids but also detect plant pathogens. We have also discussed critical challenges and future perspectives of disease diagnostic tools for plant pathogens’ detection. In summary, advanced plant disease diagnostic tools can be helpful for routine monitoring of plant pathogens toward improving crop productivity and yield that can be used for improving the financial status of farmers.

In the era of advancing technology, unmanned inspection robots have become indispensable for their efficiency, precision, and safety. Key to their autonomous operation is Simultaneous Localization and Mapping (SLAM) technology, which allows robots to navigate and create maps of unknown environments in real-time. This article explores the integration of SLAM with artificial intelligence, highlighting its role in robotic navigation, localization, and obstacle avoidance. Specifically, we delve into SLAM's principles, its implementation with LiDAR technology, and its application in autonomous robot localization. Furthermore, we introduce a collaborative mapping algorithm based on ORB-SLAM3, enhancing map construction efficiency and real-time performance. Through our exploration, we illustrate the transformative potential of SLAM technology, paving the way for safer and more efficient robotic inspection systems across various industries.

Edge Computing emerges as a solution that overcomes some obstacles of traditional central data centers to support the performance-critical Internet of Things applications. However, a challenge therein is the resource allocation for heterogeneous applications at a network edge composed of distributed and resource-restricted nodes. In this paper, we investigate how to place replicas of applications, and distribute requests among these replicas to optimize multiple objectives. We propose a genetic algorithm based on Pareto fronts as a problem-solving meta-heuristic to prioritize latency-sensitive applications and optimize conflicted objectives. Evaluation results show that our proposal outperforms other benchmark algorithms in terms of response deadline violation, as well as terms of other important and sometimes conflicting objectives, such as cost and availability.

Pattern Recognition (SIPR 2016) and The Second International Conference of Networks, Communications, Wireless and Mobile Computing (NCWC 2016) were collocated with the CSEN-2016.The conferences attracted many local and international delegates, presenting a balanced mixture of intellect from the East and from the West.The goal of this conference series is to bring together researchers and practitioners from academia and industry to focus on understanding computer science and information technology and to establish new collaborations in these areas.Authors are invited to contribute to the conference by submitting articles that illustrate research results, projects, survey work and industrial experiences describing significant advances in all areas of computer science and information technology.The CSEN-2016, SIPR-2016, NCWC-2016 Committees rigorously invited submissions for many months from researchers, scientists, engineers, students and practitioners related to the relevant themes and tracks of the workshop.This effort guaranteed submissions from an unparalleled number of internationally recognized top-level researchers.All the submissions underwent a strenuous peer review process which comprised expert reviewers.These reviewers were selected from a talented pool of Technical Committee members and external reviewers on the basis of their expertise.The papers were then reviewed based on their contributions, technical content, originality and clarity.The entire process, which includes the submission, review and acceptance processes, was done electronically.All these efforts undertaken by the Organizing and Technical Committees led to an exciting, rich and a high quality technical conference program, which featured high-impact presentations for all attendees to enjoy, appreciate and expand their expertise in the latest developments in computer network and communications research.In closing, CSEN-2016, SIPR-2016, NCWC-2016 brought together researchers, scientists, engineers, students and practitioners to exchange and share their experiences, new ideas and research results in all aspects of the main workshop themes and tracks, and to discuss the practical challenges encountered and the solutions adopted.The book is organized as a collection of papers from the CSEN-2016, SIPR-2016, NCWC-2016.We would like to thank the General and Program Chairs, organization staff, the members of the Technical Program Committees and external reviewers for their excellent and tireless work.We sincerely wish that all attendees benefited scientifically from the conference and wish them every success in their research.It is the humble wish of the conference organizers that the professional dialogue among the researchers, scientists, engineers, students and educators continues beyond the event and that the friendships and collaborations forged will linger and prosper for many years to come.

microRNAs (miRNAs) are small non-coding RNAs regulating gene expression. They have attracted significant interest as biomarkers for early diagnosis, prediction and monitoring of treatment response in many diseases. As individual miRNAs often lack the required sensitivity and specificity, miRNA signatures are developed for clinical applications. Digital PCR (dPCR) is a sensitive fluorescent-based quantification method, that can be used to detect the expression of miRNAs in patient samples. Our study presents the first proof-of-concept of a multiplexed dPCR assay for the simultaneous analysis and quantification of multiple miRNAs. After reverse transcription (RT) using a pool of miRNA-specific stem-loop primers, dPCR was performed with a universal reverse primer and miRNA-specific forward primers along with fluorescently-labelled hydrolysis probes. Multiple experimental parameters were evaluated and strategies for modulating the observed signals were devised. The optimised assay was applied to the analysis of miRNAs from cell lines and biological samples. Although absolute quantification was lost, due to the reverse transcription step, quantification was linear for the dilution series and results were highly reproducible for independent dPCR and RT reactions. Our results confirmed the high sensitivity of dPCR for patient samples. We demonstrate the feasibility and reliability of multiplexed detection and quantification of miRNAs by dPCR that can be applied in a clinical setting to evaluate miRNA signatures. • MiRNA signatures represent promising biomarkers for clinical applications. • First proof-of-concept of a multiplexed digital PCR assay for miRNA analysis. • Combination of miRNA-specific stem-loop primers and dPCR with hydrolysis probes. • Linear and reproducible quantification results for up to six miRNAs. • Optimised protocol can be applied to different types of biological samples.

With the rapid development of artificial intelligence and robot technology, SLAM technology, as a key component, has been paid more and more attention. SLAM technology enables robots to autonomously navigate, build maps, and achieve accurate positioning in unknown environments, providing strong support for the autonomy and intelligence of robots and unmanned vehicles. In this paper, the position prediction method of flying object based on SLAM technology and the application of EvolveGCN model in behavior prediction are introduced. First, through the fusion of SLAM technology and liDAR data, we can accurately predict the position and movement trajectory of flying objects, thereby improving the safety and efficiency of the system. Secondly, with the EvolveGCN model, we are able to capture dynamic changes in the environment and achieve accurate predictions of the behavior of flying objects. Through experimental verification, the prediction accuracy of our method has been significantly improved in both simulation and real environment, which indicates the feasibility and effectiveness of the method in practical application, and provides an important reference and technical support for the development of autonomous navigation, aerial surveillance and other fields.

Indoor localization has recently witnessed an increase in interest due to its wide range of potential services. Further, the location information is very important in many applications, such as the Internet of Things, logistics, library management and so on. Hence, different technologies and techniques have been proposed in the literature for indoor localization systems. Most of these systems present the disadvantages of a poor performance, low accuracy and high cost. However, thanks to its low cost, high accuracy and non-line-of-sight detection, radio frequency identification (RFID)-based localization has increasingly become the most used technology for indoor localization. In this paper, we propose an innovative approach based on the multiple input single output (MISO) protocol to improve the accuracy of a low-cost RFID localization system. Whereas most traditional systems use a single tag for localization, the proposed architecture encourages the use of a group of RFID tags named as a constellation. According to experimental results and based on the signals' diversity, the location accuracy is improved to get an estimated position error of 81 cm at the cumulative distribution function of 90%.

International audience

Programmed death-ligand 1 (PD-L1) plays a key role in maintaining immune tolerance and also in immune evasion of cancers and pathogens. Though the identity of stimuli that induce PD-L1 in various human innate cells and their function are relatively well studied, data on the basophils remain scarce. In this study, we have identified one of the factors, such as IFN-γ, that induces PD-L1 expression in human basophils. Interestingly, we found that basophil priming by IL-3 is indispensable for IFN-γ-induced PD-L1 expression in human basophils. However, priming by other cytokines including granulocyte-macrophage colony-stimulating factor (GM-CSF) and thymic stromal lymphopoietin (TSLP) was dispensable. Analyses of a published microarray data set on IL-3-treated basophils indicated that IL-3 enhances IFNGR2, one of the chains of the IFNGR heterodimer complex, and CD274, thus providing a mechanistic insight into the role of IL-3 priming in IFN-γ-induced PD-L1 expression in human basophils.

Deep Learning architectures, albeit successful in most computer vision tasks, were designed for data with an underlying Euclidean structure, which is not usually fulfilled since pre-processed data may lie on a non-linear space. In this paper, we propose a geometry aware deep learning approach using rigid and non rigid transformation optimization for skeleton-based action recognition. Skeleton sequences are first modeled as trajectories on Kendall’s shape space and then mapped to the linear tangent space. The resulting structured data are then fed to a deep learning architecture, which includes a layer that optimizes over rigid and non rigid transformations of the 3D skeletons, followed by a CNN-LSTM network. The assessment on two large scale skeleton datasets, namely NTU-RGB+D and NTU-RGB+D 120, has proven that the proposed approach outperforms existing geometric deep learning methods and exceeds recently published approaches with respect to the majority of configurations.