Back-end Part Research Articles

Extended Reality Telemedicine Collaboration System Using Patient Avatar Based on 3D Body Pose Estimation.

Recent advances in extended reality (XR) technology have opened the possibility of significantly improving telemedicine systems. This is primarily achieved by transferring 3D information about patient state, which is utilized to create more immersive experiences on VR/AR headsets. In this paper, we propose an XR-based telemedicine collaboration system in which the patient is represented as a 3D avatar in an XR space shared by local and remote clinicians. The proposed system consists of an AR client application running on Microsoft HoloLens 2 used by a local clinician, a VR client application running on the HTC vive Pro used by a remote clinician, and a backend part running on the server. The patient is captured by a camera on the AR side, and the 3D body pose estimation is performed on frames from this camera stream to form a 3D patient avatar. Additionally, the AR and VR sides can interact with the patient avatar via virtual hands, and annotations can be performed on a 3D model. The main contribution of our work is the use of 3D body pose estimation for the creation of a 3D patient avatar. In this way, 3D body reconstruction using depth cameras is avoided, which reduces system complexity and hardware and network resources. Another contribution is the novel architecture of the proposed system, where audio and video streaming are realized using WebRTC protocol. The performance evaluation showed that the proposed system ensures high frame rates for both AR and VR client applications, while the processing latency remains at an acceptable level.

Сross-platform application for NFT tokens

A software application has been developed, which is a platform for creating, storing, viewing and selling NFT tokens. The object of research is the process of managing NFT tokens, the subject of research is the development and implementation of the NFT token marketplace. The developed application is a micro service that includes an API for working with NFT tokens and a client part that provides a convenient way to use the marketplace for users. The backend part was developed in the C# programming language, and the frontend part uses React and Redux. OOP methods, the Git version control system, namely Github, work with multithread, asynchrony and implemented SOLID principles were used during development. The application allows users to access NFT tokens, perform various operations with them and manage tokens. The libraries used in the application ensure stability and security in the process of working with tokens. This application allows for efficient and secure exchange of NFT tokens between different systems and users. The developed software product is built according to the principle of Clean Architecture, which will provide easy further expansion of the scope of the project.

Concept of Delivery System in the Smart City Environment

Regarding to the smart city infrastructures, there is a demand for big data processing and its further usage. This data can be gained by various means. There are many IoT devices in the city, which can communicate and share the information about the environment which they are situated in. Moreover every personal mobile device can also participate in this process and help to gain data via various applications. Every app provides the useful data, enabling the location and data sharing. This data can be further processed and used for improving the city infrastructure, transport or other services. We designed the system for shared delivery process, which can help to achieve the described situation. It consists of frontend and backend part. The frontend part, multiplatform mobile app, represents the graphical interface and the backend part represents the database for the gained data.

Learning the Optimal Partition for Collaborative DNN Training With Privacy Requirements

With the growth of intelligent Internet of Things (IoT) applications and services, deep neural network (DNN) has become the core method to power and enable increased functionality in many smart IoT devices. However, DNN training is difficult to carry out on end devices because it requires a great deal of computational power. The conventional approach to DNN training is generally implemented on a powerful computation server; nevertheless, this approach violates privacy because it exposes the training data to curious service providers. In this article, we consider a collaborative DNN training system between a resource-constrained end device and a powerful edge server, aiming at partitioning a DNN into a front-end part running on the end device and a back-end part running on the edge server to accelerate the training process while preserving the privacy of the training data. With the key challenge being how to locate the optimal partition point to minimize the end-to-end training delay, we propose an online learning module, called learn-to-split (L2S), to adaptively learn the optimal partition point on the fly. This approach is unlike existing efforts on DNN partitioning that relies heavily on a dedicated offline profiling stage. In particular, we design a new contextual bandit learning algorithm called LinUCB-E as the basis of L2S, which has provable theoretical learning performance and is ultralightweight for easy real-world implementation. We implement a prototype system consisting of an end device and an edge server, and experimental results demonstrate that L2S can significantly outperform state-of-the-art benchmarks in terms of reducing the end-to-end training delay and preserving privacy.

Currency Denomination Identifier Application

Abstract: For visually challenged people, distinguishing between different denominations of cash is a difficult process. Even though unique symbols are engraved on various currencies in India, the work is still difficult for blind individuals. The inadequacy of identifying devices prompted the development of a portable gadget for denominational segregation. This project aims to create an Android application that will assist visually and hearing-impaired people in detecting Indian cash denominations by putting a banknote in front of the camera. The work uses machine learning and Android programming approaches and is based on real-world applications. The android application uses text to speech concept to read the value of note to the user and then it converts the text value into speech. To harness the power of them all, we are leveraging the Keras, TensorFlow, Fastai, and PyTorch libraries, as well as different machine learning techniques like ResNet and MobileNet. Various technologies like machine learning models, python and many more libraries are used for the backend part of application. And for front end, java concepts and android development techniques are employed. Altogether they are integrated into a single platform which is highly user-friendly and makes it easy to use and implement in our daily life. Keywords: Denomination, Hearing-impaired, Android, Keras, TensorFlow, Fastai, PyTorch, ResNet, MobileNet

MS-SLAM: Motion State Decision of Keyframes for UAV-Based Vision Localization

To improve the robustness of UAVs in rotating and motion scenes, we propose a stereo simultaneous localization and mapping (SLAM) for UAVs with a new keyframe strategy that differs from current SLAM systems. Moreover, it has two strategies. 1) In the dominant strategy based on image quality, we filter and retain the strong feature points in each image frame by our own defined rules; this leads to longer survival time and attribute invariance in image tracking, and we save the image frames containing more strong feature points as keyframes. 2) In the secondary strategy based on motion state, we quantify the motion state during the UAV’s motion (called the compound rotation amount), and characterize the intensity of the motion. Since we want the UAV to have better robustness in the rotating scene, this strategy generates keyframes when the compound rotation amount meets the threshold. These two strategies are used to cope with gentle motion scenes and rotational motion scenes, respectively. Thus, the insertion of our keyframes is determined by the motion state; our SLAM system is proposed based on the motion state (MS-SLAM). In the back-end part of the system we construct a new weighted cost function to optimize the pose. Finally, through comparison experiments on the public dataset EuRoc, we demonstrate that our algorithm is more advantageous than some current mainstream algorithms. In difficult sequences, our algorithm compares the absolute trajectory error with ORB-SLAM2, SVO+gtsam, and VINS-Mono, the absolute trajectory error of our algorithm can be reduced by 87%.

FrailSafe: An ICT Platform for Unobtrusive Sensing of Multi-Domain Frailty for Personalized Interventions.

The implications of frailty in older adults' health status and autonomy necessitates the understanding and effective management of this widespread condition as a priority for modern societies. Despite its importance, we still stand far from early detection, effective management and prevention of frailty. One of the most important reasons for this is the lack of sensitive instruments able to early identify frailty and pre-frailty conditions. The FrailSafe system provides a novel approach to this complex, medical, social and public health problem. It aspires to identify the most important components of frailty, construct cumulative metrics serving as biomarkers, and apply this knowledge and expertise for self-management and prevention. This paper presents a high-level overview of the FrailSafe system architecture providing details on the monitoring sensors and devices, the software front-ends for the interaction of the users with the system, as well as the back-end part including the data analysis and decision support modules. Data storage, remote processing and security issues are also discussed. The evaluation of the system by older individuals from 3 different countries highlighted the potential of frailty prediction strategies based on information and communication technology (ICT).

A Novel RGB-D SLAM Algorithm Based on Cloud Robotics.

In this paper, we present a novel red-green-blue-depth simultaneous localization and mapping (RGB-D SLAM) algorithm based on cloud robotics, which combines RGB-D SLAM with the cloud robot and offloads the back-end process of the RGB-D SLAM algorithm to the cloud. This paper analyzes the front and back parts of the original RGB-D SLAM algorithm and improves the algorithm from three aspects: feature extraction, point cloud registration, and pose optimization. Experiments show the superiority of the improved algorithm. In addition, taking advantage of the cloud robotics, the RGB-D SLAM algorithm is combined with the cloud robot and the back-end part of the computationally intensive algorithm is offloaded to the cloud. Experimental validation is provided, which compares the cloud robotic-based RGB-D SLAM algorithm with the local RGB-D SLAM algorithm. The results of the experiments demonstrate the superiority of our framework. The combination of cloud robotics and RGB-D SLAM can not only improve the efficiency of SLAM but also reduce the robot’s price and size.

Sensorial Network Framework Embedded in Ubiquitous Mobile Devices

: Today’s digital society is interconnected and networked, with modern smart devices ubiquitously built into and embedded within smart environments and other environments, where people (their users) typically live. It is very important to mention that sensorial awareness of an environment depends on one’s current location and equipment, as well as the equipment’s real-time capabilities. Personal sensorial information is considered to be the key factor for progress in the improvement of the productivity of everyday life and creation of a smart surrounding environment. This paper describes the design, implementation, and testing process of a new sensorial framework based on the current possibilities created by ubiquitous smart mobile devices with sensors, which involves computing power and battery power issues. The two parts of the proposed framework have been designed, implemented, and tested. The client part is represented by a front-end mobile application, and the back-end part is represented by a server-side application. The analysis of the data, captured during the testing phase, involves the analysis of the processing time, battery consumption, and transmitted data amount. This analysis reveals that Transmission Control Protocol (TCP) and user datagram protocol (UDP) protocols have a comparable performance, although TCP is preferable for use in local networks. In comparison to other solutions such as MobiSense or Feel the World framework, the final solution of the proposed and developed sensorial framework has two main capabilities, which are the security support and social networking possibilities. The advantage of the MobiSense platform is the existence of several real-world applications, whereas the proposed sensorial framework needs to be verified in the massive context of many users in real time.

A mixed-signal large dynamic range front-end ASIC for high capacitance detectors

A 64-channel mixed-mode ASIC, suitable for particle detectors of large dynamic range and high capacitance up to hundreds of pF, is presented here. Each channel features an analogue front-end for signal amplification and filtering, and a mixed signal back-end to digitise and store the signal information. The analogue part consists of a low input-impedance programmable gain pre-amplifier based on a regulated common-gate (RCG) input stage, two shapers optimised for time and energy measurements. The back-end part mainly includes discriminators, TDCs and ADCs, which are used to process the signal and encode both the time of arrival and the charge in the input signal with a fully digital output. The programmable gain of the front-end (up to 400 fC input dynamic range) and the versatile back-end allow the readout of different gaseous detectors like GEM, MicroMEGAS and MWPC. The ASIC is designed for an event rate up to 100 kHz per channel and a power consumption less than 9 mW/channel, has been fabricated in a 110 nm CMOS technology.

Feedwaterheating in wasteheat boiler with exhaust gases from gas-turbine power unit

In order to increase cost effectiveness of steam-gas power station, it was suggested to place heat exchange surface in backend part of wasteheat boiler downstream of condensate gas pre-heater to heat feedwater before it is transferred to chemical treatment. Estimations have been made for steam-gas unit PGU-450. In process conditions of steam-gas unit operation, wasteheat boiler energy conversion efficiency increases from 82.65% to 84.97% (by 2.32%) when gas feedwater heater is installed in its low-temperature area. In addition to this, steam-gas unit efficiency increases from 52.99% to 53.53% (by 0.54%) and specific consumption of reference fuel for power generation is reduced by 1.01% (from 232.12 to 229.78 g/(kWt⋅h)). Savings of reference fuel in monetary terms for steam-gas unit with the capacity of 452.35 MWt and running time of 6500 h/year and reference fuel priced at 3000 rub/t of reference fuel are amount to 20 million 640 thousand rubles per year.

다음 장소 예측을 위한 맵리듀스 기반의 이동 패턴 마이닝 시스템 설계

본 논문에서는 모바일 기기 사용자들의 다음 방문 장소를 효율적으로 예측할 수 있는 맵리듀스 기반의 이동 패턴 마이닝 시스템을 소개한다. 이 시스템은 대용량의 사용자 이동 궤적 데이터 집합으로부터 은닉 마코프 모델로 표현되는 각 사용자의 이동 패턴을 학습해내고, 이 모델을 현재 이동 궤적에 적용함으로써 다음 방문 장소를 예측한다. 본 시스템은 사용자별 이동 패턴 모델을 학습하는 후단부와 실시간으로 다음 방문 장소를 예측하는 전단부 등 크게 두 부분으로 구성된다. 이 중에서 후단부는 주요 장소 추출, 이동 궤적 변환, 이동 패턴 모델 학습 등 총 3개의 맵리듀스 작업 모듈들로 구성된다. 이에 반해, 본 시스템의 전단부는 이동 경로 후보군 생성, 다음 장소 예측 등 총 2개의 작업 모듈들로 구성된다. 그리고 본 시스템을 구성하는 각 작업 모듈의 맵과 리듀스 함수들은 하둡 인프라를 효과적으로 활용하여 병렬 처리를 극대화할 수 있도록 설계하였다. 대용량의 공개 벤치마크 데이터 집합인 GeoLife를 이용하여 본 논문에서 소개한 시스템의 성능을 분석하기 위한 실험들을 수행하였고, 실험 결과를 통해 본 시스템의 높은 성능을 확인할 수 있었다. In this paper, we present a MapReduce-based mobility pattern mining system which can predict efficiently the next place of mobile users. It learns the mobility pattern model of each user, represented by Hidden Markov Models(HMM), from a large-scale trajectory dataset, and then predicts the next place for the user to visit by applying the learned models to the current trajectory. Our system consists of two parts: the back-end part, in which the mobility pattern models are learned for individual users, and the front-end part, where the next place for a certain user to visit is predicted based on the mobility pattern models. While the back-end part comprises of three distinct MapReduce modules for POI extraction, trajectory transformation, and mobility pattern model learning, the front-end part has two different modules for candidate route generation and next place prediction. Map and reduce functions of each module in our system were designed to utilize the underlying Hadoop infrastructure enough to maximize the parallel processing. We performed experiments to evaluate the performance of the proposed system by using a large-scale open benchmark dataset, GeoLife, and then could make sure of high performance of our system as results of the experiments.

Database Backend as a Service: Automatic Generation, Deployment, and Management of Database Backends for Mobile Applications

Managing data in the Cloud is a challenging task, especially scaling resources in order to prevent under- and over-provisioning. In this paper, we consider a specific domain of applications, namely mobile applications for events like conferences or festivals, where automatic managing and scaling the backend part of the application would be beneficial in terms of efficient resource utilization as well as a good end-user experience. In order to achieve this, we make the following contributions. We automate generation, deployment, operation, and management of the backend in an Infrastructure as a Service Cloud. Thereby, we address the fluctuating load characteristics of mobile applications for events by applying our monitoring and autoscaling framework based on data stream processing and complex event processing.

Design of a Remote Wireless Surveillance System Based on 3G

The remote wireless surveillance system based on 3G consist of the front-end image capture module, data transmission module, and the back-end monitoring part. The basis elements of the system and the principle is discussed .The system of software and the hardware is introduced. The system is high-compressed, of high quality ,easily wiring, and of high efficiency. The best use of the system is to surveillance the complicated enviroment of a large scale project.

The NA62 Liquid Krypton calorimeter readout module

The NA62 experiment [1] at CERN SPS (Super Proton Synchrotron) accelerator will be focused on precision tests of the Standard Model via studies of ultra-rare decays of charged kaons. The high resolution Liquid Krypton (LKr) calorimeter of the former NA48 experiment [2], together with other detectors, will provide a photon-veto with hermetic coverage from zero out to large angles from the decay region. The old backend electronics [3] does not satisfy the NA62 specifications and the study of a new readout system began in 2008.This paper presents the Calorimeter REAdout Module (CREAM), an upgrade project for the backend part of the LKr data acquisition chain [3]. The CREAMs will provide 40 MHz sampling of 13248 calorimeter channels, data buffering during the SPS spill, zero suppression, and programmable trigger sums for the experiment trigger processor.

A Robust Frequency Tracking Loop for Energy-Efficient Crystalless WBAN Systems

This brief presents a frequency tracking loop (FTL) to realize a crystalless wireless sensor node (WSN) for wireless body area network (WBAN). By tracking a remote wireless RF reference for system clock calibration, the proposed FTL allows WSNs to tolerate a large-frequency error from on-chip CMOS oscillators. Moreover, to achieve energy-efficient transmissions in crystalless, a sufficiently accurate convergence clock is required to enable burst overmegabits-per-second system throughput with minimized operation duty cycle. For the dedicated purpose, a comparison-based binary-search tracking scheme, which ensures accurate and robust convergence against noisy wireless channel, is further developed to manage the operation of FTL. The intermediate frequency back-end part of FTL is implemented in 90-nm CMOS process. Measurement results show that the FTL extends an initial tolerance of system clock error to ±3% and achieves a final quartz-crystal comparable ±50-ppm accuracy. This enables 4.85-Mb/s wireless links and improves 79% energy efficiency by RF operation-time reduction, giving a power-saved and miniaturized WSN device for WBAN applications.