Reduce Communication Time Research Articles

The application of improved densenet algorithm in accurate image recognition

Image recognition technology belongs to an important research field of artificial intelligence. In order to enhance the application value of image recognition technology in the field of computer vision and improve the technical dilemma of image recognition, the research improves the feature reuse method of dense convolutional network. Based on gradient quantization, traditional parallel algorithms have been improved. This improvement allows for independent parameter updates layer by layer, reducing communication time and data volume. The introduction of quantization error reduces the impact of gradient loss on model convergence. The test results show that the improvement strategy designed by the research improves the model parameter efficiency while ensuring the recognition effect. Narrowing the learning rate is conducive to refining the updating granularity of model parameters, and deepening the number of network layers can effectively improve the final recognition accuracy and convergence effect of the model. It is better than the existing state-of-the-art image recognition models, visual geometry group and EfficientNet. The parallel acceleration algorithm, which is improved by the gradient quantization, performs better than the traditional synchronous data parallel algorithm, and the improvement of the acceleration ratio is obvious. Compared with the traditional synchronous data parallel algorithm and stale synchronous parallel algorithm, the optimized parallel acceleration algorithm of the study ensures the image data training speed and solves the bottleneck problem of communication data. The model designed by the research improves the accuracy and training speed of image recognition technology and expands the use of image recognition technology in the field of computer vision.Please confirm the affiliation details of [1] is correct.The relevant detailed information in reference [1] has been confirmed to be correct.

DTQFL: A Digital Twin-Assisted Quantum Federated Learning Algorithm for Intelligent Diagnosis in 5G Mobile Network.

Smart healthcare aims to revolutionize med-ical services by integrating artificial intelligence (AI). The limitations of classical machine learning include privacy concerns that prevent direct data sharing among medical institutions, untimely updates, and long training times. To address these issues, this study proposes a digital twin-assisted quantum federated learning algorithm (DTQFL). By leveraging the 5G mobile network, digital twins (DT) of patients can be created instantly using data from various Internet of Medical Things (IoMT) devices and simultane-ously reduce communication time in federated learning (FL) at the same time. DTQFL generates DT for patients with specific diseases, allowing for synchronous training and updating of the variational quantum neural network (VQNN) without disrupting the VQNN in the real world. This study utilized DTQFL to train its own personalized VQNN for each hospital, considering privacy security and training speed. Simultaneously, the personalized VQNN of each hospital was obtained through further local iterations of the final global parameters. The results indicate that DTQFL can train a good VQNN without collecting local data while achieving accuracy comparable to that of data-centralized algorithms. In addition, after personalized train-ing, the VQNN can achieve higher accuracy than that with-out personalized training.

Dominating Set Model Aggregation for communication-efficient decentralized deep learning

Decentralized deep learning algorithms leverage peer-to-peer communication of model parameters and/or gradients over communication graphs among the learning agents with access to their private data sets. The majority of the studies in this area focus on achieving high accuracy, with many at the expense of increased communication overhead among the agents. However, large peer-to-peer communication overhead often becomes a practical challenge, especially in harsh environments such as for an underwater sensor network. In this paper, we aim to reduce communication overhead while achieving similar performance as the state-of-the-art algorithms. To achieve this, we use the concept of Minimum Connected Dominating Set from graph theory that is applied in ad hoc wireless networks to address communication overhead issues. Specifically, we propose a new decentralized deep learning algorithm called minimum connected Dominating Set Model Aggregation (DSMA). We investigate the efficacy of our method for different communication graph topologies with a small to large number of agents using varied neural network model architectures. Empirical results on benchmark data sets show a significant (up to 100X) reduction in communication time while preserving the accuracy or in some cases, increasing it compared to the state-of-the-art methods. We also present an analysis to show the convergence of our proposed algorithm.

Efficient Collaboration Between Radiologists Using the PACS-Integrated Refer Function to Reduce Communication Times.

The purpose of this study was to assess the utility of a picture archiving and communication systems (PACS)-integrated refer function for improving collaboration between radiologists and radiographers during daily reading sessions. Retrospective analysis was conducted on refers sent by radiologists using a PACS-integrated refer system from March 2020 to December 2021. Refers were categorized according to receiver: radiologists in the same division (intra-division), radiologists in a different division (inter-division), and radiographers. The proportions of answered refers, content of refers, and timing of refer posts were evaluated. Additionally, time intervals in minutes from initial refer post to refer response were assessed to assess the efficiency of the refer system and compared according to receivers using the Mann-Whitney U test. Among a total of 691 refers posted by radiologists, 579 (83.8%) were answered directly using the refer function in PACS. Of the answered refers, 346 refers (59.8%) were made between radiologists, and 173 (50%) were intra-division refers. About the content of refers, about 82.6% of radiologists' refers were about imaging interpretation consultation, and about 98.9% of refers from radiologists to radiographers were for image quality control. The median time interval until refer response was 9min, and this response time did not differ between intra-division and inter-division refers (p = 0.998). Of the refers that got responses, 74.3% (257/346) were sent among radiologists before official reports were made, and the median time until refer response was 9-10min. The proportion of refers answered by radiographers was 85.7% (233/272). The median time interval until refer response by radiographers was 87min for all refers, and 63% were made within 6h. Therefore, the PACS-integrated refer function can facilitate communication between radiologists for image interpretation and quality control.

PHY: A performance-driven hybrid communication compression method for distributed training

Distributed training is needed to shorten the training time of deep neural networks. However, the communication overhead often hurts performance efficiency, especially in a distributed computing environment with limited network bandwidth. Hence, gradient compression techniques have been proposed to reduce communication time. But, compression also has the risk of causing lower model accuracy and longer training time due to compression loss and compression time. As a result, compression may not consistently achieve desired results, and there are limited discussions on when and which compression should be used. To address this problem, we propose a performance-driven hybrid compression solution. We make three main contributions. (1) We describe a hybrid compression strategy that chooses the compression method for individual model gradients. (2) We build an offline performance estimator and an online loss monitor to ensure the compression decision can minimize training time without sacrificing mode accuracy. (3) Our implementation can be imported to existing deep learning frameworks and applicable to a wide range of compression methods. Up to 3.6x training performance speedup was observed compared to other state-of-the-art methods.

Attack‐resilient event‐triggered distributed control strategy of microgrid based on multi‐objective optimization

AbstractAs an important form of renewable energy utilization, microgrid (MG) is considered to be the main bearing form of distributed generation in the future. One of the most concerning issues in the operation of MG is how to realize its economic dispatch (ED). Nowadays, distributed algorithms have been increasingly used to solve the economic dispatch problem of MG. However, the MG based on the distributed optimization architecture must bear higher cyber‐attack risks. To address this issue, this paper investigates the distributed robust ED problem of MG. Firstly, a multi‐objective dispatch model of MG using a linear weighted sum (LWS) algorithm is developed, which considers the environmental and economic costs. On this basis, an event‐triggered fully distributed algorithm is proposed, which can effectively reduce communication times. Furthermore, an attack resilient strategy against false data injection (FDI) attacks is implemented in the proposed fully distributed algorithm, which has strong robustness against various colluding attacks and non‐colluding attacks, and can eliminate incorrect measurement of incremental cost and power generation data. Finally, the effectiveness of the proposed distributed control strategy is demonstrated through case studies in this paper.

Power Limiting for Battery Manage System based on Event-triggered mechanism with stationary leaders

In this paper, the power constraints of battery energy storage systems(BESS) are considered by designing a cooperative consensus algorithm utilizing Soc containment control. For batteries of different capacities, power constraints and Soc balancing can be achieved under undirected topology and containment control. The distributed event-triggered mechanism is adopted to reduce communication time and consumption of communication resources. In the last section, the effectiveness and stability of the designed control strategy are verified by MATLAB simulation.

DeepGOMIMO: Deep Learning-Aided Generalized Optical MIMO with CSI-Free Detection

Generalized optical multiple-input multiple-output (GOMIMO) techniques have been recently shown to be promising for high-speed optical wireless communication (OWC) systems. In this paper, we propose a novel deep learning-aided GOMIMO (DeepGOMIMO) framework for GOMIMO systems, wherein channel state information (CSI)-free detection can be enabled by employing a specially designed deep neural network (DNN)-based MIMO detector. The CSI-free DNN detector mainly consists of two modules: one is the preprocessing module, which is designed to address both the path loss and channel crosstalk issues caused by MIMO transmission, and the other is the feedforward DNN module, which is used for joint detection of spatial and constellation information by learning the statistics of both the input signal and the additive noise. Our simulation results clearly verify that, in a typical indoor 4 × 4 MIMO-OWC system using both generalized optical spatial modulation (GOSM) and generalized optical spatial multiplexing (GOSMP) with unipolar nonzero 4-level pulse-amplitude modulation (4-PAM) modulation, the proposed CSI-free DNN detector achieves near the same bit error rate (BER) performance as the optimal joint maximum-likelihood (ML) detector, but with much-reduced computational complexity. Moreover, because the CSI-free DNN detector does not require instantaneous channel estimation to obtain accurate CSI, it enjoys the unique advantages of improved achievable data rate and reduced communication time delay in comparison to the CSI-based zero-forcing DNN (ZF-DNN) detector.

A pressure pulse recognition method based on flow-adaptive double threshold for pressure pulse telemetry

Pressure pulse telemetry as a novel and effective method of communication has received more and more attention in the field of separate layer water injection for the intelligent oilfield, and identification of the effective pulses and their positions is one of the key technologies. For the logging while drilling process, the fixed threshold method is widely used to identify the pressure pulse. However, in the process of water injection, the amplitude of pressure pulse changes with the change of flowrate controlled by process, which will lead to low recognition rate of traditional fixed threshold pressure pulse. Therefore, a pressure pulse recognition method based on flow-adaptive double threshold with fixed time window is proposed. The packet length of instructions and responses is designed to be short and fixed in order to reduce communication time and failure cost, and the fixed time window method is adopted to improve the decoding efficiency of for a frame of data. The pressure pulse recognition method based on flow-adaptive double threshold is used to identify the effective pulses and their positions to adapt to the change of pressure pulse amplitude induced by different flowrates. The first threshold which is set to fixed and lower than the pulse peak with minimum possible flowrate, is used to filter the most of noise and catch the effective pulses with little noise pulses. The second threshold which is adaptive to changes of the flowrates, is determined by average of three maximum peaks of caught pulses from the first threshold, and is used to identify the effective pulses and their positions. In the experimental well with 2000 m deep, the test results show that the error rate of pulse recognition reduce to 0.003% and communication success rate significantly increase from 50% to more than 96.5%. It means that the proposed double threshold method can be adaptive to identify effective pressure pulse under different flowrates, and then can significantly reduce the error rate of pulse recognition and improve two-way communication performance between the wellhead controller and downhole distributors.

Scaled consensus of second-order multi-agent systems based on edge-event driven control

In this paper, the scaled consensus of resource-limited multi-agent systems with second-order integrator dynamics and undirected topologies is investigated. In order to reduce bandwidth and computation requirements, a scaled consensus protocol based on periodic edge-event driven control is proposed. It is proven that all the agents could converge to a scaled consensus state while the interaction topology is connected. Moreover, a self-triggered scheme is proposed so as to further reduce communication times between agents. Notably, the event-detecting period is introduced so that Zeno behavior could be excluded in our model. Finally, simulations are given to demonstrate the effectiveness of our theoretical results.

Row Allocation Negotiation for a Fleet of Strawberry Harvesting Robots

Abstract Cooperative small harvesting robots, mimicking a group of human pickers, have the potential to significantly reduce labor dependence in strawberry production. A tri-layered algorithm is investigated to effectively assign rows to robots with each robot incentivized to maximize its total number of picked strawberries within the fleet’s harvesting time. The proposed algorithm consists of a decentralized local auction and negotiation strategy as the primary phase with a centralized fallback algorithm that guarantees an assignment. The salient features of the algorithm are reduced communication time, scalability, constant time complexity in the decentralized phase, and ease of implementation. The proposed algorithm is evaluated in a Monte Carlo simulation and the superior performance (e.g., significantly reduced computational time) is observed when compared with a centralized approach. It is expected that this row negotiation algorithm can address an important gap in strawberry harvesting via cooperative, small harvesting robots.

Resource-aware synchronized path following of multiple unmanned surface vehicles with experiments: A cooperative vector field approach

This paper presents a cooperative vector field approach for synchronized path following of multiple autonomous surface vehicles (USVs) with limited communication resources. First, a cooperative vector field is proposed by introducing a path coordination design into a three-dimensional vector field. Based on the cooperative vector field, concise guidance laws of surge speed and angular rate are derived for synchronized path following. In order to reduce communication times among USVs, an event-triggered cooperative vector field is next proposed. Specifically, a predictor is devised to predict the current path variables by using the last transmitting ones, and an event-triggered mechanism is developed based on the predicted path variables. Then, a self-triggered cooperative vector field is developed to avoid continuous listening. A self-triggered mechanism is designed such that each USV predicts its next triggering time and broadcasts it to its neighbors at the current triggering time. The stability of the three cooperative vector fields is analyzed, and Zeno behavior is excluded. Finally, both simulation and experiment results are provided to illustrate the efficacy of the proposed continuous, event-triggered, and self-triggered cooperative vector fields for resource-aware synchronized path following.

Sampled-data-based dynamic event-triggered formation control for nonlinear multi-agent systems

This paper studies the distributed formation control of multi-agent systems with nonlinear dynamics. In view of practical digital microprocessor and limited network resources, the sampled-data-based dynamic event-triggered control strategy is developed. First, each agent synchronously samples its states and monitors the event-triggered function periodically. Each agent broadcasts its states to neighbors only when the function is triggered which can greatly reduce communication times. Meanwhile, the Zeno behavior is excluded due to periodic sampling. Moreover, the dynamic parameter in event-triggered function updates in accordance with a dynamic rule which helps achieve a trade-off between communication frequency and formation performance. The formation problem is transformed into the stability analysis of a time-delay system. Finally, numerical simulations are shown to illustrate the effectiveness of the proposed strategy.

Fuzzy-based escape route fire-vulnerability assessment model for indoor built environment

Fire safety evaluation is rarely involved in the iterative building design process except for legislative approval phases compared to other building objectives. However, regardless of architectural design priorities, all buildings should have adequate fire protection. This research develops a fire vulnerability assessment model based on the impact of architects on fire protection and effects of building design characteristics on fire safety. Inherent to the uncertain nature of fire safety evaluation practice, this study proposes a fuzzy vulnerability decision-making methodology to detect and visualize escape route vulnerabilities, which have the highest impact on the interoperability of fire safety and architectural design practices. The model was validated in an opera house building since the assembly occupancies have specific importance due to the significant number of fire causalities. The escape routes of the case study building were evaluated for materials’ fire reaction, route flow, route equipment, means of egress, dimensions and layout input variables. The output vulnerability levels were discussed to enhance the understanding of critical building design factors that contribute to fire vulnerability. The results confirm that the model is fairly effective in the detection and visualization of vulnerability sources by reducing communication time delays and preventing human-induced mistakes.

CrowdManager: An Ontology-Based Interaction and Management Middleware for Heterogeneous Mobile Crowd Sensing

With the enrichment of types and numbers of sensing terminals, more and more devices such as mobile phones, smart wearables, mobile robots, drones have appeared in our life, enabling the development of Mobile Crowd Sensing (MCS) technology. MCS systems has gradually changed from isomorphic sensing to heterogeneous collaborative sensing, and finally evolved into a heterogeneous multi-source sensing mode of the fusion of humans, machines and objects (things). However, state-of-the-art systems/frameworks do not well support efficient interactions and Heterogeneous Crowd Agents (HCA) management in Heterogeneous MCS (H-MCS) systems. With this in mind, this article aims at two major gaps: Efficient interaction and collaboration of HCA, automated modeling and flexible management of HCA. To deal with the challenges, we design an ontology-based interaction and management middleware (CrowdManager). Three core modules that constitute the middleware: HCA Information Extraction and Representation (HER), Ontology-based HCA Construction & Management (OCM), and Communication and Interaction Module (CIM) are well demonstrated. Extensive comparative evalution suggests that our approach not only brings rich and efficient HCA management and interactive functions to H-MCS systems, but also reduces communication time and various resource occupancy rates by more than 50%.

Preserving Data Confidentiality in Association Rule Mining Using Data Share Allocator Algorithm

These days, investigations of information are becoming essential for various associations all over the globe. By and large, different associations need to perform information examinations on their joined data sets. Privacy and security have become a relentless concern wherein business experts do not desire to contribute their classified transaction data. Therefore, there is a requirement to build a proficient methodology that can process the broad mixture of data and convert those data into meaningful knowledge to the user without forfeiting the security and privacy of individuals’ crude information. We devised two unique protocols for frequent mining itemsets in horizontally partitioned datasets while maintaining privacy. In such a scenario, data possessors outwork mining tasks on their multiparty data by preserving privacy. The proposed framework model encompasses two or more data possessors who encrypt their information and dispense their encrypted data to two or more clouds by a data share allocator algorithm. This methodology protects the data possessor’s raw data from other data possessors and the other clouds. To guarantee data privacy, we plan a proficient enhanced homomorphic encryption conspire. Our approach ensures privacy during communication and accumulation of data and guarantees no information or data adversity and no incidental consequences for data utility. Therefore, the advantages of data mining have remained redesigned. To approve the exhibition of our protocols, we implemented the protocols through broad experiments, where the assessment outcome showed that the mined results obtained by our protocols are reliable to those obtained by a traditional sole-machine approach. Meanwhile, the findings of our performance assessment have shown that our methodology is very efficient, with reasonably reduced communication time and computation costs.

Efficient experimental quantum fingerprinting with channel multiplexing and simultaneous detection

Quantum communication complexity explores the minimum amount of communication required to achieve certain tasks using quantum states. One representative example is quantum fingerprinting, in which the minimum amount of communication could be exponentially smaller than the classical fingerprinting. Here, we propose a quantum fingerprinting protocol where coherent states and channel multiplexing are used, with simultaneous detection of signals carried by multiple channels. Compared with an existing coherent quantum fingerprinting protocol, our protocol could consistently reduce communication time and the amount of communication by orders of magnitude by increasing the number of channels. Our proposed protocol can even beat the classical limit without using superconducting-nanowire single photon detectors. We also report a proof-of-concept experimental demonstration with six wavelength channels to validate the advantage of our protocol in the amount of communication. The experimental results clearly prove that our protocol not only surpasses the best-known classical protocol, but also remarkably outperforms the existing coherent quantum fingerprinting protocol.

A scheduling policy to save 10% of communication time in parallel fast Fourier transform

SummaryThe fast Fourier transform (FFT) has applications in almost every frequency related study, for example, in image and signal processing, and radio astronomy. It is also used as a Poisson operator inversion kernel in partial differential equations in fluid flows, in density functional theory, many‐body theory, and others. The three‐dimensional FFT has large time complexity . Hence, parallelization is used to compute such FFTs. Popular libraries perform slab division or pencil decomposition of data. None of the existing libraries achieve perfect inverse scaling of time with cores because FFT requires all‐to‐all communication and clusters hitherto do not have physical all‐to‐all connections. Dragonfly, one of the popular topologies for the interconnect, supports hierarchical connections among the components. We show that if we align the all‐to‐all communication of FFT with the physical connections of Dragonfly topology we will achieve a better scaling and reduce communication time.

Network-Aware Locality Scheduling for Distributed Data Operators in Data Centers

Large data centers are currently the mainstream infrastructures for big data processing. As one of the most fundamental tasks in these environments, the efficient execution of distributed data operators (e.g., join and aggregation) are still challenging current data systems, and one of the key performance issues is network communication time. State-of-the-art methods trying to improve that problem focus on either application-layer data locality optimization to reduce network traffic or on network-layer data flow optimization to increase bandwidth utilization. However, the techniques in the two layers are totally independent from each other, and performance gains from a joint optimization perspective have not yet been explored. In this article, we propose a novel approach called NEAL (NEtwork-Aware Locality scheduling) to bridge this gap, and consequently to further reduce communication time for distributed big data operators. We present the detailed design and implementation of NEAL, and our experimental results demonstrate that NEAL always performs better than current approaches for different workloads and network bandwidth configurations.

«Цифровое слабоумие» молодежи как проблема современного образования

We consider the issues on the causes of the decline in the intellectual abilities of modern youth. We describe the correlation of this phenomenon with the wide-spreading and starting from an early age use of various kinds of digital devices. We briefly note the origin history of the term “digital dementia” and the author's position regarding this term perception. We describe a number of factors that contribute to the decline in the intellectual development of young people, including: a decrease in the quality of secondary education, deterioration of preschool training of children, simplification of animation and film production, reduction of communication time between parents and children, lack of funds for parents to organize high-quality regular leisure time for children. Negative effects on the intellectual abilities of a young person are described: digital devices can form only a fragmentary idea of reality, without forming a complete idea of a real object; a full-fledged experience of social interaction is not acquired; memory deterioration occurs; there is an appearance of mental dependence on digital devices. Examples from the author's pedagogical practice are given, confirming the manifestation of the effects of “digital dementia” among students: memory impairment in students, undeveloped critical thinking, poorly developed speech (terminologically poor, inconsistent, unsubstantiated), demonstration of misunderstanding of the meaning of the heard or read text.