Reduce Communication Latency Research Articles

Optimized Power Control for Over-the-Air Federated Averaging With Data Privacy Guarantee

This paper investigates the privacy-preserving power control problem for <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">over-the-air federated edge learning</i> (Air-FEEL), which features “one-shot” aggregation and significantly reduced communication latency. In Air-FEEL, the inevitable random perturbation encountered in the aggregation process due to the corruption by channel fading and noise poses a fundamental trade-off between the privacy and accuracy, as larger perturbation may harm the accuracy but benefit the privacy, and vice versa. Therefore, power control in Air-FEEL, as the main approach to regulate the said random perturbation, needs to be judiciously designed for balancing the accuracy-privacy trade-off. This, however, is a largely uncharted area. To bridge the research gap, we first analyze the convergence behavior (in terms of the optimality gap) and the <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">differential privacy</i> (DP) performance of Air-FEEL with respect to the power control policy at different iterations. Then, to achieve the maximized training accuracy under given DP guarantee requirement, we minimize the optimality gap by jointly optimizing the power control at edge devices and the denoising factors at the edge server, subject to a set of power constraints at individual edge devices and the DP constraint. Experimental results validate that the proposed power control policy for Air-FEEL achieves a better privacy-accuracy trade-off compared with the benchmarks.

Development of machine tool communication method and its edge middleware for cyber-physical manufacturing systems

ABSTRACT By enabling the fusion of cyber and physical world, Cyber-physical systems (CPSs) have emerged as a key technology of the fourth industrial revolution – Industry 4.0. One of the major challenges for developing manufacturing CPSs is establishing communication between physical resources and Internet/cloud-hosted applications. Recently, MTConnect has become a widely popular communication method for monitoring machine tools in cyber manufacturing. However, it does not allow remote operations of machine tools over the Internet, which is essential in many manufacturing CPSs. This paper presents an Internet-scale service-oriented communication method named Machine Tool Communication (MTComm), which provides capabilities of both monitoring and operating heterogeneous machine tools over the Internet. MTComm allows machine tools to exchange manufacturing services and associated information in XML format through RESTful web services. This paper describes design and development of MTComm components, ontology, and methodology of data acquisition and direct operation through five types of services. To reduce communication latency and improve performance, three optimization strategies were developed and used in an edge-based middleware. Experiments were conducted in a testbed with five machines at different stages of MTComm development and results demonstrated excellent feasibility for efficiently monitoring and operating heterogeneous machine tools in a scalable cyber-physical manufacturing environment.

Machine Learning for Service Migration: A Survey

Future communication networks are envisioned to satisfy increasingly granular and dynamic requirements to accommodate the application and user demands. Indeed, novel immersive and mission-critical services necessitate increased computing and network resources, reduced communication latency, and guaranteed reliability. Thus, efficient and adaptive resource management schemes are required to provide and maintain sufficient levels of Quality of Experience (QoE) during the service life-cycle. Service migration is considered a key enabler of dynamic service orchestration. Indeed, moving services on demand is an efficient mechanism for user mobility support, load balancing in case of fluctuations in service demands, and hardware failure mitigation. However, service migration requires planning, as multiple parameters must be optimized to reduce service disruption to a minimum. Recent breakthroughs in computational capabilities allowed the emergence of Machine Learning as a tool for decision making that is expected to enable seamless automation of network resource management by predicting events and learning optimal decision policies. This paper surveys contributions applying Machine Learning (ML) methods to optimize service migration, providing a detailed literature review on recent advances in the field and establishing a classification of current research efforts with an analysis of their strengths and limitations. Finally, the paper provides insights on the main directions for future research.

LFPR: A Lazy Fast Predictive Repair Strategy for Mobile Distributed Erasure Coded Cluster

Mobile distributed erasure coded Internet of Things (IoT) clusters store popular data, reducing communication latency, and ensuring data reliability while requiring low storage overhead. However, it suffers a high repair overhead to ensure data reliability and availability due to mobile device failures or leaving the cluster. Predictive repair is an effective strategy for reducing repair overhead that has gained attention with the development in accurate failure and mobile node movement trajectory prediction technologies in recent years. We propose LFPR, a hybrid lazy fast predictive repair strategy that combines two baseline predictive repair approaches (reconstruction and migration), including LFPRH and LFPRC for a hot and cold data distributed cluster, respectively. LFPRC and LFPRH adopt different mechanisms to determine whether a block should perform predictive repair immediately. The predictive repair mechanisms of LFPR couples migration and reconstruction in parallel to reduce average repair time per block. LFPR significantly reduces average repair time per block via large-scale simulation and local cluster experiments, compared with existing predictive repair solutions, such as FastPR and the two baselines.

Context-Aware Routing in Fog Computing Systems

Fog computing enables the execution of IoT applications on compute nodes which reside both in the cloud and at the edge of the network. To achieve this, most fog computing systems route the IoT data on a path which starts at the data source, and goes through various edge and cloud nodes. Each node on this path may accept the data if there are available resources to process this data locally. Otherwise, the data is forwarded to the next node on path. Notably, when the data is forwarded (rather than accepted), the communication latency increases by the delay to reach the next node. To avoid this, we propose a routing mechanism which maintains a history of all nodes that have accepted data of each context in the past. By processing this history, our mechanism sends the data directly to the closest node that tends to accept data of the same context. This lowers the forwarding by nodes on path, and can reduce the communication latency. We evaluate this approach using both prototype- and simulation-based experiments which show reduced communication latency (by up to 23 percent) and lower number of hops traveled (by up to 73 percent), compared to a state-of-the-art method.

Maritime terminals’ cargo handling equipment cooperation leveraging IoT and edge computing: The ASSIST-IoT approach

Port container terminals with their logistic infrastructure are essential nodes for a functional economic activity in the developed world. Hence, innovation in efficiency of the port-container industry is a fundamental issue. This paper presents a Next Generation IoT architecture inspired by the merging of commercial terminal-oriented ICT solutions with the results of H2020 research. The architecture is designed to overcome the constraints and meet requirements of a real scenario in maritime ports where different Container Handling Equipment must share heterogeneous data in a decentralized way, leveraging edge computing. This allows for their easier cooperation and avoiding unproductive movements and central communication, while reducing communication latency related to terminal operations. The paper describes a usage example, accompanied with evidence of the first stages of deployment.

D2D-Assisted Multi-User Cooperative Partial Offloading in MEC Based on Deep Reinforcement Learning.

Mobile edge computing (MEC) and device-to-device (D2D) communication can alleviate the resource constraints of mobile devices and reduce communication latency. In this paper, we construct a D2D-MEC framework and study the multi-user cooperative partial offloading and computing resource allocation. We maximize the number of devices under the maximum delay constraints of the application and the limited computing resources. In the considered system, each user can offload its tasks to an edge server and a nearby D2D device. We first formulate the optimization problem as an NP-hard problem and then decouple it into two subproblems. The convex optimization method is used to solve the first subproblem, and the second subproblem is defined as a Markov decision process (MDP). A deep reinforcement learning algorithm based on a deep Q network (DQN) is developed to maximize the amount of tasks that the system can compute. Extensive simulation results demonstrate the effectiveness and superiority of the proposed scheme.

Trust management for service migration in Multi-access Edge Computing environments

Multi-access Edge Computing relies on the concept of moving part of the cloud resources closer to the users to address limitations of the traditional cloud in order to reduce communication latency and increase security. This makes Multi-access Edge Computing (MEC) suitable for time-critical applications such as autonomous vehicles where they can support connectivity with a safe and more efficient experience. Nevertheless, MECs are generally deployed on constrained devices with limited resources, which may affect the infrastructure reliability and thus its trustworthiness.In this paper, we focus on a trust mechanism based on the interactions between MECs to increase reliability in the context of a service migration scenario, where MEC nodes can decide based on a trust score to which node to migrate their services and user information. Our proposed algorithm leverages ideas of the EigenTrust and RLIoT reputation system and combined with a novel correlation concept and a dynamic distance algorithm. From our simulations, our model shows better results in different scenarios and communication ranges. This work provides the core component of a complete trust management system for MECs and could be applied in various use cases.

Enabling Edge Computing in 5G for Mobile Augmented Reality

Mobile Augmented Reality is scoring more reputation due to the progression in the clever smartphones that include characteristics and hardware needs for integrating AR. MAR offers an exclusive knowledge in which the physical world is enhanced by virtual remarks. It also incorporates computationally intensive techniques that might be divested to edge servers on 5G systems, improving MAR by reducing communication latency and ensuring more robust network links, resulting in unified mobile augmented reality user experiences. Edge computing is one of the probable 5G technologies that could deliver content and processing resources familiar to the users, lowering latency and backhaul load. We presented a MAR structure situated on a 5G edge computing and latency results. Offer higher bandwidth while edge servers completely scale down the network restrictions to reduce the bandwidth severity of the network by operating closer to the user. This paper aims to address the 5G edge computing arrangement to aid consistent MAR and finding protocol for enhanced quality of experience (QoE) for MAR applications.

IoT-Enabled Energy-efficient Multipath Power Control for Underwater Sensor Networks

Aims &amp; Background: Energy saving or accurate information transmission within resource limits were major challenges for IoT Underwater Sensing Networks (IoT-UWSNs) on the Internet. Conventional transfer methods increase the cost of communications, leading to bottlenecks or compromising the reliability of information supply. Several routing techniques were suggested using UWSN to ensure uniform transmission of information or reduce communication latency while preserving a data battery (to avoid an empty hole in the network). Objectives &amp; Methodology: In this article, adaptable power networking methods based on the Fastest Route Fist (FRF) method and a smaller amount of the business unit method are presented to solve the problems mentioned above. Both Back Laminated Inter Energy Management One (FLMPC-One) networking method, that employs 2-hop neighborhood knowledge, with the Laminated Inter Energy Management Two (FLMPC-Two) networking procedure, which employs 3-hop neighborhood data, were combined to create such innovative technologies (to shortest path selection). Variable Session Portion (SP) and Information Speed (IS) were also considered to ensure that the suggested method is flexible. Results &amp; Conclusions: These findings show that the suggested methods, Shortest Path First without 3-hop Relatives Data (SPF-Three) or Broadness Initial Searching for Shortest Route. Breadth First Search to 3-hop Relatives Data (BFS-Three) was successfully developed (BFS-SPF-Three). These suggested methods are successful in respect of minimal Electric Cost (EC) and Reduced Transmission Drop Rates (RTDR) given a small number of operational sites at a reasonable latency, according to the simulated findings.

Drive-by-Wi-Fi: Model-Based Control Over Wireless at 1 kHz

In this work, we address the problem of remotely controlling possibly unstable systems using wireless channels with communication and control rates on the order of 1 kHz. While remote control over wireless has been used for applications in the kHz control rate range via low-bitrate communication protocols such as IEEE 802.15.4 (Zigbee) and IEEE 802.15.1 (Bluetooth), this represents the first successful demonstration of remotely controlling an unstable system, namely a balancing robot, using the high-bitrate IEEE 802.11 standard (Wi-Fi) with a control/communication rate in the kHz range, thus being suitable for time-critical industrial applications. This is achieved through a suitable design of both the communication and the control layer; the Wi-Fi protocol parameters are modified to reduce communication latency while still abiding to the standard and a model-based controller is implemented using a time-varying buffered Kalman filter to compensate for the sensor-to-controller lossy communication and using a packetized predictive controller to compensate for the controller-to-actuator lossy communication. The validity of this architecture is experimentally tested on a balancing robot, which adopts off-the-shelf embedded hardware (Rasperry Pi) and software (Arduino) within an industrial-like environment where substantial channel interference is present due to coexisting Wi-Fi networks, showing major improvements compared to more traditional emulation-based architectures.

Truncated Channel Inversion Power Control to Enable One-Way URLLC With Imperfect Channel Reciprocity

We propose to use channel inversion power control (CIPC) to achieve one-way ultra-reliable and lowlatency communications (URLLC), where only the transmission in one direction requires ultra reliability and low latency. Based on channel reciprocity, our proposed CIPC schemes guarantee the power of received signal that is used to decode the information to be a constant value <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$Q$ </tex-math></inline-formula> , by varying the transmit signal and power, which relaxes the assumption of knowing channel state information (CSI) at the user. Thus, the CIPC schemes eliminate the overhead of CSI feedback, reduce communication latency, and explore the benefits of multiple antennas to significantly improve transmission reliability. We derive analytical expressions for the packet loss probability of the proposed CIPC schemes, based on which we determine a closed interval and a convex set for optimizing <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$Q$ </tex-math></inline-formula> in CIPC with imperfect and perfect channel reciprocity, respectively. Our results show that CIPC is an effective means to achieve one-way URLLC. The tradeoff among reliability, latency, and required resources (e.g., transmit antennas) is further revealed, which provides novel principles for designing one-way URLLC systems.

Hardware Limitations to Secure C-ITS: Experimental Evaluation and Solutions

Cooperative Intelligent Transportation Systems (C-ITS) improve driving experience and safety through secure Vehicular Ad-hoc NETworks (VANETs) that satisfy strict security and performance constraints. Relevant standards, such as the IEEE 1609.2, prescribe network-efficient cryptographic protocols to reduce communication latencies through a combination of the Elliptic Curve Qu-Vanstone (ECQV) implicit certificate scheme and the Elliptic Curve Digital Signature Algorithm (ECDSA). However, literature lacks open implementations and performance evaluations for vehicular systems. This paper assesses the applicability of IEEE 1609.2 and of ECQV and ECDSA schemes to C-ITSs. We release an open implementation of the standard ECQV scheme to benchmark its execution time on automotive-grade boards. Moreover, we evaluate its performance in real road and traffic scenarios and show that compliance with strict latency requirements defined for C-ITS requires computational resources that are not met by many automotive-grade embedded hardware platforms. As a final contribution, we propose and evaluate novel heuristics to reduce the number of signatures to be verified in real C-ITS scenarios.

A Low Latency Secure Communication Architecture for Microgrid Control

The availability of secure, efficient, and reliable communication systems is critical for the successful deployment and operations of new power systems such as microgrids. These systems provide a platform for implementing intelligent and autonomous algorithms that improve the power control process. However, building a secure communication system for microgrid purposes that is also efficient and reliable remains a challenge. Conventional security mechanisms introduce extra processing steps that affect performance by increasing the latency of microgrid communication beyond acceptable limits. They also do not scale well and can impact the reliability of power operations as the size of a microgrid grows. This paper proposes a low latency secure communication architecture for control operations in an islanded IoT-based microgrid that solves these problems. The architecture provides a secure platform that optimises the standard CoAP/DTLS implementation to reduce communication latency. It also introduces a traffic scheduler component that uses a fixed priority preemptive algorithm to ensure reliability as the microgrid scales up. The architecture is implemented on a lab-scale IoT-based microgrid prototype to test for performance and security. Results show that the proposed architecture can mitigate the main security threats and provide security services necessary for power control operations with minimal latency performance. Compared to other implementations using existing secure IoT protocols, our secure architecture was the only one to satisfy and maintain the recommended latency requirements for power control operations, i.e., 100 ms under all conditions.

Empirical Performance Evaluation of Communication Libraries for Multi-GPU based Distributed Deep Learning in a Container Environment

Recently, most cloud services use Docker container environment to provide their services. However, there are no researches to evaluate the performance of communication libraries for multi-GPU based distributed deep learning in a Docker container environment. In this paper, we propose an efficient communication architecture for multi-GPU based deep learning in a Docker container environment by evaluating the performances of various communication libraries. We compare the performances of the parameter server architecture and the All-reduce architecture, which are typical distributed deep learning architectures. Further, we analyze the performances of two separate multi-GPU resource allocation policies ― allocating a single GPU to each Docker container and allocating multiple GPUs to each Docker container. We also experiment with the scalability of collective communication by increasing the number of GPUs from one to four. Through experiments, we compare OpenMPI and MPICH, which are representative open source MPI libraries, and NCCL, which is NVIDIA’s collective communication library for the multi-GPU setting. In the parameter server architecture, we show that using CUDA-aware OpenMPI with multi-GPU per Docker container environment reduces communication latency by up to 75%. Also, we show that using NCCL in All-reduce architecture reduces communication latency by up to 93% compared to other libraries.

A Survey and Quantitative Evaluation of Integrated Circuit-Based Antenna Interfaces and Self-Interference Cancellers for Full-Duplex

Full-duplex (FD) wireless - simultaneous transmission and reception on the same frequency - is a promising technique that can significantly improve spectrum efficiency and reduce communication latency in future wireless networks. To enable FD operation, the powerful self-interference (SI) signal leaking from the transmitter (TX) into the receiver (RX) needs to be suppressed and canceled to an extreme degree at the antenna interface as well as in the RF/analog and digital domains. Various approaches to achieve TX-RX isolation at the antenna interface and SI cancellation (SIC) in the RF/analog and digital domains have been demonstrated, with a focus on using bench-top off-the-shelf components. Recent advances in integrated circuit (IC) implementations of various types of shared antenna interfaces and cancellers in the RF and/or analog baseband (BB) domains have further pushed the frontier of realizing FD wireless in small-form-factor/hand-held devices. Moreover, it is still important to fundamentally study the SIC performance that can be achieved by different types of cancellers. In this paper, we present a first-of-its-kind comprehensive overview on the implementation and comparison of various state-of-the-art integrated shared antenna interfaces and SI cancellers in both the RF and analog BB domains. We define two figures of merit (FOM) for the IC-based shared antenna interfaces and SI cancellers, which capture various design considerations and performance tradeoffs including the achievable isolation/SIC, bandwidth, noise figure degradation, TX/SI power handling, and power consumption. In particular, we focus on two types of integrated shared antenna interfaces including electrical-balance duplexers and circulators. We also discuss two types of SI cancellers based on the time-domain and frequency-domain approaches, which respectively employ parallel delay lines and bandpass filters to emulate the SI channel. Based on realistic measurements, we perform extensive numerical evaluations to illustrate and compare the achievable RF SIC performance in different scenarios, as well as discuss various design tradeoffs. Finally, we provide an overview of the IC-based implementations and performance evaluations of both types of cancellers based on our recent work.

FogFL: Fog-Assisted Federated Learning for Resource-Constrained IoT Devices

In this article, we propose a fog-enabled federated learning framework—FogFL—to facilitate distributed learning for delay-sensitive applications in resource-constrained IoT environments. While federated learning (FL) is a popular distributed learning approach, it suffers from communication overheads and high computational requirements. Moreover, global aggregation in FL relies on a centralized server, prone to malicious attacks, resulting in inefficient training models. We address these issues by introducing geospatially placed fog nodes into the FL framework as local aggregators. These fog nodes are responsible for defined demographics, which help share location-based information for applications with similar environments. Furthermore, we formulate a greedy heuristic approach for selecting an optimal fog node for assuming a global aggregator’s role at each round of communication between the edge and cloud, thereby reducing the dependence on the execution at the centralized server. Fog nodes in the FogFL framework reduce communication latency and energy consumption of resource-constrained edge devices without affecting the global model’s convergence rate, thereby increasing the system’s reliability. Extensive deployment and experimental results corroborate that, in addition to a decrease in global aggregation rounds, FogFL reduces energy consumption and communication latency by 92% and 85%, respectively, as compared to state of the art.

A controller deployment scheme in 5G-IoT network based on SDN

Due to the wide spread of 5G networks, network users have higher requirements for communication delays. Software-defined network is a new paradigm of decoupling the control logic from packet forwarding devices. We can reduce communication latency in the network by optimizing the location of the controller and improve the communication performance of the network. In this paper, the controller deployment problem of multi-network area is studied in order to reduce the average communication latency. In order to solve this problem, we proposed an optimized DPC algorithm. Specifically, on the basis of DPC algorithm we quoted the idea of triangular stability from the BeeDPC algorithm, and introduced SC measurement indicators, the degree of separation between sub-network regions and the degree of aggregation within the sub-network region, so as to divide the network region more reasonably. At the same time, we have break out of the constraints in the DPC algorithm and introduced closeness centrality to get a more reasonable placement scheme and reduced the limitations of the DPC algorithm. Simulation results have shown that the optimized DPC algorithm can effectively reduce the average delay between the control layer and the data layer, improve the network performance, and enhance network stability and reliability.

Designed Features for Improving Openness, Scalability and Programmability in the Fog Computing-Based IoT Systems

Global-scale Internet of Things (IoT) applications commonly process a huge amount of data and quickly provide useful information (to users) and instructions to monitor or control the physical world; most of these services are required to be of low latency while satisfying resource constraints (computing, storage, networking). In order to reduce communication latency, network load, energy consumption and operational cost which are essential for real-world IoT applications, fog computing is a promising technology as it allows utilizing virtualized resources already available on the network edges which are close to IoT data generators (IoT devices) and consumers (users or actuators). However, there are still many challenges in realizing this computing paradigm as it is still in the infancy stage. This paper investigates and proposes designed features for the “openness,” “scalability” and “programmability” in fog computing to which IoT services/applications can be flexibly implemented in the fog landscape. We apply the proposed mechanism to a real-world application, namely the traffic light optimization problem, to validate its effectiveness and feasibility. The evaluation results reveal the effectiveness of the proposed approach.

IF-RANs: Intelligent Traffic Prediction and Cognitive Caching toward Fog-Computing-Based Radio Access Networks

The rapid development of wireless communication technology and smart devices has made the traditional cloud-based Internet of Things architecture unable to meet the stringent requirements of 5G mobile communication networks. In order to realize high-reliability and low-latency communication for 5G, F-RANs have become a potential evolution path. However, F-RANs still face several challenges regrading infrastructure, network traffic, and caching mechanism. From the point of view of integrating AI, F-RANs, and cloud, this article proposes intelligent traffic prediction and cognitive caching toward IF-RANs. First, a traffic flow prediction algorithm is developed that is based on LSTM with an attention mechanism. The algorithm can effectively predict real-time traffic of different data types. Then, for caching policy, cognitive caching based on LSTM and collaborative filtering is proposed to reduce the total communication delay. The experimental results demonstrate the effectiveness of the proposed IF-RANs in improving the accuracy of real-time prediction and reducing communication latency.