Smart Edge Research Articles

Developing smart devices with automated Machine learning Approach: A review

In the current era, Artificial Intelligence is playing a major role in smart devices. Apart from incorporating intelligence in the devices, it has extended its range beyond that. Now new AI development tools such as automated machine learning (AutoML) approach are increasing rapidly because of its performance on smart devices. AutoML is used to design a machine learning model with self-adapting capability which can perform the task in less time with advanced skills. This approach does not need any explicit coding and the data science experts but this can build smart sensor codes that will get executed automatically on the microcontrollers embedded within the IoT devices. Due to these capabilities, the developers can build efficient smart devices with real-time responsiveness, security, and data privacy without the help of machine learning experts. In this paper, we will show how AutoML will be used to create smart de-vices. The objective of this paper is to develop the skills that will help to develop smart edge devices which include planning, design-ing, implementation, and the evaluation of the smart IoT applications.

Silicon Photonics in Optical Access Networks for 5G Communications

Only radio access networks can provide connectivity across multiple antenna sites to achieve the great leap forward in capacity targeted by 5G. Optical fronthaul remains a sticking point in that connectivity, and we make the case for analog radio over fiber signals and an optical access network smart edge to achieve the potential of radio access networks. The edge of the network would house the intelligence that coordinates wireless transmissions to minimize interference and maximize throughput. As silicon photonics provides a hardware platform well adapted to support optical fronthaul, it is poised to drive smart edge adoption. We draw out the issues in adopting our solution, propose a strategy for network densification, and cite recent demonstrations to support our approach.

Ameliorate Performance of Memristor-Based ANNs in Edge Computing

Energy efficiency and delay time in the Internet of Things (IoT) system are becoming increasingly significant, especially for the emerging memristor-based crossbar arrays for smart edge computing. This article aims to find a solution for increasing energy efficiency and reducing the delay time, thereby improving the performance of ANNs in edge computing systems. The Number of Pulses Compression (NPC) method is proposed to optimize pulse distribution, energy consumption, and latency by compressing the number of pulses in every weight update step. The NPC method is implemented and verified in a memristor-based hardware simulator based on the MNIST and CIFAR-10 dataset under different circumstances of variations, failure rates, aging effects, architectures, and algorithms. The experimental results show that the NPC method can not only alleviate the uneven distribution of writing pulses but also save the writing energy of the crossbar array by 7.7--26.9 percent and reduce the writing latency by 30.0--50.0 percent. Additionally, the timing regularity of the system is enhanced by the NPC method.

Quantum-Inspired Blockchain-Based Cybersecurity: Securing Smart Edge Utilities in IoT-Based Smart Cities

Blockchain plays a vital task in cybersecurity. With the exerted efforts for realising large-scale quantum computers, most current cryptographic mechanisms may be hacked. Accordingly, we need a quantum tool utilised for designing blockchain frameworks to have the ability to be executed in the level of digital computers and resist the probable attacks from both digital and quantum computers. Quantum walks may be utilised as a quantum-inspired model for designing new cryptographic algorithms. In this paper, we present a new authentication and encryption protocol based on quantum-inspired quantum walks (QIQW). The proposed protocol is utilized to build a blockchain framework for secure data transmission among IoT devices. Instead of using classical cryptographic hash functions, quantum hash functions based on QIQW are employed for linking blocks of the chain. The main advantages of the presented framework are helping IoT nodes to effectively share their data with other nodes and full control of their records. Security analysis demonstrates that our proposed protocol can defend against message attack and impersonation attacks, thus ensuring secure transmission of data among IoT devices.

Heterogeneous Optical Access Networks: Enabling Low-Latency 5G Services With a Silicon Photonic Smart Edge

In the 5G era, optical fronthaul is a major challenge in meeting growing demand. Edge computation and coordinated multipoint for 5G have stringent requirements for high throughput and low latency, either in single-wavelength or wavelength-division-multiplexing fronthaul. We propose a new silicon photonic solution to deliver 5G services on existing optical access networks with colorless optical network units, such as passive optical networks. The newly added 5G services form a heterogeneous optical access network. Using the existing fiber infrastructure, broadband services coexist with new 5G signals that can densify 5G coverage. The proposed scheme is both wavelength-selective (in the distribution network) and colorless (at the end user site). We use silicon microring modulators to create subcarriers slaved from the broadband service distributed carrier; additional microring modulators generate 5G signals exploiting those subcarriers. We experimentally validated the successful coexistence of 5G signals (various formats) with a broadband signal (various formats).

CoLEAP: Cooperative Learning-Based Edge Scheme With Caching and Prefetching for DASH Video Delivery

The outstanding increase in video traffic, puts increasing pressure on network transmission. Since the Dynamic Adaptive Streaming over HTTP (DASH) adjusts the delivery to the dynamic network conditions, it has emerged as a popular approach for video transmissions. However, bitrate switching and video rebuffering may still occur and influence negatively quality of experience (QoE). Additionally the popular videos are transmitted multiple times, which leads to high bandwidth consumption, despite large transmission redundancy. In this context, we propose a Cooperative Learning-based scheme for the smart Edge servers with cAching and Prefetching (CoLEAP) to improve the QoE of adaptive video streaming. CoLEAP employs edge servers which cache the most beneficial contents to reduce redundant video transmissions and prefetches content to decrease network transmission delay. Considering user-related information and the state of network, CoLEAP intelligently makes the most advantageous decisions of caching and prefetching by employing a novel QoE-oriented deep neural network model. To demonstrate the performance of our scheme, we test the proposed solution in comprehensive simulated scenarios and against four alternative solutions. When compared with the existing schemes, CoLEAP increases average bitrate by up to 181.8%, reduces video rebuffering by up to 70.8% as well as decreases response time by up to 28.0%. These values result in minimum improvements of 57.4% and 29.0%, respectively in terms of cache hit rate and QoE.

Secure Data Sharing with Efficient Key Update for Industrial Cloud-based Access Control

Sharing data in industrial cloud requires both secure and flexibility management of data shared among users in the industry. Users should have ability to access authorized portion of data from smart edge devices with integrated secure protocol for interacting with cloud storage. Among cryptographic-based access control solutions, ciphertext-policy attribute-based encryption (CP-ABE) is recognized as a suitable solution for supporting secure and fine-grained data access control for outsourced data. However, the attribute revocation is a major drawback of CP-ABE since it introduces subsequent costs such as ciphertext re-encryption, user key re-generation, and key re-distribution. Existing revocable CP-ABE based access control models have focused on minimizing the cost of communication and computation cost for ciphertext re-encryption. Unfortunately, the key update issue is subject to re-key generation as traditional CP-ABE does. In essence, key update becomes crucial when there are a high number of users accessing shared data in the cloud. In this paper, we propose a lightweight access control model featured with the efficient key updating scheme. Finally, the performance evaluation is conducted to demonstrate the efficiency of our proposed scheme.

EffCNet: An Efficient CondenseNet for Image Classification on NXP BlueBox

Intelligent edge devices with built-in processors vary widely in terms of capability and physical form to perform advanced Computer Vision (CV) tasks such as image classification and object detection, for example. With constant advances in the field of autonomous cars and UAVs, embedded systems and mobile devices, there has been an ever-growing demand for extremely efficient Artificial Neural Networks (ANN) for real-time inference on these smart edge devices with constrained computational resources. With unreliable network connections in remote regions and an added complexity of data transmission, it is of an utmost importance to capture and process data locally instead of sending the data to cloud servers for remote processing. Edge devices on the other hand, offer limited processing power due to their inexpensive hardware, and limited cooling and computational resources. In this paper, we propose a novel deep convolutional neural network architecture called EffCNet which is an improved and an efficient version of CondenseNet Convolutional Neural Network (CNN) for edge devices utilizing self-querying data augmentation and depthwise separable convolutional strategies to improve real-time inference performance as well as reduce the final trained model size, trainable parameters, and Floating-Point Operations (FLOPs) of EffCNet CNN. Furthermore, extensive supervised image classification analyses are conducted on two benchmarking datasets: CIFAR-10 and CIFAR-100, to verify real-time inference performance of our proposed CNN. Finally, we deploy these trained weights on NXP BlueBox which is an intelligent edge development platform designed for self-driving vehicles and UAVs, and conclusions will be extrapolated accordingly.

FerroElectronics for Edge Intelligence

The future data-centric world demands edge intelligence (EI) - the ability to analyze data locally and to decide on a course of action autonomously. Challenges with Moore's Law scaling and limitations of von Neumann computing architectures are limiting the performance and energy efficiency of conventional electronics. Promising new discoveries of advanced CMOS-compatible HfO2-based ferroelectric devices open the door for FerroElectronics; electronics based on ferroelectric building blocks integrated on advanced CMOS technology nodes. It will enable much needed improvement in computing capabilities making EI a reality. In-memory computing in data-flow architectures is at the core of FerroElectronics. This approach will enable building 1000X more compute-energy-efficient small-system AI engines needed for EI. Smart edge intelligent IoT devices enable new applications, for example, micro Drones(uDrones), that demand higher performance to support local embedded intelligence, real-time learning, and autonomy. They will drive the next phase of growth in the semiconductor industry.

Biometric Gloves for Health Monitoring

Global health, defined by the access to healthcare in every region, is symbolized by its demands to be economical and easily accessible. IOT based remote health monitoring system, advancing in the health care sector, is one of the most up-and-coming technological interventions. A serious concern world over, has been perilous work environment that pose health and safety hazards. With the advancements in technology, it is now possible to supervise individual health parameters and provide comprehensive information on health conditions. Users vital information can be monitored from anywhere with access to a specific control centre where the information is stored in real time. In this paper we present a remote health monitoring system that uses an IOT based smart edge technology, where wearable vital sensors transmit data to the cloud using a Wi-Fi module.

Benchmark of the Compute-in-Memory-Based DNN Accelerator With Area Constraint

Compute-in-memory (CIM) is a promising computing paradigm to accelerate the inference of deep neural network (DNN) algorithms due to its high processing parallelism and energy efficiency. Prior CIM-based DNN accelerators mostly consider full custom design, which assumes that all the weights are stored on-chip. For lightweight smart edge devices, this assumption may not hold. In this article, CIM-based DNN accelerators are designed and benchmarked under different chip area constraints. First, a scheduling strategy and dataflow for DNN inference is investigated when only part of the weights can be stored on-chip. Two weight reload schemes are evaluated: 1) reload partial weights and reuse input/output feature maps and 2) load a batch of input and reuse the partial weights on-chip across the batch. Then, system-level performance benchmark is performed for the inference of ResNet-18 on ImageNet data set. The design tradeoffs with different area constraints, dataflow, and device technologies [static random access memory (SRAM) versus ferroelectric field-effect transistor (FeFET)] are discussed.

Smart Edge Caching-Aided Partial Opportunistic Interference Alignment in HetNets

The development of the next-generation wireless networks are regarded as the essentials to embrace of Internet of Things (IoT) and edge computing in heterogeneous networks (HetNets). Due to the the spectrum scarcity problem and the large number of connectivity demand of IoT users, intelligent interference management for IoT is worthy of thorough investigation and should be well discussed with consideration on edge computing in heterogeneous networks (HetNets). Two crucial challenges in the context are: 1) placing edge cache based on dynamic request of IoT users, and 2) cache-enabled interference management with time-varying wireless channels. In this paper, we proposed smart edge caching-aided partial opportunistic interference alignment(POIA) with deep reinforcement learning for IoT downlink system in HetNets. Towards this end, the proposed scheme can update the base station (BS) cache dynamically, and then select the optimal cache-enabled POIA user group considering the time-varying user’s requests and time-varying wireless channels. To solve this problem efficiently, the reinforcement learning is exploited that can take advantage of a deep Q-learing to replace the system action. Extensive evaluations demonstrate that the proposed method is effectiveness according to sum rate and energy efficiency of IoT downlink transmission for HetNets.

The Potential of RES in the Reduction of Air Pollution: The SWOT Analysis of Smart Energy Management Solutions for Krakow Functional Area (KrOF)

Health impacts and a decrease in the quality of life caused by air pollution is a major problem worldwide. Krakow is one of the most affected cities in the EU by air pollution mostly caused by burning solid fuels in households’ furnaces. It is considered that the most effective remedies would be adequate spatial planning solutions and application of low-emission sources including renewable energy sources (RES). This article draws from the analysis of the use of RES as a means for reducing harmful emissions in Krakow Functional Area (KrOF). The inventories of renewable energy sources and systems were compiled by the authors as a part of the EU project “Smart Edge—Sustainable Metropolitan Areas and the Role of The Edge City”. Using the data from the inventories, a SWOT analysis has been carried out to identify factors that determine the smart management of the RES potential, particularly the decisions of households on the transition towards RES. The results of the analysis have shown that many actions have been taken at the national, regional, and local levels but the greatest influence bear the solutions initiated and implemented at the communal level. The conclusion is that legislative regulations should be combined with locally tailor-made instruments. The proposed method of analysis can be applied in other metropolitan areas as a diagnostic procedure supporting action planning to solve air-quality problems caused by distributed emission sources.

Smart Street Litter Detection and Classification Based on Faster R-CNN and Edge Computing

Cleanliness of city streets has an important impact on city environment and public health. Conventional street cleaning methods involve street sweepers going to many spots and manually confirming if the street needs to be clean. However, this method takes a substantial amount of manual operations for detection and assessment of street’s cleanliness which leads to a high cost for cities. Using pervasive mobile devices and AI technology, it is now possible to develop smart edge-based service system for monitoring and detecting the cleanliness of streets at scale. This paper explores an important aspect of cities — how to automatically analyze street imagery to understand the level of street litter. A vehicle (i.e. trash truck) equipped with smart edge station and cameras is used to collect and process street images in real time. A deep learning model is developed to detect, classify and analyze the diverse types of street litters such as tree branches, leaves, bottles and so on. In addition, two case studies are reported to show its strong potential and effectiveness in smart city systems.

Bidirectional Transformation of MES Source Code and Ontologies

Future production environments must be flexible and reconfigurable. To achieve this, the devices and services to fulfill the different steps of a production order (PO) should not be selected in the manufacturing execution system (MES), but in an edge component close to the shop floor. To enable this, abstract services in the PO and concrete services provided by the field devices on the shop floor need to refer to a production ontology. The creation of this ontology is a challenge of its own. This research proposes a pragmatic automation of an encoding of a primary and light weight production ontology based on the source code of MES. The transformation procedure of source code to resource, product and generic concepts of the manufacturing plant ontology is described. To this end, the knowledge of OPC UA collaborations are also exploited during the creation of resource ontologies. Due to a fundamental difference between source code implementation (imperative paradigm) and ontology representation (declarative paradigm), the problem of information loss is inevitable. This problem is overcome by formulation of production and business rules that encapsulate the logic of the MES. The foundation of ontology is exploited to formulate these rulesets using OWL based constructs and OWL based rule languages such as Semantic Web Rule Language (SWRL), Semantic Query-Enhanced Web Rule Language (SQWRL) and SPARQL Protocol and RDF Query Language (SPARQL) based on feasibility and requirements of specific rules. Further, these rulesets are either run on the automatically generated ontology at design time with an intention to enrich the knowledge base, or production runtime to validate the pre-defined business rules between the production steps. The generated ontology also acts as basis for automatically generating the OWL-S ontologies for the OPC UA application methods for the purpose of dynamic manufacturing service discovery and orchestration. The generated ontology and an abstract PO hooked with formulated rules are cached to the shop-floor network for consequent production control to enable smart edge production. An implementation is conducted on an industrial use-case demonstrator to evaluate the applicability of the proposed approach.

Energy-efficient Task Offloading Using Hybrid Particle Swarm Optimization with Genetic Operations in Smart Edge

Wireless capacities and battery energy of smart mobile devices (SMDs) are constrained, and therefore, only limited tasks of applications can be executed in SMDs. To solve this problem, some computation tasks in SMDs can be partially offloaded to edge servers with larger processing capacities. Nonetheless, communication latency is caused for offloaded tasks because of channel bandwidth limits between SMDs and edge servers due to the task offloading. This work designs an energy-efficient task offloading approach to achieve energy consumption minimization for edge servers and SMDs by comprehensively specifying a task offloading ratio, SMDs’ processing speeds and transmission power, and channel bandwidth allocation. Specifically, a mixed-integer nonlinear programming problem is formulated for a smart edge provider. Then, it is solved by using a hybrid particle swarm optimization algorithm with genetic operations for obtaining a close-to-optimal task offloading strategy for edge servers and SMDs. It is evaluated by adopting real-life data from Google production cluster, and simulation results demonstrate that it achieves less consumption of energy for the smart edge provider in a faster way compared with its two state-of-the-art benchmark algorithms.

Software-Defined Edge Computing (SDEC): Principle, Open IoT System Architecture, Applications, and Challenges

Edge computing is a bridge for realizing the convergence between physical space and cyber space in the Internet of Things (IoT) paradigm. Large numbers of physical objects produce a huge amount of data that needs to be efficiently processed in the edge side. This situation urgently requires novel ideas and framework in the design and management of edge computing to improve and enhance its performance. In this article, we propose an approach and principle of software-defined edge computing (SDEC) from the perspective of cyber-physical mapping, where the ultimate goal is to achieve a highly automatic and intelligent edge computing system. The SDEC can also help realize flexible management and intelligent collaboration among various edge hardware resources and services by way of software. To this end, we design an SDEC-based open IoT system architecture which decouples upper level IoT applications from the underlying physical edge resources and builds dynamically reconfigurable smart edge services. The software-definition mechanism of the SDEC platform is proposed to introduce the detailed processes that the underlying physical devices are defined in the form of software. We also describe an illustrative application case about smart factory to present the practical effectiveness of the proposed scheme. Finally, we outline several challenges which are worthy of in-depth study and research. The SDEC paradigm can share, reuse, recombine, and reconfigure edge resources and services so that the overall service capability of the edge side can be improved.

Smart Edge: The Effects of Shifting the Center of Data Gravity Out of the Cloud

A confluence of macrotrends is pushing data, computing, and analytics into the rapidly expanding world at the edge of the connected ecosystem. What is shifting, what is causing it, and what are the overall effects?

DPAM: A Demand-Based Page-Level Address Mappings Algorithm in Flash Memory for Smart Industrial Edge Devices

Edge computing brings data storage closer to the location where it is needed. Therefore, the edge devices, especially smart industrial edge devices, require higher storage systems. NAND flash memory has the advantages of small size, high speed, and strong shock resistance, which is widely used in various storage systems, providing a good choice for edge devices. NAND flash has unique physical characteristics, such as “out-of-place updates” and “prewrite erasure,” therefore, the traditional address mapping methods require improvement. This article presents a novel demand-based page-level address mapping algorithm called DPAM. The goal of DPAM is to provide efficient address translation by using a smaller address mapping table. Due to the high service cost of block-level address mapping and hybrid address mapping, a page-level address mapping scheme is proposed. The algorithm is implemented and tested on the flash simulation platform FlashSim. The results indicate that our algorithm provides improvements of 7.11% for the hit ratio and 7% for the number of block erasures compared with other approaches.

Ecosystem of Things: Hardware, Software, and Architecture

Edge computing is a continuum that includes the computing resources from cloud to things. Ecosystem of things (EoT) is a subsystem of the ecosystem of edge computing, which potentially contains trillions of devices of things and directly interacts with the physical world. This paper surveys the state of the art of EoT by focusing on the computing infrastructure aspect with a forward-looking perspective. We point out a trend of smart edge computing with four types of smartness and intelligence. We address three fundamental questions. 1) What capabilities and how much energy efficiency are the hardware providing? What is the future growth potential? 2) What abstractions are provided by the system software? Are they adequate to support smart edge computing? 3) What ecosystem architectures have been proposed for the coordination of things, the edge, and the cloud? Are they meeting the needs to encourage innovation but avoid unnecessary ecosystem fragmentation? We examine advances from both industry and academia, including research results, visions, and project concepts. We also point out future research directions.