AU2021101131A4 - Qos‑aware optical fog‑assisted cyber‑physical system in 5g ready heterogeneous network and method for the same - Google Patents

Abstract

The present disclosure relates to a QOS-aware optical fog-assisted cyber-physical system in 5G ready heterogeneous network and method for the same. NFV, ONV, and SDN is used to create a computational OpticalFog node in the middle- ware of cloud and edge that provides an ecosystem to facilitate an ultra-fast infrastructure in the 5G network. The system exploit the computational potential of optical resources to create an OpticalFog node to be integrated into the 5G network infrastructure for CPS-based applications, presents an technique for optimum placement that provide service assurance to various CPS-based applications, and enables CPS-based applications, which are able to run on top of the OpticalFog and 5G infrastructure. iFogSim toolkit is used to implement the proposed OpticalFog-based deployment for the surveillance cameras. The results show the significant advantages of the proposed system for CPS-based applications in the 5G network. 28 o 001 0 Q) 0. U-U LU ~0 I4- CCL 0

Description

o 001

Q) 0.

U-U LU

~0

I4-

CCL

QOS-AWARE OPTICAL FOG-ASSISTED CYBER-PHYSICAL SYSTEM IN 5G READY HETEROGENEOUS NETWORK AND METHOD FOR THE SAME

FIELD OF THE INVENTION

The present disclosure relates to a QOS-aware optical fog-assisted cyber-physical system in 5G ready heterogeneous network and method for the same.

BACKGROUND OF THE INVENTION

Over the past two decades, mobile and wireless networks have enabled smart mobile devices that have facilitated the implementation of the Internet of Things (IoT) in 5G cellular networks with good-quality higher data speeds, lower cost, lower end-to-end delay. International Data Corporation (IDC) estimates that there will be 41.6 billion IoT devices to be deployed by 2025 generating 79.4 zettabytes (ZB) of data. Cyber-physical system (CPS) is the next digital communication technology that is capable to host smart real-time applications on a wide range of devices in various IoT sectors like smart healthcare, smart agricultural, smart education, smart surveillance, autonomous driving, augmented reality, disaster management, smart industry 4.0, etc. Eventually, CPS-based systems depend on the functionality that 5G can provide to connect with the physical environment as well as process and exchange information among cloud computing, fog computing, or virtual network technologies.

5G technology improves networking, spectral, and networking performance to allow ubiquitous computing for CPS-based applications. It provides a centralized and cost effective networking infrastructure for various IoT sectors at the edge of the emerging 5G network. Hence, the IoT technology is emerging for the next technological revolution called G ready CPS. According to the NIST, "Cyber-physical systems are highly interconnected systems that will provide new functionalities to improve quality of life and enable technological advances in critical areas. CPS contains co-engineered, interacting networks of physical and computational components from many different technology domains". The term cyber-physical suggests that it has components from the cyber world and the physical world. The cyber component represents the simulation model of a computer pro- gram describing physical objects. Data can be obtained from multiple sources using web services technology, or a combination of physical and virtual sensors.

Presently, loT-enabled CPS solutions move data from IoT devices to the cloud for processing long-term decision-making tasks. Although, the majority of existing data analytics methods are designed to manage massive data volumes on the cloud data center only. Besides cloud computing, the growing growth of the IoT leads to the introduction of fog computing, utilizing more edge resources closer to IoT devices/end-users. The dynamic complexity of IoT environments and their associated requirements of real-time data processing, dispatching, and growing edge computing capability has contributed to the fog computing paradigm evolution. Cisco pioneered the idea of fog computing that extends computing at the network's edge with essential cloud computing characteristics such as unified management, ubiquity, and collaboration. Essentially, fog computing pro- vides computations over the fog nodes at the network edge. Here, fog devices utilize their computing/storage capability to support cloud infrastructure by reducing its storage and processing burden. Furthermore, placing services closer to end-users can reduce delay effectively. Fog nodes are the self-managed "micro clouds" that provide the computing capabilities near the edge device rather than the

cloud.

The present computing scenario uses virtualization technology that has introduced a novel paradigm in telecommunications namely, Network Functions Virtualization (NFV), Optical Network Virtualization (ONV) and Software-Defined Networking (SDN). In the NFV, VMs can be dynamically scaled, delivered, and migrated to clouds efficiently. Telecommunications companies build their cloud data centers within their NFV networks and place their data centers to control their computing and networking resources elastically. It is a network architecture framework for 5G technology that uses virtualization technologies to virtualize entire classes of the network node. It offloads network functions into industry standard hardware-software which can be managed from anywhere. Moreover, Optical network resources consists of Passive Optical Network (PON), Optical Line Terminals (OLTs), optical splitters, and Optical Network Units (ONUs). Thus, CPS-based applications can also be distributed across the 5G-enabled optical network and embedded in an environment of multiple connected optical resources of heterogeneous capabilities. Therefore, it is necessary to exploit the computational and storage capabilities of optical resources. ONV enables dedicated networks to be dynamically supplied across the same network infrastructures to be converted into virtual resources, followed by abstraction, partition, and aggregation. SDN has introduced a revolutionary paradigm change in computer networks with the use of an open protocol programmable software controller. Hence, NFV, ONV and SDN are the best options for future-ready CPS-based systems in 5G ready networks using cloud/fog computing.

Effective optical fog-assisted system design is one of the key challenges of using in network services of the 5G era. Thus, the system should be designed to run CPS-based applications on a novel fog layer so that it can exploit edge/optical devices's capability to provide real-time response and use their immense resource availability simultaneously.

Already, several authors have contributed to incorporate CPS in the cloud/fog environment. Furthermore, our research concentrate on deploying CPS-based applications in the 5G network through optical resources as a novel fog layer. We believe that this the first time that optical resources have been used in successfully incorporating CPS in the 5G network. However, no attention has been paid to implement CPS-based applications over the optical resources in the cloud/fog environment. Hence, different aspects have been explored in this computing context. Also, a more detailed analysis of CPS-based applications by using fog computing in the 5G era is being conducted.

Table 1: List of acronyms IoT Internet of Things CPS Cyber-physical system SDN Software-defined networking ONV Optical network virtualization NFV Network function virtualization PON Passive optical network OLT Optical line terminal ONU Optical network unit DMIPS Dhyrstone million instructions per second C-RAN Cloud radio access network F-RAN Fog radio access network QoS Quality of service

Hung et al. presented a comprehensive 5G Cloud Radio Access Network (CRAN) assessment and examined the harmonization desired to incorporate 5G. Ultimately, a fog network was introduced to promote innovation to meet newly emerging traffic demands. Peng et al. proposed an innovative communications network based on Fog Radio Access Network (F-RAN) technology for providing high spectral and energy efficiency. Moreover, they also discussed open edge caching challenges related to integrating new technologies such as SDN, and NFV. Ku et al. implemented a hybrid cloud/fog-based architecture for implementing low-latency operations in the 5G network. Vilalta et al. introduced a new architecture, called Telco Fog for providing hybrid cloud/fog infrastructure to implement NFV and IoT services. They introduced distributed, programmable fog concept to improve the mobile network's role in the 5G network era. Ruffini introduced a single, cohesive, SDN based network architecture that provided an insight into net- work integration and application-centered architecture structure for the various technologies. Markakis et al. proposed a framework that optimized edge to allow low latency and high QoS ecosystem. Also, they suggested heterogeneous nodes that enhanced computer processing capabilities at the edge of the networks. Yang et al. introduced an SDN-based architecture that promoted cloud-fog interoperation as well as increased the QoS and enhanced network resource utilization.

To investigate optical network functionality in the 5G network, several authors represented their research towards integrating fog/cloud-based applications. Kitayama addressed the unified role of the optical and wireless network to implement real-time applications in the 5G network. Author offered a solution that reduced latency over the high speed optical and radio connections. Lingen et al. implemented a transparent and con- verged architecture to provide unified IoT services management in the 5 G network. They also developed the first YANG models using fog nodes to run IoT services. Giuntini et al. introduced carrier ethernet forwarding scheme to develop IoT applications using optical network resources dedicated to 5G infrastructure.

To address the research challenges to implement CPS in the present computing context, Gu et al. presented a fog-based framework that implements CPS related to the medical sector. They developed their framework's cost-efficiency by researching the distribution of machines, base station association methods, and virtual machine positioning. Moreover, a linear two-phase heuristic technique was proposed that formulated the proposed method. Their findings from comprehensive studies confirmed the technique's high-cost performance. To monitor the activities of CPS, Chen and Rui-Yang suggested a distributive and central traceable stream method for an intelligent device that can determine the most important traceable CPS-based events using the fog/cloud computing. Also, an intelligent predictive technique is implemented to show the efficiency of the proposed method and obtained better experimental results. Through the evolving fog computing model, Donovan et al. introduced a CPS for Industry 4.0. It allowed tightly coupled industrial artifacts and processes to realize self-configuring operations which were tested, simulated in the virtual world, using sophisticated machine learning models. Zeng et al. developed a CPS based on the fog model to achieve energy-efficiency during the load balancing service composition. A heuristic technique is presented to solve this NP-hard energy efficiency problem by using linear mixed-integer programming. Extensive simulation results demonstrated high energy efficiency. Sood and Mahajan developed a novel method for identifying, classifying, and tracking patients with Mosquito-Borne diseases. They proposed a fog/cloud-based CPS that integrates IoT sensors to assimilate, analyze, and exchange medical information among patients and healthcare providers.

In order to overcome the aforementioned drawbacks, there exists a need to develop a QOS-aware optical fog-assisted cyber-physical system in 5G ready heterogeneous network and method for the same.

SUMMARY OF THE INVENTION

The present disclosure seeks to provide a system and method to create a computational OpticalFog node in the middle- ware of cloud and edge that provides an ecosystem to facilitate an ultra-fast infrastructure in the 5G network.

In an embodiment, a QOS-aware optical fog-assisted cyber-physical system in 5G ready heterogeneous network. The system comprises:

a device layer consists of IoT devices such as sensors & actuators to gather and interact with a physical space and act as the source or sink of data; an edge fog layer equipped with a plurality of fog devices such as switches, routers, ONUs to host the application modules, wherein the application module is responsible to process all the data generated and send them to the next layer for performing machine learning tasks; an optical fog layer consists a SDN controller along with infrastructure management (ONV, 5G, virtualization) that is capable to create the OpticalFog node with the computing capability to scale up or down optical resources by dynamically aggregating or de aggregating PON resources, wherein all PON resources are grouped together and synthesized with SDN and ONV techniques for creating OpticalFog node with high configuration and interconnection capabilities as Infrastructure-as-a-Service (IaaS) model; and a cloud layer having various mangers to manage resources, scheduling, making policy in such a way that application level QoS constraints are met and resource optimization is achieved, wherein the cloud layer provides functions to the cloud for the CPS- based modules.

In an embodiment, the fog layer is introduced to exploit the computational capability of the optical resources rather than edge devices for handling real-time and CPS-based applications.

In an embodiment, in Cyber Space, the first layer of fog is the traditional EdgeFog layer (have edge devices eg. routers, switches, hubs, etc.) and the second layer is the proposed OpticalFog layer. Apart from this, this layer also consists of the proposed OpticalFog node (created using optical resources), integrated with the SDN controller, and ONV in the 5G network(small cells and macrocells).

In an embodiment, all PON resources are grouped together and synthesized with SDN and ONV techniques for creating OpticalFog node with high configuration and interconnection capabilities as Infrastructure-as-a-Service (IaaS) model, wherein the OpticalFog node node can easily scale up or down optical resources by dynamically aggregating or de-aggregating PON resources.

In an embodiment, the PON provides residual computing capabilities of optical resources to create an OpticalFog node by using the SDN and ONV techniques. In an embodiment, the PON-OLT network having 16 OLT chassis creates an OpticalFog node with the computing capability of 41,287,680 DMIPS (Dhyrstone million instructions per second) for processing, 4,194,304 MB for storage, and 2.5Gb/s interconnection speed, wherein the OpticalFog node process real-time tasks at the OpticalFog layer rather than the cloud layer. In another embodiment, a QOS-aware optical fog-assisted cyber-physical method in G ready heterogeneous network. The method comprises: gathering signals and interacting with a physical space by a device layer consists of IoT devices such as sensors & actuators which act as the source or sink of data; hosting application modules through an edge fog layer equipped with a plurality of fog devices such as switches, routers, ONUs, wherein the application module is responsible to process all the data generated and send them to the next layer for performing machine learning tasks; creating OpticalFog node with the computing capability consisting a SDN controller along with infrastructure management (ONV, 5G, virtualization) to scale up or down optical resources by dynamically aggregating or de-aggregating PON resources, wherein all PON resources are grouped together and synthesized with SDN and ONV techniques for creating OpticalFog node with high configuration and interconnection capabilities as Infrastructure as-a-Service (IaaS) model; and managing resources, scheduling, making policy in such a way that application level QoS constraints are met and resource optimization is achieved using a cloud layer having various mangers, wherein the cloud layer provides functions to the cloud for the CPS- based modules.

An objective of the present disclosure is to support the CPS-based applications with ultra-low delay, more bandwidth efficiency, and significantly low energy consumption of the edge devices. Another object of the present disclosure is to provide the shortest path through the optical network resources. Another object of the present disclosure is to create a computational OpticalFog node in the middle- ware of cloud and edge that provides an ecosystem to facilitate an ultra-fast infrastructure in the 5G network. Yet another object of the present invention is to deliver an expeditious and cost effective QOS-aware optical fog-assisted cyber-physical method in 5G ready heterogeneous network.

To further clarify advantages and features of the present disclosure, a more particular description of the invention will be rendered by reference to specific embodiments thereof, which is illustrated in the appended drawings. It is appreciated that these drawings depict only typical embodiments of the invention and are therefore not to be considered limiting of its scope. The invention will be described and explained with additional specificity and detail with the accompanying drawings.

BRIEF DESCRIPTION OF FIGURES

These and other features, aspects, and advantages of the present disclosure will become better understood when the following detailed description is read with reference to the accompanying drawings in which like characters represent like parts throughout the drawings, wherein:

Figure 1 illustrates a block diagram of a QOS-aware optical fog-assisted cyber-physical system in 5G ready heterogeneous network in accordance with an embodiment of the present disclosure; Figure 2 illustrate a flow chart of a QOS-aware optical fog-assisted cyber-physical method in 5G ready heterogeneous network in accordance with an embodiment of the present disclosure; Figure 3 illustrates a cyber space and physical space in the proposed CPS in accordance with an embodiment of the present disclosure; Figure 4 illustrates a layered view of the proposed system architecture in accordance with an embodiment of the present disclosure; Figure 5 illustrates a proposed architecture of ONV and SDN based OpticalFog Node in accordance with an embodiment of the present disclosure; Figure 6 illustrates a creating pool of resources for the proposed OpticalFog node in accordance with an embodiment of the present disclosure; Figure 7 illustrates a proposed architecture of ONV and SDN based OpticalFog Node in accordance with an embodiment of the present disclosure; Figure 8 illustrates realizing the proposed system for an academic institute in accordance with an embodiment of the present disclosure; Figure 9 illustrates a comparison graph of delay in the proposed and cloud-based deployment scenario in accordance with an embodiment of the present disclosure;

Figure 10 illustrates a comparison graph of network consumption in the proposed and cloud-based deployment scenario in accordance with an embodiment of the present disclosure; and Figure 11 illustrates comparison graphs of energy consumption in the proposed and cloud-based deployment scenario in accordance with an embodiment of the present disclosure. Further, skilled artisans will appreciate that elements in the drawings are illustrated for simplicity and may not have been necessarily been drawn to scale. For example, the flow charts illustrate the method in terms of the most prominent steps involved to help to improve understanding of aspects of the present disclosure. Furthermore, in terms of the construction of the device, one or more components of the device may have been represented in the drawings by conventional symbols, and the drawings may show only those specific details that are pertinent to understanding the embodiments of the present disclosure so as not to obscure the drawings with details that will be readily apparent to those of ordinary skill in the art having benefit of the description herein.

DETAILED DESCRIPTION

For the purpose of promoting an understanding of the principles of the invention, reference will now be made to the embodiment illustrated in the drawings and specific language will be used to describe the same. It will nevertheless be understood that no limitation of the scope of the invention is thereby intended, such alterations and further modifications in the illustrated system, and such further applications of the principles of the invention as illustrated therein being contemplated as would normally occur to one skilled in the art to which the invention relates.

It will be understood by those skilled in the art that the foregoing general description and the following detailed description are exemplary and explanatory of the invention and are not intended to be restrictive thereof.

Reference throughout this specification to "an aspect", "another aspect" or similar language means that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment of the present disclosure. Thus, appearances of the phrase "in an embodiment", "in another embodiment" and similar language throughout this specification may, but do not necessarily, all refer to the same embodiment.

The terms "comprises", "comprising", or any other variations thereof, are intended to cover a non-exclusive inclusion, such that a process or method that comprises a list of steps does not include only those steps but may include other steps not expressly listed or inherent to such process or method. Similarly, one or more devices or sub-systems or elements or structures or components proceeded by "comprises...a" does not, without more constraints, preclude the existence of other devices or other sub-systems or other elements or other structures or other components or additional devices or additional sub-systems or additional elements or additional structures or additional components.

Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. The system, methods, and examples provided herein are illustrative only and not intended to be limiting.

Embodiments of the present disclosure will be described below in detail with reference to the accompanying drawings.

Referring to Figure 1, a block diagram of a QOS-aware optical fog-assisted cyber-physical system in 5G ready heterogeneous network is illustrated in accordance with an embodiment of the present disclosure. The system 100 includes a device layer 102 consists of IoT devices such as sensors & actuators to gather and interact with a physical space and act as the source or sink of data.

In an embodiment, an edge fog layer 104 is equipped with a plurality of fog devices such as switches, routers, ONUs to host the application module. The application module is responsible to process all the data generated and send them to the next layer for performing machine learning tasks.

In an embodiment, an optical fog layer 106 consists a SDN controller along with infrastructure management (ONV, 5G, virtualization) that is capable to create the OpticalFog node with the computing capability to scale up or down optical resources by dynamically aggregating or de-aggregating PON resources. All PON resources are grouped together and synthesized with SDN and ONV techniques for creating OpticalFog node with high configuration and interconnection capabilities as Infrastructure-as-a-Service (IaaS) model.

In an embodiment, a cloud layer 108 having various mangers to manage resources, scheduling, making policy in such a way that application level QoS constraints are met and resource optimization is achieved. The cloud layer 108 provides functions to the cloud for the CPS- based modules.

In an embodiment, the fog layer is introduced to exploit the computational capability of the optical resources rather than edge devices for handling real-time and CPS-based applications.

In an embodiment, in Cyber Space, the first layer of fog is the traditional EdgeFog layer (have edge devices eg. routers, switches, hubs, etc.) and the second layer is the proposed OpticalFog layer 106. Apart from this, this layer also consists of the proposed OpticalFog node (created using optical resources), integrated with the SDN controller, and ONV in the 5G network(small cells and macrocells).

In an embodiment, all PON resources are grouped together and synthesized with SDN and ONV techniques for creating OpticalFog node with high configuration and interconnection capabilities as Infrastructure-as-a-Service (IaaS) model, wherein the OpticalFog node node can easily scale up or down optical resources by dynamically aggregating or de-aggregating PON resources.

In an embodiment, the PON provides residual computing capabilities of optical resources to create an OpticatFog node by using the SDN and ONV techniques. In an embodiment, the PON-OLT network having 16 OLT chassis creates an OpticalFog node with the computing capability of 41,287,680 DMIPS (Dhyrstone million instructions per second) for processing, 4,194,304 MB for storage, and 2.5Gb/s interconnection speed, wherein the OpticalFog node process real-time tasks at the OpticalFog layer rather than the cloud layer 108.

The emergence of Internet of things and cyber-physical system provide a proactive and efficacious solution to enable remote monitoring, machine learning-based analytics, and quick decision making for real-time applications. Cloud supported systems still face Quality of Service issues for implementing time-sensitive applications. Introducing 5G network services by revolutionizing existing technologies can increase communication efficiency and provide optimum trade-offs for connected IoT devices. The system includes, NFV, ONV, and SDN used to create a computational OpticalFog node in the middle- ware of cloud and edge that provides an ecosystem to facilitate an ultra-fast infrastruc- ture in the 5G network. The main aim behind the proposed system is to (1) exploit the computational potential of optical resources to create an OpticalFog node to be integrated into the 5G network infrastructure for CPS-based applications, (2) present an technique for optimum placement that provide service assurance to various CPS-based applications, and (3) enable CPS-based applications, which are able to run on top of the OpticalFog and 5G infrastructure. Therefore, a case study based on the deployment surveillance system in an academic institute using the proposed system is presented. iFogSim toolkit is used to implement the proposed OpticalFog-based deployment for the surveillance cameras. The results show the significant advantages of the proposed system for CPS-based applications in the 5G network.

The system provides the appropriate Quality-of-Service (QoS) criteria of CPS-based applications. An OpticalFog node is proposed that handle application mod- ule processing in the pool of optical fog resources stretching from the edge to the cloud through the optical resources across the 5G network. A popular 5G technology, called NFV management and orchestration is utilized while designing OpticalFog node along with the SND controller. Moreover, an algorithm is proposed to provide Quality of service (QoS) by reducing delay, energy usage, network congestion. We also present a case study of surveil- lance in the academic institution using the iFogSim toolkit to show the effectiveness of the proposed system.

Figure 2 illustrate a flow chart of a QOS-aware optical fog-assisted cyber-physical method in 5G ready heterogeneous network in accordance with an embodiment of the present disclosure in accordance with an embodiment of the present disclosure. At step 202, the method 200 includes gathering signals and interacting with a physical space by a device layer 102 consists of IoT devices such as sensors & actuators which act as the source or sink of data.

At step 204, the method 200 includes hosting application modules through an edge fog layer 104 equipped with a plurality of fog devices such as switches, routers, ONUs. The application module is responsible to process all the data generated and send them to the next layer for performing machine learning tasks.

At step 206, the method 200 includes creating OpticalFog node with the computing capability consisting a SDN controller along with infrastructure management (ONV, 5G, virtualization) to scale up or down optical resources by dynamically aggregating or de aggregating PON resources. All PON resources are grouped together and synthesized with SDN and ONV techniques for creating OpticalFog node with high configuration and interconnection capabilities as Infrastructure-as-a-Service (IaaS) model.

At step 208, the method 200 includes managing resources, scheduling, making policy in such a way that application level QoS constraints are met and resource optimization is achieved using a cloud layer 108 having various mangers. The cloud layer 108 provides functions to the cloud for the CPS- based modules.

The method facilitates a 5G-ready CPS using optical fog/cloud computing to overcome these challenges. To provide CPS-based services in the 5G network, the processing of real-time applications must be flexible in the 5G system and optimized using advanced network technologies such as NFV,ONV, and SDN. They can offer a novel way to create an OpticalFog node with the abstract view by orchestrating the multi-layer control over the optical resources.

Figure 3 illustrates a cyber space and physical space in the proposed CPS in accordance with an embodiment of the present disclosure. Figure 3 shows the proposed 5G ready CPS that utilizes the optical resources to implement CPS-based applications effectively. It has two spaces namely, Cyber Space and Physical Space. Cyber Space handles the processing of coming tasks and perform both machine learning and data analytics functions. The Cyber Space of the proposed system mainly consists of two layers namely, the OpticalFog layer and the cloud layer 108. whereas, Physical Space consists of physical devices (usually IoT devices) operated by software systems to carry out complex tasks under defined real-time and physical-resource constraints. IoT devices collect data from the Physical Space through sensors and forward them to the Cyber Space for further analysis and processing of the tasks. The OpticalFog layer is an additional proposed layer over the traditional cloud architecture that offers ultra-low delay, optimum network usage, and less energy consumption.

Figure 4 illustrates a layered view of the proposed system architecture in accordance with an embodiment of the present disclosure. The architectural view of the proposed system in the present computing environment is depicted in Figure 4. The architecture has multiple layers, with each layer responsible for doing specific tasks to facilitate operation of the higher layers. The bottom-most layer con- sists of IoT devices(sensors & actuators) that gathers and interacts with the Physical Space and act as the source or sink of data.

The upper layer consist fog devices (switches, routers, ONUs) to host the application modules. This application module is responsible to process all the data generated and send them to the upper layer for doing machine learning tasks. Further, the next upper layer con sists SDN controller along with infrastructure management (ONV, 5G, virtualization) that is capable to create the proposed OpticalFog node with the computing capability. Next, this layer have various mangers to manage resources, scheduling, making policy in such a way that application level QoS constraints are met and resource optimization is achieved. Here, the proposed placement technique play a key role to assign the best suitable task to the pro posed OpticalFog node. Now the upper layer provides functions to the cloud for the CPS based modules. The uppermost layer, has different application modules. This application model helps us to view an application as a directed graph, with vertices representing application modules and directed edges showing data flow between modules.

A novel fog layer is introduced to exploit the computational capability of the optical resources rather than edge devices for handling real-time and CPS-based applications. In Cyber Space, the first layer of fog is the traditional EdgeFog layer (have edge devices eg. routers, switches, hubs, etc.) and the second layer is the proposed OpticalFog layer. Apart from this, this layer also consists of the proposed OpticalFog node (created using optical resources), integrated with the SDN controller, and ONV in the 5G network(small cells and macrocells). All PON resources are grouped together and synthesized with SDN and ONV techniques for creating OpticalFog node with high configuration and interconnection capabilities as Infrastructure-as-a-Service (IaaS) model. The proposed node can easily scale up or down optical resources by dynamically aggregating or de-aggregating PON resources.

Figure 5 illustrates a proposed architecture of ONV and SDN based OpticalFog Node in accordance with an embodiment of the present disclosure. The proposed OpticalFog node enables low-cost and on-demand access to the 5G net- work in the optical network. As shown in Figure 5, SDN, NFV, and ONV offer a novel way to create OpticalFog node with the abstract view by orchestrating the multi-layer con- trol over the optical resources.

Figure 6 illustrates a creating pool of resources for the proposed OpticalFog node in accordance with an embodiment of the present disclosure. As shown in Figure 6, all residual available resources are grouped together to form a pool of resources. Hence, PON provides residual computing capabilities of optical resources to create an OpticalFog node by using the SDN and ONV techniques.

A typical PON-OLT network having 16 OLT chassis can create an OpticalFog node with the computing capability of 41,287,680 DMIPS (Dhyrstone million instructions per second) for processing, 4,194,304 MB for storage, and 2.5Gb/s interconnection speed. Therefore, the proposed OpticalFog node is enough capable to process real-time tasks at the OpticalFog layer rather than the cloud layer 108. Furthermore, to utilize the opti- cal resources effectively, an technique is proposed that assign the CPS-based/non-CPS- based tasks on the proposed OpticalFog node efficiently.

Figure 7 illustrates a proposed architecture of ONV and SDN based OpticalFog Node in accordance with an embodiment of the present disclosure. Figure 7 presents the architectural view of the proposed OpticalFog node. The OpticalFog node is created by utilizing all freely accessible computing resources of optical resources. ONV transforms optical network elements' free physical resources into virtual resources asIaaS.

To improve QoS, the SDN controller optimizes the flow delivery across the various paths and find the shortest path with less delay. The OpticalFog node has a flow table where the route information is matched with the outing information of the received packet(task). If there is no entry found in the flow table, then the received packet is sent to the SDN con troller to fmd the path with minimum delay. A new entry is made to the flow table wherever a path is selected. It uses an open-loop congestion control principle on the available buffer and previous historical information to employ a congestion-conscious direct routing protocol. Thus, the SDN controller determines the shortest path with the least congestion.

The OpticalFog node configuration manager manages the model infrastructure for net- work configuration, address assignment, and physical path. Towards the northbound, the typical Resource manager handles the creation, removal of VM instances the underlying infrastructure.

Topology manager work as an NFV orchestrator that models various policies as per the input requirements, and virtual network configurations in the 5G scenario. Flow/Link Manager designs and incorporates edge computing simulation. Also, an edge data center is modeled with interconnected physical infrastructure (hosts, switches, and links). Edge data centers distinguish with resource efficiency and data center size. Every data center operates its network and resource management policies with various network policies for communi cating among data centers. VM location information and physical hosts are available glob ally to simplify the search for and management of simulation.

Algorithm 1: Proposd atlgtiithm for pIlilnit of CPS-bafsed task in t popojiset

Data: T whilo cAcrss tdllS)N yath/ do List AliUunnq TskA: while(Opwsstode P do Lisr CP-basask/T/mue:while tsk4T CTSdo if A/Uprd/ of T anP-bstdiskirw)dithen I Atid T (I CPSbassdTA:sTPlartA:

end

while ihsk T CPS-bajadais!Uiacrd do if(&susys < tsourceed then Al1ct T'1n Opticail oy' -de if(!ipti 1Node then AlloCALe T uni Optyic~id~boyNoe end else AIl7cate T tidi pt/fljNa "

if ()dpliiqoyNtac then C(ittu 7 suTch.a (T T AND7T'= -CP) if (INULL)then Assign T ' t a lid ata cti Assign7Tt0 Opticuis FyNodp3; else I Assign TI to I-tr: daa -cloud end enld end end end

The primary focus of the proposed system is to run the CPS-based applications on the OpticalFog node. Technique 1 shows that the resources provided by the new task are evalu ated and then allocated to the OpticalFog node. Initially, the OpticalFog node categorizes the CPS and the non-CPS based tasks on resource requirements such as processing power, demand for bandwidth and the protection level needed, etc. The OpticalFog node recon figures dynamically and provides the CPS the desired stability and QoS. The submitted task is represented as T which can be either a CPS-based or non-CPS-based task. Run- ningTask represents the already running tasks. OpticalFognode is a virtual, dynamically configurable node using the concept of SDN and ONV at the optical fog layer. TaskTo- Placedrepresents all coming tasks to be allocated to the OpticalFog node for further pro- cessing. OpticaFogNode"") is free available residual resources and OpticaFogNodeai is free available resources that are dynamically configurable by the OpticalFog node. SDN controller uses the proposed technique to chooses the best optimum path and assign CPS based tasks to the proposed OpticalFog node. Whereas, non-CPS-based tasks are sent to the cloud layer 108 only if OpticalFog node resources are not accessible. Initially, the Optical Fog node categorizes the CPS-based and non-CPS-based roles based on resource require ments such as processing capacity, bandwidth demand, and required security level, etc. The OpticalFog node dynamically reconfigures itself, giving the CPS the desired reliability and QoS.

CPS-based systems need higher computing capability in the 5G network edge. The delay- sensitive traffic of CPS-based tasks requires more bandwidth with ultra-low delay. The pro- posed OpticalFog layer effectively resolves the problem by supplying OpticalFog node that is capable of performing tasks with less delay, network usage, and energy consumption. Subsequently, the QoS is strengthened by measuring the delay, network usage and energy consumption related to CPS implementation problems in the 5G network.

To strengthen QoS, the delay is one of the most important aspects. The low-latency polling method that reduces the delay for bandwidth-intensive applications is followed. Addition- ally, the overall delay and average delay are determined. The delay in receiving rj in ONUi is determined by the following equation

tD(rj)=ui+tp + TC, (1) where ii is the propagation time, tp is the time between the request arriving at ONUi and entering the OLT, and TC is the time between the request. tp and TC are still less than the average cygle time TC max andpi is a constant. As well as the constraints are:

0 ! tp 5 Tc..x; 0 Tc< Tc.ax and pi = constant (2)

Therefore, the maximum delay for request rj under high network load can be computed as

max (tD(r)) = pi + 2TCmax (3)

To measure the average delay, the above limitations are considered as the ONU request arrival time is independent of network load. Finally, the average waiting period (t,) is called half, while each frame must wait another complete cycle time for transmission to the OLT. Therefore, the average network loads are measured as:

avg (tD(r)) =pi +1L5Tcma (4)

The effects of fewer delays in performance improve more less delay for cloud-based applications.

ObjectDelection and ObjectTracking applications modules capture too much amount of data to be processed. Therefore, the device imposes a considerable strain on communication bandwidth, which can increase intolerable transmission delay and reduce QoS. Here, the heterogeneous network has specific bandwidth limits over a wide geographic region for both EdgeFog layer and OpticalFog layer. The proposed OpticalFog layer effectively ease the bandwidth burden and reduce transmission delay. The communication bandwidth constraint over the traffic rate Al sending from the i fog node located at the EdgeFog layer to thej server located at the cloud data center through the transmission path. There's a con- straint f, on ij on each path's bandwidth capacity. Bandwidth constraint is determined as

0 o i,j ! fmax (5)

To determine the energy consumed by the proposed system, the energy consumption at each layer is calculated. All the edge devices and cloud data centers are taken into account in the context of energy evaluation. Optical fibers and optical splitters are used to inter connect ONUs by providing bidirectional transmission of information and consume almost zero energy, so they are not taken into consideration. Hence, the following equation shows the energy consumption by the proposed system:

AE = EEdge-layer + EopticalFog-layer + ECloud-layer (6)

Here, EEdge-layer represents the energy consumed by the surveillance cameras as

X(ECameras), EOpticalFog-layer represents the energy consumed by all ONUs as

(EONUs), and Ecloud-layer is the energy consumed by cloud data centers.

Figure 8 illustrates realizing the proposed system for an academic institute in accordance with an embodiment of the present disclosure. In an academic institute, deploying large-scale cameras provides a higher level of security and a strong sense of assurance. Moreover, monitoring classrooms or labs helps management to observe students' learning behavior and monitor teachers' performance. The goal of using the proposed system in an academic institution is not only to prevent illegal and improper behavior but also to provide a state-of-the-art education system that can support educators and students globally in advanced leading. Cameras surveillance in academic institutions earned great attention by deploying real-time CPS-based applications. In traditional systems, computations are performed on cloud data centers which are not preferable because it leads to high delays and consumes more bandwidth/energy. Nonetheless, CPS- enabled video streaming monitoring faces challenges like more delay, the higher energy usage of IoT/surveillance devices, and restricted bandwidth.

As illustrated in Figure 8, the proposed system will automatically analyze and process data from various distributed CCTV cameras installed across the academic units in an academic institution. The major challenge is handling the significant traffic of continuously produced video frames by CCTV cameras which must be treated without congestion. Moreover, implementing CPS-based applications on the proposed system improves QoS and meets delay-sensitive communication requirements.

Furthermore, the related literature on educational CPS has been studied. Functionality comparison of the proposed system with existing CPS-based educational systems is being carried out. Table 2 shows the novelty aspect of the proposed system by comparing the existing published work on the following attributes such as major contribution, applica- tion domain, software-defined networking (SDN), 5G network (5G), Fog Computing (FC), Cloud Computing (CC), Cyber-Physical System (CPS), Internet of Things (IoT), Real-time applications (RTA), and Optical Resources (OR).

In the experimental simulation, the proposed CPS-based camera surveillance application is simulated at the OpticalFog layer that effectively model and calculate the performance of the proposed system. A fog simulation toolkit called iFogSim is used for experiments that is the extension of CloudSim simulator and uses its essential functionality. It supports custom configuration to incorporate IoT devices, OpticalFog devices, and cloud data cent- ers hierarchically.

Therefore, the CloudSim layer manages the operations of fog and cloud devices in the context of opical resources. It uses the class FogDevice, which sets out the functionality and compatibility of the proposed OpticalFog nodes with other devices and IoT sensors. Furthermore, class Sensor represents sensor functionality with different sensor attributes. The link time among OpticalFog node and other devices is predefined to realize the pro- posed system in the context of optical resources in 5G network. Class Actuator is pre- defined concerning ONV and SDN specification for implementing the functionality the proposed OpticalFog node. Subsequently, class Tuple also specify the ONV concept and used to enforce communication of the OpticalFog nodes with other devices.

The physical topology is based on the apex cloud data center configuration, OpticalFog is located in the middle and CCTV cameras are taken at the territory of the 5G network. The PTZ control of the intelligent cameras are modeled as an actuator for live video streaming in the form of tuples for performing motion detection. Therefore, the proposed placement strategy is used with the default placement strategy of the toolkit to position application tasks.

Table 2: Comparison with existing CPS-based educational systems

Lightgray Authors Major contributions SDN 5G FC CC CPS IoT RTA 0 R Tomgren and Herzog Enhancing complementary skills among students N N N Y Y Y Y N Isakovic et al. Research and Education N N N Y Y Y Y N Sood and Singh Learning through educational games N N N Y Y Y Y N Sood and Singh Enhancing students's employability N N N Y Y Y Y N Ghosh et al. Testbed for educationsl CPS N N N Y Y Y N N Konstantinou cybersecurity for educational CPS N N N Y Y Y Y N Singh and Sood Monitoring Education Institutes Y N Y Y Y Y Y Y Amor et al. Cloud-based educational e-learning N N N Y Y Y Y N Kausar et al. Examination system in e-learing N N N Y Y Y Y N Yadav et al. Affective Learning for children N N N Y N Y Y N Amad et al. Promoting Sustainable Education N N Y Y N N N Ashtari and Eydgahi Post technology adoption behavior N N N Y Y N N N Proposed system 5G ready Cyber-physical system Y Y Y Y Y Y Y Y

In the experiment, the proposed system is simulated using CCTV cameras on distinct physical network configurations. It is assumed that there are distinct numbers of monitoring blocks present in the campus that varies from 2 to 32. Here, each block has eight CCTV cameras those are continuously monitoring the academic unit. All CCTV cameras are linked to the respective gateway of each block to provide internet access and to control the monitoring activities. Table 3 determines the delay association to different categories of devices at different layers.

The physical topology configurations of the varying blocks are subsequently set as M Block1 with 2, M-Block2 with 4, M-Block3 with 8, M-Block4 with 16, and M-Block5 with 32 monitoring blocks. The computation of all monitoring program modules is per- formed in the cloud-based implementation via the cloud data centres. As devices at net- work edge (default fog) are not computationally capable of hosting all surveillance application modules, ONV and SDN support the dynamic reconfiguration to the OpticalFog node. The proposed technique efficiently computes the resources required by the CPS-based task and assign it to the OpticalFog node. Overall efficiency can be measured by comparing the both scenario outcomes.

Likewise, two task placement strategies, traditional cloud-based and the proposed (OpticalFog-based) are used to place surveillance camera's modules on the physical network. In the traditional cloud-based deployment scenario both CPS-based and non-CPS-based application tasks are placed on the cloud data center, except for MotionDetection task. The MotionDetection task is assumed to be a camera-bound task which is performed by the camera itself. In case of the OpticalFog-based deployment, the OpticalFog node is used to carry out ObjectDetection and ObjectTracking tasks rather than WiFi gateways (default). The entire simulation is performed for 20 minutes.

Table 3 Delay association to different categories of devices at different layers Source device Destination device Delay in milliseconds CCTV camera Monitoring block 5 Monitoring block Optical gateway 2 Optical gateway ISP gateway 3 ISP gateway Cloud data-center 125

Figure 9 illustrates a comparison graph of delay in the proposed and cloud-based deployment scenario in accordance with an embodiment of the present disclosure. Finally, the results presented below show that the proposed OpticalFog node placed in the middle of network hierarchy has efficiently reduced the delay, network usage and energy consumption of surveillance cameras. The average processing delay faced by the sensing- actuation control loop of the surveillance cameras is taken into account.

Delay is measured in milliseconds for each topology configurations of surveillance block. Figure 9 shows that there is very less delay experienced in the proposed OpticalFog based scenario as compared to the traditional cloud-based scenario. A bottleneck is identified in the traditional cloud-based approach while managing all activities, which greatly increases the delay. Whereas, in the proposed OpticalFog-based scenario successfully retain less delay near the network edge because CPS-based tasks which are essential to the control loop are carried out at the proposed OpticalFog node.

Figure 10 illustrates a comparison graph of network consumption in the proposed and cloud-based deployment scenario in accordance with an embodiment of the present disclosure. Similarly, different surveillance blocks with increasing numbers of surveillance cameras have been deployed to monitor the efficiency of the network. Both placement strategies are evaluated to show the network usage under the increased load. Figure 10 clearly shows that the traditional cloud-based deployment increases additional load on the network in contrast with the proposed approach. It is observed the proposed OpticalFog node is capable to host the CPS-based tasks. Since these tasks are require most of the data-intensive com munication which are handled at the OpticalFog layer rather than the cloud layer 108. Thus, the ObjectDelection and ObjectTracking tasks are placed at the OpticalFog node which significantly reduces the data to be sent.

Figure 11 illustrates comparison graphs of energy consumption in the proposed and cloud-based deployment scenario in accordance with an embodiment of the present disclosure. Finally, in the overall simulation, the energy consumption by the various categories of devices used at different layers is taken into consideration. Here, the key challenge is to reduce the energy consumption of the surveillance cameras because the camera-bounded task i.e. Motion Detection gathers video frames drains out more energy. A comparison of the energy consumed by the CCTV cameras, and cloud data centers for distinct surveillance blocks is shown in Figure 1la-d. It is clearly observed that with increasing monitoring blocks, the energy consumption by the cameras and data center also increases but it is always less in the proposed scenario than the traditional cloud-based scenario. It indicates that the OpticalFog node handles most CPS-based tasks at the OpticalFog layer.

The proposed system emerges as attractive solutions for real-time CPS-based tasks in the heterogeneous 5G network. Instead of outsourcing all activities to the cloud, it utilizes OpticalFog node resident at the OpticalFog layer that has more computing power than the edge devices. Hence, minimizing delay, network congestion, and energy consumption. The proposed system significantly reduces the said challenges and increases the application's efficiency. Further, it has two advantages. First, the SDN controller can make resource allocation and routing decisions based on requirements for QoS. The second benefit is that nearby, resource-rich OpticalFog nodes can provide real-time video frame processing and submit notification when any anomaly is detected.

The proposed OpticalFog node supported the CPS-based applications with ultra-low delay, more bandwidth efficiency, and significantly low energy consumption of the edge devices. A placement strategy is one of the main concerns when carrying out CPS-based activities in the middleware of the cloud and the EdgeFog layer. The proposed technique uses the SDN concept to provide the shortest path through the optical network resources. Moreover, 5G enabled NFV, and ONV supported dynamically reconfiguration OpticalFog node efficiently. Hence, the proposed OpticalFog node has succeeded in fulfilling the requirements of CPS-based applications in the 5G network. The experimental assessment using the case study of surveillance systems in an academic institute evaluated delays, network usage, and energy consumption efficiently. Results show the effectiveness of the proposed system in the OpticalFog-based deployment scenario.

The drawings and the forgoing description give examples of embodiments. Those skilled in the art will appreciate that one or more of the described elements may well be combined into a single functional element. Alternatively, certain elements may be split into multiple functional elements. Elements from one embodiment may be added to another embodiment. For example, orders of processes described herein may be changed and are not limited to the manner described herein. Moreover, the actions of any flow diagram need not be implemented in the order shown; nor do all of the acts necessarily need to be performed. Also, those acts that are not dependent on other acts may be performed in parallel with the other acts. The scope of embodiments is by no means limited by these specific examples. Numerous variations, whether explicitly given in the specification or not, such as differences in structure, dimension, and use of material, are possible. The scope of embodiments is at least as broad as given by the following claims.

Benefits, other advantages, and solutions to problems have been described above with regard to specific embodiments. However, the benefits, advantages, solutions to problems, and any component(s) that may cause any benefit, advantage, or solution to occur or become more pronounced are not to be construed as a critical, required, or essential feature or component of any or all the claims.

Claims (7)

WE CLAIM

1. A QOS-aware optical fog-assisted cyber-physical system in 5G ready heterogeneous network, the system comprises:

a device layer consists of IoT devices such as sensors & actuators to gather and interact with a physical space and act as the source or sink of data; an edge fog layer equipped with a plurality of fog devices such as switches, routers, ONUs to host the application modules, wherein the application module is responsible to process all the data generated and send them to the next layer for performing machine learning tasks; an optical fog layer consists a SDN controller along with infrastructure management (ONV, 5G, virtualization) that is capable to create the OpticalFog node with the computing capability to scale up or down optical resources by dynamically aggregating or de-aggregating PON resources, wherein all PON resources are grouped together and synthesized with SDN and ONV techniques for creating OpticalFog node with high configuration and interconnection capabilities as Infrastructure-as-a-Service (IaaS) model; and a cloud layer having various mangers to manage resources, scheduling, making policy in such a way that application level QoS constraints are met and resource optimization is achieved, wherein the cloud layer provides functions to the cloud for the CPS- based modules.

2. The system as claimed in claim 1, wherein the fog layer is introduced to exploit the computational capability of the optical resources rather than edge devices for handling real-time and CPS-based applications.

3. The system as claimed in claim 1, wherein in Cyber Space, the first layer of fog is the traditional EdgeFog layer (have edge devices eg. routers, switches, hubs, etc.) and the second layer is the proposed OpticalFog layer. Apart from this, this layer also consists of the proposed OpticalFog node (created using optical resources), integrated with the SDN controller, and ONV in the 5G network(small cells and macrocells).

4. The system as claimed in claim 1, wherein all PON resources are grouped together and synthesized with SDN and ONV techniques for creating OpticalFog node with high configuration and interconnection capabilities as Infrastructure-as-a-Service (IaaS) model, wherein the OpticalFog node node can easily scale up or down optical resources by dynamically aggregating or de-aggregating PON resources.

5. The system as claimed in claim 1, wherein the PON provides residual computing capabilities of optical resources to create an OpticalFog node by using the SDN and ONV techniques.

6. The system as claimed in claim 5, wherein the PON-OLT network having 16 OLT chassis creates an OpticalFog node with the computing capability of 41,287,680 DMIPS (Dhyrstone million instructions per second) for processing, 4,194,304 MB for storage, and 2.5Gb/s interconnection speed, wherein the OpticalFog node process real time tasks at the OpticalFog layer rather than the cloud layer.

7. A QOS-aware optical fog-assisted cyber-physical method in 5G ready heterogeneous network, the method comprises:

Gathering signals and interacting with a physical space by a device layer consists of IoT devices such as sensors & actuators which act as the source or sink of data; hosting application modules through an edge fog layer equipped with a plurality of fog devices such as switches, routers, ONUs, wherein the application module is responsible to process all the data generated and send them to the next layer for performing machine learning tasks; creating OpticalFog node with the computing capability consisting a SDN controller along with infrastructure management (ONV, 5G, virtualization) to scale up or down optical resources by dynamically aggregating or de-aggregating PON resources, wherein all PON resources are grouped together and synthesized with SDN and ONV techniques for creating OpticalFog node with high configuration and interconnection capabilities as Infrastructure-as-a-Service (IaaS) model; and managing resources, scheduling, making policy in such a way that application level QoS constraints are met and resource optimization is achieved using a cloud layer having various mangers, wherein the cloud layer provides functions to the cloud for the CPS- based modules.

Device Layer 102 Edge Fog Layer 104

Optical Fog Layer 106 Cloud Layer 108

Figure 1 gathering signals and interacting with a physical space by a device layer consists of IoT devices such as sensors & 202 actuators which act as the source or sink of data hosting application modules through an edge fog layer equipped with a plurality of fog devices such as switches, routers, ONUs, wherein the application module is responsible to process all the data generated and send them to the 204 next layer for performing machine learning tasks creating OpticalFog node with the computing capability consisting a SDN controller along with infrastructure management (ONV, 5G, virtualization) to scale up or down optical resources by dynamically aggregating or de- aggregating PON resources, wherein all PON resources are grouped together and synthesized with SDN and ONV 206 techniques for creating OpticalFog node with high configuration and interconnection capabilities as Infrastructure-as- a-Service (IaaS) model managing resources, scheduling, making policy in such a way that application level QoS constraints are met and resource optimization is achieved using a cloud layer having various mangers, wherein the cloud layer provides 208 functions to the cloud for the CPS- based modules

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11