Network slicing (NS), a key enabler, plays a crucial role in implementing tactile Internet (TI). This TI promises scalable, customisable, isolated, and logical end-to-end networks (i.e., network slices) by offering dedicated and tailored networking solutions on demand to meet the customer’s need. Research in NS for TI is essential to realise haptic communications to achieve ultra-low latency, high throughput, and high reliability for real-time immersive and interactive environments in fifth-generation (5G) and beyond 5G (B5G). This article proposes a novel network slicing mechanism based on TI communication infrastructure for three different scenarios (TINetS3) leveraging software-defined networking (SDN) with Open vSwitch (OVS) to resolve the issue of provisioning and controlling the network slices on demand. This communication infrastructure is sliced under various traffic/application loads using pre-designed slice configurations. We develop algorithms for three scenarios, i.e., topology slicing, service slicing, and emergency slicing, to have a tailored slice for the physical infrastructure according to the network requirements. We validate the system performance by measuring the key performance indicators (KPIs), such as throughput and round-trip time (RTT), to optimise the network resources by network emulator. These KPIs show the expected experimental results aligned with the preset allocated spectrum for slices and RTT constraint values in various scenarios. The simulation results show the efficacy of our NS algorithms in designing tailored and customised slices as per the network-required KPIs. Finally, we validated our algorithms by comparing them with state-of-the-art algorithms.

Vehicular network (VN) is considered as one of the enabling technologies that requires stable connectivity among the vehicles. The significant dynamics in vehicular structure on the road necessitate the rapid information flow in managing crucial events. Dynamically, each network has a given lifetime from the initial formulation of a networked group to the final diminishment of the group. Throughout this lifecycle, a group of vehicles form a network and open the door for further vehicles to join, or existing vehicles to leave. Vehicles departing the group may rupture the data network and cause fragmentation or isolation of part of the network structure that raises the need for smart self-healing. The response to these events may cause latency in handling the flow of data, and hence efficient operation with minimal latency is required. Such network demands a flexible and autonomous self-structure approach with real time computational capabilities. Realizing this vision requires a network that supports low latency for interaction among the members as with the edge, fog and/or cloud. It is anticipated that driving towards the novel concept of vehicular network digital twin will offer the platform with backup intelligence and support for designing the scenarios that can have better interaction with road dynamic events with minimal latency. The objective of the paper is modelling the latency associated with the dynamics of the vehicular grouping approach to investigate the performance of a given dynamic group running on the road. This is aligned with road model boundary and limitations. Herein, the applicability of the grouping approach is tested based on the road capacity supported by the road boundary and number of lanes. Testing using Contiki-Cooja simulator has been implemented focusing on the latency and packet loss performance measures for the self-healing and self-formation phases.

With more than a decade of research and development, the Internet of Things (IoT) and Cyber-Physical Systems (CPS) are both promising solutions for various novel healthcare applications. This proves significant for supporting data collection on human movement detection and recognition. The paper discusses a novel concept of the IoT for healthcare applications. The focus is on the follow-up of prehabilitation programs that are essential for pre-operative patients. The solution caters for a flexible approach that benefits both supervised and unsupervised prehabilitation programs. This paper describes the overall system architecture considering a distributed computation model that includes the wearable sensors, the access points and the edge or cloud-related computing. The model can act as a baseline for a digital twin that could follow up the prehabilitation program implementation of a preoperative patient and offer the necessary information relevant to recommendations and alerts.

Using UAVs as mobile data ferries can amplify ground network performance by mitigating the impact of horizontal radio pollution on the environment. Designing an energy-efficient UAV communication with a randomly distributed ground sensors via enhancing the ground network structure dependent to the UAV path has been one of the challenges for the UAV-assisted WSN data gathering efforts. This paper aims at developing an approach called ‘UAV Fuzzy Travel Path’ that supports UAV smooth trajectory planning in gathering data of distributed wireless sensors over a large space. The dynamic orchestration of ground wireless sensors grouping has been considered in this paper for improving the network performance in data collection. This offers a more dynamic and flexible network that interact with the UAV planned path. Preliminary results of testbed simulations reflect that the proposed software define ground network system has presented promising outcomes in terms of network performances such as network latency and packet delivery rate. Further work is required for testing the scheme on a wider range of software defined WSN and UAV flight formations.

Light may be suitable for underwater wireless communication of multimedia data. However, current optical solutions face obstacles due to directionality and wavelength-dependent absorption and scattering. Here, we propose a conceptual design using differential signalling and a concave, multiple-aperture array of photodetectors. This combination could increase the field of view, enhance link adaptability, and introduce redundancies, supporting remote monitoring and data-collection with mobile robots in offshore operations.

Empowered by sufficient robotics and haptic hardware at the edges of a communication network system, tactile Internet (TI) is the upcoming transformation that will empower the control of Internet of Things in real time and facilitate a true paradigm shift in making a dream into a reality. However, one of the most crucial challenges in realizing TI is to achieve round-trip time (RTT) of 1 ms or less. This RTT includes transmission, queuing, processing (operator’s end), and acknowledgement times (controlled end). Another challenge is to achieve ultrahigh reliability for establishing haptic communications for TI. This article proposes a novel fog-based traffic flow framework employing software-defined networking (SDN) and fog computing (FC) to address these challenges. An efficient traffic flow algorithm is developed to effectively manage complex and critical traffic in the network to reduce extra processing and waiting times at each level of clouds. The SDN and FC approaches provide an effective solution to route the traffic efficiently in the system. Hence, the traffic flow paths from the master to slave sections are reduced, and consequently, RTT is also reduced. The performance of the proposed system is evaluated by an iFogSim simulator in terms of throughput, RTT, energy consumption, and reliability. The simulation results obtained show that the proposed system outperforms the existing edge, cloud, and cellular networks.

Recent advances in Unmanned Aerial Vehicle (UAV) technologies make their use as data collection sinks feasible for practical applications of Wireless Sensor Networks (WSN). Using UAVs as mobile data collection agents can improve network performance by reducing the horizontal radio pollution in animal sanctuaries or conservations, for example. This may also positively improve energy consumption of both UAV and sensor nodes (SN) on the ground. Designing an efficient UAV trajectory which has the capability of data gathering from randomly distributed large number of SNs has been one of the challenges for the UAV-aided WSN data gathering solutions. This paper aims at conceptualizing a solution for UAV path planning in gathering data of distributed wireless sensors over a large space. Both efficient flight path and dynamic orchestration of sensor grouping are considered as two key factors for improving the flight performance in data collection. Herein, a classical Travel Salesman Problem (TSP) solution supported by the ground group offerings are used for manipulating a simplified centroid-based trajectory and turning it into a more relaxed fuzzy route. This in turn would offer a ground for smoother and more efficient flight paths. While preliminary modelling of the approach offers encouraging results, the demand for fluidity in dynamically re-orchestrating the ground sensor networks must be carefully considered. Successful implementation of this concept would support a wide range of applications such as the farming and conservation used cases.

Connecting wireless sensor networks (WSNs) to remote servers or clouds through the Internet of Things inspires the formation of a highly complex system. This has significant potential to encourage WSN softwarization. Such a network enables interaction between the physical and virtual sensor network, allowing for flexibility within the physical system to adapt with the dynamics of the service requirements. An architecture of a test environment with adaptive quality of service (AQoS) has been presented. The concept provides an avenue for a flexible system that is capable of reacting to dynamic changes of process demands. Physical network performance can be predicted by analyzing the historical data in the background on a network simulator or virtual network. This in turn allows for estimation of the necessary adjustments needed to improve the network performance without disturbing the physical system. This paper reports the experimentation and applicability of userdriven AQoS models on the system. Early stages of testing the organization, taking into account the system behavior and subsequent reaction to necessary adjustment of the WSN operational configuration, has shown encouraging results.

Motor current signature analysis is a modern approach to fault diagnose and classification for induction motors. Many studies reported successful implementation of MCSA in laboratory situations whereas the method was not so successful in real industrial situation due to propagation of neighbor faults and unwanted noise signals. This paper investigate the correlation between different observations of events in order to provide a more accurate estimation of behavior of electrical motors at a given site. An analytical framework has been implemented to correlate and classify independent fault observations and diagnose the type and identify the origin of fault symptoms. The fault diagnosis algorithm has two layers. Initially outputs of all sensors are processed to generate fault indicators. These fault indicators then are to be classified using an Artificial Neural Network. A typical industrial site is taken as a case study and simulated to evaluate the concept of distributed fault analysis.

Unlike schemes based on public key cryptography requiring mathematically expensive operations on large integers, our scheme uses simple modulo arithmetic operations on small integers making it very fast and efficient and requiring a small amount computing resources. This makes it practical for use in networks using devices like sensors nodes. We showed some results of performance in the MICAz mote. Our scheme has an interesting side-effect - the private-public-master-key association (PPMka) obscurity such that the keying material cannot be used directly to attack the system. This means that even if sufficient number of nodes are compromised and their keys obtained, the attacker would have to first expand a considerable amount of effort to discover the public key and master key associated with each private key. This allows the scheme to be used on a large scale in hostile environments.

Data centricity is a key characteristic that distinguishes wireless sensor networks (WSNs) from other types of wireless networks. In traditional WSNs, sensors detect environmental attributes or events, which are transmitted to and stored on dedicated base stations or sinks at the network edge for further processing. With advances in data storage technologies, it becomes possible for embedded sensor devices to cooperatively store and maintain data within the WSN for later retrieval by in-network users such as humans who interact with their environments which have been smartified by embedded intelligence. In this paper, an in-network storage scheme is proposed for a virtual broker (VB) based pub/sub system in WSNs. The proposed scheme is shown to achieve good performance in terms of storage load balancing, communication cost, and energy efficiency.

Brokers are critical elements for many publish and subscribe (pub/sub) systems. They are dedicated devices with large computation and storage capacities, but they could suffer from scalability problems when comes to actual large-scale implementations in wireless sensor networks. Inspired by the recent virtual MIMO concept for wireless sensor networks, this paper proposes the idea of `virtual brokers' for large-scale pub/sub systems in sensor networks, and presents in detail the design of a three-stage algorithm for forming a virtual broker around multiple co-located sensor nodes using only network connectivity information. Using computer modeling, we conduct an investigation of our virtual broker formation algorithm and preliminary results demonstrate the potential of our algorithm for scalable formation of virtual brokers.

The recent technological advances have entrenched the potential benefits, when large population of wireless sensor nodes deployed in agricultural, industrial and environmental areas to predict the behavioral analysis of physical attributes such as temperature or gas. This work mainly focuses on the three dimensional temperature distribution of a specified field based on virtually deployed sensor nodes in a simulation environment. The parameters temperature and location are considered in the simulation model. In this work, we have evaluated the minimum number of nodes that are required to map the given space. Modeling & simulation has been dealt with in testing the network density on the space coverage. This work exploits a spatial correlation of temperature data in a given space. Finally the paper discusses the extension of approaches that leads to new research challenges due to the relationships between the obstacles within the environment.

This paper presents a Valve Regulated Lead Acid (VRLA) battery state of health (SOH) estimation model. The model is based on a region of the battery discharge voltage response known as the coup de fouet. Two implementation approaches are considered. The first approach utilises soft computing techniques in the form of an Adaptive Neuro-Fuzzy Inference System known as ANFIS [1], [2]. The second approach utilises the least squares estimator (LSE) hard computing technique. The comparison of approaches has highlighted the potential of soft computing for providing accurate results. It, however, also raises the need for caution in employing these techniques, as the additional accuracy obtained comes at the expense of increased complexity which may not be justified in practice.

The design and implementation of a low-cost robot model suitable for items retrieval in a lightweight inventory system are discussed. A Cartesian-type guided manipulator is constructed. The robot controller design is based on the utilization of a low-cost PC supported by a front-end microcomputer. Dedicated, simple, store-operator menu-based commands are developed for robot programming. The design has resulted in a compact, low-cost, and moderate performance robot system. The use of a single-chip controller to replace the front end computer with a considerable reduction in the required hardware is suggested. The design also lends itself to both software and hardware expansion.>