**1. Introduction**

The rapid development of communication technologies from First-Generation (1G) to Sixth-Generation (6G) has gained enormous attention due to its emerging services like human-to-human (H2H), machine-to-machine (M2M), and human-to-machine (H2M) communication. These emerging services are induced by drivers like mobile Internet, Internet of Things (IoT), and tactile Internet (TI). The IoT envisions to fill the gap between the cyber and physical world [1]. It is defined as to interrelate every existing computing object around us such as, mobile devices, sensors, and actuators, over the Internet. Moreover, IoT technology provides data sharing and communication in the M2M environment. Recently, the TI, with the aim to enable haptic communications, has shifted the IoT paradigm to real-time interaction between H2M and revolutionized the next-generation communication technologies [2,3]. The TI is envisioned to empower H2M communication where a human can interact with machines in a virtual and physical environment, while experiencing the haptic sensations (touch and forces) in addition to traditional audio-video data [4]. Figure 1 depicts the technological evolution of the communication trends from 1G to 6G wireless communication.

Several international standard organizations, such as the international telecommunication union, the Third-Generation Partnership Project (3GPP), and the Institute of Electrical and Electronics Engineering (IEEE), are working to enable the existing and develop new network architectures to carry haptic data over the communication in real-time. The TI standard working group IEEE P1819.1 has already initiated and defined reference architecture, technical functions, and the definition of the TI [5]. Moreover, it also described

**Citation:** Zubair Islam, M.; Shahzad; Ali, R.; Hadier, A.; Kim, H.S. IoTactileSim: A Virtual Testbed for Tactile Industrial Internet of Things Services. *Sensors* **2021**, *21*, 8363. https://doi.org/10.3390/s21248363

Academic Editors: Zihuai Lin and Wei Xiang

Received: 10 November 2021 Accepted: 13 December 2021 Published: 15 December 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

standard use cases of the TI and corresponding strict requirements, including teleoperation, automotive, immersive virtual/augmented reality, internet of drones, interpersonal communication, live haptic broadcast, and cooperative automated driving. However, these use cases demand near real-time connectivity (ultra-reliable and ultra-responsive) for M2M and H2M communication. This type of real-time connectivity is termed as ultra-reliable and low latency communication (URLLC). The URLLC is one of the key services of the Fifth-Generation (5G) networks, along with enhanced mobile broadband and massive machine-type communication. Moreover, 3GPP has introduced the 5G new radio to increase reliability and minimize end-to-end (E2E) communication latency for the URLLC services. In Release 15, 3GPP describes the URLLC requirement with the reliability of 99.9% for a single 32-byte packet under 1ms latency [6]. Conclusively, 5G URLLC services are one of the potential enablers for the extreme requirements of the TI.

**Figure 1.** A taxonomy of the different emerging communication trends.

Moreover, these requirements become more critical for loss-intolerant and delaysensitive TI industrial and medical applications. For example, remote industrial management and the automation of industrial robots (sensitivity of control circuits) demand latency between 0.25–10 ms with a packet loss of ≤10−<sup>9</sup> [7]. Therefore, supporting next-generation industrial applications, including immersive reality, holographic, and haptic/tactile communication, demands a 5G network with new physical and upper layer techniques to guarantee quality of service (QoS) and quality of experience (QoE) provisioning. Furthermore, the 6G technology paradigm promises to break the 5G network limitations and enable them to virtualize human skills and transfer them from one place to another within 1ms through 6G native artificial intelligence (AI) network architecture. In-depth work on 5G URLLC services, beyond the 5G and 6G communication network, is presented in these articles [8–12]. Table 1 compares the connectivity requirements of the traditional and emerging tactile IIoT applications (adapted from [7]). The relationships between emerging technologies such as IoT, IIoT, Industrial Internet, Internet of Everything (IoE), TI, tactile IoT, tactile IIoT, Industry 4.0 and 5.0 are presented in Figure 2.

An in-depth discussion on conventional and emerging industrial is presented in [7], where the authors investigated the role of TI in the industrial environment, along with technical connectivity requirements of the TI industrial services. One of the vital use cases of the TI in the industrial domain is the bilateral/multilateral haptic-driven teleoperation systems. A teleoperation system consists of a human operator (master), teleoperator (slave), and a communication network that link the master to a slave domain, and enable the operator to interact with the teleoperator in the distant and inaccessible remote environment to perform complex tasks. The TI-based network provides bilateral communication to manage touch and actuation in real-time between the master and slave domain with a

focus to ensure QoS and QoE requirements. Haptic-enabled teleoperation systems have numerous applications in Industry 4.0, such as robotic automation, smart manufacturing, smart logistic, the mining industry, food industry, healthcare industry, and industrial managemen<sup>t</sup> (controlling and monitoring). Contrary to the traditional application, Hapticenabled industrial applications demand high QoS and QoE, and depend on the nature of the application.


**Table 1.** Summary of the connectivity requirements for traditional IIoT and emerging tactile IIoT services.

**Figure 2.** An overview of the relation between IoT, IIoT, tactial IoT, tactile IIoT, Industry 4.0, and Industry 5.0.

One of the effective ways to investigate the tactile IIoT application requirements, performance, and testing the new solutions to ensure QoS and QoE, is to set up a virtual testbed similar to the real network. The testbed must allow us to utilize and maintain hardware and software virtually on a standard computer without purchasing them. In the literature, several recent articles have proposed testbeds to overcome the above-mentioned challenges. The work in [13] proposed a haptic system testbed to characterize and validate E2E haptic communication of different use cases of TI. The authors introduce a framework comprised of multiple sub-blocks that can be re-configured based on the nature of use cases, with a focus on minimizing cost and evaluation time. It also provides an option to integrate the testbed with the simulation platform through a connector interface to perform testing. Commonly, it is intended to offer an extensive range of haptic hardware, including sensors, actuators, and tactile interface boards. A testbed for tactile and kinesthetic data coding was proposed in [14] aligned with IEEE P1918.1 TI standard working group to improve and standardize haptic codec. The proposed haptic coding testbed is considered as a reference testbed with the aim to develop optimal data compression schemes to exchange tactile and kinesthetic information and enable human-in-the-loop TI services. The authors

also provide some reference tactile data traces, software, and hardware to evaluate newly developed kinesthetic and tactile codecs.

In [15], a framework for tactile cyber physical systems was proposed, which is specifically for physical remote environments and based on network simulator NS3. It provides an interface for robotic experiments, along with haptic communication modules. However, the authors ignored the extensibility of the proposed testbed for other haptic-driven applications. Similarly, the authors in [16,17] proposed a generic testbed framework for different TI use cases. A data-driven experiment setup was proposed in [16] to provide a common playground for testing haptic applications. The proposed haptic communication testbed at the Otto-von-Guericke University of Magdeburg (OVGU-HC) focused on providing experiment testbed for long-distance haptic-enabled teleoperation systems, in addition to small scale wireless haptic-driven applications. The OVGU-HC presents experiment automation and data collection utilizing experiment description language (DES-Cript). The proposed OVGU-HC did not work standalone, is a part of the MIoT-Lab, and is just used to gather hepatic experiment information. Moreover, it utilized domain-specific language DES-Cript [18], and did not provide an open-source facility to the research and development community.

The study in [17] presents a two-level classification of the TI applications based on controlled environment and master-slave integrations to develop a generic testbed, with a focus to ensure compatibility for all these classified applications, which is named as TI- eXtensible Testbed (XT). To demonstrate the potential of the TIXT, they discuss H2M haptic communication in the virtual and physical environment. However, they ignored the explanation on how to characterize the network impairments (delay, jitter, and packet losses) and investigate the performance of the haptic-driven IIoT application. Therefore, there is a strong need for a testbed that offers flexibility, scalability, open-source availability, tailored to examine network impairments, communication flow, and extensible for TI IIoT use cases. In this regard, we proposed a virtual testbed called IoTactileSim to investigate tactile IIoT services from QoS and QoE perspectives. The IoTactileSim employs Software Define Network (SDN) and edge computing at the core network to tactile industrial application. The following section presents the main contribution of the proposed IoTactileSim testbed.
