Precise clock synchronization underpins deterministic operation in wireless systems spanning industrial automation, vehicular networks, distributed extended reality (XR), smart infrastructure, and wide-area precision agriculture. Wireless links introduce variable propagation delays, channel asymmetry, interference, clock drift, and scalability constraints that make sub-microsecond alignment difficult. This article provides a comprehensive survey and tutorial on wireless clock synchronization. We introduce a five-dimension taxonomy covering: system architecture, synchronization mechanism, correction strategy, delay and uncertainty modeling, and resource and deployment class, and apply it to eight canonical protocol families and to synchronization as realized across IEEE 802.15.4, ZigBee, Bluetooth low energy (BLE), LoRa, Wi-Fi, Ultrawide band (UWB), and 4G/5G/6G systems. We examine solutions across five application domains: industrial automation and Industrial Internet of Things (IIoT), vehicular V2X, distributed XR and metaverse, infrastructure monitoring, and wide-area Internet of Things (IoT) and precision agriculture; alongside tools, testbeds, and datasets supporting evaluation. Open challenges addressed include scalability and mobility, ultralow-jitter determinism, robust clock parameter estimation under non-Gaussian delay distributions, distributed and consensus-based synchronization for infrastructure-free networks, secure and resilient synchronization against wireless-specific threats, and cross-domain convergence encompassing time-sensitive networking (TSN)–5G/6G interoperability and joint communication, sensing, and timing as an emerging 6G design paradigm. Together, these contributions provide the first unified cross-technology framework connecting fundamentals, protocol families, application domains, and open research challenges in wireless clock synchronization.

This paper highlights the value of Digital Twins (DTs) in optimizing and maintaining Cyber-Physical Systems (CPSs) across domains like manufacturing and communication. It proposes the interaction between an industry DT and a 6G DT and identifies Situational Awareness (SA) as a key step toward supporting critical services. A use case involving 6G communication with mobile User Equipment (UE) in manufacturing is analyzed to identify critical operating scenarios and how SA can predict and mitigate the risk of failure in these scenarios. Based on this use case, relevant parameters for improving the SA of the entire CPS are identified in both DTs and an example of interaction is given for optimizing the joint task planning and execution. Additionally, the benefits of cross-domain SA are demonstrated through 5G testbed measurements. The use case illustrates the broader applicability of our proposal.

Reliable time synchronization is of utmost importance within 6G networks, particularly when envisioning their role in future industrial automation scenarios that demand deterministic communication. In particular, we focus on the integration of 6G with time-sensitive networking (TSN). Traditionally, the best timeTransmitter clock algorithm (BTCA) is used to dynamically select a grandmaster (GM) for TSN nodes in a 6G-TSN network. However, BTCA’s performance in case of link or device failures is categorized to be slow. The IEEE 802.1ASdm standard addresses this limitation by introducing a hot standby GM in the network. In this paper, we extend the hot standby amendment to 6G-TSN networks to support continuous time synchronization. For the analysis, we develop a simulation framework in OMNeT++. We analyze the performance in terms of clock drift and out-of-sync time for different time synchronization scenarios in case of failures. Having two synchronized GMs in the network reduces the out-of-sync time in case of failure when compared to BTCA.

We demonstrate a private 5G network integration with precise ultra-wideband (UWB)-based localization for enabling a collaborative robots use case in an industrial environment. The results confirm 5G’s ability to provide high throughput, low latency, and reliable connectivity, even in complex industrial environments. UWB provides real-time decimeter-level wireless localization accuracy, critical for reliable and precise object localization. These combined capabilities are essential for real-time monitoring, automation, and operational efficiency. The developed demonstrator serves as proof of concept, showcasing how 5G, when coupled with advanced localization technologies, can optimize industrial processes and improve safety. This study offers a valuable system performance evaluation under real-world conditions for researchers and industry, underscoring the transformative potential of 5G in future industrial systems and its integration with precise wireless localization technology.

The rise of industrial applications such as autonomous systems, industrial automation, and precision manufacturing has increased the need for resilient and continuous time synchronization. Time-sensitive networking (TSN) addresses these requirements using the IEEE 802.1AS standard, which implements the generalized precision time protocol (gPTP) for sub-microsecond accuracy. However, best timeTransmitter clock algorithm (BTCA) used for grandmaster (GM) selection is slow to recover from failures and cannot detect transient faults, resulting in unstable synchronization. The IEEE 802.1ASdm standard addresses these limitations by disabling BTCA and introducing a static configuration with a hot standby GM. This paper presents a first analysis of time synchronization in TSN networks with hot standby. Using OMNeT++, we compare BTCA and hot standby performance in terms of out-of-sync time and clock offset during failures. The results demonstrate a close to zero out-of-sync time with hot standby as compared to BTCA.

Industry 4.0 promises the increase of productivity and efficiency within manufacturing and industrial processes. A key milestone in this evolution is the 3GPP specification on time-sensitive communication and integration of 5G with time-sensitive networking (TSN), thus enabling the required flexibility for future time-sensitive industrial applications. Although 3GPP introduces ultra-reliable low latency communication (URLLC) features to meet stringent timing requirements, the notable challenge persists in resource overprovisioning, limiting the capacity of the next-generation radio access network (NG-RAN). Our work addresses this by focusing on the sources of packet delay and packet delay variation within the 6G system. Through analysis and proposed enhancements, supported by simulation and measurement results, we aim to minimize queuing delay and optimize 6GS capacity. Our study not only explores standardized 5GS features but also proposes extensions to support periodic deterministic traffic, offering insights to enhance 6GS performance and capacity in real-world scenarios.

5G is considered a promising wireless communication technology to fulfill the high-demanding communication requirements of many Industry 4.0 applications. This work evaluates a 5G campus network for indoor factory floor scenarios using the latest commercially available 3GPP release-16 developments. The measurement campaign is conducted to obtain detailed coverage maps with reference signal received power, channel quality indicators, and downlink and uplink throughput (TP). Moreover, end-to-end delay measurements with varying channel conditions, 5G quality of service (QoS) priorities, and traffic loads were evaluated. It was concluded that even without ultra-reliable low-latency features, the TP and latency performance could be controlled by configuring QoS parameters. The evaluations suggest scenarios where QoS allocation and retention priority levels can be used in order to ensure the required performance of a specific QoS flow within the 5G system.

Functional safety protocols provide automatic protection against system failures. Safety will be of critical importance for future industrial systems since they are undergoing automation and digital transformation under the Industry 4.0 revolution. However, functional safety depends on the underlying communication system employed in the manufacturing area. An extensive evaluation of the communication system and the functional safety protocol is needed before deploying it to the production environment. To facilitate the evaluation process of a fail-safe communication protocol, we integrate openSAFETY in the OMNeT++ simulation framework. We validate its integration by comparing the simulation results with an existing experimental work where the results show similar trends. Furthermore, we simulate an industrial closed-loop control use case with IEEE 802.11n being the communication technology. The results obtained in terms of safe-state duration and packet error rate demonstrate that 802.11n with its default configuration is not suitable for a high demanding industrial application.

Functional safety has become a crucial part of industrial automation. With industry environments being connected more than ever, achieving required functional safety depends on establishing safety-critical communication. Numerous application layer safety protocols have been developed to ensure compliance with functional safety standards for wired communications. However, with all the benefits wireless communications entails, wireless communication is pushing for an important place in the future of industrial automation. It is important that safety protocols also follow this transition and are able to operate as intended using a wireless channel. By default, openSAFETY frames are exchanged using UDP broadcast, whereas TCP provides additional reliability features. Thus, the aim of this paper is to analyze if safety-critical communication using a wireless channel can benefit from the additional reliability features provided by TCP. To answer this, we experimentally analyze the performance of openSAFETY protocol providing functional safety over the IEEE 802.11 standard for three different test cases. We analyze the performance in terms of median end-to-end delay and time spent by a safety node in a safe state for both protocol stacks. Results show that UDP provides lower median end-to-end delay, whereas TCP is able to achieve less time spent in a safe state under bounded delay constraint.