TIME-SENSITIVE NETWORKING: Exploring the Potential Aviation and Aerospace Use Cases
By Adeyinka Olumuyiwa Osunwusi, PhD
For the league of industry sectors whose operations revolve around constant and reliable control of physical processes and devices, the efficient delivery of data traffic across network elements is a critical task. For these and quite a good number of other industry players particularly in the utilities and automotive industrial sectors, communication across network devices needs to be carried out in a speedy, timely, predictable, cost-effective, reliable and efficient manner to facilitate high level of Quality of Service (QoS). Traditionally, critical industry players have been having to have recourse to the use of traditional Ethernet, typically based on IEEE 802.1Q and IEEE 802.3 standards, when the game comes down to facilitating cost-effective, reliable and speedy communication across a network.
The rationale behind this choice may not be far-fetched after all. Standard Ethernet is known to be a technology that combines affordability, speed and versatility with high throughput. It is also a robust technology in terms of flexibility and the capability to deliver increased data. In addition, Ethernet typically features non-centralized network with open and widely supported protocols.
Notwithstanding these advantages and the ambitious evolution of Ethernet standards over the years, traffic flows across traditional Ethernet networks are known to be susceptible to latency and jitter whilst guaranteeing just the best effort behavior of the networks. There are also challenges revolving around safety requirements, bandwidth reservation limitations, packet buffering and the incompatibility of available network solutions. More importantly, traditional Ethernet lacks determinism – a critical performance factor for real-time applications in industrial sectors such as the aerospace, transportation, utilities, manufacturing and automotive industries – where time synchronization and robust controls of traffic flows across network elements are a critical performance factor. This critical performance factor, to be sure, is what has been driving the increasing shifts towards Time-Sensitive Networking (TSN) technology.
TSN is clearly becoming the technology of choice when the needs are all about providing the determinism, scalability and reliability required for the provision of time- and safety-critical communication between bridges or switches and end devices (sender and receiver(s)) in standard Ethernet networks. As a matter of fact, the central objective of TSN is to add Layer 2 deterministic capability to Ethernet-based networks such that data traffic can be delivered much more efficiently from one point to another in a deterministic manner with all network elements sharing a common concept of time. This underscores the fact that TSN is primarily a time-focused technology, a feature that enables the predictability and efficiency of TSN flows. With its time synchronization and traffic shaping features, TSN defines limits for traffic travel time, prioritizes the delivery of time-critical traffic and provides temporal isolation from non-critical traffic, thus guaranteeing ultra-high reliability, low bounded end-to-end latency, low bounded jitter and minimal or zero congestion loss. As part of the time synchronization mechanism of TSN, the IEEE 1588 standard specifies a set of Precision Time Protocol (PTP) profiles for a specific set of TSN applications, thus ensuring the interoperability of TSN networks and network elements in spite of the multi-vendor product nature of the TSN technical terrains.
In terms of versatility, TSN is also a multi-environment technology given the non-limitation of the payloads of the Ethernet frames (stream) of a TSN flow. This capability enables traffic convergence as multiple traffic – both time sensitive and non- time sensitive data traffic – can share a common network and a common sense of time. Aside from this, TSN-enabled networks are famed for enhanced security as they incorporate a number of security mechanisms such as encryption, authorization, and authentication. Scalability is another unique feature of TSN, enabling the robust optimization of topologies and the non-limitation of TSN applications to a specific transmission rate. Unlike traditional Ethernet, a TSN communication network necessarily guarantees a real-time behavior, ensuring that time-critical traffic is delivered within the bounds of a specified deadline whilst also managing the delivery of best-effort, non-time critical traffic. Additionally, TSN features, unlike traditional Ethernet, a fully centralized network configuration implementing, among other components, a central network controller (CNC) and a centralized user configuration (CUC) components. A TSN network, though, can also be configured either as a hybrid (centralized/ distributed configuration) or a fully distributed network to provide ultra-reliable, time-sensitive and low latency communication over standard Ethernet.
Today, the use of TSN for applications in both wired and wireless communication networks scenarios is being explored by industry stakeholders, although wired networks are much more ideal for supporting real-time applications that guarantee predictability as well as low bounded latency, low bounded jitter and minimal congestion loss. This notwithstanding, standards have been developed to enable reliable wireless TSN, particularly from the vantage point of challenges related to time synchronization performance requirements.
TSN – TECHNICAL DESCRIPTION
Time-sensitive networking (TSN) technology represents not only an enhancement on the IEEE 1588 standards but also a set of standard modifications on the IEEE 802.1Q standards. It is a reflection of the work of the Time-Sensitive Networking (TSN) Task Group of the IEEE 802.1 Working Group1 established in 2012 for the purpose of developing, among other things, real-time Ethernet solutions and standards that would provide the level of reliability, predictability and time synchronization required for real-time deterministic applications. It is a technology that is based on open standard interfaces and is also largely an Ethernet standard as opposed to an Internet Protocol Standard (IPS). In addition, TSN is an Open Systems Interconnection (OSI) Model Layer 2 – Data Link Layer – technology. The implication, therefore, is that TSN flows are effectively limited to data packets encapsulated into frames. Aside from this, TSN application does not extend to higher level application functionalities.
Technically speaking, TSN is not based on a single specific standard. Rather, it is a set of IEEE 802.1 standards that define guidelines for facilitating reliable and time-sensitive communication over an Ethernet network using a profiles-based mechanism that defines specific protocols, configurations, and features that are appropriate for particular TSN applications. Typical TSN standards include: IEEE 802.1AS/ASRev (Timing and Synchronization for Time-Sensitive Applications), IEEE 802.1Qbv (Enhancements for Scheduled Traffic), IEEE 802.1Qcc (Stream Reservation Protocol (SRP) Enhancements and Performance Improvements), and IEEE 802.1Qav (Forwarding and Queuing Enhancements for Time-Sensitive Streams).
TSN FEATURES
Given the multi-standard characteristics of TSN applications, typical TSN features are usually partitioned in line with specifications and features that are specific to the set of standards developed for TSN applications. The set of TSN standards address the following four features:
- Time Synchronization: The time synchronization feature of TSN is central to enabling networking elements implementing time-sensitive networking to share a common concept of time with an accuracy of just a few nanoseconds. It is also the primary enabler of that core performance factor of the TSN technology – determinism. This synchronization is the key to ensuring reduced and deterministic latency as well as the timely delivery of critical data traffic.
TSN accomplishes precise timing synchronization using the IEEE 1588 standard-defined Precision Time Protocol (PTP). The time synchronization mechanism is based on a Master-Slave clock relationship whereby distributed Ethernet frames are used to synchronize the clocks of all network elements – the slave nodes – to a common base sub-microsecond time reference clock domiciled in the master element (also known as grandmaster) of the network. This synchronization process is accomplished with the use of the IEEE standard 1588 PTP profile that meets the requirements of a given TSN application. Typical TSN PTP profiles include IEEE 802.1AS and IEEE 802.1ASRev. The IEEE 802.1ASRev essentially reflects changes in performance characteristics and standard requirements, particularly when transitioning to much more safety-critical and time-sensitive applications where instantaneous responses to time synchronization challenges due to network elements failure are imperative.
- Traffic Prioritization: The traffic prioritization features of TSN provide for the allocation of the necessary bandwidth, allowing for traffic flows on the network while prioritizing time-sensitive traffic flows over less critical packets and guaranteeing low bounded end-to-end latency with zero congestion loss.
The TSN prioritization capability is accomplished using a set of IEEE 802.1 standards such as IEEE 802.1Qbv (Enhancements for Scheduled Traffic), IEEE 802.1Qav (Forwarding and Queuing Enhancements for Time-Sensitive Streams), IEEE 802.1Qcr (Asynchronous Traffic Shaping) and IEEE 802.1Qch (Cyclic Queuing and Forwarding) to achieve TSN mechanisms such as traffic shaping and traffic scheduling. IEEE std. 802.1Qbv, which provides specifically for deterministic traffic scheduling, defines time-aware shapers (TAS), allowing for the assignment of specific time-defined slots to different traffic flows (streams) on the network with priority assigned to critical traffic flows over the rest of the network traffic. Allied to the traffic shaping features are TSN’s traffic policing mechanisms using IEEE 802.1Qci (Per-Stream Filtering and Policing), thus ensuring that network elements are sufficiently protected from failures or damage through checks on frame timing and content. The IEEE 802.1Qav standard controls traffic flows and allocates bandwidth, thus guaranteeing reduced latency bounds by allowing for the prioritization of real-time traffic over non-real time best-effort traffic using a Credit-Based Shaper (CBS) mechanism. To further improve network efficiency and minimize latency times and jitter, frame pre-emption enhancement for Ethernet MAC (Medium Access Control) outlined in IEEE 802.1Qbu (Frame Preemption) and IEEE 802.3br (Specification and management Parameters for Interspersing Express Traffic) can also be implemented. The IEEE 802.1Qbu standard’s role is very significant when it comes to efficiency optimization as the protocol is responsible for how higher-priority traffic flows can take prominence over lower-priority traffic.
- Reliability: The capability to offer redundancy is central to the ultra-high reliability features of TSN-enabled networks. The necessary redundancy mechanisms are defined around the IEEE 802.1Qca (Path Control and Reservation), and the IEEE 802.1CB (Frame Replication and Elimination for Reliability) standards. The IEEE 802.1Qca protocol provides packet routing functions as well as establishes different paths and allocates resources for sending packets to the receiving network element. The IEEE 802.1CB, which essentially aggregates a number of redundancy frame transmission functions, ensures redundancy by replicating and sending packets belonging to a stream along multiple paths whilst deleting the duplicate packets near or at the point of reception. In this way, packets are delivered efficiently and reliably without any delays or congestion loss. This packet duplication and transmission along multiple paths on the network is crucial to ensuring the eventually delivery of the packet at the point of reception even in the event of the failure of a network element.
- Resource Reservation and Management: The capability to guarantee the optimal utilization and sharing of resources is one of the unique features of TSN-enabled networks. The resource reservation, resource management and robust network configuration features of TSN constitute a crucial element for accomplishing this capability and for achieving deterministic networking as well as guaranteeing ultra-high reliability and low latency. The IEEE 802 standards that are crucial to guaranteeing these TSN features include the IEEE 802.1Qat (Stream Reservation Protocol – SRP), the IEEE 802.1Qcc, the IEEE 802.1Qca, and the IEEE 802.1CS (Link-local Registration Protocol).
The IEEE 802.1Qat protocol paves the way for the allocation and reservation of network resources – such as buffers – between the sender and the receiver(s), thus providing end-to-end Quality of Service (QoS), guaranteeing latency, and achieving the desired bandwidth. The IEEE 802.1Qcc, which is an integration of a wide variety of specifications, defines management interfaces and protocols for central network configuration. It is essentially an amendment of the IEEE 802.1Qat standard, providing crucial SRP enhancements and performance improvements to meet the growing needs of new and emerging time- and safety-critical applications.
The IEEE 802.1Qca, on its part, defines the capabilities for stream reservation, including bandwidth reservation, while the IEEE 802.1CS protocol takes care of resource management tasks with the central objective of developing application protocols that distribute information across a network. Essentially, the 802.1CS standard defines methods and procedures for replicating a large registration database from one end of a link to the other end.
AVIATION AND AEROSPACE USE CASES
The prospect for the applicability of Time-Sensitive Networking in civil aviation, military aviation and the aerospace cum space sectors is becoming brighter and brighter by the day, particularly in areas such as airborne Ethernet communication networks and for related onboard Ethernet network applications regarding avionics systems, sensor systems, passenger aircraft cabin management systems, passenger aircraft flight management/control systems and the electronic flight instrument system (classed under the Aircraft Control networking Domain – ACD), on-board warning and crew alert systems (under ACD), military aircraft mission/flight control systems, aircraft-ground communication systems (under the Airline Information Services networking Domain, and in-flight entertainment and connectivity (IFEC) systems situated in the Passenger Information and Entertainment Services networking Domain (PIESD).
A further boost for this prospect, of course, consists of the security requirements as well as the safety- and time-criticality of operations in the aviation, space and aerospace sectors. For example, the emerging application of TSN technology in airborne flight management system (FMS) comes with stringent requirements regarding the precise synchronization of FMS with GPS time, although proprietary applications do not necessarily specify this requirement. To be sure, the traditional communication protocols hitherto in use in these critical sectors, such as Ethernet-based ARINC-664 and MIL-STD 1553, cannot guarantee the safety, security, reliability and timeliness attributes for which TSN has become known.
Safety and timing considerations are a primary driver of the integration of TSN-enabled Ethernet networks into the aviation and aerospace techno-operational environments. To guarantee the seamless integration of TSN into the aerospace terrain, particularly in the area of on-board Ethernet communication networks, work is already at an advanced stage for the development of an aerospace-specific TSN profile – the IEEE P802.1DP standard2 – under a cooperative IEEE P802.1DP/SAE AS6675 project involving IEEE 802.1 TSN Task Group, SAE Avionics Networks AS-1 A2, and other aerospace industry stakeholders. The project has progressed from the collection of use cases and the collection/description of requirements phases to the actual TSN profile specification stage. Frontline industry players, such as Germany-based Fraunhofer IPMS, New Jersey USA-headquartered Computer Aided Software Technologies, Inc. (CAST) and GE Aerospace, for example, are also expending tremendous energy to deliver cutting-edge TSN solutions for the aerospace industry.
The IEEE P802.1DP/SAE AS6675 aerospace-specific standard under development is being tailored towards the integration of two TSN profiles based on both synchronous and asynchronous operations with the synchronous operation focusing on the IEEE 802.1AS-2020, which offers support for multiple and redundant clocks. Although, the final goal is to develop a single-specification protocol, the standard will be focused on defining IEEE 802.1 TSN profiles for aerospace onboard IEEE 802.3 Ethernet networks hinged on critical TSN requirements and functionalities such as time synchronization (with IEEE 802.1AS), traffic shaping (with IEEE 802.1Qav/1Qbv), per-stream traffic filtering and policing (with IEEE 802.1Qci), management and configuration (with IEEE 802.1Qcc), and redundancy (with IEEE 802.1CB).
In recent times, there has been a bourgeoning of research aimed at exploring prospective use cases for TSN in aerospace. Steiner et al have explored the use of IEEE 802-enabled TSN networks as cabin backbone bus (CBS) for on-board aircraft communication, leveraging upon IEEE 802 standards such as IEEE 802.1AS, IEEE 802.1Qbv, and IEEE 802.1CB3. A seminal document4 in support of IEEE P802.1DP and SAE AS6675 has also explored the whole spectrum of potential aircraft flight network applications covering small and large civil passenger aircraft, military fixed wing aircraft, and military rotary wings aircraft. From the perspective of the space sector, Fiori et al. also proposed and investigated a lite TSN solution to support real-time services in space launcher networks5.
_________________________________________________________________
1IEEE 802.1 Task Group website: https://1.ieee802.org/tsn/
2IEEE 802.1DP. https://1.eee802.org/tsn/802.1dp/
3W. Steiner, P. Heise, and S. Schneele, “Recent IEEE 802 Developments and their Relevance for the Avionics Industry,” in Digital Avionics Systems Conference 2014, 33rd IEEE/AIAA, pp. 2A2-1 (2014).
4W. Fischer, J. Gelish, M. Hegarty, A. Jabbar, G. Lawton, B. Nelson, J. Rang, G. Scanlon, L. Santinelli, and J. Zaehring, “Aerospace TSN Use Cases, Traffic Types, and Requirements – Document in support of IEEE 802.1DP and SAE AS6675: TSN Profile for Aerospace”, Rev. 0.6 (2021).
5T. Fiori, F.G. Lavacca, F. Valente, and V. Eramo, “Proposal and Investigation of a Lite Time Sensitive Networking Solution for the Support of Real Time Services in Space Launcher Networks”, Vol.12, 2024, pp 10664-10680. https://doi.org/10.1109/ACCESS.2024.3353466.
All rights reserved. No part of this material or the content of this website may be reproduced or published in any form or by any means without the prior written permission of the publisher.
Contact: atsei@dextermarie.com
- INTERVIEWS
