DIS Integration or Disintegration

With defence in an unheralded period of uncertainty it is always pleasant to find some firm ground. To that end the notion that the use of simulation in military training will increase dramatically over the next 5 years will not cause much disagreement.

Against a training burden that has never been so intensive the MoD is faced with equipment costs increasing, availability of ammunition ¹  decreasing and the complexity of training clashing with emerging restrictions. Simulation is a critical tool in tackling these problems.

Given the scale of this challenge, coupled with the breadth (more on this later) of the UK defence industry, integrating multiple systems into a single virtual² battlefield (or Single Synthetic Environment) demands simple, understandable, interoperable and effective standards. We are not there.

Where are we?

The world of defence simulation is sufficiently opaque. This piece does not intend to add to that technical layer of fog. Rather it intends to explain the current issues as simply as possible, before offering three potential solutions.

In 2024 we are collectively spoiled. We are accustomed to the concept of “plug in and play” across our lives with HDMI cables, USB plugs, QI charging among others. So much so that one might casually assume that the same level of standardisation would be found in defence simulation hardware.

DIS³ (or the Distributed Interactive Simulation standard) was created in 1992 from work done with SimNet (created in 1987). Despite cancellation by NATO in 2010 it is still in use within the MoD. DIS’s successor HLA⁴ (High Level Architecture), formed from a blend of DIS and ALSP (Aggregate Level Simulation Protocol) in 1996, is still 1 year older than the Nintendo64 game Goldeneye, at 28.

Despite iteration, both standards are outdated and limiting. This limitation is exacerbated by the number of adaptations being made with lenient, or in some cases no, centralised oversight (despite the valiant efforts of some in the UK through the Defence Policy for Modelling and Simulation – DMASC).  Experimental Protocol Data Units (PDUs for short)⁵are network messages created by all parties to overcome the standard’s shortfalls. Their variance, lack of regulation and lack of standardisation have created a situation akin to the simulation Tower of Babel.

An unwillingness to conduct wholesale change, combined with a broader lack of understanding is now leading to the creation of an entirely unexpected problem – the launch of new products to overcome the shortfalls of DIS and HLA. Far from solving the problem by filling the gaps, this is adding to both the complexity and now the cost of new capabilities. Imagine the entire country having to use plug adaptors, just because manufacturers were unwilling to adhere to the Type G standard. Nobody would tolerate it there: so why do we tolerate it in defence?

Solution 1: Unilateral not collaborative development.

There are 3 potential solutions to this issue and these will be looked at in increasing order of feasibility.

The MoD likes collaboration and partnerships and with the technological breadth and challenges on the global stage this has it’s place. That being said, some of the biggest technological leaps of the last 20 years have been made by singular organisations headed up by empowered and focused leaders.

Nowhere would this difference in approach be more apparent than when comparing Project Purple (the 2005-2007 £120M development of the first generation iPhone) and Morpheus (the now cancelled 2017-2024 £690M component of the development of the next generation of tactical communications)⁶

Collaboration is critical to development but when that approach drifts into “design by committee”, both from MoD and industry, things go wrong.

Especially when said collaboration is not being done to ensure best in class but to prop up a British defence industry landscape that is overburdened ⁷, when adjusted to a like for like comparison with the US DoD.

One way to simplify integration would be to offer larger slices of the simulation pie to fewer companies. While this by no means advocates the monopolisation of simulation it would be simpler if the Army, Navy and Air Force had 1 or 2 simulation specialists each as would largely limit the requirement for standards to the service, or strategic, level.

In simulation, the loss of fidelity inevitably caused by standardisation matters less at the higher level, where the data is often icons on maps. Where it causes considerable frustration is when a single soldier’s equipment has simulation capability from 2+ companies interoperating poorly.

Solution 2: New Standard.

A bold approach would be the creation of an up-to-date and more prescriptive standard, designed with the premise of empowering contemporary engines (such as Unreal or Unity)⁸

This approach could rebaseline defence simulation in 2024 rather than 1996. HLA, through it’s UDIs (user defined interactions), has the same issues as DIS and is only 4 years younger. That simply isn’t good enough for this century, and it is high time for a change.

This approach could shake the current paralysis but a decision would need to be made as to whether this is done at a national (MoD) or international (NATO) level.

Real lessons could be learnt from the HDMI standardisation of 2002. Lessons such as pan-industry engagement (representing a large portion of the market), it being led by industry vice Government/Academia and regular committee meetings with strong expectations have all ensured the HDMI standard remains at the forefront of visual connectivity. All this, despite being conceived 5 years before the first iPhone was launched.

Solution 3: Adherence.

While the first two solutions would create the most dramatic change, they would both take more time than is available to benefit operational performance. As such a more tactical, and timely, approach is needed.

The core of this solution is the need for a developed and enforced common behaviour. There must be a unilateral arrangement of a multilateral agreement as to permitted/unregulated PDUs, namely those that enhance the individual simulator, such as the enhancement of avatar emotion, but do not interfere with the aggregated simulation, such as weapon effects. The MoD should be strict and industry should be candid, that the intent of all simulation capability should be to be standardised. Vacuous statements such as “DIS compatible” must become a thing of the past from industry and toothless statements like “wherever there is a clear operational benefit and Value for Money” must disappear from policy. Integration must lead to the sum being greater than the parts, or it is no integration at all.

A well-resourced, funded, and intentioned committee must be established with a frequent drumbeat and empowered to act decisively. This may need to be re-regionalised to reinvigorate much-needed dynamism, as both HLA and DIS are currently managed by the US based Simulation Interoperability Standards Organisation (SISO). The funding is critical, as ensuring PDU compliance will require otherwise unincentivised industry to be pushed to keep their software up to date. It is naïve to assume this will be done without proper motivation.

One aspect of solution 1 that could be leveraged is the fact that adherence is easier if industry have larger slices of the pie, and thus greater incentives. One way to look at this would be to differentiate products (simulators) and federations (simulations).

For example a policy that could facilitate rapid improvements could be one that single simulators, or experiences, should, where possible, be the sole responsibility of a single commerical entity. This would allow the development of proper international standardisation whilst also leveraging short term ehanced integration that is only realistic when development is owned by a single entity.

Summary

Simulation is here to stay, and is in fact destined to grow considerably. If the UK, and it’s Allies, are to leverage this to find operational differentiators there is a need for all simulators to truly integrate into centralised simulations. Army JTACs should be talking to RAF and Navy Pilots daily through simulation. Combined arms should have improved dramatically in the last decade, it hasn’t and the lack of proper standardisation is heavily to blame for this.

U-turns are challenging and are even more so in defence. As such it is often preferable to utilise the existing rules better than try to create new ones. In order to make this a reality the MoD should consider both supplier and data standardisation at the core of future acquisitions.

If this is achieved it could facilitate the federated and aggregated approach to simulation that both industry and defence want and avoid the need to create, or acquire, expensive temporary adaptors. The MoD has a limited budget, it should be spending more on hardware and simulators that enhance and improve training rather than middleware that, with some clearer thought, should not be required.

The alternative is that we continue with an experimental PDU free-for-all, with multiple users all trying to capture the exact same data in a myriad of different ways. Which, in turn, means simulators cannot talk to each other in any sensible or painless manner, leading DIS integration to become disintegration.

1. https://www.telegraph.co.uk/news/2023/03/07/dangerously-low-uk-ammunition-stocks-put-ukraines-resupply-risk/
2. https://en.wikipedia.org/wiki/Distributed_Interactive_Simulation
3. https://en.wikipedia.org/wiki/High_Level_Architecture
4. What Distributed Interactive Simulation (DIS) Protocol Data Units (PDU) Should My Australian Defence Force Simulator Have? Zaicman – DSTO-TR-1565.
5. Distributed Interactive Simulation (DIS) Protocol Data Units (PDUs, Implemented Into a Combat Model (A Case Study of the Direct Fire Module II) – Sauerborn. https://en.wikipedia.org/wiki/IPhone_(1st_generation) vs https://euro-sd.com/2023/12/major-news/35641/uk-mod-kills-morpheus-contract/
6. Morpheus: The British Army’s next Procurement Failure – Clark and Armstrong – HJS https://www.gov.uk/government/statistics/mod-trade-industry-and-contracts-2022/mod-trade-industry-and-contracts-2022
7. Achieving Interoperability Between Gaming Engines by Utilizing Open Simulation Standards – Weblin, Srinivasan and Wang – Pitch Blog page, a great source on this subject
8. Defence policy for modelling and simulation (JSP 939).