As the initial headline-grabbing moment of the AUKUS announcement has faded, the discussion in Australia has transitioned to the nitty ^[1] gritty ^[2] of getting nuclear submarines, in parallel with the pressing work of the nuclear-powered submarine taskforce ^[3]. ‘Advanced capabilities’ ^[4]—an all-pervading catch-all term—have received less attention. The defence strategic review ^[5] will have to, among other things, carefully unpick their significance.

Truly integrated command-and-control systems are one priority that might be considered in the various assessment processes now underway.

The Australian Defence Force is having a parallel discussion to the US military and other allied forces about future operating concepts. It has approved a concept for multi-domain strike ^[6], in which ‘effects, enablers and information from the five domains of land, air, maritime, space and cyber will be synchronised’. The individual services are going about related work, with much discussion about how the army might integrate long-range missiles ^[7], for example.

Clearly, the future operating visions of the ADF and its allies are predicated in large part on being able to realise a next-generation form of command and control, of which cloud ^[8] and edge ^[9] computing are likely a necessary component.

First to developments in the US. In 2021, the Department of Defense cancelled ^[10] the US$10 billion Joint Enterprise Defense Infrastructure project, known as JEDI. At the same time, it announced that the cloud computing services sought by JEDI were still required, and launched the JWCC ^[11], the Joint Warfighting Cloud Capability. A departmental spokesperson at the time summed up the ambition of the new project as an ‘enterprise cloud spanning the entire department in all three security levels with availability from CONUS [the continental United States] to the tactical edge, at scale’.

US efforts on this front fall under the moniker JADC2 ^[12] (joint all-domain command and control), a ‘concept to connect sensors from all of the military services … into a single network’. As the Congressional Research Service explains ^[12]:

Traditionally, each of the military services developed its own tactical network that was incompatible with those of other services … JADC2 envisions providing a cloud-like environment for the Joint force to share intelligence, surveillance, and reconnaissance data, transmitting across many communications networks, to enable faster decisionmaking.

The JWCC project is a major part of how the US military aims to achieve this goal, but there’s a gamut of programs nested with this capstone concept. Key programs appear to be single-service lines of effort, among them the US Air Force’s Advanced Battle Management System (ABMS), the US Army’s Project Convergence and the US Navy’s Project Overmatch. Making sense of this array of work from a non-technical and open-source perspective is difficult to say the least.

Publicly acknowledged exercises paint a picture of what this means in tactical terms. A December 2019 exercise in Florida reportedly ^[12]:

represented the first demonstration of ABMS. Air Force and Navy aircraft (including F-22 and F-35 fighter jets), a Navy destroyer, an Army Sentinel radar system, a mobile artillery system, plus commercial space and ground sensors demonstrated being able to collect, analyze, and share data in real-time to provide a fuller picture of the operating environment.

In simplistic terms, we are talking about mature, tactical applications of cloud ^[8] and edge ^[9] computing. These are critical parts of the next-generation command-and-control architecture, necessary to fully realise the US’s aspirational operating concepts like multi-domain operations ^[13] and distributed maritime operations ^[14]. The common themes linking these operational concepts are that they respond in large part to so-called anti-access/area-denial threats in the Western Pacific and attempt to process high volumes of sensor data at speed. They demand that this information be seamlessly linked to various ‘shooters’, across all domains and dispersed over great distances.

The authors of one paper ^[15] offer a useful analogy with civil mobile phone networks. Imagine a fighter jet or naval platform as a mobile phone. Each logs on to a network and accesses various services over that network. At different points in space and time, the level of functionality varies. A user might shift from their home internet connection to a rural area with poor reception; an aircraft or ship might move from a part of the battlespace with very good, uncontested connectivity to one in which connectivity is heavily degraded. That network, agnostic about what device or weapons platform is connecting to it, ‘will automatically push and pull mission-relevant and timely information to assist the user’.

Like with mobile phone networks, the major problems are network strength and bandwidth. Buffering might be a mild inconvenience when you’re watching a Netflix series, but on the battlefield it is deadly. Solutions, such as sharing data with satellites, just add additional links between sensor, transmitter and recipient. Because poor network strength can be a feature of any communication channel, the proper vision for interconnectedness is not any sensor to any shooter, but some sensors to some shooters, some of the time ^[16]. Optimal choices about what computing is done ‘at the edge’ with the battlefield user and what is communicated up the chain are key, and hugely complicated by the large volume of sensor data.

These are no longer frontier technologies in any general sense; they have been widely adopted across civil and national security applications. The technical and organisational challenges to realising the military command-and-control applications of these technologies remain significant despite their being old news in many ways. A force made up of traditionally designed platforms with sporadically interoperable command-and-control technology, like radios, is one. Many of these key systems don’t have much room for change. An F-35, for example, comes with a set of software tightly coupled with its hardware, and there isn’t spare space for adding new systems.

Getting the security arrangements right is another key challenge. Excessive restrictions militate against the very data sharing that cloud and edge technologies aims to realise, but insufficient caution might expose a veritable treasure trove of data and communications. This difficulty is more acute still in multinational environments, and full realisation of the potential in many of these technologies also presents a range of potential cyber vulnerabilities.

The ADF has projects in progress related to these advanced command-and-control capabilities. For example, Joint Project 2289 ^[17] is about delivering ‘a single information environment for the joint force’. One of Defence Science and Technology Group’s ‘STaR Shots ^[18]’ research priorities is ‘agile command and control’ ^[19]. Given the technical and no doubt in places sensitive nature of these projects, it’s tough to get a good read on their status and extent, but they are happening. ASPI’s Ulas Yildirim ^[20] has recently observed that Australia and other allies are already participating in US efforts on this front.

It’s also important to acknowledge differences of scale and resourcing. I certainly don’t want to suggest that the ADF should be doing all the same things in this area that the US military behemoth is doing. Yildirim’s emphasis on prudently picking and choosing rather than replicating US scale is well placed. Getting the balance right between ambitious integration and buying existing capability quickly is another ongoing friction point in this area.

Notwithstanding these caveats, there is clearly space for a little more informed public discussion of these command-and-control capabilities and whether there might be opportunities presented by AUKUS and the forthcoming defence review. There needs to be substantial discussion of compatibility with our key partners as well as the associated standards and sharing arrangements. And what about the long-term cost and capacity issues related to data storage for these systems? Perhaps there are ways to make both data collection and storage complementary rather than duplicated, for example. These opportunities might well be more consequential than any individual weapons platform.