The most recent exercise in pursuit of a friendly-fire fix was Bold Quest Plus (BQ+), which was hosted by US Joint Forces Command (USJFCOM) at Eglin AFB in Florida in July 2008. BQ+ was the latest instalment of the Quest testing series, the formal goal of which is to improve coalition combat identification (CCID).
As an extension of the first Bold Quest exercise (BQ), held in September 2007, BQ+ focused on the ability of air and ground units to identify each other, although a USJFCOM website noted that ‘secondary ground-oriented test objectives are achievable based on the status of participant technical programmes’.
Emphasising intermodal air-ground CID capability is certainly understandable considering the history of fratricide accidents during the first few years of military operations in Afghanistan and Iraq by western nations. Early fratricide lowlights include the following:
- In April 2002, at Tarnak Farms near Kandahar, a USAF F-16 kills four Canadian Army soldiers
- On three occasions US Army air defence units targeted Patriot missiles against friendly aircraft. Each incident involved a different service and aircraft (a US Navy Hornet, a RAF Tornado and a USAF F-16) and in the first two incidents aircraft and aircrews were lost
- In March 2003, vehicles of the British Army’s Blues and Royals regiment take fire from USAF A-10s, which kills two soldiers
At the time of these accidents the common factor most emphasised was that US forces, and particularly USAF units, were doing most of the erroneous shooting. To a significant extent, however, this trend was a statistical artefact of disproportionate US (and air force) participation combined with specialised mission allocation practices.
Less obviously but more critically, the broader underlying problem behind these and the plurality of other contemporary incidents was the lack of interoperability not only between nations but also between services. As the Patriot incidents demonstrated, the air / ground disconnect was a two-way street.
Next-generation warfare ups the CID ante
More recent incidents illustrate that CID problems are like weeds – they pop up wherever pesticides haven’t been applied:
- In August 2007, under fire from Taliban insurgents, a British Army patrol requests close air support from USAF F-15s, which get too close killing three UK soldiers
- In September 2007 British Army units firing missiles at Taliban guerrillas overshoot their targets and hit a Danish unit on the far side of the Taliban force, killing two
- In January 2008, during a night-time patrol in bad weather, a Dutch infantry unit fires on another Dutch unit working with Afghan National Army soldiers, killing six
Clearly, these incidents differ from the earlier accidents in several respects, the most important of which is that an ongoing engagement with real enemy forces at fairly close range acted as the catalyst for the fratricide. On the bright side, the increasing relative prevalence of these events suggests that US and allied forces are not making truly stupid mistakes (i.e. those without any mitigating circumstances).
On the other hand, the nature of these incidents demonstrates that the fratricide problem will become more rather than less challenging in the future.
First, the increasing prevalence of coalition operations, both with allied nations and friendly local forces, means that CCID efforts will suffer from the mathematics of network complexity. In other words, as coalitions expand the number of CID interfaces expands factorially; for example if a coalition that expands from three nations to four (a 33% increase), the number of two-way CCID relationships increases from six to 12 (a 100% increase).
Second, preventing fratricide requires both node-to-node CID technology and collective situational awareness (SA). As battlefields evolve from linear scrums into mosaics featuring mixtures of allied, adversary and neutral entities in checkerboard patterns, traditional free-fire vectors and zones will become the exception rather than the rule. In particular, the lethal range of weaponry is increasing faster than the effective range of local CID technologies.
Third, the likelihood of protracted campaigns featuring insurgent and guerrilla tactics increases both the number and tactical ambiguity of missions, and thus raises the bar for cumulative systems reliability. In this context the common notion of the ‘two standard deviation’ accident would probably be unacceptable: over 10,000 missions, for example, a 2.5% error rate means 250 potential friendly fire incidents. A paradigm shift to something like the Six Sigma objective (less than four errors per million events) in commercial manufacturing may not be possible in warfare but it’s worth thinking about.
Finally, efforts to solve the CCID problem must not lose sight of ‘type B’ errors. If a type A error is applying firepower in error against friendly forces, a type B error is withholding firepower in ambiguous or volatile situations, only to have the adversary you’ve spared kill the friendly forces you were trying not to hit in the first place. Psychologically, the mind assigns far more weight to losses incurred than losses foregone but at the end of the day, dead is dead regardless of which bullets do the killing.
Building it is half the battle
The real challenge, however, may not be creating a solution but rather implementing it. Consider the following timeline:
In August 1991 Pentagon officials disclose that Operation Desert Storm produced several friendly-fire accidents that collectively killed 35 American servicemen and wounded 72 more. Aggregate US battle deaths totalled 148. After doing the math the US government prioritises development of new automated IFF systems.
In April 1994 two USAF F-15Cs shoot down two US Army UH-60s in northern Iraq, killing 29 servicemen in what is henceforth called the Blackhawk Incident.
In 1995 the US Army announced that the battlefield combat identification system (BCIS) has met, and in some cases exceeded, its performance requirements. Developed by Raytheon, BCIS is a dedicated query / response (QR) radio system operating at millimetre-wave frequencies, and as such, represents a watershed generational advancement over the original WWII-era IFF technologies, which used coded signals embedded in generic search radar to automatically trigger transponders on friendly aircraft.
Raytheon continued to enhance BCIS through an ongoing advanced concept technology demonstration (ACTD) programme sponsored by the Communication and Electronics Command of the US Army. As the identity of the client suggests, however, this programme is a service-specific arrangement.
In April 2003, Raytheon announced that the successor to the BCIS – the battlefield target identification device (BTID) – was ready to roll. A Raytheon press release stated that BTID proved capable of making positive identifications ‘of friendly forces vehicles in less than a second, at more than three miles distance, in all weather and environmental conditions, and with more than 98% accuracy’.
According to Pete Glikerdas, technical manager for the US Army’s coalition combat identification programme, the demonstration “validates the maturity of BTID’s design architecture and its readiness to support [the Army’s FCS programme] and other weapon platform[s].” Significantly, in light of contemporaneous inter-alliance fratricidal incidents, Army officials also noted that BTID met NATO’s combat identification performance standards.
From 2003 to 2005, the Urgent Quest programme examines the military utility of emerging CID technologies (such as, presumably, BTID). This programme emphasised cooperative target identification (CTI) capabilities permitting both ground and air units to positively identify friendly forces using local QR solutions. Nine nations participated in the final operational demonstration, Exercise Urgent Quest, which was held in the UK’s Salisbury Plain training area during September and October of 2005. Among these nations are the US, the UK and Denmark.
Because of the ‘strides made’ (i.e., problems highlighted) in Urgent Quest, the coalition of the willing extended the CCID ACTD programme into the Bold Quest series. According to USJFCOM, Bold Quest’s specific objective (within its general focus on the air / ground interface) was to ‘assess the military utility of two designated non-cooperative target identification (NCTI) technologies for coalition operations and further inform US and allied investment in the optimal combat identification capability. Bold Quest also provided the first opportunity to push [situational awareness] from ground [units] to close air support aircraft cockpits’.
CCID is a difficult problem, but why have existing technological fixes taken so long to propagate from the lab to the front lines? A report from the British government may provide a clue.
The art of (mis)management
In May 2007, a report from the Public Accounts Committee (PAC) of the House of Commons asserted that British soldiers were at risk from ‘friendly fire’ because the Ministry of Defence had mismanaged technology. According to committee chair Edward Leigh, “Over half of the programmes promising technological solutions for [identifying] friend and foe have been delayed, deferred or rescoped.” In particular, the MoD failed to implement an acquisition programme ‘despite the development of a successful [domestic] prototype in 2001’.
Among other executive blunders, the report cited the MoD’s appointment in 2004 of Air Vice Marshal Stephen Dalton as CID tsar – but with neither budgetary control nor any direct authority. Reading between the lines, the lack of progress has resulted from a handful of the ‘usual suspects’ of mismanagement, such as:
- NIH: the ‘not invented here’ syndrome
- Allowing better to be the enemy of good
- Catering to parochial service agendas
A comprehensive and enduring solution to the CCID problem could take a book, but a few principles would appear to be in order.
First, design CID capability into weapons and platforms from the get-go; don’t wait to retrofit solutions on to already built systems.
Second, impose CID standards and priorities from the top down. ‘Black boxes’ such as BTID are surprisingly compact, even after accommodating telescoping antennas for robust reliability, so finding room ought not to be an onerous problem, all things considered.
Third, build as much capability as possible into software rather than the hardware. As current events in Afghanistan and Pakistan suggest, equipment in the hands of allies today might be in enemy hands tomorrow; indeed, today’s allies themselves might become enemies tomorrow. CCID is intimately connected with cryptography anyway, so a turnkey philosophy toward hardware would not only save time, but would probably prove operationally necessary anyway.
Finally, converting information flow from ‘stovepipe’ architectures to distributed-network models would facilitate CCID, but only if everyone is on board. Given that everyone is never on board at the same level of advancement, networks should be built to accommodate back-compatibility; in other words, units shouldn’t need to run the latest version to get benefits from the system.
The foregoing principles seem simple enough. As Clausewitz once said, though, “in war, everything is simple, but even the simplest things are difficult.” Let’s hope that the front office doesn’t turn difficulty into impossibility.