Air combat and stealth


For a long time, visual detection was the only type of detection possible. But in World War II, two significant advances appeared: radar, as well as radar- and IR- -guided missiles. Until 1970s, these were defeated through jamming and decoys. Early German attempts at building LO aircraft – Ho-229 – never got anywhere, albeit their RAM paint utilization at snorkels and aircraft was noted. In United States, first attempt at reducing the radar signature of aircraft was on U-2, by utilizing RAM paint, but it was not very successful. First actual stealth aircraft appeared in early 1960s – SR-71 Blackbird, which utilized shaping such as canted surfaces to reduce radar signature. In 1970s a second generation of stealth aircraft appeared with B-1A, and also began a programme of development of VLO aircraft. Result of that was diamond-shaped F-117, to soon be followed by B-2. All these aircraft successfully performed against enemy air defenses, but in the case of B-1A and later aircraft, their performance against air defenses was similar or identical to performance of conventional aircraft they were deployed alongside. Fourth generation of stealth aircraft are F-22, F-35, PAK FA, J-20 and J-31. While still stealth aircraft, they sacrifice stealth characteristics for the sake of better flight characteristics, allowing them to match conventional fighters in terms of maneuverability. However, their stealth requirements make them larger and heavier than comparable conventional aircraft, thus sacrificing kinetic performance for the sake of stealth.^1

Limitations of stealth

Main areas where aircraft’s signature have to be reduced are radar reflection, IR emissions, visual signature and electronic emissions. Radar reflection (RCS) reduction requires careful shaping of the aircraft, so that the return signal becomes weak enough to be lost in the electromagnetic noise.^2 However, a shape that is conductive to low radar signature does not correspond well to aerodynamic configuration optimized for flight performance, even though many problems had been solved through fly-by-wire systems. As a result, low-observable air superiority or multi-role fighters have reduced maneuverability compared to what would have been possible without stealth optimization. RAM coating and internal weapons bays are the main reason for this; RAM coating has to be of certain thickness to work against radar (0,6-1 cm for X-band), and weapons bays increase aircraft cross-section and complexity.

Aircraft shaping and RAM are effective against centimetre band radars. However, radars that work in metre-band area (such as early warning radars, and other air defense search radars) can cause resonance in certain parts of the aircraft – stabilizers, wing leading edges – and thus receive strong radar reflections. Further, modern aircraft have many maintenance panels, doors and access points, which are ideal radar reflectors, which means that extreme tolerances have to be observed during the production. This in turn raises price of both production, and later of logistical support.

Main sources of IR radiation in aircraft are engine emissions and airframe heating during the flight, which produce IR radiation of medium and long wavelengths. These are the same bands as used by modern dual-band IRST systems. Skyward G can pick up targets flying at 300-400 knots solely through skin heating. Skin heating cannot be significantly reduced, and neither can heating of air due to compression (especially noticeable at supersonic speeds). IR signature of engines can be significantly reduced, but only by sacrificing engine performance – flat nozzles lead to thrust loss in 14-17% bracket, meaning that F-22s nominal TWR of 1,35 might be less than 1,16. In F-22 itself, usage of thrust vectoring, also intended to reduce radar signature by allowing more precise control of aircraft during the flight as well as reduced need for aerodynamic control surface movement, has meant increasing the weight of the nozzle, greater maintenance requirements, reduced lifetime, and increased rear aspect RCS.


When it comes to visual detection, stealth aircraft are at disadvantage because they tend to be larger than comparable conventional fighters – even though extreme dimensions may be the same, stealth aircraft’s internal carriage requirement results in “fatter” body of the aircraft. This factor is typically underestimated, despite the proliferation of optical and electrooptical systems in both aircraft and surface units (even with IRST, at high altitude physical size is also a factor in detection, since modern IRSTs can detect temperature gradients of less than 1* Celsius, and can in fact detect fighter aircraft merely through sun glint from the canopy glass).

Radar is still used as a main sensor by air forces despite its tactical disadvantages, because it offers longer theoretical detection range and better bad-weather performance than IRST (even though, in reality, a fighter using radar would be detected first). Stealth aircraft, which have to remain completely passive to prevent being discovered, have to rely on either offboard sensors or on their own passive sensor suites. However, due to sensitivity of uplinks to jamming as well as vulnerability of AWACS to long-range missile systems such as Ks-172 and Meteor, AWACS is not a good solution.


Appearance of stealth fighters caused a necessity for defining new tactical approaches. In the last few decades of the 20th century, radar had become the main sensor of a fighter aircraft, for both long-range combat with radar guided missiles and for gunlaying in visual-range dogfight. Radar is also the main sensor for control of airspace by ground or air based radar platforms. Appearance of LO aircraft means reduction in radar detection range, which reduces the ability to cover the whole defended area with radar surveillance. This leads to appearance of “passages” that allow strikes against targets deep within the defended territory. Likewise, low-level strike aircraft also find themselves in danger from surprise attacks. However, in either case, stealth should be combined with supercruise to fully exploit its potential.

Stealth aircraft vs conventional aircraft

Stealth aircraft will be primarily used to disable high-value ground systems, through usage of stealth ground attack aircraft such as F-35 and various UCAVs in the future. This way, risk would be reduced for conventional aircraft, such as CAS and conventional strike aircraft. To defend against this, it is possible to analyze enemy forces (tactical and technical capabilities of stealth aircraft, force organization etc.) to determine likely targets and flight corridores. This can then be combined with deployment of sensory systems optimized for detection of stealth aircraft, such as HF and VHF band radars as well as ground and airborne IR systems. Since stealth aircraft can only be stealthy in a single band, combination of sensors optimized for various frequency bands can provide an effective countermeasure.

Gaps can also be covered with fighter aircraft. In that case, there are two options. One is usage of small, fast cruising single-engined aircraft equipped with passive sensors. These can then be combined with other systems or used individually to search for stealth aircraft, since IR signature cannot be significantly reduced. Second one is usage of large aircraft with powerful AESA radars (e.g. F-15, Typhoon, Su-35). These would fly in a widely separated formations, using datalinks to exchange and fuse sensory data. By comparing the data from different aircraft, false returns could be eliminated, significantly lowering the detection treshold and allowing early detection of a stealth aircraft, which would then be attacked by the closest fighter. This is also important since stealth fighters’ RCS is lowest nose-on; from sides and rear it is significantly higher, and even more so if a fighter using radar is not at the same altitude as the stealth aircraft. By using datalinks, such groups could form a flying multistatic radar system, improving the chances of detecting stealth aircraft. However, since twin-engined fighters with large radars tend to be expensive, best option would be combining them with smaller single-engined fighters such as ones described (example here), with twin-engined fighters serving as command aircraft for groups of fighters. In this case, taking out command aircraft would affect effectiveness of a group, but that could be mitigated by using first approach. Also, since stealth aircraft are built around the notion of radar-based air combat, usage of radar jammers, decoys or jammer-decoys would reduce their effectiveness by forcing them to come close to utilize the radar, as it would need to burn through the jamming; this would increase probability of detection. Effective enough countermeasures – such as DRFM jamming which jams active radar missiles and even fighters’ AESA radars – might even force visual-range combat, where stealth fighters are at disadvantage due to their larger size and inferior kinematic performance.^1 While this problem can be relatively easily countered by installing IRST systems, and relying on them for beyond-visual-range combat, this negates and even reverses the advantage that radar stealth offers to LO aircraft.


Combat between stealth fighters

When both sides have stealth fighters, detection in air combat will rely solely on TV and IR systems, in which case missile launches can provide significant cue as to the opponent’s position. When penetrating enemy’s air defense, stealth fighters could launch small decoys such as US TALD in order to force the enemy’s reaction and discover his stealth fighters. Also, air-to-air missiles can only carry small radars, which can only be solved through two possibilities. One is semi-active radar homing, where aircraft’s own radar sends out emissions whose reflection is caught by the missile. However, this necessitates the fighter itself to give away its position by constantly radiating, and also makes it vulnerable to anti-radiation missiles. Further, due to stealth fighters not having “swashplate” radars, they would need to continually close in onto the target, risking their destruction by IR missiles (this happened in AIMVAL/ACEVAL tests, where F-14/F-15 would get destroyed by F-5s; because they had to constantly illuminate the target, F-5 could close in to small distance and launch fire-and-forget IR missile before being destroyed). One way out of the problem is to combine radar and IR seeker in the missile; other possibility is to rely only on IR missile, with guidance being done via fighter-missile data link.

If radar missiles are made useless by stealth, jamming, or combination of two, and neither side has a combination of QWIP IRST and IR BVRAAM (example of that is Rafale’s OSF + MICA IR), both sides will have to engage in close-range maneuvering combat (“dogfight”). In this case, victory comes down to pilot training, numerical balance of forces, and technical factors such as aircraft maneuverability, agility, endurance, usage of HMD and HOBS missiles, and eventually DIRCM. Further, in such a case stealth fighters will only be able to detect each other from short range, as will other units not equipped with IRST. If no air units have IRST, stealth fighters will have problems finding each other and will go after highly visible targets – AWACS, tankers and air bases, which will force enemy stealth fighters to defend them. This will largely negate the advantage of stealth, as it will result in either mutual destruction of support assets, or close-range dogfight. Designers have accounted for this, and all stealth fighters that are actually intended to fight other aircraft are highly maneuverable.

Strategic considerations

Main problem with stealth aicraft is that they are expensive. USAF has a fleet of 186 F-22 fighters, of which only 123 are combat-coded, with additional 20 being backup aircraft inventory machines, and remaining being test and training assets. With availability rate of 63% and sortie rate of one sortie every two days, this means that the entire F-22 fleet can generate 40 sorties per day. Regardless of how good F-22 may be – and that too is up in the air – one aircraft can only be in one place at one time. USAF has only six understrength F-22 squadrons in operational status. While F-15/16 units have 24 primary authorized aircraft and 2 backup inventory jets, F-22 units have 21 PAA aircraft plus 2 BAI machines; the exception is Air National Guard’s sole F-22 squadron at Hickam AFB at Hawaii, which has 18 PAA and 2 BAI aircraft. But even to achieve that, USAF had to cut its test and training force to the bone. Pilots of Air Force Weapons School at Nellis AFB have to share their 13 F-22s with the 53rd Test and Evaluation Group; two squadrons sharing half-a-squadron worth of aircraft between them.


Overall, it is not impossible to create a stealth air superiority fighter, as demonstrated already by F-22, PAK FA and J-20. However, doing so is not worth the money invested, and countries that do invest money in such projects are generally doing so as statements of a prestige. Stealth aircraft are a fad, much like battleships used to be, and wasting money on such ego trip projects may display one’s power and technological prowess, but is ultimately counterproductive. Even United States cannot afford the expenses of development, deployment and support of stealth fighters, but continue to do so at great cost (failling infrastructure included), simply to show off their power, and provide profit for overly influential corporations.


1^F-15C has similar dimensions to F-22, yet F-22 is 55% heavier (19.700 vs 12.700 kg). F-35 also has similar dimensions to F-16, yet it is 54% heavier (13.199 vs 8.573 kg). Su-35 has significantly larger dimensions than PAK FA (21,9×15,3×5,9 vs 19,8×13,95×4,74 m), yet is 500 kg lighter.

2^This is also why LPI techniques do not work (for long). Only about 1% of the emissions which reach the target aircraft get reflected towards the source. If, by chance, a technique could be discovered to near-perfectly reliably separate radar’s return emissions from the background noise, radars would indeed become almost undetectable to RWR, but such an advancement would also make any RCS reductions in aircraft meaningless as any return signal – no matter how meagre – would be correctly identified as belonging to radar itself.

Further reading

Hrvatski Vojnik, Broj 8., Godina VI, Veljača 1996. (Croatian Soldier, No.8, Year VI., February 1996.)

Airborne IRST properties and performance

Value of stealth aircraft

Stealth – evolution of justification

20 thoughts on “Air combat and stealth

  1. “Overall, it is not impossible to create a stealth air superiority fighter, as demonstrated already by F-22, PAK FA and J-20.”

    So far the J-20 has not displayed any super-maneuverability traits and it’s Air-Show performances have been underwhelming demonstrating inferior maneuverability to F-16 let alone F-22 and Rafales. It’s seems to be optimized for range and super cruise so most analysts agree J-20 is not an air superiority fighter but an old-fashioned long range, long loiter area defense interceptor with some stealth. For Air superiority the Chinese will probably really on their Flankers and J-10C’s (which is the closest I’ve seen so far to your FLX ) deployed probably similar to the way you suggested. The mission of the J-20 during day one of a war would be take-off early with its four massive drop tanks, move to a remote waiting area were it would drop the tanks and stealthy loiter expecting the first wave of US Stealth fighters to pass them. They would then have free reign to engage support assets such as tankers and AWACS. Its main weapon would probably be a long range AA missile that China just unveiled. The Chinese also developed a very interesting system to hold the missiles in the bay which allows for the missile to be pushed into the air-stream and remain attached to the aircraft while the weapon bays close behind it, which is a very elegant solution that allows for the missile to search for targets before launch, without incurring a drag penalty by keeping the bay doors open, the way the F-22 does. Here is an article explaining the mechanism: . If this tactics succeeds it does not matter if US F-22 and F-35 get a 10 to 1 kill ration on day one against Chinese fighters. Without tankers and AWACS in the Pacific they would be blind and stranded on return and USAF will probably have to send them to remote airstrips, scattered on islands, behind the advance front of the J-20s and hope they can refuel them with MC-130. The fact that USAF practiced just such a scenario the other days means they are actually aware of the high probability such a thing could happen:

    PAK-FA on the other hand seems to be a air-superiority fighter, but because the Russians are developing new long range anti-radiation AA missiles for it, it would probably be used initially in the same way. Still conditions over Europe are not the same as over the Western Pacific. Ground troops would be much closer and there would be much more alternate landing sites and forward locations available. So such a tactic would not have the same efficiency as over the Pacific and PAK-FA would eventually have to engage other fighters hence it’s optimization for both maneuverability and, range and speed.

    Liked by 1 person

    1. It is probably meant to stop enemy AWACs and perhaps as a counter for the B1 series of bombers.

      Alternatively, it could be a Chinese tactical bomber.


      1. More likeally it’s secondary mission is Aircraft Carrier hunting. On day two of the war if the day one scenario I presented above is successful it would be one of the multiple forms of attack used against US Carrier Strike Groups. J-10C and Flankers would launch Anti-shipping missiles in a frontal assault. Ground forces would launch the anti-shipping ballistic missiles that the Chinese have been developing. And the J-20 would again stealthily sneak using it’s huge range and super-cruise behind the US “front” attacking the Carrier Strike groups with Anti-ship cruise missiles from a completely different angle and would offer final guidance for the ballistic anti-ship missiles.

        If these scenarios are successful on day three, with its Air Force neutered on day one and its Navy on day two, the United States would either have to sue for peace or start along attrition war in which in the begging it could really only on its submarine force.


  2. Now that I think about it, what would be the best combination, a FLX 2 seater as a command aircraft and a lot of FLX or a Su-27 derived aircraft that doubled up as a command aircraft due to size?

    One option is to delete the radar and put a huge IRST aperture in, allowing very long range detection.


    1. Actually, I’d probably use F-15 or F-22 derivative with IRST and additional radar apertures as a command aircraft, and F-22 version would be able to serve as AWACS hunter as well.

      Huge IRST aperture is not realistic, you still need nose cone for performance unless you want a subsonic aircraft.


      1. Actually better then one IRST with a huge aperture, which has the same disadvantages as a radar antenna imposing size requirements of the air-frame, a better approach would be to use multiple standard size IRSTs. In such a way that in every direction a least 2 IRST can be used to scan. The advantages are multiple:
        1) Interferometry works just as well in Infrared as in Radio Frequency domain, so the signal composed from two IRSTs is the same as the signal fro one as big as the two put together. So one gets the same resolution and detection range as the large aperture IRST without the associated size requirements.
        2) Much more easily achieved passive telemetry. On big aperture IRST can determine range by varying the aperture. This requires a complicated internal mechanism of the IRST that increases mechanical complexity and precludes the usage of all-staring arrays that have a large field of vision. With two IRST one can use binocular effect to measure range
        3) Redundancy. This is self explanatory. If ones huge IRST gets broken one looses all it’s capabilities. If one of 2 smaller IRSTs get broken one maintains degraded capabilities: lower resolutions, lower range, less efficient passive telemetry (it would have to use shape recognition)

        I can think of only one disadvantage:
        Increased initial cost. Two or more smaller IRSTs, giving together the same resolution as on big one, will cost more then one bigger IRST no matter how complicated the bigger one is. But this disadvantage will be offset over the life of the aircraft with lower maintenance costs for the smaller IRSTs and higher reliability (no complicated aperture control mechanism)


      2. One in the nose covering 180 degrees front (I think they got to 180 degres – the DDM-NG uses only two sensors and offers almost spherical coverage the only blind spot is directly bellow, and the DDM-NG is basically as powerful as the IR component of the OSF ). One in each wingtip covering 180 degrees left and right and one on the top of the tail covering 180 degrees back. One looses the ability to use tip missiles but alternatives could be found like lets say a missile pod caring one or multiple IR Missile directly above of bellow the wing tip designed in such a way as to behave like a wingtip missile.


      3. In FLX I used four UV MAWS and three standard IRSTs. But using DDM-NG, with 180* instead of 90* FOV, you can place them on tail fin, plus two on outermost underwing rails; no need to sacrifice missile carriage. But you still need classical IRST as wide-FOV MAWS cannot have the range of a telescopic IRST.


  3. I’d advise the following for the command aircraft then:

    Right below front canopy (as normal)
    Two near the wing tips
    Maybe one on the vertical stabilizer

    Those 4 should be able to give excellent resolution when combined together for rangefinding. You could combine with a laser rangefinder for dogfighting if you wanted (for preparing a firing solution). There will be a penalty in drag and mass, but it won’t be anywhere near say, a 20Kw radar (that’s how powerful the Su-35’s radar is).

    For my command type aircraft, I’d go with something about the size of an Su-27. It would be a large canard tailless delta.

    No radar
    Go with 1 vertical stabilizer and engines put a bit closer together (will need to relocate parachute); trying to minimize parasitic drag here (already enough with 2 engines). If there is pod armor, maybe a bit to split the engines apart
    You’d want 2 classic IRST at the rear too – a second one facing the rear on the vertical stabilizer and another near the fuselage’s bottom (with 2 IRST it is possible to get binocular vision and very hard to “sneak” up for enemies
    The fuel fraction would be quite high with the delta wing and even with the extra IRSTs, no radar

    There would be 2 basic variants
    – Command – 2 seater that doubles up as a trainer
    – Bomber Interceptor – 1 seater (could use the extra seat space for a more powerful gun or more fuel)

    It would complement in small numbers an FLX type of fighter.


  4. “But you still need classical IRST as wide-FOV MAWS cannot have the range of a telescopic IRST.”

    Thru interferometry one might get as good a range thru brute force: the resolution would be so good as to offset the lack of telescopic lenses. But in this case one would need a very good signal processing software. The advantage again would be lower maintenance and initial cost as the lack of telescopic lenses would reduce both.

    Another option would be to cover all angles with multiple cheap all-staring IR and UV MAWS such that every direction has at least two IR and UV sensors (practically the disposition I gave above but with both IR and UV sensors). This would allow multi spectral detection, would increase resolution by fusing signal from four sensors and two sensor types and would offer pretty good range as well as binocular vision for ranging. Added to this would be two smaller telescopic IRST mounted on gimbals such that they can both be pointed forward so that they give binocular vision for 60 degrees front and each one independently could be pointed left and right one covering 180 degrees to port the other to starboard. When pointed sideways the signal from one IRST would be fused with signals coming from the MAWS on that angle thus offering increased resolution and telescopic vision. This configuration would offer great detection range forward (2 IRST +4 MAWS) and somewhat smaller but still significantly larger detection on all other angles (1 IRST + 4 MAWS ). All this would be prerequisite on using cheap sensors both MAWS and IRST and fusing their data together such that one obtains the same resolution, range and detection capabilities as fewer but more expensive sensors at equal and smaller cost and with the advantage of smaller maintenance cost and smaller reliability issues.


  5. J-20’s canards are in the same plane as the plane’s wings meaning that the lift isn’t additive and the wings are biting into clear air.


    1. If I remember correctly, studies on canard position have shown that even coplanar canards increase wing lift, albeit much less than close-coupled high canards. Been a while since I’ve read them, though.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s