The AIM-54 Phoenix missile - Technology, History and Performance

CW or HPRF + FM modulation allows both resolution in range and radial speed so targets flying with the same speed and at different ranges will be separated by such seeker.

It also increases target return - to - MBC ratio due the MBC spot size in range.

But it also increases MBC clutter area in the frequency scale due the MBC spot size in range. For example a tail-on target (with radial speed smaller than the minimal MBC radial speed) at a range larger than the MBC min range appear in the MBC clutter because frequenct = speed + range. Without FM this target will appear in MBC-free region.

There is no obvious advantage of having FM in order to see through the notch.

I have already explained this in the topic about Skyflash missile.

2 Likes

Do we have any sources/data even saying what prf it uses? (The missile’s on board radar) It seems like they went through a lot of trouble developing ECCM techniques for it if it is only effective in tracking head-on targets.

This is a bad example

Hf = 10 km
Ht = 5 km
Dh = 40 km - horizontal
D = 40.3 km - slant

Elevation to the target is atan((5 km - 10 km) / 40 km) = -7 deg

Beam width for AIM-7 seeker is 7 deg * 2 = 14 deg - this is why 7 deg is here

MBC starts in range from 10 km / sin (7 + 14 / 2 deg) = 41.1 km to 10 km / sin(7 - 14 / 2 deg) = infinity

Only 41.1km - 40.3 km = 0.8 km separation between target and MBC in range.

But this is not the worst news.

If HPRF is used that means for example 2 km unambiguous range and range gate 0.2-0.4 km will fit all returns from 2.2-2.4 km, 4.2-4.4 km, e.t.c. The range gate will receive all returns from MBC behind the airplane becase MBC is longer than 2 km.

2 Likes

If HPRF+FM is used that means for example frequency shift for 200 m in range is the same as for to 2 m/s in radial speed.

Let’s assume that fighter speed is 300 m/s.
MBC starts in radial speed from 300 m/s * cos(7 + 14 / 2 deg) = 290 m/s to 300 m/s * cos(7 - 14 / 2 deg) = 300 m/s.
Target radial speed is 300 ms * cos(7 deg) = 298 m/s.
Without FM target radial speed is in MBC area.

Target frequency shift is equals to 298 m/s + 40300 m / (200m / 2 m/s) = 298 m/s + 4030 m/s = 4328 m/s
MBC starts from 290 m/s + 41100 m / (200m / 2 m/s) = 4400 m/s to infinity

They are separated well by 72 m/s - this is wider than normal speed gate width.

Ok

Now assume that the notching target with speed of 300 m/s turns slightly towards to the fighter. The angle changes from 90 deg to 75 deg.

Target radial speed changes from 298 m/s to (300 m/s + 300 m/s * sin(90 - 75 deg)) * cos(7 deg) = 375 m/s.

The frequency shift changes from 4328 m/s to 375 m/s + 40300 m / (200m / 2 m/s) = 4405 m/s. The target now doesn’t notch, but target return appears within the notch area from 4400 m/s to infinity.

Yes, spectrum density of MBC with FM is smaller, but if it was 1000 times stronger than the target return w/o FM and becomes 1000 times better because MBC gets 1000 times wider - it still not strong enough, but MBC is much wider. Doesn’t look a good deal.

2 Likes

I know what you mean, but you are forgetting range in the doppler range map. MBC spot size in F increases but it’s center is on the (large) range gate(s) created where the center of the beam is. As you get farther from it, in the noise is reduced and with enough SNR, a traget can be found.

Look at this map. Range is 0 to N range bins(created by the FM) and doppler from 0 to M Doppler bina
Screenshot_20230920_033740

You say spot size increases, I agree. But on the range axis the MLC ia centered in the beam center at range 5th bin. If the target is getting tracked and is beaming, near the ground. The target will be in the 4th range bin, its signal return will be lower than the clutter. Now lets go on my first example. The target is on the 3rd range bin. If the side aspect RCS is big enough and SNR is sufficient track can be kept.

Why is the beamwidth 14 degrees? I’m not getting this or misunderstanding what you want to say here.
Of course if the BW is as large as 14° the area illuminated will offer a larger return bandwidth.

Sin(7-7) is infinity, I don’t get your math or notation. But if I’m understanding what you say, a 14° beam centered at -7° in elevation will look from angle (-14° to 0°[horizon]). If its for example ok.

Of course you get a an integrated look area of infinity. But the signal return is not infinity. There’s a point the noise is bliss. And simply ia ignored due to the SNR.
On the doppler range map MBC at f0 is from the altitude return to infinity. But the intensity is not uniform but centered on the beam. Even if you might argue that the MBC is not on the horizon( which is what I think you want to point out), the sidelobes do and they don’t get infinity return because the intersected area goes to infinity.

Think of which ground return is stronger, that at 40km or the sum from near horizon to some x distance the signal isn’t even picked by the receiver!

When you range gate HPRF, you take into account the range bins created by the FM. Look at the doppler map again, if we use these (FM) range bins, the target(if enough SNR) will pop out between the rangebin(s) the MBC is and 0.

Let me answer the other post later, is 4h30am here

1 Like

Radial Speed, is not the only discriminator, that would be relevant. If sufficiently separated by distance, a similar RCS targets will reflect less power and so the resulting convolved guidance commands would favor the closer target with the falloff aligning with the reflected power formula.

and if separated in angle the conical scan seeker will limit the duty cycle(due to nutation of the scan)of the off axis target, on top of observed power losses due to HPBW of the illuminator (the target would need to be ~84% the range of the referenced target at the -3db angle)

The AIM-7 and it’s derivitives should be able to due to the use of an FM-CW waveform, so range is additionally recoverable so could be further used to filter out non target returns, and arm the warhead when close to intercept.

1 Like

training was good , it was just that they were using old soviet school technics

You can tell that this doesn’t really work well by taking the F-14 into Air Assault and trying to use the AIM-54 with TWS on targets there.

maverick will work better

it really does which is so weird, it makes me wonder what and AGM-65 armed F-14 would have been like.

As I mentioned previously Skyflash uses purely unmodulated CW (so much so that the radar requires a special unit to remove any modulation present in the CW emission):

It is able to differentiate very close targets using a phase lock loop (and also has excellent clutter resistance)

4 Likes

The problem is this functionality is found on basically all ARM’s. If there is coding limitations or the dev’s aren’t sure how it works then with AIM-120 and R-77 round the corner that may present issues. I don’t see a reason to withhold it from Phoenix, especially if it gets correct G-Pull.

3 Likes

If few targets are within the radar main beam and they are not separated in range or speed the radar tracks the average weighted (based on return signal strength) direction to these targets. In WT this have worked for all radars, radar seekers and even IR seekers (with FoV instead radar beam) since their introduction.
Each target return signal strength depends on RCS, range and its position within the radar main beam.

If one target is slightly closer to the antenna than another, but it is further away from the main beam centerline than the another one, it may happen that return signal from this more distant target is stronger and the radar will track direction closer to this target than to the closer one. Until the range decreses and makes range ratio between them more favourable to the closer target.

Yes, there is no such thing as side aspect RCS in the game. RCS is more or less the same from all aspects.

Are there plans to improve this?

1 Like

Sorting solely on return strength is not a universal characteristic of missiles though take for example the AIM-9C a 2nd target could be illuminated with any level of power but it would be filtered by the range / range rate gate and so not be a factor in the guidance of the missile unless the distance was very close to the tracked target (at most 300 meters resolution based off the mention of a microsecond).

So the current implementation doesn’t account for the reduction in received energy due to the nutation of a conical scan seeker, centered on the reference target impacting the duty cycle of a sufficiently off axis target, or the Seeker’s Fov / sidelobes?

I guess this does also explain the performance of rosette scanning type IR seekers.

Target may be found or may be not found. It strongly depends on MBC clutter strength.

The picture illustrates how BMC expands if FM ranging is used. SNR gets better, but if it still not good enough FM ranging makes the problem even worser.
HPRF search radars in RWS usually apply small depth of FM, which gives them very bad precision and resolution in range, but doesn’t make MBC such wide. They also use unmodulated HPRF and than HPRF with ascending and descending FM to find out both speed and range of each target.

Skyflash seeker still being incorrectly modelled after all this time makes me tired

I mean, at this point its largely pointless in even bothering to argue. We KNOW radar missiles are underperforming in-game, particularly in the low altitude intercept capacity, with multiple sources stating improvements in low latitude target intercepts for things like the AIM-7M (which I’ve submitted a bug report on regarding its actual min altitude of 5m irl, while in-game its min altitude is 95m for the missile to have little enough error vs a non-maneuvering target to pass within the proxy fuse range).

There are some rather basic and clearly problematic modelling issues with the AIM-54C. The smokeless motor being omitted when it was not only added to the new AIM-9M, but retroactively added to the AGM-65D is tantamount to spitting in the F-14 communities face, the admittance that the modelling decision behind the 17G max pull on the AIM-54C is intentional is another egregious situation, both of which would have a major positive affect on the missile.

The low altitude intercept problem is just a widespread modelling problem for radar missiles which at this point is just outright horrendous, particularly when we’ve entered an era of high performance and flare resistant IR missiles for only a handful of planes putting all other planes at a notable disadvantage in close range without any real way to provide effective spacing using radar missiles as a credible threat as multipath is HIGHLY abusable in WT, not only due to how high the multipath effect is, but also from the simple reality that 3rd person flying makes abusing ground hugging tactics easy, safe and reliable. Contrails are also modelled excessively, making abusing spotting mechanics by low altitude flying even more advantageous.

Of this list:

for which @k_stepanovich argues almost all points of, 3 are substantial issues well within the ability of the devs to correct with little work (max G-load, low alt intercept, smokeless motor). Arguing NCTR capability, beam target tracking in lock down scenarios, stream raid ability, etc… although interesting, and possibly helpful are all things that could/would take more time to fix and are more debatable.

Fixing simple issues such as max G load and the lack of smokeless motor should be prioritized as first steps for the AIM-54C specifically, with low altitude intercept ability of late radar missiles using PD/monopulse seekers being something that should just be fixed across the board at this point. The AIM-7F has been in-game for just under 2 years at this point and remains the “baseline” for radar missiles at top tier, with the 7M being a copy paste, and radar mechanics flip flopping widely ever since, to the point where relatively low performance radar missiles such as the AIM-7E/E-2 were more reliable weapons than current top missiles are now.

Even IF we want to sit on the argument of “we dont have enough info to ascertain exact capabilities o X radar missile” which is going to become more and more of an issue as the game progresses, gameplay very much SHOULD take priority over that. As it stands, there’s no reason to believe the upcoming AIM-120/R-77 and other ARH’s will be any functionally better radar-wise to things such as the AIM-7M/R-27ER, since the multipath effect will still be highly abusable, and the missiles themselves dont offer particularly game changing kinematic performances either.

Radar missiles at top tier should be a credible threat, not a dice roll at best and a completely irrelevant threat at worse as they are currently.

2 Likes

You don’t need to dive into the radar physics to understand that it doesn’t mean that the seeker can track notching targets in all situations. Relatively modern (in comparision to AIM-54C introduction time) fighter radars like AN/APG-63, 65, 68 can’t detect notching targets in look down scenarios in search modes and may lose track as well. AIM-54C seeker can’t be better then these radars.

  1. As I just stated, beam aspect tracking is one of the issues thats quite frankly low on the priority list of issues with the AIM-54C. The fact its maneuvrability is nerfed by a MASSIVE 32%, or that its much more visible than it has any right to be, with a massive smoke trail despite bug reports indicating it used a low smoke motor and a massive RCS which makes the AIM-54 missile detectable on radar at over 90km in-game are clear issues which harm the missiles usability substantially and are EASY to fix.

  2. I dont expect the missile to reliably track targets through beam aspect in any and all situations and afaik that isnt what anyone here is arguing. What I am arguing is that the missiles ability to track beam aspect targets which is stated to be improved over that of the AIM-54A is NOT improved in-game, which is a clear discrepancy. The fact that the AIM-54C’s seeker ingame offers literally no advantage over that of the AIM-54A despite a rather substantial overhaul in real life is laughable.

  3. Radar missiles at top tier shouldn’t be so easily rendered useless by multipath, regardless of the argument being that multipath has simplistic modelling in-game, the degree of multipath error currently present is absurd and just bad game design. To take the 7M as an example, a 95m min altitude against a non-maneuvering target in-game vs a stated 5m min altitude irl represents a whopping 1900% increase in min altitude. Not only is the modelling itself questionable, but as previously stated, in a 3rd person game where low altitude flying is made easy by improved spatial awareness is just bad game design. Current top radar missiles should be threatening and a reliable tool for spacing. The fact most radar missiles are fired sub 10km and that games just turn into low alt (like sub 1000m) furballs all the time is further proof of this.

1 Like