Jump to content

F-14 Update: ALR-67 RWR Development Snapshot


Cobra847

Recommended Posts

Radar warning receivers of this type are simple devices regarding direction finding. There are four antennas. For each radar pulse received, the amplitude of the signal received by each antenna is measured. Then the results are compared, and the two strongest signals are used to reconstruct the direction. Effectively, out of that, you can only reconstruct the total signal strength and one angle - you can't even tell if the pulse came from above or from below. That is also why applying stabiliser position correction doesn't even make any sense, because instead of improving direction reconstruction accuracy, you can make it worse under certain conditions. Finally, even with fixed antennas, a typical error in direction finding for this type of RWR is 5-10° root mean square. Hence for most of the time, the stabiliser-induced error is much less significant than the other types of errors.

 

We simulate different effects leading to those errors. Nevertheless, the impact of the moving stabiliser was the easiest to show and the most spectacular one smile.gif.

 

 

Regarding the performance - the RWR computations aren't CPU heavy. Of course, we are monitoring the impact of each system on the overall performance. However, optimisation and profiling is a vast topic, so this topic is not a right place to discuss it smile.gif.

 

This is weird.. why are they using the signals from only 2 of the antennas? The only reason I can think of is the airframe blocking reception at certain angles. Normally you can reconstruct 3d geometry with 4 receivers using advanced correlation or equivalent techniques. I'm no expert of RF processing but I did exactly this with audio signals and it works just fine. I would expect with RF the ambient noise is not as much of a bitch to deal with and compressed signals should be easier to detect. I'm not all too familiar with ground based radars but most airborne mechanical doppler radars use chirping and/or various compressing techniques which should make it easier to detect them. Now doing this against an ESA is a different topic.. :)

 

The error issue is there like you said, what I found is that it's especially high when the source direction is lined up with 2 receivers.

Link to comment
Share on other sites

This is weird.. why are they using the signals from only 2 of the antennas? The only reason I can think of is the airframe blocking reception at certain angles. Normally you can reconstruct 3d geometry with 4 receivers using advanced correlation or equivalent techniques. I'm no expert of RF processing but I did exactly this with audio signals and it works just fine. I would expect with RF the ambient noise is not as much of a bitch to deal with and compressed signals should be easier to detect. I'm not all too familiar with ground based radars but most airborne mechanical doppler radars use chirping and/or various compressing techniques which should make it easier to detect them. Now doing this against an ESA is a different topic.. :)

 

The error issue is there like you said, what I found is that it's especially high when the source direction is lined up with 2 receivers.

 

 

I'm not an expert in radar, but in astronomy (and I guess telecommunications? that one is not my field though) reconstructing signals from multiple radio antennae requires more and more computational power as you add more stations, especially if you need to do it in real time like with an RWR. It could just be that the Tomcat didn't have enough on-board computational power to use more than two. Also selecting just the two highest signal-to-nose ratio antennae probably gives a clearer and more reliable signal, which may be worth it even if the resolution of the signal degrades.

Link to comment
Share on other sites

If a better accuracy were required, the engineers probably would use more antennas, six or eight.

 

Regarding using two antennas to reconstruct the signal, please, take a look at the pattern diagram from this post: https://forums.eagle.ru/showpost.php?p=2636058&postcount=1 . You will see that the antenna is almost "blind" in the rear half. Plus, as you mentioned, the airframe may obstruct the view. I'm not saying that in all cases only two antennas register signal from an electromagnetic pulse, but I doubt that using the additional two antennas would improve direction reconstruction accuracy.

 

Finally, I haven't seen the code of the actual ALR-67 (if anybody reading it did see the code and want to share some knowledge with me, please don't hesitate :) ), so I can't guarantee that there is no third, or fourth antenna correction applied. But I can assure you that even without that knowledge, we did our best to gather all available resource to build a device as real as it is possible in this virtual environment.

Krzysztof Sobczak

 

Heatblur Simulations

https://www.facebook.com/heatblur/

Link to comment
Share on other sites

I'm not an expert in radar, but in astronomy (and I guess telecommunications? that one is not my field though) reconstructing signals from multiple radio antennae requires more and more computational power as you add more stations, especially if you need to do it in real time like with an RWR. It could just be that the Tomcat didn't have enough on-board computational power to use more than two. Also selecting just the two highest signal-to-nose ratio antennae probably gives a clearer and more reliable signal, which may be worth it even if the resolution of the signal degrades.

 

I was thinking about this after I posted too. Working on a 2d field with 3 or more antennas instead of 3d would save a lot of resources while not really affecting the results from the perspective of the end user. Also like you say if even that requires too much resources then the approach of 2 strongest receivers would work I guess. Normally 2 receivers cannot determine a 2d direction of source, but using the stronger signals you could narrow down the field, which would solve the ambiguity problem.

 

If a better accuracy were required, the engineers probably would use more antennas, six or eight.

 

Regarding using two antennas to reconstruct the signal, please, take a look at the pattern diagram from this post: https://forums.eagle.ru/showpost.php?p=2636058&postcount=1 . You will see that the antenna is almost "blind" in the rear half. Plus, as you mentioned, the airframe may obstruct the view. I'm not saying that in all cases only two antennas register signal from an electromagnetic pulse, but I doubt that using the additional two antennas would improve direction reconstruction accuracy.

 

Finally, I haven't seen the code of the actual ALR-67 (if anybody reading it did see the code and want to share some knowledge with me, please don't hesitate :) ), so I can't guarantee that there is no third, or fourth antenna correction applied. But I can assure you that even without that knowledge, we did our best to gather all available resource to build a device as real as it is possible in this virtual environment.

 

I'm not questioning the accuracy of your simulation as I'm sure it'll be great. I'm merely interested in the background of system design.

 

Thanks for the link, that was an interesting read.


Edited by <Blaze>
Link to comment
Share on other sites

Hehe, and I didn't want to sound like I thought you were questioning smile.gif .

I can try explaining a little bit more.

 

 

This is the gain pattern for four antennas (as in the F-14) in the horizontal plane, assuming that the airframe doesn't interact with the signal. I created it today from the model we use. Each antenna is represented by a different colour.

attachment.php?attachmentid=189699&stc=1&d=1531311383

As you can see, there are some overlapping regions, but they are not huge.

 

 

 

Let's take a look at two extreme situations.

 

 

1. The signal is coming from the straight ahead.

 

attachment.php?attachmentid=189700&stc=1&d=1531311383

Let's look at the point marked as A). The red antenna and the blue antenna are the same sensitive to that direction, and the gain value is good. The other two antennas, point B), are almost completely blind to that direction. Effectively, we can use only two signals - red and blue.

 

 

2. The signal from 45°.

attachment.php?attachmentid=189701&stc=1&d=1531312026

Now, three antennas can see the signal. The blue antenna is at its maximum. The red and the green - C) - are of the same magnitude, although much lower than the blue (the plots are in db, so log scale... the red and green are ~50 times lower than the blue). The fourth antenna is completely useless in that direction.

It could be possible to use the blue, and both the red and the green for the grey area. However the area is small and adding another signal of such small strength, and thus low signal to noise ratio, may not necessarily improve the precision significantly. On the other hand, as mentioned above, it would require more CPU power. Concluding, the result may be not worth the effort.

 

 

Radar warning receivers aren't sophisticated ELINT devices. They are to warn the crew that there's a threat that can kill them, and inform about where the crew should roughly look to find a bandit or incoming missile. There is a trade-off between the precision and a set of factors as costs, weight, structural demands, etc. We may question the choices of the engineers and the customers (NAVY), but we have to live with what we have smile.gif.

 

 

Side note: The world isn't 2d, each antenna has a similar pattern to the patterns from the included graphics in its vertical axis (both in real life and in our model).

gains_320.png.aa260c28a6e0682ae727a360b5a23ecc.png

gains2_320.png.b545592df8468cc280a9bf7af4475041.png

gains3_320.png.e2a28856d3c779780a6060304a2ee647.png


Edited by Super Grover

Krzysztof Sobczak

 

Heatblur Simulations

https://www.facebook.com/heatblur/

Link to comment
Share on other sites

Is everything on schedule according to y'alls plan?

i9 9900k @5.1GHz NZXT Kraken |Asus ROG Strix Z390 E-Gaming | Samsung NVMe m.2 970 Evo 1TB | LPX 64GB DDR4 3200MHz

EVGA RTX 3090 FTW3 Ultra | Reverb G1  | HOTAS Warthog | Saitek Flight Pedals

Link to comment
Share on other sites

Might have missed this in an earlier answer, but what is the angular resolution of the display? With regard to angular uncertainty in signal processing, the data truncation or display resolution may make such signal processing uncertainties moot.

Link to comment
Share on other sites

.... we did our best to gather all available resource to build a device as real as it is possible in this virtual environment.

 

Boom - and that is awesome and all anyone can really ask for. Great job guys - this is looking really good! :thumbup:

Asus Maximus VIII Hero Alpha| i7-6700K @ 4.60GHz | nVidia GTX 1080ti Strix OC 11GB @ 2075MHz| 16GB G.Skill Trident Z RGB 3200Mhz DDR4 CL14 |

Samsung 950 PRO 512GB M.2 SSD | Corsair Force LE 480GB SSD | Windows 10 64-Bit | TM Warthog with FSSB R3 Lighting Base | VKB Gunfighter Pro + MCG | TM MFD's | Oculus Rift S | Jetseat FSE

 

[sIGPIC][/sIGPIC]

Link to comment
Share on other sites

For the F-14's ALR-67 (-B) and ALR-45 (-A), we're working on a new, in-depth simulation of RWR antennas and how they're affected by various factors. This also includes recreating the way the cockpit systems interpret and process the data received by the system.

 

AN/ALR-67 is the radar warning receiver (RWR) system used in the F-14B. The eyes of the system are four spiral high-band wide-field-of-view antennas looking front right (45°), back right (135°), back left (225°), and front left (315°). The two front antennas are located on the sides of the air intakes, and the two rear antennas are attached to the horizontal stabilisers. When the aircraft is pictured by a radar beam, the RWR antennas receive the emission. The closer the beam direction is to the antenna centre of the view, the stronger the registered signal is. The AN/ALR-67 electronics compares signal amplitudes from the antennas and uses the strongest two to reconstruct the incoming signal direction.

 

In the video, the simulated radar location is to the aircraft rear and below. When the left stabiliser rotates and moves the trailing edge up, the antenna rotates up too, and the incoming radar signal shifts away from its centre of view - thus the registered signal becomes weaker. At the same time, the signal in the front left antenna doesn't change. The electronics don't know about the horizontal stabiliser deflection and interpret the change as the emitter moving away from the rear left antenna field of view.

 

Heatblur AN/ALR-67 will simulate: radar wave attenuation, signal reception for each antenna independently, antenna condition (damage), signal amplification and threat direction reconstruction from the received signal amplitudes. Just as a real unit does, no faking or RWR-magic.

 

Here's a quick video from the Chromecat branch showing how the location of the RWR antennaes influences signal processing and display in the F-14.

 

 

 

I used to work on those while I was in the Navy.. 77-97.. started on ALR 45/50, ALQ-126A the went to ALR-45F, then ALR-67, ALQ-126B... also ALE-29/39/41.... Ive been retired for 21 years, and still work on F/A-18 Avioncs systems at Lemoore Naval Air Station. Here is some good information on the ALR-67 suite.

 

http://www.dote.osd.mil/pub/reports/FY1999/pdf/navy/99alr67.pdf

Sempre Fortis

Link to comment
Share on other sites

after losing f-4 for an unknown period, i really want to get this bird more than before

FC3 | UH-1 | Mi-8 | A-10C II | F/A-18 | Ka-50 III | F-14 | F-16 | AH-64 Mi-24 | F-5 | F-15E| F-4| Tornado

Persian Gulf | Nevada | Syria | NS-430 | Supercarrier // Wishlist: CH-53 | UH-60

 

Youtube

MS FFB2 - TM Warthog - CH Pro Pedals - Trackir 5

Link to comment
Share on other sites

Hehe, and I didn't want to sound like I thought you were questioning smile.gif .

I can try explaining a little bit more.

 

Considering the antenna characterstics this makes sense, thanks for explaining. I was working with omnidirectional microphones where this wasn't really an issue :)

Link to comment
Share on other sites

I did not want to write any of this, but ever since it was announced that the F-4U1 Corsair was "six months away" and us finding out it never left the research phase, along with 3 years of delays for the F-14, there seems to be quite the discrepancy in what is being said and what's actually being done.

I think he was talking about the F-4E Phantom from BST, that got pushed all the way back the line after the merge with ED.

 

https://forums.eagle.ru/showthread.php?t=215599

While the F-4E is still planned, we have other modules in higher priority like the Hornet, Yak-52, Mi-24, F-16, and AH-1. Focus on the F-4E will come once we clear several of these projects.

 

So, I would not expect any F-4E news for quite some time.

 

Thanks

Modules: KA-50, A-10C, FC3, UH-1H, MI-8MTV2, CA, MIG-21bis, FW-190D9, Bf-109K4, F-86F, MIG-15bis, M-2000C, SA342 Gazelle, AJS-37 Viggen, F/A-18C, F-14, C-101, FW-190A8, F-16C, F-5E, JF-17, SC, Mi-24P Hind, AH-64D Apache, Mirage F1

System: Win 11 Pro 64bit, Ryzen 3800X, 32gb RAM DDR4-3200, PowerColor Radeon RX 6900XT Red Devil ,1 x Samsung SSD 970 EVO Plus 2TB NVMe, 2 x Samsung SSD 2TB + 1TB SATA, MFG Crosswind Rudder Pedals - VIRPIL T-50CM and VIRPIL MongoosT-50 Throttle - HP Reverg G2, using only the latest Open Beta, DCS settings

 

Link to comment
Share on other sites

Sorry for your health problems SVTONY. I am hoping for an announcement soon but realistically we are talking 2019. The year is almost half over already and we still have a Chromecat.

 

At least they have been showing us some of the work lately and I have to give credit when credit is due.

____________________________________________________

PC: ASROCK Z370 Gaming K6 | Intel i7 8700K | GeForce 2080TI | 32GB GeSkill 3200 RAM | GeForce 2080TI | 500GB Samsung 850 EVO M.2 | 1TB Samsung 860 EVO M.2

____________________________________________________

FLIGHT STUFF: Rift S | Warthog Base | Virpil Base | Hornet Grip | A-10 Grip | Cougar Grip | Virpil F-14 Grip | Cougar MFD's | A-10C UFC | Saitek Flight Panels | MFG Crosswind Rudder Pedals

Link to comment
Share on other sites

yeah i don't wait it to be released in this year either. despite it was said that it is almost ready for ea at the begining of the 2018

FC3 | UH-1 | Mi-8 | A-10C II | F/A-18 | Ka-50 III | F-14 | F-16 | AH-64 Mi-24 | F-5 | F-15E| F-4| Tornado

Persian Gulf | Nevada | Syria | NS-430 | Supercarrier // Wishlist: CH-53 | UH-60

 

Youtube

MS FFB2 - TM Warthog - CH Pro Pedals - Trackir 5

Link to comment
Share on other sites

 

First it was 2015, then it was 2016, then it was 2017, now it's 2018 and it was actually available for purchase 8 months ago but it doesn't look like it's coming anytime soon.

 

Actually thats not quite true. The only "semi firm" release window that we had was end of 2017, and Heatblur apologized and explained why they missed it.

Current specs: Windows 10 Home 64bit, i5-9600K @ 3.7 Ghz, 32GB DDR4 RAM, 1TB Samsung EVO 860 M.2 SSD, GAINWARD RTX2060 6GB, Oculus Rift S, MS FFB2 Sidewinder + Warthog Throttle Quadrant, Saitek Pro rudder pedals.

Link to comment
Share on other sites

Sorry for your health problems SVTONY. I am hoping for an announcement soon but realistically we are talking 2019. The year is almost half over already and we still have a Chromecat.

 

At least they have been showing us some of the work lately and I have to give credit when credit is due.

Chromecat means nothing, I am sure they have several development branches and there is already a texture.

But it makes more sense to use a rather basic version (the chromecat branch) for development, no reason to load all the textures and visual stuff which consumes time and resources (and maybe introduces bugs) when there is no need for.

Also last statement (iirc) was 2018. So no reason to think about 2019 yet :)

Link to comment
Share on other sites

Back on topic (for a few more seconds at least) the ALR-67 RWR system looks very interesting and will take some time to get use to it.

 

But it's a old bird with old systems compared to the new jets *swings fly swatter at the bug flying past*.

 

Thanks for the update.

Link to comment
Share on other sites

Back on topic (for a few more seconds at least) the ALR-67 RWR system looks very interesting and will take some time to get use to it.

 

But it's a old bird with old systems compared to the new jets *swings fly swatter at the bug flying past*.

 

Thanks for the update.

 

It's really not much older than the F-15C (1979). The RWR signals move a lot when you roll in the F-15, and if you're not paying attention, it'll look like the signal is coming from the opposite direction. The main difference here is the signal moving with stabilator deflection... but I don't think that will impact fights too much, just something to keep in mind.

Link to comment
Share on other sites

Regarding the hubbub of 'RWRgate' and the pros and cons of having critical either on the outside or the inside band, as well as this Natops / manual saying this and the other saying that, could a request / remark be made for having this as a 'special' option so that each can adhere to his / her preference (ie similar to the takeoff assist in the P-51 or the cockpit shake in the Mig-21bis)?

 

As much as I admire the hardcore and real life accurate simulation aspect of sticking solely to the explanation in the manual, we can equally make such links to playability (ie familiarity with other aircraft) and the obfuscated switching of the RWR logic by the navy: we still do not have a solid when and why on the change (and I suspect we never will :P). Could it thus also make sense that if this logic change was applied during the F14's service life they also would have received this 'software patch' (as clearly hte ALR-67 can operate in both Critical Band Outside and Critical Band Inside)?

 

Again, I am just pitching an idea and would love to hear opinions from HB as well as anyone else on this. :pilotfly:

[sIGPIC][/sIGPIC]

 

Groundpounder extraordinaire

 

 

SPECS: i7-4790K, MSI Z97 Gaming 7, 16 GB RAM, MSI GTX 980ti, Thrustmaster WARTHOG HOTAS, Saitek Pro Combat Rudder pedals, TrackIR 5

 

Link to comment
Share on other sites

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...