Notching SAM radars\missiles in DCS World?

[Lurker]

Currently it is possible to notch SAM radars in DCS World. My question is, is this realistic behavior? Since these radars are almost always in a look-up situation why would entering "the notch" work against them? Wouldn't they turn off their ground return filters or 0-gate filters, wouldn't they switch to some kind of pulse mode in this case? 

 

It is very likely that I am misunderstanding something here so please feel free to explain this to me. Thank you. 

[Northstar98]

Something I've wondered as well.

 

If you look through db_sensors.lua which is found inside your main DCS installation -> Scripts -> Database, you can see all the parameters for each RADAR. They all have a 'radial velocity min', our notch filter in this case.

 

Just going by 2 examples; the I-Hawk acquisition RADARs; the MPQ-50 PAR (Pulse Acquisition RADAR) and the MPQ-55 ICWAR (Improved Continuous Wave Acquisition RADAR).

 

As the name suggests the MPQ-50 is a more simple pulse RADAR; with no doppler filtering, measurement or analysis - it should only have MTI at most to deal with ground clutter. The MPQ-55 is a continuous-wave RADAR, which requires a doppler shift (i.e from a moving target to detect targets (so it should have a notch filter, and I'm guessing it should always be active, as it is a continuous wave RADAR, not a pulse-doppler RADAR). 

 

Both are modelled as such, yes it treats them both as being exactly the same RADAR.

 

["Hawk sr"] =
        {
            type = RADAR_AS,
            scan_volume =
            {
                azimuth = {-180.0, 180.0},
                elevation = {-15.0, 60.0}
            },
            max_measuring_distance = 90000.0,
            detection_distance =
            {
                [HEMISPHERE_UPPER] =
                {
                    [ASPECT_HEAD_ON] = 90000.0,
                    [ASPECT_TAIL_ON] = 90000.0
                },
                [HEMISPHERE_LOWER] =
                {
                    [ASPECT_HEAD_ON] = 90000.0,
                    [ASPECT_TAIL_ON] = 90000.0
                }
            },
            lock_on_distance_coeff = 0.85,
            velocity_limits =
            {
                radial_velocity_min = 15,
            },
            scan_period = 1.0,
        },

 

What I get from this, is that if there's a target within 90km of the RADAR, within the scan zone and with a radial speed above 30 knots, it will be detected. The RADAR updates every second, and when the target is at 85% of the range, it will transition to a tracking mode (this is confusing, these RADARs are basically acquisition only - the ICWAR might not even have ranging capability if it's just a CW RADAR, without FM ranging).

 

There is no 0-doppler filter, there is only a ~ ±30 knot notch filter, which for the MPQ-50 is completely wrong - it's a pulse RADAR, not a doppler RADAR, and as above it shouldn't be doing any doppler analysis and should only have MTI at most.

 

With pulse-doppler RADARs I would've thought SAM operators would be smart enough to disable notch filters in look-up situations.

 

 

In general though, AI RADARs in DCS are very, very, very simplified.

 

Edited March 18, 2021 by Northstar98

[Lurker]

So if Im following you correctly in your example the I-HAWK has two radars. Even if one is somehow notched (let's say the operator "Forgot" to turn off the notch filter in a look-up situation) the other radar being a pulse radar should still detect the target? 

[Noctrach]

1 hour ago, Lurker said:

Currently it is possible to notch SAM radars in DCS World. My question is, is this realistic behavior? Since these radars are almost always in a look-up situation why would entering "the notch" work against them? Wouldn't they turn off their ground return filters or 0-gate filters, wouldn't they switch to some kind of pulse mode in this case? 

 

It is very likely that I am misunderstanding something here so please feel free to explain this to me. Thank you. 

I'm no expert on the subject but a similar discussion popped up in the F-14 forum with regards to the MLC and ZDF functioning. I started looking a bit deeper into it and found this excellent MIT series on radar signal processing on youtube.

 

A radar beam is not a perfect cone. While the main power output (the mainlobe) is directed as such, there is a ton of radiated power in all directions of the antenna (sidelobes) that will cause reflections by bouncing off of surrounding objects. Even a stationary radar will get ground reflections in the form of this "sidelobe clutter", so your 0-gate filter will not be as effective as you might think. Especially this close to the ground you will have a LOT of echoes from objects near the radar antenna (trees, buildings, rocks). Since the radar itself is stationary, there hardly any way to differentiate a notching target with near-zero relative closure from this clutter.

 

Aside from that, if the antenna has a rotational motion this will also contribute to clutter. In any given timeframe, multiple pulses will be returning to a moving dish, causing a slight doppler shift at the receiver end. Your doppler filtering gates will have to take these echoes into account as well and will have to be broader as a result.

 

As you can see, despite advantages in power output and antenna size, stationary radars have some inherent problems caused by environmental and mechanical clutter sources. They are not able to turn off their filters to mitigate this like a moving airborne radar would.

 

As a result, I would say the ability to notch ground radar is very much realistic behaviour.

Edited March 1, 2021 by Noctrach

[Lurker]

So ground based look-up radars have to deal with significantly more ground clutter than air based look-down radars? Is that what you're saying Nocratch? 

[Noctrach]

2 minutes ago, Lurker said:

So if Im following you correctly in your example the I-HAWK has two radars. Even if one is somehow notched (let's say the operator "Forgot" to turn off the notch filter in a look-up situation) the other radar being a pulse radar should still detect the target? 

 

A pulse radar would detect the target, but inherently suffers much more from clutter sources than the pulse-doppler one. Depending on the circumstances it would have its own struggles to deal with (clutter from, but not limited to:  surroundings, weather effects, electronic warfare).

As an example: while chaff is almost pointless against PD, it will royally screw up the picture quality of a pulse radar.

[Lurker]

4 minutes ago, Noctrach said:

 

A pulse radar would detect the target, but inherently suffers much more from clutter sources than the pulse-doppler one. Depending on the circumstances it would have its own struggles to deal with (clutter from, but not limited to:  surroundings, weather effects, electronic warfare).

As an example: while chaff is almost pointless against PD, it will royally screw up the picture quality of a pulse radar.

 

That makes sense I guess, but only if the radars are still in search mode. Once a target is acquired wouldn't the radar operators focus all their energy on the enemy aircraft, especially when guiding a missile? Wouldn't that significantly reduce any clutter, since both the pulse and doppler radars would be focusing on a single target in a look up situation?

Edited March 1, 2021 by Lurker

[Noctrach]

1 minute ago, Lurker said:

So ground based look-up radars have to deal with significantly more ground clutter than air based look-down radars? Is that what you're saying Nocratch? 

 

Yessir, they are closer to the ground than airborne radars, so the signal return from ground clutter in the sidelobes is exponentially stronger.

Not to say it's a cakewalk for air-borne look-down radars because if the mainlobe is getting ground reflections that is an entire separate ballpark of pain in the behind.

 

Merely saying, stationary ground-based radars have good reasons to have strong clutter filtering in place that they cannot easily turn off even in look-up mode.

[Lurker]

5 minutes ago, Noctrach said:

 

Yessir, they are closer to the ground than airborne radars, so the signal return from ground clutter in the sidelobes is exponentially stronger.

Not to say it's a cakewalk for air-borne look-down radars because if the mainlobe is getting ground reflections that is an entire separate ballpark of pain in the behind.

 

Merely saying, stationary ground-based radars have good reasons to have strong clutter filtering in place that they cannot easily turn off even in look-up mode.

 

I guess that makes sense. It's still feels kind of weird that all of these expensive sophisticated radar systems are so easily spoofed simply by flying perpendicular to their search beam. Especially once they've acquired and locked a target. Seems like we're still missing something here, especially considering how many SAM sites have multiple radar types and antennas working in concert for target acquisition and tracking. 

Edited March 1, 2021 by Lurker

[Northstar98]

21 minutes ago, Lurker said:

So if Im following you correctly in your example the I-HAWK has two radars. Even if one is somehow notched (let's say the operator "Forgot" to turn off the notch filter in a look-up situation) the other radar being a pulse radar should still detect the target? 

 

My main point is that whether or not a RADAR has a notch filter depends on what kind of RADAR it is, however, DCS doesn't model the difference - all the RADARs are identical just with different scan zones, update rates and probabilities.

 

The MPQ-50 PAR shouldn't be able to be notched - ever. It's a pulse RADAR that has nothing to do with doppler anything (think of the pulse search mode in the tomcat). It should however be susceptible to ground clutter (though it might have MTI to maybe work around that).

 

The MPQ-55 ICWAR should only be a doppler-only RADAR, it's CW (not pulsed at all AFAIK, though it might be modulated), and it would be reliant on a doppler shift (i.e a target with a sufficient enough closure rate) to be able to see it. The reason why this RADAR was developed was for detecting low-level targets, which the PAR would have difficulty doing (being a pulse RADAR).

 

 

In DCS however, ground RADARs don't suffer from ground clutter at all; so long as there's LOS, and the target is inside the scan zone and range, with enough radial velocity, the RADAR can see it.

 

@Noctrach Thanks for the explanation, it all makes sense, though it should be said that DCS definitely doesn't model sidelobes at all. But wouldn't a pulse RADAR not have doppler filtering at all? And instead use a MTI to filter out ground returns.

Edited March 1, 2021 by Northstar98

[Noctrach]

6 minutes ago, Lurker said:

 

That makes sense I guess, but only if the radars are still in search mode. Once a target is acquired wouldn't the radar operators focus all their energy on the enemy aircraft, especially when guiding a missile? Wouldn't that significantly reduce any clutter, since both the pulse and doppler radars would be focusing on a single target in a look up situation?

 

 

The reality is that you cannot simply focus all the energy in a single spot. Even in locked mode the radar will still experience significant clutter from side-lobes.

 

The "lock" means you have found a signal of interest that has a strong enough signal-to-noise ratio (SNR) to keep track of it.

This mainly brings different factors into play like velocity and range gating to keep distinguishing this identified signal as the object moves through your line of sight.

However, the more its radial velocity drops to near-zero, the more that SNR will start dropping, increasing the risk of losing the lock to clutter.

 

Quote

I guess that makes sense. It's still feels kind of weird that all of these expensive sophisticated radar systems are so easily spoofed simply by flying perpendicular to their search beam. Especially once they've acquired and locked a target. Seems like we're still missing something here. 

 

Yeah I getcha, but the reality is that it's just the nature of the physics involved. I do recommend you give the youtube lectures a watch. It's good to see how astronomically large the effect of all the different sources of clutter is on all the other signals. Frankly, I think it's more that DCS makes it seem like getting a good lock is much more straightforward than it is in real life.

 

Modern radars have less inherent noise, better materials, better electronics, higher frequencies, larger power output, better signal processing techniques... and they still struggle in a lot of ways. Only once AESA gets involved and you can listen to an entire spectrum of frequencies at once can you really say things change fundamentally.

1 minute ago, Northstar98 said:

In DCS however, ground RADARs don't suffer from ground clutter at all; so long as there's LOS, and the target is inside the scan zone and range, with enough radial velocity, the RADAR can see it.

 

Yep, very true. A target overflying at high altitude is treated the same in DCS as a target flying treetop level, even though the complexity of detecting the latter is in an entirely different league.

3 minutes ago, Northstar98 said:

Thanks for the explanation, it all makes sense, though it should be said that DCS definitely doesn't model sidelobes at all. But wouldn't a pulse RADAR not have doppler filtering at all? And instead use a MTI to filter out ground returns.

 

Yeah they'd filter in different ways but I presume the concept is much the same. It's all SNR in the end

[Stearmandriver]

6 hours ago, Noctrach said:

I'm no expert on the subject but a similar discussion popped up in the F-14 forum with regards to the MLC and ZDF functioning. I started looking a bit deeper into it and found this excellent MIT series on radar signal processing on youtube.

 

A radar beam is not a perfect cone. While the main power output (the mainlobe) is directed as such, there is a ton of radiated power in all directions of the antenna (sidelobes) that will cause reflections by bouncing off of surrounding objects. Even a stationary radar will get ground reflections in the form of this "sidelobe clutter", so your 0-gate filter will not be as effective as you might think. Especially this close to the ground you will have a LOT of echoes from objects near the radar antenna (trees, buildings, rocks). Since the radar itself is stationary, there hardly any way to differentiate a notching target with near-zero relative closure from this clutter.

 

Aside from that, if the antenna has a rotational motion this will also contribute to clutter. In any given timeframe, multiple pulses will be returning to a moving dish, causing a slight doppler shift at the receiver end. Your doppler filtering gates will have to take these echoes into account as well and will have to be broader as a result.

 

As you can see, despite advantages in power output and antenna size, stationary radars have some inherent problems caused by environmental and mechanical clutter sources. They are not able to turn off their filters to mitigate this like a moving airborne radar would.

 

As a result, I would say the ability to notch ground radar is very much realistic behaviour.

 

Disclaimer: I know absolutely nothing about radar designed to track aircraft, other than that I can stop making position reports when they can see me ;).

 

But regarding ground clutter from a stationary radar, I do have some familiarity with this as it applies to weather radar. 

 

Basically, since a stationary radar site is seeing the same ground clutter targets at the same azimuths and ranges with every sweep of a given elevation, they are just memorized by the computer and then filtered off the display.  Mount Baker is always in the same place, for instance, so the Seattle NWS radar just doesn't display it. 

 

I would think a stationary air traffic radar would / could do the same thing, but this could mean that a low target would be caught in the filtering algorithm and not displayed, so... dunno what tactics would be used.  But it seems likely that some form of ground return filtering could be applied that relies on memory of some low-elevation calibration sweeps, vs just doppler filtering.  Maybe operator-selectable filtering algorithms/techniques?

[Lurker]

10 hours ago, Stearmandriver said:

Disclaimer: I know absolutely nothing about radar designed to track aircraft, other than that I can stop making position reports when they can see me ;).

 

But regarding ground clutter from a stationary radar, I do have some familiarity with this as it applies to weather radar. 

 

Basically, since a stationary radar site is seeing the same ground clutter targets at the same azimuths and ranges with every sweep of a given elevation, they are just memorized by the computer and then filtered off the display.  Mount Baker is always in the same place, for instance, so the Seattle NWS radar just doesn't display it. 

 

I would think a stationary air traffic radar would / could do the same thing, but this could mean that a low target would be caught in the filtering algorithm and not displayed, so... dunno what tactics would be used.  But it seems likely that some form of ground return filtering could be applied that relies on memory of some low-elevation calibration sweeps, vs just doppler filtering.  Maybe operator-selectable filtering algorithms/techniques?

 

This is the kind of thing that I'm talking about. Sure, low flying targets have always been problematic for ground based radars, but what happens in DCS is that you're flying at 15000ft you get a launch warning from a ground based SAM site, you turn perpendicular to the beam and "poof" the missile is spoofed. Regardless of what type of SAM site is targeting you or what altitude you're flying at. What Nocratch has written so far makes sense, but if it were this easy to spoof real life SAM systems then no country in their right mind would invest into building them ever. 

Now I'm not saying that this shouldn't happen ever, but that depending on the type of search radar used, and their configuration and placement on the ground this shouldn't happen 100% of the time. 

Edited March 2, 2021 by Lurker

[DD_Fenrir]

21 minutes ago, Lurker said:

 

This is the kind of thing that I'm talking about. Sure, low flying targets have always been problematic for ground based radars, but what happens in DCS is that you're flying at 15000ft you get a launch warning from a ground based SAM site, you turn perpendicular to the beam and "poof" the missile is spoofed. Regardless of what type of SAM site is targeting you or what altitude you're flying at. What Nocratch has written so far makes sense, but if it were this easy to spoof real life SAM systems then no country in their right mind would invest into building them ever. 

Now I'm not saying that this shouldn't happen ever, but that depending on the type of search radar used, and their configuration and placement on the ground this shouldn't happen 100% of the time. 

 

 

 

Altitude isn't the key metric; it's slant angle. 

 

If you're at 15,000ft and the SAM tracking Radar is only 15 miles away, you still appear to that radar at a slant angle of only 11° above the horizon. You'd have to get down to 3 miles to make that 45°.

 

I think you misunderstand just how large some of the sidelobes are and just how complex a task filtering out their noise can be.

 

That being the case, having been shot down by a great number of SAMs in DCS during the last 11 years, I find SAMs not so easy to notch as you make out, especially when you get close.

 

 

[Noctrach]

The F-14 with its mid 60's radar had a lot of advantages in situations with heavy jamming that later radars would find almost impossible to solve.

As someone put it, the power of the AWG-9 isn't the machine, but the fact that you have a human operator capable of judging and manually manipulating the raw input data.

 

How do you teach an algorithm which reflection is a mountain peak, which is a speeding car and which is the low flying jet with the tiny radar cross-section? Modern radars have ways of solving this, but we're well into the early 90's before this starts becoming somewhat reliable. It simply requires a lot of processing power.

 

Most of the SAMs in DCS are quite old. You also have to realise there's just a lot of "DCS-isms" going on around these kinds of topics.

Part of this is due to the simplistic simulation of radar, part of this is the absence of genuine human intelligence, part of this is the absence of electronic warfare.

 

Real life SAMs will not be constantly searching for targets. A known SAM location is vulnerable to SEAD/DEAD or can simply be avoided by pilots. The problems start when you get a SAM site going live right underneath you because it was networked in an IADS with the large search radars 50+ miles away. Now you don't know exactly where the missile is coming from and you are within lethal range so simple evasive manoeuvres won't save you anymore. That said, there's a great video of an F-16 over Baghdad evading six consecutive SAM shots.

 

Again, human intelligence will find ways to mitigate the inherent weakness of the systems. This goes for both the pilots and the SAM operators.

[Lurker]

Yes I'm aware of the DCS-isms, there are many in this sim. Apparently that seems to be the main reason this part of the simulation is very rudimentary, which is why I've wondered about this particular part of it and hence the topic, but if that's the take-away then I guess it is, what it is. 

I do find it interesting that we do get somewhat realistic behavior because of the inherent flaws in DCS World and this part of it's simulation. In a way, we are getting all the real-world weaknesses of SAM systems for all the wrong reasons

It would be interesting if at some point we could get SAM and EWR systems (IADS) working together in concert, whether as an add-on module and human operators or as some kind of AI system, though the latter part seems like it would probably not be very likely. 

2 hours ago, Noctrach said:

Again, human intelligence will find ways to mitigate the inherent weakness of the systems. This goes for both the pilots and the SAM operators.

 

Yeah, human intelligence working the SAM systems in place would go a long way towards mitigating their inherent weaknesses. Maybe it happens some day, who knows. 

Edited March 2, 2021 by Lurker

[Stearmandriver]

There is an IADS script that has been incorporated into DCS Liberation that will use inputs from separate early warning radar sites to suddenly make a SAM site go active right below you, if you aren't paying attention.  The SAMS will hold their fire until you're within lethal range, and surprise you ;).  You can mitigate this somewhat by destroying the EWR sites. 

 

I agree that defeating some SAMS in DCS is not as simple as notching.  That works down low or at the edge of the range (when you're using slant angle to your advantage, as mentioned).  But fly to within 15nm of an SA-10 site at 20,000ft and see if notching alone saves you. 

 

Edited March 2, 2021 by Stearmandriver

[Lurker]

On 3/2/2021 at 4:50 PM, Stearmandriver said:

I agree that defeating some SAMS in DCS is not as simple as notching.  That works down low or at the edge of the range (when you're using slant angle to your advantage, as mentioned).  But fly to within 15nm of an SA-10 site at 20,000ft and see if notching alone saves you. 

 

Yes the closer you are to the beam source the more difficult it is to notch. But I feel that in-game that is because at that point you need to fly in a more circular motion to keep the radial velocity as close to 0 as possible.

Edited March 4, 2021 by Lurker

[Dragon1-1]

SA-10 uses a phased array FCR, so realistically, notching shouldn't be possible at all. It also shouldn't give a launch warning on RWR. Another thing is that with a real SA-10, if you get a spike from FLAP LID (the "10") radar, you might as well assume a launch, because its search capability is very limited, and the site should be nearly (but not completely) blind without its EWRs. The current way the system is modeled in DCS is very lacking, to say the least.

[Northstar98]

 

Quote

It also shouldn't give a launch warning on RWR. Another thing is that with a real SA-10, if you get a spike from FLAP LID (the "10") radar, you might as well assume a launch, because its search capability is very limited, and the site should be nearly (but not completely) blind without its EWRs.

 

Here I'm not so sure on, by my understanding we have an earlier version of the S-300 (well, apart from the 64H6E - which is more associated with the S-300PMU AFAIK, though we are getting a more appropriate ST-68U) which uses SARH/DL as opposed to the later TVM guidance (though that still requires illumination of the target, at least pretty frequently).

 

As I understand it the older version of the S-300 uses SARH w. mid-course updates, SARH requires illumination (probably frequently pulsing instead of a CW mode) and RWRs might be able to detect the uplink commands - both in conjunction I would've thought would trigger a launch warning.

 

I am speculating though. 

 

Quote

The current way the system is modeled in DCS is very lacking, to say the least.

 

Absolutely, DCS doesn't even model RADAR bands or PRFs (which is why RWRs like the Viggen use an emulator) let alone phased array RADARs. 

 

The Flap Lid PESA is modelled exactly like any other RADAR including much older FCRs such as the SNR-75 'Fan Song' and SNR-125 'Low Blow', just with a different scan zone. It's all there in the db_sensors.lua.

 

Spoiler

 

Here's how the 30N6 is modelled:

["S-300PS 40B6M tr"] =
        {
            type = RADAR_AS,
            scan_volume =
            {
                azimuth = {-180.0, 180.0},
                elevation = {-15.0, 60.0}
            },
            max_measuring_distance = 260000.0,
            detection_distance =
            {
                [HEMISPHERE_UPPER] =
                {
                    [ASPECT_HEAD_ON] = 160000.0,
                    [ASPECT_TAIL_ON] = 160000.0
                },
                [HEMISPHERE_LOWER] =
                {
                    [ASPECT_HEAD_ON] = 160000.0,
                    [ASPECT_TAIL_ON] = 160000.0
                }
            },
            lock_on_distance_coeff = 0.85,
            velocity_limits =
            {
                radial_velocity_min = 10.0,
            },
            scan_period = 1.0,

 

Looks like it has a notch filter around ±10m/s radial velocity, same lock_on_distance_coeff as basically every other RADAR and the same notch filter.

 

Compare that to the SNR-75V of the SA-2d:

["snr s-125 tr"] =
        {
            type = RADAR_AS,
            scan_volume =
            {
                azimuth = {-180.0, 180.0},
                elevation = {-15.0, 60.0}
            },
            max_measuring_distance = 90000.0,
            detection_distance =
            {
                [HEMISPHERE_UPPER] =
                {
                    [ASPECT_HEAD_ON] = 90000.0,
                    [ASPECT_TAIL_ON] = 90000.0
                },
                [HEMISPHERE_LOWER] =
                {
                    [ASPECT_HEAD_ON] = 90000.0,
                    [ASPECT_TAIL_ON] = 90000.0
                }
            },
            lock_on_distance_coeff = 0.85,
            velocity_limits =
            {
                radial_velocity_min = 10,
            },
            scan_period = 1.0,
        },

 

Firstly, you might've think I've gotten it wrong, as the name of the RADAR is snr s-125 tr - referring to the SNR-125 TTR/FCR of the SA-3. But this is just a highlight of IMO a lazy implementation - all that was done to add the SNR-75, was to copy and paste the existing SNR-125 and give it a different 3D model - that's it.

 

The same is true of the AN/SPY-1 on board the Ticonderoga and Arleigh-Burke - they just reuse the existing MPQ-53 TAR/TTR/FCR of the Patriot PAC-2 system - it's again a copy and paste job of an existing RADAR..

 

But going back to the comparison between the Flap Lid; the only difference between them is the scan volume and detection distance - the performance is otherwise completely identical in terms of lock_on_distance_coeff (which is probably just the distance where it transitions from search to STT/FC), the notch filter and the scan period.

 

Edited March 7, 2021 by Northstar98
40V6M is the mast the RADAR is mounted on, not the RADAR itself.

[Skysurfer]

PD you can notch. Pulse against the sky or newer PESA/AESA radars you really cant. 

[Noctrach]

Correct, but then again... Pulse is a lot more susceptible to EW.

 

PESA/AESA is dark sorcery as far as most public sources are concerned.

 

Neither are simulated well, if at all, within DCS.

[Dragon1-1]

PESA and AESA are hardly sorcery, this technology has many civilian applications, as well. The basic principles are well known, the devil is in details. It would not be impossible nor illegal to include a simulation of a "generic" PESA radar, even if it might not be an exact match for FLAP LID.

 

Pulse is indeed more susceptible to both clutter and EW. It's mostly used by older systems.

[GGTharos]

On 3/4/2021 at 6:44 AM, Dragon1-1 said:

SA-10 uses a phased array FCR, so realistically, notching shouldn't be possible at all.

 

Why would an ESA not have a notch gate?

 

On 3/4/2021 at 6:44 AM, Dragon1-1 said:

It also shouldn't give a launch warning on RWR.

 

It produces an illumination signal at launch for the missile, which is also accompanied by the missile up/down-link.  At least the first is detectable by an RWR.

 

On 3/4/2021 at 6:44 AM, Dragon1-1 said:

Another thing is that with a real SA-10, if you get a spike from FLAP LID (the "10") radar, you might as well assume a launch, because its search capability is very limited, and the site should be nearly (but not completely) blind without its EWRs. The current way the system is modeled in DCS is very lacking, to say the least.

 

That we agree on, most guided systems are lacking in some way.

[Northstar98]

3 hours ago, Noctrach said:

Correct, but then again... Pulse is a lot more susceptible to EW.

 

PESA/AESA is dark sorcery as far as most public sources are concerned.

 

Neither are simulated well, if at all, within DCS.

 

Can confirm - not at all. The phased array 30N6 is modelled no different to an SNR-75 with a different range and scan zone - apart from that it's identical.

 

Heck the SNR-75 and SNR-125 are identical in DCS (they both share the exact same sensor according to their respective .luas - the SNR-75 of the SA-2 is just a copy and paste of the existing SNR-125 but given a different model).

 

33 minutes ago, Dragon1-1 said:

PESA and AESA are hardly sorcery, this technology has many civilian applications, as well. The basic principles are well known, the devil is in details. It would not be impossible nor illegal to include a simulation of a "generic" PESA radar, even if it might not be an exact match for FLAP LID.

 

Yes and there are a tonne of details missing to make something accurate. However, I am firmly in the camp of something being better than nothing.

 

But to be honest DCS AI RADARs are incredibly simplified - DCS doesn't even account for the rough frequency band. It simulates no limitations aside from LOS and maybe RCS AFAIK (and even then, different RADAR bands will have different RCS values).

 

Edited March 7, 2021 by Northstar98