China Flanker thread

[RedMercury]

Laugh, sounds like something invented in the 50s.

 

[daveman]

This approach to counter stealth aircrafts seems very clever on paper. Does anyone have any real world experience with this technique?

 

[challenge]

Somebody in Sweden did his Master's Thesis on "track-before-detect":

Please, Log in or Register to view URLs content!

Interesting because it's really just a well-known statistical technique, but of course difficult because you'd have to do it in real-time.

In short, if you use a single scan to decide whether or not a target is present, then you have to decide on a threshold value. If the return signal is stronger than the threshold, then there's something there, otherwise the space is empty. Of course the threshold has to be far enough above the normal noise level that you can be sure not to get a lot of false-positives.

But, you could also base your decision on multiple scans. A "slightly high" return value in a single scan might be noise, but if you get several "slightly high" return values from the same area then you can progressively increase the probability of there really being a target there. This way you can detect targets even if the return signal is weaker than the normal background noise.

The upside: this technique would work with any given radar technology. It's not dependent on increasing the signal, or separating the receiver from the transmitter. The downside: Bayesian statistics takes a lot of computer power.

But given the rate at which computational power increases, it's likely that this technique will be used widely pretty soon, if it isn't already.

it means that a radar automatically track any weak signal such as noise?

 

[crobato]

If I do google or yahoo searches, there seems to be a lot of work on this topic already. As a note it is also used for weak signal processing, noise and clutter rejection techniques.

Examples:

Please, Log in or Register to view URLs content!

Please, Log in or Register to view URLs content!

Please, Log in or Register to view URLs content!

Please, Log in or Register to view URLs content!

 

[RedMercury]

I suspect the concept itself is not very new. Just that increases in available processing power makes it feasible to implement, and perhaps the difficulty of tracking stealthy objects force the engineers to use shinier hammers. The basic idea is to entertain and maintain more hypothesis of where aircraft might be, even before telling the radar operator that it is "tracking". So in a sense it is just a user interface tweak, but what it boils down to is more processing power and memory to maintain all these potential contacts. It is sort of like a brute force approach, like using more particles in your particle filter.

 

[challenge]

can J-11B(not to mention J-10) airborne radar performed both ground mapping and air to air search-simultaneosly?
with the increase in computer power,even mechanical radar such EF-2000's Captor radar able perform both SAR/GMTI and air to air search and track .

 

[challenge]

I suspect the concept itself is not very new. Just that increases in available processing power makes it feasible to implement, and perhaps the difficulty of tracking stealthy objects force the engineers to use shinier hammers. The basic idea is to entertain and maintain more hypothesis of where aircraft might be, even before telling the radar operator that it is "tracking". So in a sense it is just a user interface tweak, but what it boils down to is more processing power and memory to maintain all these potential contacts. It is sort of like a brute force approach, like using more particles in your particle filter.

it occur to me, with track before detection technique,this allow radar to detect even large RCS target at max. range.
assume without TBD,a radar can detect 3 m.rcs at said 150km away,with TBD one can detect same 3 meter target beyond 200km.

 

[crobato]

can J-11B(not to mention J-10) airborne radar performed both ground mapping and air to air search-simultaneosly?
with the increase in computer power,even mechanical radar such EF-2000's Captor radar able perform both SAR/GMTI and air to air search and track .

I don't think so. Not with mechanical array and a single emitter. These kind of arrays do not multitask. I have doubts that Captor can do SAR/GMTI and air search and track at once.

 

[challenge]

I don't think so. Not with mechanical array and a single emitter. These kind of arrays do not multitask. I have doubts that Captor can do SAR/GMTI and air search and track at once.

according to air international,suppliment issue,claim that Captor radar the most advance mechanical radar ,and last ME radar to be enter service.in there opinion radar slightly ahead of APG-71,able to performed both high resolution mapping and air to air search-simultaneosly.

 

[crobato]

Does not matter what you mean by the "most advanced mechanical radar". Its still a mechanical radar, and the array does not multitask its scanning like phase array and AESA does. Go ahead and explain to me how a slotted array can split its beams and head them to different directions when its the mechanical arm behind the array that points the array and turns it around.