From MRCA thread
SaiK wrote:thanks for that dileep.. i also read AESA t/r modules can be divided into many roles.. a set of t/r modules for scanning, while a set does the jamming. a few of those t/r modules work as RWRs and some are used for X-band communication under a network centric system.
In what way a PESA could not perform these compared to AESA, if both has similar charecteristics? is it only because of electronic scanning that can "selectively" scan for tracking, communicating and jamming etc?
Let me go a little bit deeper into technology here.
It is already explained how the beam is formed by making a number of emitters radiate waves with specific phase relationship. That is transmission. Let us now see reception.
The beam concept is bidirectional. Just like a dish antenna can both transmit and receive, the phased array can also trasmit and receive. All you need is to add the individual received signals with proper phase relationship.
Let us see how PESA works. You have a single high power signal source feeding a number of emitter modules. The emitter module just contain a variable phase shifter and an antenna element (simplified view!). There is no amplification of the outgoing signal. It just changes the phase of the signal. the received signal also pass through the same phase shifter and gats added up at the back.
You can see that you can not split or group the individual emitters in this technique.
In the first gen AESA, you just add active circuitry (amplifiers) into the individual modules. They are still fed with a master signal, because they should be synchronous to each other PRECISELY. This master signal is phase shifted and fed to a power amplifier. the received signal is fed through the phase shifter and combined at the back. The russian AESA projects have this scheme according to available info.
This scheme also can not do multiple roles for the modules, so it do not offer any functional advantage over PESA. However, it improves reliability and power efficiency.
Next improvement is using multi-channel TX/RX modules. Each module can be fed with multiple master signals. One channel can be selected at a time for operation. So, with a four channel system (eg. APG-79), you can form four different beams at the same time. Note that one module can do only one function at a time, so you got to divide up the modules. The received signals also travel back the same way.
Now, THAT PESA can not do. But this scheme makes the TX/RX modules to cost more and ulky.
The second gen AESA (eg Vixen) uses Direct Digital Synthesis (DDS). Here the outgoing signal is NOT from a master. It is generated using DSP techniques. You still nees a master clock, so all the modules are synchronized in frequency and phase. Each module will have DSP and processors and the master RC will talk to them setting up various waveforms and phases.
This scheme can generate any combination of waves, limited only by software. But this has a major problem. that is received signal.
In the previous techniques, the received signals just passed through the same phase shifter and automatically combined with proper phase relationships. The DDS system need to do it using DSP, the same way it generated the transmit signal. So, the received signal is DSP procesed at the module itelf.
In previous models, you do DSP processing at the combined signal, which has more amplitude. Here we have to process a lower amplitude signal, so the signal to noise ratio is much lower. So, the effective sensitivity of the receiver is much lower here. That in fact offset a lot of advantages.
But there are remarkable adantages. You are no longer limited to a single frequency. You can transmit and receive a spread spectrum signal. You can do frequency hopping. Also, you can do a lot better in weapon guiding, where receiver sensitivity is not THAT important.
And the blessing and curse of this system is the dependance on software. You can smarten or dumb down the performance without anyone knowing it. the manufacturer can provide a friendly user country with a software development kit with libraries, so the user country, like India can create kick-ass modes themselves. Or, the manufacturer can do it the Microsoft Way, by exporting a dumbed down library and SDK.
I have not seen any info about any American Radars, (incl APG-77) to have this architecture, but it may be in fact.
And finally, the photos of the AESA modules of DRDO indicate they belong to the last category.