MSK Modulator

[Dr. Dhiman Kakati]

Hi all,

There is no Minimum Shift Keying design file in the samples provided along-with the software installation. So if anyone have any sample Modulator design in optical domain please upload here.
I am Not finding any MSK sequence generator in the component library, only component available is MSK pulse generator, so I need help in this context as I need MSK sequence generator.

The problem I am Finding here is if we use MSK pulse generator the constellation diagram shows no continuous phase. Rather it shows is of 4psk or 8 PSK. I am attaching here the design file. Please go through and I need your suggestion how can we make the phase continuous.
and in the constellation visualizer the extra point at the x axis coming why..?

With regards

[alistu]

Hi Dhiman,

Actually there is one MSK modulator example and a MSK pulse generator in the library, which I don’t think would be of any benefit to you, since the demodulation is not shown and the demodulation component doesn’t exist in the library. You can find the “Electrical MSK Modulator” component in the modulators section in transmitters library.

Regards

[Dr. Dhiman Kakati]

Yes I agree but I needed the MSK sequence generator because i need to generate the MSK optical modulator. so electrical msk modulator will not work for me.

Regards

[alistu]

You have mentioned that for MSK pulse generator, no phase continuity is seen in the constellation diagram. Could you please elaborate more on the “phase continuity” you are expecting to see? Are you expecting to see the constellation points being somehow dragged in the constellation diagram? If you insert an image of what you expect (from a link, etc) it would also help.

[Dr. Dhiman Kakati]

Hi Alistu,
the continuous phase here means, there should not be discrete point in the constellation, rather it is a continuous circle i want to mean. I am attaching here an image of continuous phase.
Thank you.

Regards

[alistu]

Thank you. This is not continuous (and I don’t know if continuity actually has a meaning in constellation diagram); the points have a constant regular space with one another. If you use 32-PSK in your MSK pulse generator, the constellation should be something like this. Please compare the two images attached, corresponding to phase continuous MSK and the PSK (which is not phase continuous) as they show “phase-continuity” concept (first one is PSK and the second one is MSK). So the constellation points do not increase in MSK. If you have a link for the constellation you have attached, please address it.

[Dr. Dhiman Kakati]

the image I uploaded is not of MSK, in MSK it constellation points will be connected.

The second waveform you have attached is for MSK.
there the phase is not changing.

regards

[alistu]

Indeed, Dhiman. The second image shows MSK. Can you please upload a link or introduce a paper that verifies in MSK the points are connected? As I simply think the fact that the phase is continuous in MSK does not make any difference in the constellations of PSK and MSK. In other words, that (continuous) PHASE is in signal “time” domain and this (constellation points) PHASE is in signal “space” domain.

[Dr. Dhiman Kakati]

Alistu, in each theory book there is is mentioned that the MSK is continuous phase frequency shift keying, from that only i said the phase is continuous, if at the constellation diagram the points are distinct and not connected in circular pattern then there must have been a phase shift whether it is 8PSK or 32PSK where we can see atleast some degree of phase shift, I have some doubt what kind of constellation there should come in case of MSK. whatever constellation i have seen till now has some amount of phase shift. I still confused in CPFSK what exactly the continuous phase means. from waveform point of view I am clear about that but in case of constellation I am still confused. In case of continuous phase minimum shift keying waveform we Notice that when the output frequency changes, it is a smooth, continuous transition. Consequently, there are no phase discontinuities. I also found that CPFSK has a better bit error performance than a conventional
binary FSK for a given signal to noise (SNR) ratio. so i am very much interested in MSK transmitter and receiver design.

Thanks

Regards

[alistu]

I was not sure about it at first, but thanks to your topic I looked into it and I hope the following explanation will clear up your confusion:

The “waveform” you have mentioned shows the amplitude in time, so it is the “time domain” signal. And you can see the difference between the waveforms (or the time domain signals) of PSK and MSK in the images I formerly attached (and you’ve mentioned it was clear to you).

When it is said that MSK is continuous-phase, the phase in here refers to the phase in time domain signal (or the “waveform”). This has nothing to do with the constellation points phase, as the constellation shows the signal in space domain (or in vector domain), and not in the time domain (MSK has more similarities to QPSK in terms of encoding, rather than PSK).

Regards

[Dr. Dhiman Kakati]

Oh thank you so much for making me clear. from the last paragraph u mentioned can we say that the constellation for MSK may be same as that for QPSK..?

Regards

[alistu]

You’re welcome. The similarity of MSk and QPSK is in encoding with bits alternating between quadrature components, and not every bit being decided on based on the phase it starts with (as in PSK). And the constellation is like the constellation of PSK or QPSK (whose constellations look the same).

[Dr. Dhiman Kakati]

So what are the other possibilities of similarity of MSK with any other modulation scheme… Please let me know.

Regards

[alistu]

Aside from the similarities with QPSK (or more precisely, with Offset-QPSK), MSK has an inherent similarity with Continuous Phase Frequency Shift Keyed (CPFSK), as, after all, both are continuous-phase modulation schemes. More precisely, it can be considered as a CPFSK signal which has a frequency separation of one-half the overall bit rate.

[Dr. Dhiman Kakati]

Thank you Alistu for making me clear. Hope to get help in future also.

Regards

[Dr. Dhiman Kakati]

One more thing I need to mention here that at the output osf the modulator the spectrum of the light is matched with that of Minimum Shift Keying, I have attached here please have a look, and I want everyone to comment on it.
Thank You

With Regards

[alistu]

I did not exactly understand where the attached image of the optical spectrum belongs to. The minimum shift keying takes place in the electrical domain and should be observed with a RF spectrum analyzer, whereas the image attached shows the optical spectrum. Can you elaborate more please?

[Dr. Dhiman Kakati]

I have designed the MSK modulator in optical domain by applying the electrical MSK pulse generator signal applied to LN-MZM. the graph is from the output of LN-MZM.

Thanks

[alistu]

Thank you. The spectrum reads with the radio-frequency spectrum of a MSK pulse generator (which is baseband, of course) or that of the MSk modulator. The only difference is that they have a long tail, which should be filtered and is not there in the optical spectrum the image you have attached shows.

[Dr. Dhiman Kakati]

Yes I think the word “Tail” means sideband spectrum if I am not wrong.
there is very less power at the side lobes that leads MSK to a spectrally efficient modulation Scheme, This brings my attention to go for MSK.

Thank you.

Regards

[alistu]

You’re welcome. Indeed! By “tail” I meant the tail of sidelobes that consists of many of them getting less in power. And their power is less than the main lobe, but in your case they are much less, which is a good think as long as you get the proper result (which I am guessing you are).

Regards

[Dr. Dhiman Kakati]

Alistu, that is why I am looking into this matter so seriously.

Regards

[Author]

Posts