# About combination between QAM and PAM

I have confused question regarding the possibility of combination between QAM modulation and PAM in order to improve the spectral efficiency. Is that possible?

As you see in the below Figure, it's the QAM modulation, where the four status of 1+j, 1+j, -1+j and -1-j are existed. where J represents the imaginary part.

My question, what's about using the amplitude with that? For example, in case of combination with PAM modulation, we can have eight status instead of four, which are: 1+j, 1+j, -1+j and -1-j at the same time the four added status will be 1/2(1+j), 1/2(1+j), 1/2(-1+j) and 1/2(-1-j), where the amplitude is used to represent additional bits to transmit too.

Is that reasonable and logical to be used ? Is there any articles or books which explain that ?

It is quite common to combine amplitude and phase modulation. QAM is actually a good example because generally you will have symbols with different phases and different amplitudes. Just in the special case of $$4$$-QAM (which is actually the same as QPSK), all symbols have the same amplitude. But if you take as an example $$16$$-QAM then you'll see that the information is encoded in the phase as well as in the amplitude: