By applying the Offset QAM to the transmission waveform, it can get a good frequency localization without using the cyclic prefix but with more computational complexity in comparison to the QAM.
Yes, because by employing the Offset QAM to the waveform like OFDM, there is no need to occupy an extra intervals time for protection, hence, the unused guard band frequencies can be utilized to increase the transmission channel capacity/speed of transmission.
To the best of my knowledge, apply the Offset QAM to the MIMO is not applicable as it's quite hard to ensure a well-localized frequency in presence of multiple transmitted waveforms, as a result, the QAM with the cyclic prefix is preferred.
The offset QAM is accomplished by delaying the Q signal by a time equal half symbol time to avoid the the zero crossings of the modulated signal wave form which leads to more amplitude variations of waveform. More amplitude variations leads to more out of band emission. OQAM helps reduce the amplitude variations
of the waveform.
On the other side OFDM itself because of the statistical variations of the amplitudes and phases of the subcarriers, the amplitude of the waveform suffers from high signal peaks leading to much greater peak to average power ratio.
Because that OQAM needs delay blocks to implement it is more complicated than the ordinary QAM.
There are two things that are important for the performance metrics; the bandwidth of the waveform, its variation in terms of the P/A power ratio and the out of band emission i n addition to immunity against noise or the bit error rate.
Variations in amplitude could increase or decrease the OOB based on the level of injected power whether it's high or low So, did you mean the OQAM can control the OOB? Any reference? What about the Cyclic Prefix do you think it's required with the OQAM? Why?