my second question is concerning Figure 12. the figure shows that a clear winner in this comparison is using NOMA although DSSS-PDMA increases over it a little for U-31. So what about the BER comparison between FDMA, NOMA and DSSS-PDMA? would your proposed MA achieve better performance than the others?
Many thanks for your questions, it shows that when U is greater than 31, more users than NOMA can be communicated with efficiently. This matches the use case of 5G NR technology, mMTC. On the other hand, as answered in the first question, the larger the number of users, the larger the BER in demodulating users with lower power levels.
Assuming perfect CSI, the BER of the three types of multiple access schemes would be similar. However, if we assume imperfect CSI, the BER of the DSSS-PDMA scheme will be worse.
One way to improve the BER of DSSS-PDMA is to set the number of power levels and power difference appropriately. This is a multiple access method that we have just started to verify, and many verifications and experiments such as the optimal setting values are urgently needed.
