Considering that there are no other side peaks in NCCFR reconstruction for low order BOC (2, 2), the MFSPR is analyzed for both BOC (8, 4) and BOC (6, 1), which is shown in Figures 11 and 12.
The results show that MFSPR for BOC (8, 4) in NCCFR is the biggest, indicating that the algorithm has good performance in restraining the first side peaks.
Furthermore, in order to validate the acquisition performance of NCCFR, setting the parameter of false alarm probability, pfa = 0.01, which is a constant value, comparison analysis has been done from the correlation peak detection probability according to several typical signals in different SNR conditions, which mainly focuses on the main peak detection probability (MPDR), the side peaks detection probability (SPDR), and the first side peaks detection probability (FSPDR).
The results show that MPDR performance for both (2, 2) and BOC (8, 4) in NCCFR is better than others and MPDR performance of NCCFR is weaker than RQCC for BOC (6, 1), but in general the new algorithm has a better MPDR performance.
The results show that SPDR for NCCFR in different SNR conditions is the smallest one, indicating that the algorithm has good performance in restraining the side peaks for BOC signals.