Fig.1

Model of the PCMA signal"

Fig.2

Square spectrum of signals"

Fig.3

Relative prominence of the square spectrum of signals"

Table 1

Spectral line characteristic parameter ${\mathrm{L{N}}_s}$ of the relative prominence of the square spectrum of the signals"

Fig.4

Curve of variation of ${H_2}$ with signal-to-noise ratio for some signals"

Fig.5

Curve of variation of ${H_4}$ with signal-to-noise ratio for some signals"

Table 2

Theoretical values of cumulants of timed sampling sequence ${x_k}$"

Fig.6

Value of ${F_1}$ with the signal-to-noise ratio"

Table 3

Theoretical values of cumulants of difference sequence ${y_k}$"

Fig.7

Value of ${F_2}$ with the signal-to-noise ratio"

Fig.8

Value of ${F_3}$ with the signal-to-noise ratio"

Fig.9

Signal set types"

Fig.10

Decision tree modulation recognition flowchart"

Fig.11

Variation curve of recognition rate with signal-to-noise ratio for each signal of the proposed algorithm"

Fig.12

Variation curve of recognition rate with signal-to-noise ratio for QPSK and 8PSK mixed signals"

Fig.13

Variation curve of recognition rate with amplitude ratio for each signal of the proposed algorithm"

Fig.14

Variation curve of recognition rate with amplitude ratio for QPSK and 8PSK mixed signals"

Fig.15

Variation curve of recognition rate with normalised frequency offset for each signal of the proposed algorithm"

Fig.16

Variation curve of recognition rate with normalised frequency offset for QPSK and 8PSK mixed signals"

Fig.17

Variation curve of recognition rate with roll-off coefficient for each signal of the proposed algorithm"