In the problem of the parameter estimation of frequency-hopping (FH) signals, most of existing works can only provide the unauthorised detection of the FH signals or estimate the part of parameters of FH signals, therefore cannot provide sufficient information to demodulate signals for message deciphering applications in a non-cooperative communications. This study proposes a new method based on reassigned smoothed pseudo Wigner-Ville distribution (SPWVD) and maximum-likelihood estimation for joint signal parameter estimation of FH communications with M-ary frequency-shift-keyed (MFSK) orthogonal modulation. With the good time frequency concentration and restraining cross-term ability of reassigned SPWVD, this algorithm can efficiently estimate the parameters of FH signals which include hopping frequencies, hopping rate, hopping sequence and modulation type without making any assumption about the alphabet of hopping frequencies or the synchronisation. The algorithm has improved estimation accuracy, reduced estimation root-mean-squared error and obtained better performance than that of the approach based on conventional time-frequency distribution. The simulation results are presented to evaluate the performances of the proposed algorithm. The root-mean-squared error of hopping frequency estimation is <;10-3 for signal-to-noise ratio (SNR) of >-2 dB. The percentage of correct modulation recognition goes to 90% when SNR >3 dB.