The identification of the modulation type of an arbitrary noisy signal is necessary in various applications including signal confirmation, interference identification, spectrum management, electronic support systems in warfare, electronic counter-counter measures etc. In this paper a novel classification scheme based on the variance of instantaneous frequency is proposed to discriminate between noisy M-ary Phase Shift Keyed (MPSK) and M-ary Frequency Shift Keyed (MFSK) signals. In the proposed method, the received signal is passed through a pair of band pass filters and the ratio of variances of instantaneous frequency of the filter outputs is used as a decision statistic. Analytic expressions are developed for the decision statistic. These expressions show that a high degree of discrimination is possible between PSK and FSK signals even at a carrier-to-noise ratio (CNR) of 0 dB. Simulation studies have been carried out and the theoretical predictions are validated.