New structures for modulation classification and SNR estimation with applications to Cognitive Radio and Software Defined Radio

Modulation classification structures for M-PSK (M-ary Phase Shift Keying) and D-MPSK (Differential M-ary Phase Shift Keying) are presented. The modulation classifiers estimate the most likely value of the modulation index M that is present at the input of the receiver. The modulation classifiers are NDA (Non Data Aided) and are shown to have the following advantages: (1) they do not require prior carrier synchronization; (2) they have a compact fixed-point hardware implementation suitable for implementation in devices such as FPGAs (Field Programmable Gate Arrays) and ASICs (Application Specific Integrated Circuits); (3) they require only 1 sample/symbol; (4) relatively few symbol intervals are needed in order to achieve good detection certainty; (5) the decision thresholds for the modulation classifiers are not dependent upon the signal amplitude and are resilient to imperfections in the AGC (Automatic Gain Control) circuit; and (6) parts of the circuits can be used concurrently to perform SNR (Signal to Noise Ratio) estimation. Applications of the proposed structures to CR (Cognitive Radio) and SDR (Software Defined Radio) are discussed.