Blind phase-shift keying (psk) and quadrature amplitude modulation (qam) identification

Abstract

Technology for blind phase-shift keying (PSK) and quadrature amplitude modulation (QAM) identification of a received radio frequency (RF) signal is disclosed. One method can include: uniform sub-sampling the received RF signal to eliminate a phase contribution from a carrier frequency of the received RF signal; and computing a likelihood function of observed phase differences of the sub-sampled received RF signal of a phase sequence for each PSK modulation type. Another method can include: non-uniformly sub-sampling the received RF signal for a distribution of signal amplitudes of the received RF signal; and computing a likelihood function of the signal amplitudes of the sub-samples of the received RF signal for each modulation type.

What is claimed is: 
     
         1 . A method for blind identification of a received phase-shift keying (PSK) modulated radio frequency (RF) signal, comprising:
 uniform sub-sampling the received RF signal to reduce a phase contribution from a carrier frequency of the received RF signal;   computing a likelihood function calculation of observed phase differences of the sub-sampled received RF signal of a phase sequence for each PSK modulation type;   finding a maximum value of the likelihood function calculation; and   identifying a PSK modulation type of the received PSK modulated RF signal based on the maximum value of the likelihood function calculation.   
     
     
         2 . The method of  claim 1 , further comprising:
 estimating a baud rate prior to uniform sub-sampling the received RF signal, wherein uniform sub-sampling uses the estimated baud rate to generate an uniform sub-sampling rate.   
     
     
         3 . The method of  claim 1 , wherein the PSK modulation type includes binary PSK (BPSK), quadrature PSK (QPSK), and 8PSK. 
     
     
         4 . The method of  claim 1 , further comprising:
 estimating a baud rate or a symbol period;   using the estimated baud rate to generate a non-uniform sub-sampling rate;   non-uniformly sub-sampling the received RF signal at the non-uniform sub-sampling rate for a distribution of signal amplitudes of the received RF signal;   computing a likelihood function of the signal amplitudes of the sub-samples of the received RF signal for each modulation type; and   match filtering the received RF signal using a square-root raised-cosine filter to differentiate between quadrature amplitude modulation (QAM) and PSK modulated RF signals.   
     
     
         5 . The method of  claim 4 , wherein the modulation type includes phase-shift keying (PSK), 16QAM, 32QAM, 64QAM, 128QAM, 256QAM, 512 QAM, 1024 QAM, 2048 QAM and 4096 QAM. 
     
     
         6 . The method of  claim 1 , wherein uniform sub-sampling the received RF signal further comprises:
 computing phase differences between adjacent sub-samples of the received RF signal to generate a phase difference signature for inputs of the likelihood function.   
     
     
         7 . The method of  claim 1 , wherein computing the likelihood function further comprises:
 computing a conditional probability density function (PDF) of a phase difference for each modulation type.   
     
     
         8 . The method of  claim 1 , wherein computing the likelihood function further comprises:
 estimating a phase θ c  contributed by a carrier frequency with a modulation type by maximizing a log-likelihood function over the modulation type and the phase θ c .   
     
     
         9 . The method of  claim 1 , wherein blind identification provides modulation identification without a priori knowledge of a carrier frequency, a symbol rate, a baud rate, or a pulse shape of a signal transmission. 
     
     
         10 . The method of  claim 1 , wherein the received RF signal is a pulsed shaped signal with additive white Gaussian noise. 
     
     
         11 . A blind phase-shift keying (PSK) and quadrature amplitude modulation (QAM) identification detector, having computer circuitry configured to:
 uniformly sub-sample a received RF signal to reduce a phase contribution from a carrier frequency of the received RF signal;   compute a likelihood function calculation of observed phase differences of the sub-sampled received RF signal of a phase sequence for each modulation type;   find a maximum value of the likelihood function calculation; and   identify a modulation type of the received RF signal based on the maximum value of the likelihood function calculation.   
     
     
         12 . The computer circuitry of  claim 11 , wherein the modulation type includes phase-shift keying (PSK), 16QAM, 32QAM, 64QAM, 128QAM, 256QAM, 512 QAM, 1024 QAM, 2048 QAM and 4096 QAM. 
     
     
         13 . The computer circuitry of  claim 11 , further configured to:
 estimate a baud rate or a symbol period;   use the estimated baud rate to generate a non-uniform sub-sampling rate;   non-uniformly sub-sample the received RF signal at the non-uniform sub-sampling rate for a distribution of signal amplitudes of the received RF signal;   compute a likelihood function of the signal amplitudes of the sub-samples of the received RF signal for each modulation type; and   match filter the received RF signal using a square-root raised-cosine filter to differentiate between quadrature amplitude modulation (QAM) and PSK modulated RF signals.   
     
     
         14 . The computer circuitry of  claim 13 , further configured to:
 identify a PSK modulation type of the received RF signal based on the maximum value of the likelihood function calculations, wherein the PSK modulation type includes binary PSK (BPSK), quadrature PSK (QPSK), and 8PSK.   
     
     
         15 . A method for identification of a received quadrature amplitude modulation (QAM) radio frequency (RF) signal, comprising:
 non-uniformly sub-sampling the received RF signal for a distribution of signal amplitudes of the received RF signal;   computing a likelihood function calculation of the signal amplitudes of the sub-samples of the received RF signal for each modulation type;   finding a maximum value of the likelihood function calculation; and   identifying a modulation type of the RF signal based on the maximum value of the likelihood function calculation.   
     
     
         16 . The method of  claim 15 , further comprising:
 estimating a baud rate or a symbol period prior to non-uniform sub-sampling the received RF signal, wherein non-uniform sub-sampling uses the estimated baud rate to generate a non-uniform sub-sampling rate; and   match filtering the received RF signal using a square-root raised-cosine filter to decrease an effect of pulse shaping or differentiate between QAM and phase-shift keying (PSK) modulated signals.   
     
     
         17 . The method of  claim 15 , wherein non-uniform sub-sampling the received RF signal further comprises at least one of:
 a. coarse estimating the symbol period from an estimated spectrum of the received RF signal; and   b. sub-sampling the received RF signal at an approximate mid-point of each baud.   
     
     
         18 . The method of  claim 15 , wherein the distribution of signal amplitudes of the received RF signal includes a weighted sum of distributions corresponding to different amplitude values. 
     
     
         19 . The method of  claim 15 , wherein computing the likelihood function further comprises:
 computing a conditional probability density function (PDF) of the signal amplitudes for each modulation type, wherein the modulation type includes phase-shift keying (PSK), 16QAM, 32QAM, 64QAM, 128QAM, 256QAM, 512 QAM, 1024 QAM, 2048 QAM and 4096 QAM.   
     
     
         20 . The method of  claim 15 , wherein identification provides modulation identification without a priori knowledge of a carrier frequency, a symbol rate, a baud rate, or a pulse shape of a signal transmission.