초록 ▼

Nonlinear target detection for the detection of extremely weak signals when buried in very high levels of noise is performed with improvement of the quality of the signals received in the data by the amplification of the signal to noise (S/N) ratio. The S/N ratio increases dramatically after the ori

Nonlinear target detection for the detection of extremely weak signals when buried in very high levels of noise is performed with improvement of the quality of the signals received in the data by the amplification of the signal to noise (S/N) ratio. The S/N ratio increases dramatically after the originally measured signal is passed through a nonlinear stochastic resonance filter.

대표청구항 ▼

What is claimed is: 1. A highly sensitive, nonlinear target detection device comprising: a target signal; means for decomposing said target signal into Fourier components; an adder for summing a noise signal with said target signal; a nonlinear filter for transforming an output of said adder; mea

What is claimed is: 1. A highly sensitive, nonlinear target detection device comprising: a target signal; means for decomposing said target signal into Fourier components; an adder for summing a noise signal with said target signal; a nonlinear filter for transforming an output of said adder; means for analyzing output of said filtering step; and means for comparing output of said analyzing step with target parameters. 2. The highly sensitive, nonlinear target detection device of claim 1 wherein said means for decomposing further comprises means for decomposing said target signal into Fourier components 3. The highly sensitive, nonlinear target detection device of claim 1 wherein said means for comparing further comprises means for comparing using Bayesian testing. 4. The highly sensitive, nonlinear target detection device of claim 1 wherein said means for comparing further comprises means for comparing using Neyman-Pearson testing. 5. The highly sensitive, nonlinear target detection device of claim 1 wherein said nonlinear filter further comprises a nonlinear filter exhibiting stochastic resonance in its response output. 6. The highly sensitive, nonlinear target detection method of claim 1 wherein said nonlinear filter comprises description="In-line Formulae" end="lead"{dot over (x)}=(x)[a+x][a-x] where a>0.description="In-line Formulae" end="tail" 7. The highly sensitive, nonlinear target detection device of claim 1 wherein said nonlinear filter comprises description="In-line Formulae" end="lead"{dot over (x)}=f(x)=-(x+a1)(x+a2) (x-a3)(x-a4)(x-a5).description="In-line Formulae" end="tail" 8. The highly sensitive, nonlinear target detection device of claim 1 wherein said filter step further comprises filtering output of said summing step through a nonlinear filter description="In-line Formulae" end="lead"{dot over (x)}=x-x3+S(t)+η(t).description="In-line Formulae" end="tail" 9. The highly sensitive, nonlinear target detection device of claim 1 wherein said target signal comprises a sum of two sine waves with unity amplitude and having frequencies of 1 and 4 Hertz description="In-line Formulae" end="lead"S( t)=sin(2π(1)t)+sin(2π(4)t).description="In-line Formulae" end="tail" 10. The highly sensitive, nonlinear target detection device of claim 1 wherein said noise signal comprises white a Gaussian noise. 11. A highly sensitive, nonlinear target detection method comprising the steps of: providing a target signal; decomposing said target signal into Fourier components; summing a noise signal with said target signal; filtering output of said summing step through a nonlinear filter transformation; analyzing output of said filtering step; and comparing output of said analyzing step with target parameters. 12. The highly sensitive, nonlinear target detection method of claim 11 wherein said decomposing step further comprises decomposing said target signal into Fourier components: 13. The highly sensitive, nonlinear target detection method of claim 11 wherein said comparing step further comprises using Bayesian testing. 14. The highly sensitive, nonlinear target detection method of claim 11 wherein said comparing step further comprises using Neyman-Pearson testing. 15. The highly sensitive, nonlinear target detection method of claim 11 wherein said filter step further comprises filtering output of said summing step through a nonlinear filter transformation exhibiting stochastic resonance in its response output. 16. The highly sensitive, nonlinear target detection method of claim 11 wherein said filter step further comprises filtering output of said summing step through a nonlinear filter description="In-line Formulae" end="lead"{dot over (x)}=(x)[a+x][a-x] where a>0.description="In-line Formulae" end="tail" 17. The highly sensitive, nonlinear target detection method of claim 11 wherein said filter step further comprises filtering output of said summing step through a nonlinear filter description="In-line Formulae" end="lead"{dot over (x)}=f(x)=-(x+a1)(x+a2) (x-a3)(x-a4)(x-a5).description="In-line Formulae" end="tail" 18. The highly sensitive, nonlinear target detection method of claim 11 wherein said filter step further comprises filtering output of said summing step through a nonlinear filter description="In-line Formulae" end="lead"{dot over (x)}=x-x3+S(t)+η(t).description="In-line Formulae" end="tail" 19. The highly sensitive, nonlinear target detection method of claim 11 wherein said providing step further comprises providing a target signal of the sum of two sine waves with unity amplitude and having frequencies of 1 and 4 Hertz description="In-line Formulae" end="lead"S( t)=sin(2π(1)t)+sin(2π(4)t).description="In-line Formulae" end="tail" 20. The highly sensitive, nonlinear target detection method of claim 11 wherein said summing step further comprises summing a noise signal of white Gaussian noise.