System for the suppression of noise and its variations for the detection of a pure signal in a measured noisy discrete signal
First Claim
1. A system for the suppression of noise and its variations for the detection of a pure signal in a measured discrete noisy signal, comprising:
- a detector for converting radiation received into electrical pulses;
a shaping means coupled to the detector for shaping signals produced at an output of the detector;
an analog-digital converter coupled to said shaping means for converting analog output signals from said shaping means to digital signals;
a processor means coupled to the analog-digital converter for processing said digital signals; and
a signaling means for signaling results of processing performed by said processing means;
said processing means comprising a processor, at least one memory and an input-output circuit;
wherein the processor sequences the selection of two identical time durations Δ
t in a selected range of frequential and spatial spectrums of corpuscular radiation, one range where radiating signal is present constituting a signal channel, the other range where there is only noise constituting a reference channel;
wherein the detector receives the radiation signal from the signal channel having a mean noise level M2 and delivers a first pulse at each detection of radiation signal, each said pulse being transformed and converted by the analog-digital converter and counted by the processor, to produce a sum which constitutes a counting rate N2, wherein N2 gives an estimation of the mean noise level M2, N2 being stored in the memory;
wherein the detector receives radiation in the reference channel having a mean noise level M3 and delivers a second pulse at each such detection, each said second pulse being transformed and converted by the analog-digital converter and counted by the processor to produce a sum which constitutes a counting rate N3, wherein N3 gives an estimation of the mean noise level M3, N3 being stored in the memory;
wherein the processor performs the following steps;
computation of the ratio λ
E =N3 /N2 which represents an estimation of λ
defined as the ratio of M3 /M2,computation of the false detection rate TF, according to the following relationships, ##EQU6## computation of a threshold Im where
space="preserve" listing-type="equation">I.sub.m =-log (TF/(1-TF));
computation of the probability of false detection PF according to the following relationships, ##EQU7## computation of a threshold I, where
space="preserve" listing-type="equation">I=-log (PF/(1-PF)), andcomparison of I and Im,control of the signaling means as a function of the result of said comparison of I and Im.
1 Assignment
0 Petitions
Accused Products
Abstract
A system for the suppression of noise and its variations for the detection of a pure signal in a measured noisy discrete signal. The system measures a counting rate N2 in a signal channel and a counting rate N3 in a reference channel, evaluates the estimation value λE of the ratio λ between the mean noise levels in the channels and evaluates the associated false detection probability I as a function of (N2, N3 λE). The system then performs a comparison of I with at least one information threshold Im, each associated with a predetermined false detection level. Comparison can take place indirectly on a function of the probability and can use tabulated values. The system has particular application to the detection of corpuscular radiation.
-
Citations
7 Claims
-
1. A system for the suppression of noise and its variations for the detection of a pure signal in a measured discrete noisy signal, comprising:
-
a detector for converting radiation received into electrical pulses; a shaping means coupled to the detector for shaping signals produced at an output of the detector; an analog-digital converter coupled to said shaping means for converting analog output signals from said shaping means to digital signals; a processor means coupled to the analog-digital converter for processing said digital signals; and a signaling means for signaling results of processing performed by said processing means; said processing means comprising a processor, at least one memory and an input-output circuit; wherein the processor sequences the selection of two identical time durations Δ
t in a selected range of frequential and spatial spectrums of corpuscular radiation, one range where radiating signal is present constituting a signal channel, the other range where there is only noise constituting a reference channel;wherein the detector receives the radiation signal from the signal channel having a mean noise level M2 and delivers a first pulse at each detection of radiation signal, each said pulse being transformed and converted by the analog-digital converter and counted by the processor, to produce a sum which constitutes a counting rate N2, wherein N2 gives an estimation of the mean noise level M2, N2 being stored in the memory; wherein the detector receives radiation in the reference channel having a mean noise level M3 and delivers a second pulse at each such detection, each said second pulse being transformed and converted by the analog-digital converter and counted by the processor to produce a sum which constitutes a counting rate N3, wherein N3 gives an estimation of the mean noise level M3, N3 being stored in the memory; wherein the processor performs the following steps; computation of the ratio λ
E =N3 /N2 which represents an estimation of λ
defined as the ratio of M3 /M2,computation of the false detection rate TF, according to the following relationships, ##EQU6## computation of a threshold Im where
space="preserve" listing-type="equation">I.sub.m =-log (TF/(1-TF));computation of the probability of false detection PF according to the following relationships, ##EQU7## computation of a threshold I, where
space="preserve" listing-type="equation">I=-log (PF/(1-PF)), andcomparison of I and Im, control of the signaling means as a function of the result of said comparison of I and Im. - View Dependent Claims (2, 3, 4, 5, 6, 7)
-
Specification