Processing of a signal representing radiation

• US 8,600,694 B2
• Filed: 07/12/2006
• Issued: 12/03/2013
• Est. Priority Date: 07/12/2005
• Status: Expired due to Fees

• Current Assignee: Commissariat a L'Energie Atomique (AES Corporation)
• Original Assignee: Commissariat a L'Energie Atomique (AES Corporation)
• Inventors: Eglin, Luc, Barat, Eric
• Primary Examiner(s): Raymond, Edward
• Application Number:
  US11/995,451
• Publication Number:
  US 20080215267A1
• Time in Patent Office: 2,701 Days
• Field of Search: 702/66, 702/79, 702/106, 702/190, 702/191, 702193-197
• US Class Current: 702/106
• CPC Class Codes: G01R 35/00 Testing or calibrating of apparatus covered by the other groups of this subclass
• CPC Class Codes: G01T 1/17 Circuit arrangements not adapted to a particular type of detector pulse-selection circuits H03K, G01R

Abstract

A method for processing a noisy digital time signal y_k of digital pitch k, corresponding to an initial analogue signal s_t after being conditioned by a conditioning chain. The initial analogue signal s_t includes at least one pulse representing information concerning at least one radiation from a radiation source, the radiation and the pulse having an energy distribution. The method includes the determination of a non-noisy digital estimation signal s_k from the noisy time signal y_k by using a state model representing the conditioning imposed by the conditioning chain and in that the state model includes a Markovian variable r_k to be estimated whereof at least two values are associated with physical characteristics of at least two typical pulses constituting a possible representation, at least approximate, of the pulse in the signal s_t.

Claims

1. A method for treatment of a noisy temporal digital signal (y_k) with digital pitch (k), said noisy temporal digital signal (y_k) corresponding to an initial analogue signal (s_t) after it has been conditioned by a conditioning chain, with said initial analogue signal (s_t) including at least one pulse representative of information on radiation (30) coming from a radiation source, with this radiation (30) and the at least one pulse exhibiting an energy distribution, wherein a non-noisy estimated digital signal (s_k) is determined, by a computing unit, from the noisy temporal digital signal (y_k) by using a state model representative of said conditioning imposed by the conditioning chain, and wherein the state model includes a Markov variable (r_k) to be estimated, said Markov variable (r_k) being able to take at least two values, each of the at least two Markov variable (r_k) values being associated with physical characteristics of a corresponding model pulses, with each corresponding model pulse constituting a possible representation, or at least closely approaching a possible representation, of the at least one pulse in the initial analogue signal (S_t).

2. A treatment method according to claim 1, wherein the treatment method also includes a preliminary stage in which the noisy temporal digital signal (y_k) is segmented into successive sections of N samples.

3. A treatment method according claim 1, wherein a given one of the at least two Markov variable values corresponds to a pulse mode, known as the “

noise”

mode, for which it is considered that there exists no pulse in the noisy temporal digital signal (y_k) at a given digital pitch (k).

4. A treatment method according to claim 3, wherein the state model also includes a vector (F) with constants, each one of the constants of the vector (F)-being associated with one of the at least two Markov variable (r_k) values, and each one of said constants of the vector (F) defining, as physical characteristics, an energy range linked to the corresponding model pulse of each one of the at least two Markov variable (r_k) values.

5. A treatment method according to claim 4, wherein the constants of the vector (F) are proportional to a maximum predetermined amplitude (A_min) of the corresponding model pulse of the one of the at least two Markov variable (r_k) values, the energy range of the corresponding model pulse of the one of the at least two Markov variable (r_k) values being the lowest energy ranges defined by constants of the vector (F) that are each associated with one of the at least two Markov variable (r_k) values, wherein the one of the at least two Markov variable (r_k) values does not correspond to the “

noise”

mode.

6. A treatment method according to claim 1, wherein the state model includes a state vector (X_k) to be estimated, which includes a first component that corresponds to the non-noisy estimated digital signal (s_k), wherein a digitised representation of the initial analogue signal (s_t) at the instant (t) corresponds to the digital pitch (k), and wherein the determination, from the noisy temporal digital signal (y_k), of the non-noisy estimated digital signal (s_k) by means of the state model includes an intermediate stage of implementing a Kalman filter for observation of the noisy temporal digital signal (y_k), this intermediate stage supplying an intermediate estimate (X_k/k) of the state vector (X_k), thus including a non-noisy intermediate digital estimation signal (s_k/k) corresponding to a first determination of the non-noisy estimated digital signal (s_k).

7. A treatment method according to claim 6, wherein the determination, from the noisy temporal digital signal (y_k), of the non-noisy estimated digital signal (s_k) by means of the state model, includes a second intermediate stage of implementing a smoothing algorithm, with this second intermediate stage supplying an improved smoothed estimate (X_k/N) of the state vector (X_k), and therefore an improved smoothed estimate (s_k/N) of the non-noisy estimated digital signal (s_k).

8. A treatment method according to claim 1, wherein the at least two Markov variable (r_k) values are estimated by means of a Viterbi algorithm.

9. A treatment method according to claim 6, wherein the determination of the non-noisy estimated digital signal (s_k) is optimized based on a criterion of a Maximum A Posteriori (MAP).

10. A treatment method according to claim 9, wherein the criterion of the Maximum A Posteriori (MAP) is combined with an Expectation-Maximisation (EM) algorithm.

11. A treatment method according to claim 10, wherein the criterion of the Maximum A Posteriori combined with the Expectation-Maximisation (EM) algorithm is implemented:

for estimating the Markov variable (r_k)

for estimating the state vector (X_k), or

for estimating the Markov variable (r_k) and the state vector (X_k).

12. A treatment method according to claim 4 or 6, wherein the state model is expressed in the following form:

13. A treatment method according to claim 12, wherein the treatment method includes an initialisation stage in which the dynamic noise matrix B associated with the “

noise mode”

is equal to that associated with a least energetic pulse mode except the “

noise mode”

, namely

B(r_k=0)=B(r_k=1)

and in which, for the values of r_k equal to or greater than 1, the matrix B(r_k) is determined by

B(r_k)−

F(r_k)−

F(r<sub>k−

1</sub>), where r_k≧

1.

14. A non-transitory memory having a computer program stored thereon, the computer program being configured, upon execution, for treating of a noisy temporal digital signal (y_k) with digital pitch (k), said noisy temporal digital signal (y_k) corresponding to an initial analogue signal (s_t) after it has been conditioned by a conditioning chain, with said initial analogue signal (s_t) including at least one pulse that is representative of information on radiation (30) coming from a radiation source, with this radiation (30) and the at least one pulse exhibiting an energy distribution, characterised in that the computer program includes instructions for implementation of the method according to claim 1.

16. A spectrometer that includes means that are designed to implement a method according to claim 1.

17. A bolometer that includes means that are designed to implement a method according to claim 1.

15. A system for treatment of a noisy temporal digital signal (y_k) of digital pitch (k), said noisy temporal digital signal (y_k) corresponding to an initial analogue signal (s_t) after it has been conditioned by a conditioning chain, with said initial analogue signal (s_t) including at least one pulse that is representative of information on radiation coming from a radiation source, with this radiation and the at least one pulse exhibiting an energy distribution, wherein the system includes a detector, a conditioning chain, a computing unit and means that are designed to cooperate with a state model that is representative of said conditioning imposed by the conditioning chain in order to determine, from the noisy temporal digital signal (y_k), a non-noisy estimated digital signal (s_k), and wherein the state model includes a Markov variable (r_k) to be estimated, said Markov variable (r_k) being able to take at least two values, each of the at least two Markov variable (r_k) values being used to determine physical characteristics of a corresponding model pulses, with each model pulses constituting a possible representation, or at least closely approaching a possible representation, of the at least one pulse in the initial analogue signal (s_t).