System for estimating signals received in the form of mixed signals

• US 5,909,646 A
• Filed: 02/20/1996
• Issued: 06/01/1999
• Est. Priority Date: 02/22/1995
• Status: Expired due to Term

• Current Assignee: Future Link Systems, LLC (Vector Capital Corporation)
• Original Assignee: US Philips Corporation (Koninklijke Philips N.V.)
• Inventors: Deville, Yannick
• Primary Examiner(s): Vo, Nguyen
• Assistant Examiner(s): KINCAID, LESTER G
• Application Number:
  US08/602,535
• Time in Patent Office: 1,197 Days
• Field of Search: 455/313, 455/501, 455/296, 455/226.1, 455/67.3, 364/559, 364/574, 364/572, 364/570, 364/582, 364/728.03, 395/20, 395/23, 395/27, 381/94, 342/178, 342/153, 375/232, 375/346, 375/349, 375/350, 375/209, 375/210, 706/14, 706/22
• US Class Current: 455/313
• CPC Class Codes: G01S 3/043 Receivers
• CPC Class Codes: G01S 3/46 using antennas spaced apart and measuring phase or time difference between signals therefrom, i.e. path-difference systems
• CPC Class Codes: G01S 3/74 Multi-channel systems specially adapted for direction-finding, i.e. having a single antenna system capable of giving simultaneous indications of the directions of different signals systems in which the directions of different signals are determined sequentially and displayed simultaneously G01S3/04, G01S3/14
• CPC Class Codes: G01S 3/8006 Multi-channel systems specially adapted for direction-finding, i.e. having a single aerial system capable of giving simultaneous indications of the directions of different signals
• CPC Class Codes: G01S 3/808 using transducers spaced apart and measuring phase or time difference between signals therefrom, i.e. path-difference systems
• CPC Class Codes: G06F 18/21342 using statistical independence, i.e. minimising mutual information or maximising non-gaussianity

Abstract

Source separation system for processing input signals (E_i (t)) formed by instantaneous linear mixtures of primary signals (X_j (t)) that result from sources (S1-Sn) and for producing at least one estimated primary signal (x_k (t)) . The system includes separation apparatus (10) and characterization apparatus (15) which determine cumulants of the input signals for extracting estimated mixing coefficients (α_ij). The coefficients are transformed into separation coefficients (C_ki, d_ki) in the separation apparatus (10). The separation apparatus (10) may have a direct structure or a recursive structure.

Claims

1. A source separation system for processing input signals formed by instantaneous linear mixtures of primary signals that result from sources and for producing at least one estimated primary signal, which mixtures cause mixing coefficients to occur, the system comprising:

source separation means having a number of inputs for receiving the input signals and at least one output for producing the estimated primary signal, the source separation means adaptively determining separation coefficients which are used for extracting the estimated primary signals as a linear combination of the input signals; and

characterization means having input means coupled to receive the input signals for estimating the mixing coefficients based upon the input signals, and have at least one output for supplying estimated mixing coefficients;

and wherein the source separation means have at least one further input for receiving the estimated mixing coefficients and include means for transforming the estimated mixing coefficients into separation coefficients defining the linear combination of the input signals.

2. The system of claim 1, in which the characterization means comprise estimating means for making estimates of cumulants of a higher order than the second order based upon the input signals, and calculating means for calculating estimated coefficients for transforming the estimates of the cumulants into estimated mixing coefficients.

3. The system of claim 2, in which the estimated coefficient calculating means include a neuron network for transforming, in accordance with a predetermined transform function, the estimates of the cumulants into estimated mixing coefficients.

4. The system of claim 2, in which the separation means have a direct structure and include a transformation unit for transforming estimated mixing coefficients into separation coefficients adapted to the direct structure.

5. The system of claim 2, in which the separation means have a recursive structure and include a transformation unit for transforming estimated mixing coefficients into separation coefficients adapted to the recursive structure.

6. The system of claim 3, in which the separation means have a direct structure and include a transformation unit for transforming estimated mixing coefficients into separation coefficients adapted to the direct structure.

7. The system of claim 3, in which the separation means have a recursive structure and include a transformation unit for transforming estimated mixing coefficients into separation coefficients adapted to the recursive structure.

8. A source separation system for processing input signals formed by instantaneous-linear mixtures of primary signals that result from sources and for producing at least one estimated primary signal, which mixtures cause mixing coefficients to occur, the system comprising:

source separation means responsive to the input signals for providing the at least one estimated primary signal as a linear combination of the input signals; and

characterization means responsive to the input signals for estimating the mixing coefficients based upon the input signals so as to obtain estimated mixing coefficients;

and wherein the source separation means receive the estimated mixing coefficients and include means for transforming the estimated mixing coefficients into separation coefficients defining the linear combination of the input signals.

9. The system as claimed in claim 8 wherein the characterization means comprise estimating means for making estimates of cumulants of a higher order than the second order and based upon the input signals, and calculating means for calculating estimated coefficients for transforming the estimates of the cumulants into estimated mixing coefficients.

10. The system as claimed in claim 9 wherein the estimated coefficient calculating means include a neuron network for transforming, in accordance with a predetermined transform function, the estimates of the cumulants into estimated mixing coefficients.

11. The system as claimed in claim 9 wherein the separation means comprise a transformation unit for transforming estimated mixing coefficients into separation coefficients.

12. The system as claimed in claim 8 wherein at least one of the primary signal sources and the source separation system are located in a mobile vehicle.

13. The system as claimed in claim 8 wherein the characterization means comprise estimating means for making estimates of cumulants of the fourth order, and calculating means for calculating estimated coefficients for transforming the estimates of said cumulants into estimated mixing coefficients.