Manifold assisted channel estimation and demodulation for CDMA systems in fast fading environments
First Claim
Patent Images
1. A method of signal processing for a wireless communication system, comprising:
- receiving a set of de-spread vectors yi, wherein i=1 to N and N is the number of possible received symbols from a wireless transmission;
spatially correlating each of the de-spread vectors yi with an array manifold matrix C to produce a correlation matrix P, wherein each vector yi has M components, M being equal to the number of antenna output elements;
determining angle of arrival (AOA) information of each vector yi from matrix P;
determining angular distribution information from a collection of AOA information;
creating a subspace matrix from the matrix C and the AOA and angular distribution information;
projecting each vector yi into the subspace matrix to produce a set of N vectors zi;
finding a vector zmax from the one of vectors zi having the maximum energy;
averaging vector zmax to get a vector zavg; and
multiplying vector zavg with the subspace matrix to produce a beamforming coefficient vector w.
9 Assignments
0 Petitions
Accused Products
Abstract
Known antenna array manifold information is exploited to provide for fast and accurate channel estimation and demodulation on both the forward and reverse links, thereby increasing capacity in PCS and cellular CDMA networks that use adaptive antenna arrays. On the reverse link, an “extended” array manifold is used to assist the demodulator in maintaining a dynamic estimate of the spatial signature to use for beamforming and coherent demodulation. On the forward link, channel estimation is performed in the handset to provide a robust solution for beamforming.
57 Citations
48 Claims
-
1. A method of signal processing for a wireless communication system, comprising:
-
receiving a set of de-spread vectors yi, wherein i=1 to N and N is the number of possible received symbols from a wireless transmission;
spatially correlating each of the de-spread vectors yi with an array manifold matrix C to produce a correlation matrix P, wherein each vector yi has M components, M being equal to the number of antenna output elements;
determining angle of arrival (AOA) information of each vector yi from matrix P;
determining angular distribution information from a collection of AOA information;
creating a subspace matrix from the matrix C and the AOA and angular distribution information;
projecting each vector yi into the subspace matrix to produce a set of N vectors zi;
finding a vector zmax from the one of vectors zi having the maximum energy;
averaging vector zmax to get a vector zavg; and
multiplying vector zavg with the subspace matrix to produce a beamforming coefficient vector w. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21)
selecting the maximum-valued element from each row of the matrix P to form a K-element {tilde over (p)} vector;
averaging the {tilde over (p)} vector; and
selecting the maximum value of the averaged {tilde over (p)} vector and corresponding position.
-
-
15. The method of claim 1, wherein the system is a W-CDMA or CDMA2000 proposed system.
-
16. The method of claim 15, wherein the yi vectors are a combination of de-spread pilot vectors yp and de-spread data vectors yd.
-
17. The method of claim 15, wherein N is equal to 2.
-
18. The method of claim 16, wherein the combination comprises a real part and an imaginary part.
-
19. The method of claim 16, wherein the determining comprises finding the maximum element of the matrix P.
-
20. The method of claim 16, wherein the determining comprises finding the maximum element of the matrix resulting from averaging a set of P matrices.
-
21. The method of claim 16, wherein the determining comprises averaging over all the elements of the matrix P and finding the resulting maximum element and corresponding position.
-
22. A method of signal processing for a wireless communication system, comprising:
-
receiving, in a mobile station, a de-spread pilot signal vector yfp, wherein the vector yfp has M components, M being equal to the number of antenna output elements in the base station;
spatially correlating the vector yfp with an array manifold matrix C to produce a correlation vector pf;
determining angle of arrival (AOA) information for the vector yfp from vector pf;
determining angular distribution information from a collection of AOA information;
creating a subspace matrix from the matrix C and the AOA and angular distribution information;
projecting the vector yfp into the subspace matrix to produce a filtered version of the vector yfp;
calculating beamforming coefficients xi, i=1 to J, from the matrix C and the AOA and angular distribution information; and
determining the inner product, xiTyfp, of the vector yfp with each beamforming coefficient vector xi. - View Dependent Claims (23, 24)
-
-
25. A beamforming system for wireless communication, comprising:
-
a primary spatial correlator (PSC), wherein the PSC receives de-spread data vectors yi (i=1 to N) and signals representing an array manifold matrix C and transmits signals representing spatial information of the vectors yi;
each vector yi having M components, M being equal to the number of antenna output elements and N being equal to the number of possible symbols from a wireless transmission;
a digital signal processor (DSP) coupled to the PSC, wherein the DSP determines additional spatial information and determines a subspace containing most of the energy from the vectors yi; and
an extended manifold processor (EMP) coupled to the DSP and the vectors yi, wherein the EMP projects signals representing each vector yi into the subspace to generate projection coefficient vectors zi and outputs signals representing a beamforming coefficient vector w. - View Dependent Claims (26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45)
first selection circuitry for selecting the maximum-valued element from each row of the matrix P to form a K-element {tilde over (p)} vector;
averaging circuitry for averaging the {tilde over (p)} vector; and
second selection circuitry for selecting the maximum value of the averaged {tilde over (p)} vector and corresponding position.
-
-
41. The system of claim 28, wherein the system is a W-CDMA or CDMA2000 proposed system.
-
42. The system of claim 41, further comprising M data estimators, each data estimator receiving a de-spread pilot vector yp and de-spread data vector yd to form combinations of yp and yd to use as vectors yi.
-
43. The system of claim 42, wherein:
-
the PSC further comprises a receiver for receiving a set of vectors yi from the data estimators and a transmitter for transmitting spatial information of the vectors yi; and
the DSP is adapted to determine a subspace containing most of the energy from the vectors yi.
-
-
44. The system of claim 41, wherein N is equal to 2.
-
45. The system of claim 41, further comprising a beamformer coupled to the EMP and the de-spread data vectors yd, wherein the beamformer determines the inner product, wHyd, of each vector yd with the vector w.
-
46. A demodulation system for wireless communication comprising a plurality of manifold assisted demodulator (MAD) fingers, a first one of the plurality of MAD fingers comprising:
-
a first block for de-spreading pilot and data signals and generating a de-spread pilot vector yp and a de-spread data vector yd;
a first data estimator block that forms complex valued combination vectors, yi, of the de-spread pilot and data signal vectors;
a first primary spatial correlator (PSC), wherein the PSC receives the de-spread data vectors yi (i=1 to N) and an array manifold matrix C and transmits spatial information of the vectors yi, each vector yi having M components, M being equal to the number of antenna output elements and N being equal to the number of possible symbols from a wireless transmission;
a first extended manifold processor (EMP), wherein the first EMP receives the vectors yi and projects each vector yi into a subspace to generate projection coefficient vectors zi and outputs a beamforming coefficient vector w; and
a first beamformer coupled to the first EMP and the data signal vectors yd, wherein the first beamformer determines the inner product, wHyd, of each vector yd with the vector w. - View Dependent Claims (47, 48)
a second block for de-spreading the pilot and data signals and generating de-spread pilot vector yp and de-spread data vector yd;
a second data estimator block that forms the complex valued combination vectors, yi, of the de-spread pilot and data signal vectors; and
a second primary spatial correlator (PSC), wherein the second PSC receives the de-spread data vectors yi (i=1 to N) and the array manifold matrix C and transmits spatial information of the vectors yi, each vector yi having M components, M being equal to the number of antenna output elements and N being equal to the number of possible symbols from a wireless transmission.
-
-
48. The system of claim 47, wherein a third one of the plurality of MAD fingers consists of:
-
a third block for de-spreading the pilot and data signals and generating de-spread pilot vector yp and de-spread data vector yd;
a third data estimator block that forms the complex valued combination vectors, yi, of the de-spread pilot and data signal vectors; and
a second extended manifold processor (EMP), wherein the second EMP receives the vectors yi and projects each vector yi into a subspace to generate projection coefficient vectors zi and outputs a beamforming coefficient vector w; and
a second beamformer coupled to the second EMP and the data signal vectors yd, wherein the first beamformer determines the inner product, wHyd, of each vector yd with the vector w.
-
Specification