Estimating channel parameters in multi-input, multi-output (MIMO) systems

• US 7,310,304 B2
• Filed: 04/24/2002
• Issued: 12/18/2007
• Est. Priority Date: 04/24/2001
• Status: Expired due to Term

• Current Assignee: American Patents LLC
• Original Assignee: BAE Systems Information and Electronic Systems Integration Incorporated (BAE Systems Plc)
• Inventors: Stuber, Gordon L., Mody, Apurva N.
• Primary Examiner(s): Chan, Wing
• Assistant Examiner(s): Mais, Mark A.
• Application Number:
  US10/128,756
• Publication Number:
  US 20020181390A1
• Time in Patent Office: 2,064 Days
• Field of Search: 370204-210, 370/503, 370/342, 370/335, 370/479, 370/404, 370/480, 370/488, 370/328, 370/329, 370/330, 370/428, 370/281, 375/267, 375/135, 375144-150, 375/285, 375/130, 375/140, 375/142, 375/346, 375/348, 375/350, 375/343, 375/354, 375/355, 375/316, 375/219, 375/222, 375/283, 375/299, 375/362, 375/366, 375/229, 375/231, 375/232, 375/279, 375/308, 375/329, 375/260, 375/261, 375/295, 375/298, 375/349, 375/340, 375/347, 375/358, 375/210
• US Class Current: 370/208
• CPC Class Codes: H04B 7/0413 MIMO systems
• CPC Class Codes: H04L 1/004 by using forward error control H04L1/0618 takes precedence; coding, decoding or code conversion, for error detection or correction H03M13/00
• CPC Class Codes: H04L 1/06 using space diversity
• CPC Class Codes: H04L 1/20 using signal quality detector
• CPC Class Codes: H04L 25/0226 sounding signals per se
• CPC Class Codes: H04L 27/2602 Signal structure
• CPC Class Codes: H04L 27/26035 Maintenance of orthogonality, e.g. for signals exchanged between cells or users, or by using covering codes or sequences using different training sequence per antenna H04B7/0684; code allocation H04J13/16
• CPC Class Codes: H04L 27/2647 Arrangements specific to the receiver only equalisation H04L27/01

Abstract

In a communication system, and in particular a wireless Orthogonal Frequency Division Multiplexing (OFDM) communication system, the present invention provides systems for estimating parameters of a channel across which a signal is transmitted. The present invention may be used in a Multi-Input, Multi-Output (MIMO) system in which the data is transmitted from any number of transmitting antennas and received by any number of receiving antennas. The number of transmitting and receiving antennas does not necessarily have to be the same. Circuitry is provided for calculating parameters indicative of the characteristics of the communication channel. Estimates of channel parameters may include channel estimates and noise variance estimates.

Claims

1. A transmitter in an Orthogonal Frequency Division Multiplexing (OFDM) system, the transmitter comprising:

an encoder configured to process data to be transmitted within the OFDM system, the encoder further configured to separate the data onto one or more transmit diversity branches (TDBs);

one or more OFDM modulators, each OFDM modulator connected to a respective TDB, each OFDM modulator configured to produce a frame including a plurality of data symbols, a training structure, and cyclic prefixes inserted among the data symbols; and

one or more transmitting antennas in communication with the one or more OFDM modulators, respectively, each transmitting antenna configured to transmit the respective frame over a channel;

wherein the training structure of each frame includes a predetermined signal transmission matrix at a respective sub-channel, each training structure adjusted to have a substantially constant amplitude in a time domain, and the cyclic prefixes are further inserted within the training symbol, and wherein the cyclic prefixes within the training symbol are longer than the cyclic prefixes among the data symbols, thereby countering an extended channel impulse response and improving synchronization performance.

2. The transmitter of claim 1, wherein the transmitter is incorporated within a Multi-input, Multi-Output (MIMO) system.

3. The transmitter of claim 2, wherein the training structure includes information allowing a receiver device to perform synchronization and parameter estimation for calibrating the MIMO system.

4. The transmitter of claim 1, wherein the cyclic prefixes protect the data symbols against Inter Symbol Interference (ISI).

5. The transmitter of claim 1, wherein the encoder comprises:

a channel encoder;

a symbol mapper in communication with an output of the channel encoder;

a space-time processor in communication with output of the symbol mapper, the space-time processor configured to separate the data into the one or more TDB'"'"'s; and

a pilot/training symbol inserter, the pilot/training symbol inserter configured to insert pilot symbols and training symbols onto the TBD'"'"'s.

6. The transmitter of claim 1, wherein each of the OFDM modulators comprises:

a serial-to parallel converter;

an inverse discrete Fourier transform (IDFT) stage in communication with an output of the serial-to-parallel converter;

a cyclic prefix inserter in communication with an output of the IDFT state;

a parallel-to-serial converter in communication with an output of the cyclic prefix inserter;

a digital-to-analog converter (DAC) in communication with an output of the parallel-to-serial converter;

a local oscillator;

a mixer having a first input and a second input, the first input in communication with an output of the DC, the second input in communication with an output of the local oscillator; and

an amplifier in communication with an output of the mixer.

7. The transmitter of claim 1, wherein the training structure includes a signal transmission matrix resembling an existing space-time block code.

8. The transmitter of claim 1, wherein the training structure includes a signal transmission matrix having an existing pseudo inverse matrix.

9. The transmitter of claim 1, wherein a length of the training structure is adjustable.

10. The transmitter of claim 1, wherein a length of the training structure is equal to a single OFDM symbol using alternate tones transmitted from different antennas at substantially the same time.

11. A method comprising:

receiving one or more frames across a channel from one or more transmitting antennas, each frame including a plurality of data symbols, training symbols, and cyclic prefixes inserted among the data and training symbols;

synchronizing the one or more received frames in a time domain;

synchronizing the one or more received frames on a frequency domain; and

estimating parameters of the channel by a process involving one of calculating a channel coefficient matrix, calculating a signal to noise ratio, calculating a noise variance, calculating a time correlation, and calculating a frequency correlation, wherein estimating the parameters of the channel further comprises calculating fine channel estimates, and calculating the fine channel estimates comprises;

zero-padding in the frequency domain to eliminate certain frequency components from the OFDM signal;

calculating coarse channel estimates using a least squares method;

employing a zero-forcing method to calculate the coarse channel estimates;

interpolating the channel estimates in the frequency domain;

extrapolating frequency domain values;

reducing the mean squared error of the coarse channel estimates; and

discarding the unused tones.

12. The method of claim 11, wherein the one or more frames are transmitted in a Multi-Input, Multi-Output (MIMO) orthogonal Frequency Division Multiplexing (OFDM) system.

13. The method of claim 12, further comprising:

maintaining orthogonality by making a space-time matrix of the training symbols at each sub-carrier frequency orthogonal or unitary.

14. The method of claim 11, wherein reducing the mean squared error comprises at least one of a linear prediction procedure, a procedure for averaging the coarse channel estimates in the frequency domain, a procedure for averaging the coarse channel estimates in the time domain, a procedure for utilizing feedback equalizer, a decision directed procedure, a procedure utilizing a feedback from the channel decoder, a procedure for filtering the coarse channel estimates in the frequency domain, and a procedure for processing the coarse channel estimates in the time or frequency domain.

15. The method of claim 11, wherein estimating the parameters of the channel comprises calculating a noise variance.

16. The method of claim 15, wherein calculating the noise variance comprises calculating noise variance estimates using a unitary signal transmission matrix.

17. The method of claim 16, wherein calculating the noise variance estimates comprises calculating the noise variance estimates directly from time domain coefficients of coarse channel estimates.

18. The method of claim 15, wherein calculating the noise variance comprises calculating noise variance estimates using a non-unitary signal transmission matrix.

19. The method of claim 18, wherein calculating the noise variance estimates comprises:

calculating a noise term from coarse channel estimates and fine channel estimates;

multiplying the noise term by signal transmission matrix;

calculating time domain noise estimate vectors; and

determining a noise variance estimation from the time domain noise estimate vectors.

20. The method of claim 11, further comprising:

solving a system of linear equations; and

calculating a coarse channel estimation from the system of linear equations.

21. The method of claim 11, wherein the number of training symbols is equal to the number of transmitting antennas.

22. The method of claim 11, wherein the number of training symbols is greater than the number of transmitting antennas.