SNR Dependent Channel Tracking For SUN OFDM

• US 20140307841A1
• Filed: 12/31/2013
• Published: 10/16/2014
• Est. Priority Date: 04/10/2013
• Status: Active Grant

• Current Assignee: Texas Instruments, Inc.
• Original Assignee: Texas Instruments, Inc.
• Inventors: BATRA, Anuj, SCHMIDL, Timothy Mark, HOSHYAR, Reza
• Granted Patent: US 8,964,916 B2
• Time in Patent Office: Days
• US Class Current: 375/350
• CPC Class Codes: H04B 17/336 Signal-to-interference ratio [SIR] or carrier-to-interference ratio [CIR]
• CPC Class Codes: H04L 27/2647 Arrangements specific to the receiver only equalisation H04L27/01

Abstract

An orthogonal frequency-division multiplexed (OFDM)-based receiver for channel tracking with signal-to-noise ratio dependent parameters that includes a memory; and a signal processor, coupled to the memory. The signal processor to estimate a signal-to-noise ratio (SNR) for a received packet of OFDM symbols and determine an SNR region in which the SNR estimate falls, wherein the signal processor implements a different set of finite impulse response (FIR) filter coefficients for each SNR region.

Claims

1. An orthogonal frequency-division multiplexed (OFDM)-based receiver for channel tracking with signal-to-noise ratio dependent parameters, comprising:

a memory; and

a signal processor, coupled to the memory, to estimate a signal-to-noise ratio (SNR) for a received packet of OFDM symbols and determine an SNR region in which the SNR estimate falls, wherein the signal processor implements a different set of finite impulse response (FIR) filter coefficients for each SNR region.

2. The receiver of claim 1, wherein the signal processor derives the FIR filter coefficients for each SNR region based on a nominal SNR value for each SNR region and an SNR-dependent timing error.

3. The receiver of claim 1, wherein the signal processor implements a different forgetting factor for each SNR region, wherein the forgetting factor is implemented by the signal processor to smooth channel estimates along time-domain for the OFDM symbols.

4. The receiver of claim 3, wherein the different sets of FIR filter coefficients and the forgetting factor for each SNR region are stored in the memory.

5. The receiver of claim 1, wherein the signal processor generates an estimate for a complex gain for each tone of each OFDM symbol in the OFDM packet using a FIR filter with the corresponding set of FIR filter coefficients for an SNR region.

6. The receiver of claim 5, wherein the signal processor generates the estimate for the gain for odd subcarriers of each OFDM symbol using a first set of FIR filter coefficients.

7. The receiver of claim 5, wherein the signal processor generates the estimate for the gain for even subcarriers of each OFDM symbol using a second set of FIR filter coefficients.

8. The receiver of claim 5, wherein the signal processor uses a forgetting factor to generate a refined estimate of the gain for each OFDM symbol of the OFDM packet.

9. A smart utility network device, comprising:

a receiver configured to receive orthogonal frequency-division multiplexed (OFDM) symbols; and

a signal processor coupled to the receiver, the signal processor configured to;

estimate a signal-to-noise ratio (SNR) of a packet of OFDM signals based on a long training field (LTF) symbol included in the OFDM packet;

determine a SNR region in which the SNR estimate falls, wherein there are at least two SNR regions and each SNR region is associated with a set of finite impulse response (FIR) filter coefficients and a forgetting factor; and

based on the SNR region, generate estimates of a channel gain for each tone of each OFDM symbol in the OFDM packet with the SNR region'"'"'s associated FIR filter coefficients and the forgetting factor.

10. The device of claim 9, further comprising a memory coupled to the signal processor to store the FIR filter coefficients and the forgetting factor associated with each SNR region.

11. The transceiver of claim 9, wherein the FIR filter coefficients for each SNR region are pre-determined based on a nominal SNR value associated with each SNR region.

12. The device of claim 11, wherein the FIR filter coefficients for each SNR region are further based on an SNR-based timing error.

13. The device of claim 9, wherein the FIR filter coefficients associated with each SNR region comprise a first set of coefficients and a second set of coefficients.

14. The device of claim 13, wherein the first set of coefficients is used by the FIR filter to generate an improved estimate of a channel gain for odd subcarriers of the OFDM symbols.

15. The device of claim 13, wherein the second set of coefficients is used by the FIR filter to generate an improved estimate channel gain for even subcarriers of the OFDM symbols.

16. The device of claim 9, wherein the forgetting factor is used to improve the channel gain along a time domain of the OFDM symbols.

17. The device of claim 9, wherein there are three SNR regions.

18. A method for signal-to-noise ratio (SNR) dependent channel gain tracking of an orthogonal frequency-domain multiplexing (OFDM)-based receiver, comprising:

receiving, by a receiver, a packet of OFDM symbols, wherein a preamble of the packet includes a long training field (LTF) symbol;

estimating, by the receiver, a SNR value with the LTF symbol;

determining, by the receiver, a SNR region in which the SNR estimate falls; and

generating, by the receiver, estimates of a channel gain for tones of the OFDM symbols based on a set of finite impulse response (FIR) filter coefficients associated with the SNR region and smoothing the tones of the OFDM symbols along a time-domain using a forgetting factor associated with the SNR region.

19. The method of claim 18, further comprising:

retrieving, by the receiver, the set of finite impulse response (FIR) filter coefficients and the forgetting factor associated with the SNR region in which the SNR estimate fell.

20. The method of claim 18, wherein there are at least two SNR regions.