Methods for fast and low-power UWB IR baseband receiver synchronization

• US 8,804,880 B2
• Filed: 07/13/2009
• Issued: 08/12/2014
• Est. Priority Date: 07/11/2008
• Status: Expired due to Fees

• Current Assignee: Stichting Imec Nederland
• Original Assignee: IMEC International
• Inventors: Rousseaux, Olivier, Huang, Li, Berekovic, Mladen, Neirynck, Dries
• Primary Examiner(s): Garcia, Santiago
• Application Number:
  US13/003,378
• Publication Number:
  US 20110150044A1
• Time in Patent Office: 1,856 Days
• Field of Search: 375/343
• US Class Current: 375/343
• CPC Class Codes: H04B 1/71637 Receiver aspects H04B1/7183 takes precedence
• CPC Class Codes: H04B 1/7183 Synchronisation

Abstract

The invention relates to methods for synchronizing a device to a signal containing a train of pulses representing a programmable number of repetitions of a predetermined code. Pulse position and code phase are searched for. In a first aspect, a programmable number of samples is taken per pulse frame in function of the available number of repetitions of said predetermined code. In a second aspect, the method comprises a signal detection stage from which, after performing a confirmation stage, information can be kept for a subsequent stage. In a third aspect, only a limited number of rotated versions of the predetermined code are checked, using a presumed code phase which is kept from a preceding stage.

Claims

1. A method, comprising:

receiving a predetermined code phase from a course synchronization stage;

receiving a signal comprising at least one pulse and a plurality of pulse frames, wherein each pulse frame comprises a number L of discrete possible pulse positions and the at least one pulse is located at an actual pulse position within the plurality of pulse frames;

applying a varying sampling delay to repeatedly sample the signal to obtain a number N of samples, wherein N is less than or equal to L, and wherein the number N of samples includes at least one sample at each of the L discrete possible pulse positions;

correlating the number N of samples with at least one rotated version of a predetermined code to determine a plurality of correlations, wherein the at least one rotated version is selected from a plurality of predetermined limited rotated versions based on the predetermined code phase, and wherein the plurality of predetermined limited rotated versions include fewer than all possible rotated versions; and

based on the correlations, determining the actual pulse position and a code phase of the signal, wherein the code phase is a phase of the signal with respect to the predetermined code.

2. The method of claim 1, wherein the at least one rotated version of the predetermined code being selected based on the predetermined code phase comprises rotating the predetermined code by the code phase.

3. The method of claim 2, wherein the at least one rotated version of the predetermined code comprises the predetermined code rotated by the code phase, the predetermined code rotated by the code phase plus one, and the predetermined code rotated by the code phase minus one.

4. The method of claim 1, wherein the at least one rotated version of the predetermined code comprises a number of rotated versions, the number being programmable as a function of a channel delay spread of a signal over which the signal is received.

5. The method of claim 1, wherein correlating the number N of samples with the at least one rotated version of the predetermined code to determine a plurality of correlations comprises:

storing the number N of samples as serial data;

converting the number N of samples to parallel data; and

correlating the parallel data with the at least one rotated version of the predetermined code to determine the plurality of correlations.

6. The method of claim 5, wherein determining the actual pulse position and the code phase of the signal comprises selecting a maximum among the correlations.

7. The method of claim 6, wherein the actual pulse position comprises the maximum.

8. The method of claim 1, wherein the course synchronization stage is a signal detection stage that occurs substantially at the same time as the signal is received.

9. The method of claim 1, wherein the course synchronization stage is a synchronization stage of a previous pulse of the signal.

10. A method of synchronizing a device to a signal, comprising:

selecting a number N as a function of an available number of repetitions of a predetermined code;

receiving a signal comprising a plurality of pulse frames and a plurality of pulses that represent N, wherein each pulse frame comprises a number L of discrete possible pulse positions, wherein at least one pulse in the plurality of pulses is located at an actual pulse position within the plurality of pulse frames, and wherein N is less than or equal to L;

sampling the signal to obtain N samples, wherein the N samples comprise at least one sample at each of the L discrete possible pulse positions;

correlating the N samples with a plurality of rotated versions of the predetermined code to determine a first correlation;

comparing the first correlation value with a noise threshold to determine if a signal is present;

sampling the signal to obtain N times two samples, wherein the N times two samples comprise at least one sample at each of the L times two discrete possible pulse positions;

correlating the N times two samples with a plurality of rotated versions of the predetermined code to determine a second correlation; and

based on the second correlation, determining the actual pulse position of the at least one pulse and a code phase of the signal, wherein the code phase is a phase of the signal with respect to the predetermined code.

11. The method of claim 10, wherein correlating the N samples with the plurality of rotated versions of the predetermined code to determine the first correlation comprises:

storing the N samples as serial data;

converting the N samples to parallel data; and

correlating the parallel data with the plurality of rotated versions of the predetermined code to determine the first correlation.

12. The method of claim 10, wherein determining the actual pulse position and the code phase of the signal comprises selecting a maximum among the at least one correlations.

13. The method of claim 10, wherein the N samples are taken at non-adjacent pulse positions.

14. The method of claim 10, wherein the N samples are taken at adjacent pulse positions.

15. The method of claim 10, further comprising increasing N and repeating the sampling, correlating, and determining steps with the increased N.

16. A method, comprising:

selecting a sampling delay;

receiving a signal comprising a plurality of pulse frames and a plurality of pulses that represent a number N selected as a function of an available number of repetitions of a predetermined code, wherein each pulse frame comprises a number L of discrete possible pulse position and at least one pulse in the plurality of pulses is located at an actual pulse position within the plurality of pulse frames, and wherein N is less than or equal to L;

sampling the signal to obtain a first set of N samples covering at least the length of the predetermined code, wherein the first set of samples comprises at least one sample at each of the L discrete possible pulse positions;

correlating the first set of samples with rotated versions of the predetermined code to obtain a first set of correlation values;

comparing each correlation value in the set of correlation values with a noise threshold to determine if a signal is present;

if a signal is not present, repeating the method with a new sampling delay; and

if a signal is present, confirming the presence of the signal, wherein confirming the presence of the signal comprises;

resampling the signal with the sampling delay to obtain a second set of samples covering at least the length of the predetermined code; and

correlating the second set of values with the rotated versions of the predetermined code to determine if a signal is present.

17. A method, comprising:

selecting a sampling delay;

receiving a signal comprising a plurality of pulse frames and a plurality of pulses that represent a number N selected as a function of an available number of repetitions of a predetermined code, wherein each pulse frame comprises a number L of discrete possible pulse position and at least one pulse in the plurality of pulses is located at an actual pulse position within the plurality of pulse frames, and wherein N is less than or equal to L;

sampling the signal to obtain a first set of N samples covering at least the length of the predetermined code, wherein the first set of samples comprises at least one sample at each of the L discrete possible pulse positions;

correlating the first set of samples with rotated versions of the predetermined code to obtain a first set of correlation values,

comparing each correlation value in the set of correlation values with a noise threshold to determine if a signal is present;

if a signal is not present, repeating the method with a new sampling delay; and

if a signal is present, determining the actual pulse position and a code phase of the signal, wherein the code phase is a phase of the signal with respect to the predetermined code.

18. The method of claim 17, further comprising, if the signal is present, outputting a presumed code phase, wherein the presumed code phase is related to at least one of (i) the rotated versions of the predetermined code and (ii) the sampling delay.