Granted Patent B1

US 7,136,423 · App. 10/263,732 · Granted Nov 14, 2006

Abstract

Claims (41)

1. A method for reducing the peak-to-average ratio of a single-carrier or a multi-carrier modulated signal, the method comprising:

storing an original discrete time signal;

extending the original discrete time signal by adding a suffix to form an original discrete time extended signal;

interpolating and upsampling the original discrete time extended signal at a fractional sampling rate to form an extended interpolated signal;

generating sequences basis (Q) by partitioning available bins into a number of subsets, and taking N-dimensional inverse Fourier transforms of each subset at a rate of T;

locally generating a PAR lowering sub-sequence (J) at T,

in a time-domain, locally combining the PAR lowering sub-sequence and highest peak source at a scale of one extended symbol for multi-carrier modulation and at one or more of symbols for single-carrier modulation;

searching for a signal peak that resulted from the combination of the PAR lowering sub-sequence and highest peak source;

determining if the peak that resulted from the combination of the PAR lowering sub-sequence and highest peak source is an optimal peak that yields a low PAR; and

combining the optimal peak that yields a low PAR and the original interpolated and upsampled discrete time signal that has been stored.

2. The peak-to-average ratio reduction method of claim 1 , where the original signal is extended by appending new samples at a rate of T, such that new samples will be repeating samples that were available most recently.

3. The peak-to-average ratio reduction method of claim 1 , where the original extended signal is interpolated by a factor of 2.

4. The peak-to-average ratio reduction method of claim 1 , where a threshold is determined by taking the standard deviation of the signal samples.

5. The peak-to-average ratio reduction method of claim 1 , where peak detection and mitigation is done at an oversampling rate of T/K, where T is a regular sampling rate and K is an integer.

6. The peak-to-average ratio reduction method of claim 1 , where the locally generated PAR lowering sequence (J) at T is a subset of the sequence basis (Q) calculated as

J=2 Q .

7. The peak-to-average ratio reduction method of claim 1 , where extending the signal by adding the suffix anticipates a high PAR due to boundaries time sample values between two successive samples.

8. The peak-to-average ratio reduction method of claim 1 , where the sampling rate is at the sampling rate of T/K.

9. The peak-to-average ratio reduction method of claim 1 , where the PAR reduction is made at a transmitter.

10. An article of manufacture for enabling reducing the peak-to-average ratio of a single-carrier or a multi-carrier modulated signal, the article of manufacture comprising:

at least one processor readable carrier; and

instruction carried on the at least one carrier;

wherein the instructions are configured to be readable from the at least one carrier by at least one processor and thereby causing at least one processor to operate so as to:

store an original discrete time signal;

extend the original discrete time signal by adding a suffix to form an original discrete time extended signal;

interpolate and upsample the original discrete time extended signal at a fractional sampling rate to form an extended interpolated signal;

generate sequence basis (Q) by partitioning available bins into a number of subsets, and taking N-dimensional inverse Fourier transforms of each subset at a rate of T;

generate a PAR lowering sub-sequence (J) at T/K,

in the time-domain combine the PAR lowering sub-sequence and highest peak source at a scale of one extended symbol for multi-carrier modulation and at one or more of symbols for single-carrier modulation;

search for a signal peak that resulted from the combination of the PAR lowering sub-sequence and highest peak source;

determine if the peak that resulted from the combination of the PAR lowering sub-sequence and highest peak source is an optimal peak that yields a low PAR;

combine optimal peak that yields a low PAR and the original interpolated and upsampled discrete time signal that has been stored.

11. The article of manufacture of claim 10 , wherein the original signal is extended by appending new samples at the rate of T, such that new samples will be repeating samples that were available most recently.

12. The article of manufacture of claim 10 , wherein the original extended signal is interpolated by a factor of 2.

13. The article of manufacture of claim 10 , wherein a threshold is determined by taking a standard deviation of the signal samples.

14. The article of manufacture of claim 10 , where the peak detection and mitigation is done at an oversampling rate of T/K, where T is a regular sampling rate and K is an integer.

15. The article of manufacture of claim 10 , where the locally generated PAR lowering sequence (J) at T is a subset of the sequence basis (Q) calculated as

J=2 Q .

16. The article of manufacture of claim 10 , where extending the signal by adding the suffix anticipates a high PAR due to boundaries time sample values between two successive samples.

17. The article of manufacture of claim 10 , where the sampling rate is at a sampling rate of T/K.

18. The article of manufacture of claim 10 , where the PAR reduction is made at a transmitter.

Assignments (10)