Transmitter and the Method Therefor

• US 20100166105A1
• Filed: 12/21/2009
• Published: 07/01/2010
• Est. Priority Date: 12/31/2008
• Status: Active Grant

• Current Assignee: Industrial Technology Research Institute
• Original Assignee: Industrial Technology Research Institute
• Inventors: Wu, Jia-Hao, Lin, Ming-Che, Shih, Yun-Yi, Ting, Pang-An
• Granted Patent: US 8,385,442 B2
• Time in Patent Office: Days
• US Class Current: 375/295
• CPC Class Codes: H04L 27/262 Reduction thereof by selection of pilot symbols

Abstract

A transmitter and the method therefor are provided. The transmitter is applied in an orthogonal frequency division multiplexing (OFDM) based or other multicarrier communication systems. The transmitter uses N sub-carriers. Among the N sub-carriers, L sub-carriers (or tones) are reserved for PAPR reduction. L is less than N. The method includes the following steps. Original frequency domain data symbols are generated and fed into the reserved-tone symbol generation unit. The reserved-tone symbols are generated by using pure frequency domain signal processing. Then the reserved-tone symbols are combined with original data symbols to generate combined PAPR reduction signal. The resultant PAPR reduction signal can be fed into reserved-tone symbol generation unit iteratively to obtain updated reserved-tone symbols. For proper iterations, the original data symbols combined with L frequency domain reserved-tone symbols will result in lower PAPR transmit signal.

Claims

1. A method for reducing a peak-to-average power ratio (PAPR), the method being applied by using N sub-carriers, L sub-carriers of the N sub-carriers being reserved sub-carriers, L being less than N, the method comprising:

receiving a data signal, the envelope power of the data signal having a constant power and a fluctuant power;

generating L reserved sub-carrier symbols of the L sub-carriers located in the reserved sub-carriers;

modifying the L reserved sub-carrier symbols according to the data signal, and generating a to-be-combined signal accordingly; and

combining the to-be-combined signal into the data signal in frequency domain.

2. The method according to claim 1, wherein the data signal has N data symbols located in the N sub-carriers, the N data symbols correspond to an aperiodic auto-correlation function (AACF) vector, the L reserved sub-carrier symbols correspond to a reserved sub-carrier symbol vector, and in the step of modifying the L reserved sub-carrier symbols, the L reserved sub-carrier symbols are modified according to a gradient or any other algorithmic methods of the square of the absolute value of the AACF vector at the reserved sub-carrier symbol vector, or modified according to the absolute value for the AACF vector to the p^th power, wherein p is larger than 0.

3. The method according to claim 2, wherein a set of sub-carrier indices of the L sub-carriers of the reserved sub-carriers is denoted as U, wherein U={k₀, k₁, . . . k_L−

1}, the N data symbols of the data signal are denoted as X₀, X₁, . . . X<sub>N−

1</sub> and the data vector is defined as X=[X₀, X₁, . . . , X<sub>N−

1</sub>]^T, wherein X_i is the i^th data symbol, and i is an integer between 0 to N−

1, before the data is combined into the reserved sub-carrier symbol, if iε

U, X_i=0 the AACF vector is denoted as ρ

, the L reserved sub-carrier symbols are denoted as X_k0, X_k1, . . . , X<sub>kL−

1</sub>, the reserved sub-carrier symbol vector is denoted as X_R=[X_k0, X_k1, . . . , X<sub>kL−

1</sub>]^T, X_R⁽ⁿ⁺¹⁾ denotes an n+1^th modified X_R, and X_R⁽ⁿ⁾ denotes an n+1^th pre-modified X_R, then it is derived that

$\mathrm{\rho <br><br>}={\left[\mathrm{\rho <br><br>}<br><br>\left(1\right),\mathrm{\rho <br><br>}<br><br>\left(2\right),\mathrm{\dots <br><br>}<br><br>,\mathrm{\rho <br><br>}<br><br>\left(N-1\right)\right]}^T,\mathrm{and}$ $X_R^{\left(n+1\right)}=X_R^{\left(n\right)}-\mathrm{\mu <br><br>}<br><br>{\nabla <br><br>}_{X_R}<br><br>{<br><br>\mathrm{\rho <br><br>}<br><br>}^2<br><br>{<br><br>}_{X_R^{\left(n\right)}},<br><br>\mathrm{wherein}<br><br><br><br>\mathrm{\rho <br><br>}<br><br>\left(u\right)=\underset{k=0}{\overset{N-u-1}{\sum <br><br>}}<br><br><br><br>X_k<br><br>X_{k+u}^{*}<br><br><br><br>\mathrm{for}<br><br><br><br>1\le <br><br>u\le <br><br>N-1.$

4. The method according to claim 1, wherein the step of modifying the L reserved sub-carrier symbols comprises modifying the L reserved sub-carrier symbols by iteration means.

5. The method according to claim 1, wherein the L reserved sub-carrier symbols are divided into several groups of reserved sub-carrier symbols, and in the step of modifying the L reserved sub-carrier symbols, the several groups of reserved sub-carrier symbols are modified sequentially so as to modify the L reserved sub-carrier symbols.

6. The method according to claim 1, wherein the L reserved sub-carrier symbols are divided into several groups of reserved sub-carrier symbols, and in the step of modifying the L reserved sub-carrier symbols, the several groups of reserved sub-carrier symbols are modified in parallel so as to modify the L reserved sub-carrier symbols.

7. The method according to claim 1, wherein the transmitter comprises a filter, the filter has a frequency response function denoted as H, H_k denotes the frequency response coefficient at the k^th sub-carrier, the data signal has N data symbols located in N sub-carriers, and the step of modifying the L reserved sub-carrier symbols comprises the following steps:

multiplying the N data symbols with a corresponding H_k, respectively;

modifying the L reserved sub-carrier symbols according to the N data symbols being multiplied with H_k; and

multiplying the to-be-combined signal with 1/H_k;

wherein the to-be-combined signal being multiplied with 1/H_k is served as the to-be-combined signal for use in combining with the data signal.

8. A transmitter, applied in an orthogonal frequency division multiplexing (OFDM) based system, the transmitter using N sub-carriers, L sub-carriers of the N sub-carriers being reserved sub-carriers, L being less than N, the transmitter comprising:

a modulation unit, for outputting a data signal, the envelope power of the data signal having a constant power and a fluctuant power;

a reserved sub-carrier symbol generation unit, for generating L reserved sub-carrier symbols of the L sub-carriers located in the reserved sub-carriers, for modifying the L reserved sub-carrier symbols according to the data signal and generating a to-be-combined signal accordingly; and

a combination unit, for combining the to-be-combined signal into the data signal in frequency domain.

9. The transmitter according to claim 8, wherein the data signal has N data symbols located in the N sub-carriers, the N data symbols correspond to an aperiodic auto-correlation function (AACF) vector, the L reserved sub-carrier symbols correspond to a reserved sub-carrier symbol vector, and the modulation unit modifies the L reserved sub-carrier symbols according to a gradient or any other algorithmic methods of the square of the absolute value of the AACF vector at the reserved sub-carrier symbol vector, or according to the absolute value for the AACF vector to the p^th power, wherein p is larger than 0.

10. The transmitter according to claim 9, wherein a set of sub-carrier indices of the L sub-carriers of the reserved sub-carriers is denoted as U, wherein U={k₀, k₁, . . . , k_L−

1}, the N data symbols of the data signal are denoted as X₀, X₁, . . . X<sub>N−

1</sub> and the data vector is defined as X=[X₀, X₁, . . . , X<sub>N−

1</sub>]^T, wherein X_i is the i^th data symbol, and i is an integer between 0 to N−

1, before the data is combined into the reserved sub-carrier symbol, if iε

U, X_i=0, the AACF vector is denoted as ρ

, the L reserved sub-carrier symbols are denoted as X_k0, X_k1, . . . , X<sub>kL−

1</sub>, the reserved sub-carrier symbol vector is denoted as X_R=[X_k0, X_k1, . . . , X<sub>kL−

1</sub>]^T, X_R⁽ⁿ⁺¹⁾ denotes an n+1^th modified X_R, and X_R⁽ⁿ⁾ denotes an n+1^th pre-modified X_R, then it is derived that

$\mathrm{\rho <br><br>}={\left[\mathrm{\rho <br><br>}<br><br>\left(1\right),\mathrm{\rho <br><br>}<br><br>\left(2\right),\mathrm{\dots <br><br>}<br><br>,\mathrm{\rho <br><br>}<br><br>\left(N-1\right)\right]}^T,\mathrm{and}$ $X_R^{\left(n+1\right)}=X_R^{\left(n\right)}-\mathrm{\mu <br><br>}<br><br>{\nabla <br><br>}_{X_R}<br><br>{<br><br>\mathrm{\rho <br><br>}<br><br>}^2<br><br>{<br><br>}_{X_R^{\left(n\right)}},<br><br>\mathrm{wherein}<br><br><br><br>\mathrm{\rho <br><br>}<br><br>\left(u\right)=\underset{k=0}{\overset{N-u-1}{\sum <br><br>}}<br><br><br><br>X_k<br><br>X_{k+u}^{*}<br><br><br><br>\mathrm{for}<br><br><br><br>1\le <br><br>u\le <br><br>N-1.$

11. The transmitter according to claim 8, wherein the modulation unit modifies the L reserved sub-carrier symbols by iteration means.

12. The transmitter according to claim 8, wherein the L reserved sub-carrier symbols comprise several groups of reserved sub-carrier symbols, and the modulation unit modifies the L reserved sub-carrier symbols sequentially, so as to modify the L reserved sub-carrier symbols.

13. The transmitter according to claim 8, wherein the L reserved sub-carrier symbols comprise several groups of reserved sub-carrier symbols, and the modulation unit modifies the L reserved sub-carrier symbols in parallel, so as to modify the L reserved sub-carrier symbols.

14. The transmitter according to claim 8, wherein the data signal has N data symbols located in N sub-carriers, and the transmitter further comprises:

a filter, having a frequency response function denoted as H, wherein H_k denotes the frequency response coefficient at the k^th sub-carrier

a first multiplier, for multiplying the N data symbols with a corresponding H_k, respectively, wherein the reserved sub-carrier symbol generation unit modifies the L reserved sub-carrier symbols according to the N data symbols being multiplied with H_k, and generating the to-be-combined signal accordingly;

a second multiplier, for multiplying the to-be-combined signal with 1/H_k, and for outputting the multiplied signal to the combination unit; and

a conversion unit, for converting the data signal being processed by the combination unit into a time-domain transmitting signal, and for outputting the transmitting signal to the filter.