WO2008069392A2

WO2008069392A2 - Method and apparatus for estimating noise variance

Info

Publication number: WO2008069392A2
Application number: PCT/KR2007/003479
Authority: WO
Inventors: Gi-Yoon Park; Jin-Kyu Choi; Tae-Joong Kim; Hyeong-Jun Park
Original assignee: Electronics And Telecommunications Research Institute; Samsung Electronics Co., Ltd
Priority date: 2006-12-07
Filing date: 2007-07-18
Publication date: 2008-06-12
Also published as: WO2008069392A3

Abstract

The present invention relates to a method of and an apparatus for estimating the noise variance in a receiver. When a first symbol vector on the frequency domain is received, a windowing controller (110) sets all symbols except pilot symbols of the first symbol vector to zero to make a second symbol vector. An inverse discrete Fourier transformer (120) transforms the second symbol vector on the frequency domain into a sample vector on the time domain, a noise sample extractor (130) extracts noise samples of a block in which much reduced signals appear from the sample vector on the time domain, and a variance calculator (140) calculates the variance of the extracted noise samples.

Description

METHOD AND APPARATUS FOR ESTIMATING NOISE

VARIANCE

Technical Field

[1] The present invention relates to a method for estimating noise variance (i.e., information that is necessary to estimate transmitted symbols from received signals) in a receiver. More particularly, the present invention relates to a method for estimating noise variance in multi-carrier communication systems.

[2] This work was supported by the IT R&D program of MIC/IITA [2005-S-404- 12,

Research & Development of Radio Transmission Technology for 3G evolution]. Background Art

[3] Generally, a receiver estimates transmitted symbols from received signals and reconstructs transmitted bits from the transmitted symbols. It reduces a bit error rate to calculate reliability of estimating the transmitted symbols and use it during reconstruction of the transmitted bits.

[4] Noise variance is information that is necessary to calculate reliability of estimation of transmitted symbols.

[5] In a case where a channel between a transmitter and a receiver changes slowly, it can be assumed that channel effects on adjacent pilot symbols are similar so noise can be estimated with a difference between received pilot symbols. But in case where a channel between a transmitter and a receiver changes rapidly, the assumption is not correct so noise cannot be estimated as above.

[6] A method of calculating back noise from an estimated channel coefficient using pilot symbols and a method of calculating noise by referencing higher degree statistics of a channel coefficient/noise have been publicized.

[7] Further, at least partial information about a channel coefficient is necessary so there is problem that a result of estimating a noise variance depends on performance of channel estimation.

[8] The above information disclosed in this Background section is only for enhancement of understanding of the background of the invention and therefore it may contain information that does not form the prior art that is already known in this country to a person of ordinary skill in the art. Disclosure of Invention Technical Problem

[9] It is an object of the present invention to provide a method and apparatus for noise variance estimation having an advantage of being independent of channel estimation when a channel between transmitter and receiver changes rapidly. Technical Solution

[10] In one aspect of the present invention, a method for estimating noise variance in multi-carrier communication systems includes: receiving a first symbol vector on a frequency domain; setting all symbols except pilot symbols of the first symbol vector to zero to make a second symbol vector; transforming the second symbol vector on a fr equency domain into a sample vector on a time domain; and estimating the noise variance using the sample vector on the time domain.

[11] In another aspect of the present invention, an apparatus for estimating a noise variance in multi-carrier communication systems includes an inverse discrete Fourier transformer that transforms a symbol vector on a frequency domain into a sample vector on a time domain, a noise sample extractor that extracts noise samples from the sample vector, and a variance calculator that calculates variance of the extracted noise samples.

Advantageous Effects

[12] As described above, an apparatus for estimating a noise variance of the present invention estimates noise variance without higher degree statistics of a channel coefficient/noise. Thus, the present invention may provide a method and apparatus for estimating noise variance having advantages of being independent of channel estimation when a channel between a transmitter and a receiver changes rapidly. Brief Description of the Drawings

[13] FIG. 1 is a block diagram of an apparatus for noise variance estimation according to an exemplary embodiment of the present invention.

[14] FIG. 2 is a flow chart for describing a method of noise variance estimation according to an exemplary embodiment of the present invention. Mode for the Invention

[15] In the following detailed description, only certain exemplary embodiments of the present invention have been shown and described, simply by way of illustration. As those skilled in the art would realize, the described embodiments may be modified in various different ways, all without departing from the spirit or scope of the present invention. Accordingly, the drawings and description are to be regarded as illustrative in nature and not restrictive. Like reference numerals designate like elements throughout the specification.

[16] Throughout this specification and the claims which follow, unless explicitly described to the contrary, the word "comprise" and variations such as "comprises" or "comprising" will be understood to imply the inclusion of stated elements but not the exclusion of any other elements. [17] Method and apparatus of estimating noise variance according to an exemplary embodiment of the present invention will be described referring to FIG. 1 and FIG. 2

[18] FIG. 1 is a block diagram of an apparatus for estimating noise variance according to an exemplary embodiment of the present invention. An apparatus for estimating noise variance according to the exemplary embodiment of the present invention is located in a receiver of a communication system, and it estimates noise variance from received signals.

[19] As shown in FIG. 1, the apparatus for estimating noise variance includes a windowing controller 110, an inverse discrete Fourier transformer 120, a noise sample extractor 130, and a variance calculator 140.

[20] The apparatus for estimating noise variance receives signals on a frequency domain from a transmitter. Symbol vectors on a frequency domain are carriers of a band except a guard band of the received signals. The windowing controller 110 controls the symbol vectors on the frequency domain. That is, the windowing controller 110 sets all symbols except pilot symbols of each symbol vector on the frequency domain to zero and shares received pilot symbols at transmitted pilot symbols. The transmitted pilot symbols are known to a receiver.

[21] The inverse discrete Fourier transformer 120 transforms the controlled symbol vector on the frequency domain into a sample vector on a time domain. The noise sample extractor 130 extracts noise samples of a block in which much reduced signals appear from the sample vector on the time domain. The variance calculator 140 calculates a variance of the extracted noise samples.

[22] A method for estimating a noise variance according to an exemplary embodiment of the present invention will be described referring to FIG. 2.

[23] FIG. 2 is a flow chart for describing a method for estimating noise variance according to an exemplary embodiment of the present invention.

[24] The windowing controller 110 controls the symbol vector (S210). That is, the windowing controller 110 sets all symbols except pilot symbols of the symbol vector to zero and shares the pilot symbols at transmitted pilot symbols.

[25] With the symbol vector to

, the pilot symbols set to {*_p(r) | r = 1, 2_> - , Λ}

, and the transmitted pilot symbols set to {a \ r = L, 2, - , R} , the controlled symbol vector

is as in Equation 1. [26] (Equation 1)

[27]

y_m iχ-,R

[28] Where w m is defined as a low pass filter(LPF) coefficient. A rectangular window or a

Hamming window can be used as the low pass filter coefficient (w ). The low pass filter coefficient (w ) can be normalized to

∑_r w P(r) ^' IN

[29] The inverse discrete Fourier transformer 120 transforms the controlled symbol vector on the frequency domain into a sample vector on the time domain as IDFT (inverse discrete Fourier transform) (S220). When N is a number of carriers constituting a transmit band including a guard band, the inverse discrete Fourier transformer 120 transforms the controlled symbol vector as N-point IDFT and outputs the sample vector

Z = (Z_{0 5} Z₁ , ^{• • •} , z_Ar_₁ )^r

[30] The noise sample extractor 130 extracts noise samples of a block in which much reduced signals appear because signals are transmitted through a distant course from the sample vector on the time domain (S230).

[31] When L is defined as a length of an impulse response of a channel and P is defined as a minimum interval between pilot symbols, the noise sample extractor 130 extracts (N/P-L) samples because the length of the impulse response of the channel is smaller than a number of carriers. The noise sample extractor 130 extracts samples from z to z

N/P-l because sig ^Cnals a^Jp- ^Jp^. ear from z O to zL-I and signals are attenuated from z L to z N/P-l , and samples from z N/P to z N-I are repetitions of samples from z 0 to z N/p-1

[32] The variance calculator 140 calculates average of squared norms of the (N/P-L) samples to estimate the noise variance (S240). To estimate the noise variance more precisely, filtering that smoothes the progress of the estimated noise variance can be executed during several periods.

[33] While this invention has been described in connection with what is presently considered to be practical exemplary embodiments, it is to be understood that the invention is not limited to the disclosed embodiments, but, on the contrary, is intended to cover various modifications and equivalent arrangements included within the spirit and scope of the appended claims.

Claims

[1] A method for estimating a noise variance in multi-carrier communication systems, the method comprising: receiving a first symbol vector on a frequency domain; setting of all symbols except pilot symbols of the first symbol vector to zero to make a second symbol vector; transforming the second symbol vector on the frequency domain into a sample vector on a time domain; and estimating the noise variance using the sample vector on the time domain.

[2] The method of claim 1, wherein the estimating the noise variance comprises: extracting noise samples of a block in which much reduced signals appear from the sample vector on the time domain; and calculating a variance of the extracted noise samples.

[3] The method of claim 2, wherein a number of the extracted noise samples is (a number of carrier/a minimum interval of the pilot symbol a length of impulse response).

[4] The method of claim 2, wherein the extracting noise samples comprise extracting from (L+l)th to (N/P)th samples from the sample vector on the time domain when L is a length of an impulse response and P is a minimum interval between pilot symbols, and N is a number of carriers.

[5] The method of claim 1, wherein the setting of all symbols except pilot symbols of the first symbol vector to zero comprises sharing the pilot symbols at transmitted pilot symbols.

[6] The method of claim 5, wherein the setting of all symbols except pilot symbols of the first symbol vector to zero further comprises multiplying the pilot symbols by low pass filter coefficients

[7] The method of claim 1, further comprising filtering during at least one period.

[8] An apparatus for estimating a noise variance in multi-carrier communication systems, the apparatus comprising: an inverse discrete Fourier transformer that transforms a symbol vector on a frequency domain into a sample vector on a time domain; a noise sample extractor that extracts noise samples from the sample vector; and a variance calculator that calculates a variance of the extracted noise samples.

[9] The apparatus of claim 8, further comprising a windowing controller that sets all symbols except pilot symbols of the symbol vector on frequency domain to zero.

[10] The apparatus of claim 8, wherein the noise sample extractor extracts from (L+l)th to (N/P)th samples from the sample vector on the time domain when L is a length of an impulse response, P is minimum interval between pilot symbols, and N is a number of carriers.

[11] The apparatus of claim 8, wherein a number of the extracted noise samples is (a number of carrier/a minimum interval of the pilot symbol a length of impulse response).