CN102594761A - Method for estimating noise power of orthogonal frequency division multiplexing (OFDM) signals - Google Patents

Abstract

Provided is a method for estimating noise power of orthogonal frequency division multiplexing (OFDM) signals, which includes conducting fast fourier transform (FFT) algorithm of energy normalization after removing cyclic prefixes of the received signals; then extracting useful messages of pilot frequency or synchronous head from the FFT algorithm, and also extracting useful messages from the pilot frequency or synchronous head messages of the sending end data; and finally, calculating the noise power of signals by the method of using four points of adjacent subcarriers of the same or the adjacent two OFDM signs for offsetting the subcarrier channel frequency response error value, and combining with the useful messages extracted before. The method has good performance of resistance to frequency selective fading and time selective fading; simultaneously, the method can utilize both the pilot frequency and the synchronous head for estimating the noise power, while used for estimating the noise power, the method is applicable for situations of a plurality of synchronous heads, single symmetric synchronous head and single asymmetric synchronous head, and has good adaptability.

Description

A kind of method that is used to estimate the noise power of ofdm signal

Technical field

The present invention relates to mobile radio system parameter Estimation field, be specifically related to a kind of method of estimating the noise power of ofdm signal.

Background technology

OFDM (OFDM) is a kind of multi-carrier modulation; It turns to the high-speed bit circulation of serial the parallel bit stream of low speed; The inverse fast Fourier transform (IFFT) that utilization is simplified; With needing the information transmitted symbol-modulated to each mutually orthogonal subchannel, make the channel in broadband be transformed into more smooth narrow band channel, thereby suppressed intersymbol interference.Owing to can resist frequency selective fading channels, and have the high and balanced advantage such as simple of spectrum efficiency, therefore a lot of communication standards all adopt OFDM to modulate with system, and are used as the key technology of the 4th third-generation mobile communication.Noise power is an important parameters of ofdm system as a kind of index of tolerance channel quality.The document of research ofdm system noise power estimation at present is many, mainly can be divided into two big types, and a kind of algorithm for estimating that is based on synchronous head or pilot tone, another kind are the blind estimations technique.Though blind estimation does not need given data, its algorithm need be than intensive, and convergence rate is slow, so great majority research all is based on synchronous head or pilot tone.

The OFDM synchronous head, be present in front in each OFDM frame one or several to sending and receive all known OFDM symbol of both sides, it is mainly used in synchronously and channel estimating, also can be used for estimating noise power.During Project Realization, OFDM symbol both sides be several 0, as boundary belt, prevent adjacent band interference.Pilot tone then is the subcarrier that is distributed in some ad-hoc location in the OFDM data symbol, and the data on these aspects of transmitting terminal are known.Pilot tone is divided into Comb Pilot and block pilot tone; And synchronous head also is divided into single synchronous head and many synchronous heads, under the situation of single synchronous head, also can be divided into time domain symmetry and the asymmetric situation of time domain.

In ofdm system, the data of a frame of transmitting terminal are following at frequency domain representation:

$\left[\begin{array}{cccc}{\mathrm{<figure-callout\; id="0"\; label="c"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">c</figure-callout>}}_{\mathrm{0,0}}& \mathrm{<figure-callout\; id="0"\; label="L\; c"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">L</figure-callout>}& c_{}{}_{0,\mathrm{<figure-callout\; id="0"\; label="n\; L\; c"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">n</figure-callout>}}& L& c_{}{}_{0,N-1}\\ M& & M& & M\\ c_{m,0}& L& c_{m,n}& L& c_{m,N-1}\\ M& & M& & M\\ c_{M-\mathrm{1,0}}& L& c_{M-1,n}& L& c_{M-1,N-1}\end{array}\right]$

Wherein, N is the number of ofdm system subcarrier, also is counting of FFT, IFFT; M is the OFDM symbolic number that a frame contains; c _{M, n}Be the data on m OFDM symbol, the n number of sub-carrier.

Adding the Cyclic Prefix long enough and the fully synchronous prerequisite of system under, signal is added additive white Gaussian noise advancing multipath channel, then receives signal and can be expressed as at frequency domain:

Y _m，n＝H _m，nc _m，n+η _m，n

Wherein, H _{M, n}Be the frequency domain response of channel, η _{M, n}Be additive white Gaussian noise, c _{M, n}Signal for energy normalized.

The frequency domain response of channel is obtained by following formula:

$H_{m,n}=\underset{p=1}{\overset{P}{\mathrm{\&Sigma;}}}{h}_p\left({\mathrm{mT}}_s\right)e^{-\mathrm{<figure-callout\; id="2"\; label="j"\; filenames="HSA00000627483800024.png"\; state="\{\{state\}\}">j</figure-callout>}2\mathrm{\&pi;}\frac{{\mathrm{n\&tau;}}_p}{{\mathrm{NT}}_s}}$

Wherein, h _p(mT _s) be the gain of p paths, τ _pBe path delay, T _sBe the OFDM symbol period.

In the prior art, mainly be how in Boumard algorithm and REN algorithm, to have mentioned the noise power of estimating ofdm system based on synchronous head and pilot configuration.

The Boumard algorithm is S.Boumard the author, and thesis topic is, and " Novel noise variance and SNR estimation algorithm for wireless MIMO OFDM systems " is published in Global Telecommunications Conference; 2003.GLOBECOM ' 03.IEEE, 2003, above the paper of pp.1330-1334vol.3; And the author appoints light, Luo Meiling, Chang Yilin; Thesis topic is " an ofdm system SNR estimation new method ", is published in above the paper of Xian Electronics Science and Technology University's journal (natural science edition) in October, 2007 (34 volume, the fifth phase); Write up is all arranged, comprises following step:

Step 1:, mainly may further comprise the steps to receiving the preliminary treatment that termination is collected mail and ceased:

Step 1-1: it is N that each OFDM symbol is removed the sequence front end length _CpCyclic Prefix (CP) obtains time-domain information, and is following with matrix notation:

$X=\left[\begin{array}{ccccc}{x}_{\mathrm{0,0}}& L& x_{0,n}& L& x_{0,N-1}\\ M& & M& & M\\ x_{m,0}& L& x_{m,n}& L& x_{m,N-1}\\ M& & M& & M\\ x_{M-\mathrm{1,0}}& L& x_{M-1,n}& L& x_{M-1,N-1}\end{array}\right]$

Step 1-2: will be carried out N point FFT computing by the time-domain information that step 1-1 obtains, N is that FFT counts, and the frequency domain information that obtains is following with matrix notation:

$Y=\left[\begin{array}{ccccc}{y}_{\mathrm{0,0}}& L& y_{0,n}& L& y_{0,N-1}\\ M& & M& & M\\ y_{m,0}& L& y_{m,n}& L& y_{m,N-1}\\ M& & M& & M\\ y_{M-\mathrm{1,0}}& L& y_{M-1,n}& L& y_{M-1,N-1}\end{array}\right]$

Step 1-3: from the frequency domain information that step 1-2 obtains, take out the information in the synchronous head, be divided into following 2 steps:

Step 1-3-1: from the matrix that step 1-2 obtains, take out the N that belongs to synchronous head _pIndividual OFDM symbol:

$Y=\left[\begin{array}{ccccc}{y}_{i_0,0}& L& y_{i_0,n}& L& y_{i0,N-1}\\ M& & M& & M\\ y_{i_m,0}& L& y_{i_m,n}& L& y_{i_m,N-1}\\ M& & M& & M\\ y_{i_{N_p-1},0}& L& y_{i_{N_p-1},n}& L& y_{i_{N_p-1},N-1}\end{array}\right]$

Step 1-3-2: respectively from the N of step 1-3-1 _pIn the row sequence, remove left side boundary belt N _B1Individual point, the right boundary belt N _B2Individual point and Y _{I, n/2}(0 frequency) to one group of new sequence, is expressed as following matrix form:

$Z=\left[\begin{array}{cccccc}{y}_{i_0,N_{b1}}& L& y_{i_0,N/2-1}& y_{i_0,N/2+1}& L& y_{i_0,N-N_{b2}}\\ M& & M& M& & M\\ y_{i_m,N_{b1}}& & y_{i_m,N/2-1}& y_{i_m,N/2+1}& & y_{i_m,N-N_{b2}}\\ M& & M& M& & M\\ y_{i_{N_p-1},N_{b1}}& L& y_{i_{N_p-1},N/2-1}& y_{i_{N_p-1},N/2+1}& L& y_{i_{N_p-1},N-{\mathrm{<figure-callout\; id="2"\; label="N\; b"\; filenames="HSA00000627483800024.png"\; state="\{\{state\}\}">N</figure-callout>}}_b}\end{array}\right]=\left[\begin{array}{cccc}{\mathrm{<figure-callout\; id="0"\; label="z"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">z</figure-callout>}}_{\mathrm{0,0}}& \mathrm{<figure-callout\; id="0"\; label="L\; z"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">L</figure-callout>}& z_{}{}_{0,\mathrm{<figure-callout\; id="0"\; label="j\; L\; z"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">j</figure-callout>}}& L& z_{}{}_{0,J-1}\\ M& & M& & M\\ z_{i,0}& L& z_{i,j}& L& z_{i,J-1}\\ M& & M& & M\\ z_{N_p-\mathrm{1,0}}& L& z_{N_p-1,j}& L& z_{N_p-1,J-1}\end{array}\right]$

Equality the right Z matrix is the redetermination symbol, and J is the length of each synchronous head useful data, J=N-N _B1-N _B2-1.

Step 1-4: according to the sampling mode of step 1-3, take out the useful sequence the synchronous head, represent following matrix from known transmitting terminal data:

$D=\left[\begin{array}{cccc}{\mathrm{<figure-callout\; id="0"\; label="d"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">d</figure-callout>}}_{\mathrm{0,0}}& \mathrm{<figure-callout\; id="0"\; label="L\; d"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">L</figure-callout>}& d_{}{}_{0,\mathrm{<figure-callout\; id="0"\; label="j\; L\; d"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">j</figure-callout>}}& L& d_{}{}_{0,J-1}\\ M& & M& & M\\ d_{i,0}& L& d_{i,j}& L& d_{i,J-1}\\ M& & M& & M\\ d_{N_p-\mathrm{1,0}}& L& d_{N_p-1,j}& L& d_{N_p-1,J-1}\end{array}\right]=\left[\begin{array}{cccccc}{c}_{i_0,N_{b1}}& L& c_{i_0,N/2-1}& c_{i_0,N/2+1}& L& c_{i_0,N-N_{b2}}\\ M& & M& M& & M\\ c_{i_m,N_{b1}}& & c_{i_m,N/2-1}& c_{i_m,N/2+1}& & c_{i_m,N-N_{b2}}\\ M& & M& M& & M\\ c_{i_{N_p-1},N_{b1}}& L& c_{i_{N_p-1},N/2-1}& c_{i_{N_p-1},N/2+1}& L& c_{i_{N_p-1},N-{\mathrm{<figure-callout\; id="2"\; label="N\; b"\; filenames="HSA00000627483800024.png"\; state="\{\{state\}\}">N</figure-callout>}}_b}\end{array}\right]$

The D matrix is the redetermination symbol, the useful data of expression transmitting terminal.

Step 2: through 2 new sequences that step 1 obtains, the noise power of computing system:

$\hat{W}=\frac{1}{4N_p(J-1)}\underset{i=0}{\overset{N_p-1}{\mathrm{\&Sigma;}}}\underset{j=0}{\overset{J-1}{\mathrm{\&Sigma;}}}{|z_{i,j-1}{d}_{i,j}-z_{i,j}{d}_{i,j-1}|}^2$

The above-mentioned noise power calculation formula of Boumard algorithm has used assumed condition: adjacent sub-carrier channel frequency response approximately equal, i.e. H _{M, n-1}≈ H _{M, n}, the prerequisite that it is set up is that the time delay expansion is less.But the multipath number of a lot of wireless channels is many, and the time delay expansion is bigger, at this moment can not obtain H _{M, n-1}≈ H _{M, n}, this can cause the Boumard algorithm performance sharply to descend.

The REN algorithm is to appoint light, Chang Yilin the author, and thesis topic is " SNR estimation algorithm based onthe preamble for OFDM systems in frequency selective channels, "; Be published in Communications; IEEE Transactions on, vol.57, pp.2230-2234; 2009. paper above write up is arranged, comprise following step:

Step 1: to receiving the collect mail preliminary treatment of breath of termination, identical with aforesaid Boumard algorithm steps 1;

Step 2: two the new matrix Z and D that obtains from step 1, choose the wherein data of first row and second row, even i=0,1, the noise power of computing system:

$\hat{W}=\frac{1}{2J}\underset{j=0}{\overset{J-1}{\mathrm{\&Sigma;}}}{|{\mathrm{<figure-callout\; id="0"\; label="Y"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">Y</figure-callout>}}_{0,j}^{\&prime;}-{\mathrm{<figure-callout\; id="1"\; label="Y"\; filenames="HSA00000627483800024.png"\; state="\{\{state\}\}">Y</figure-callout>}}_{1,j}^{\&prime;}|}^2$

Wherein, work as d _{I, j}≠ 0 o'clock,

Work as d _{I, j}=0 o'clock, Y ' _{I, j}=v _{I, j}

Expression is to d _{I, j}Get conjugation.

The utilization of REN algorithm has two restrictive conditions, one of which, and it need use two synchronous heads that structure is identical, because it need subtract each other with the Y ' on the same sub-carrier of different cycles, i.e. Y ' _{I, j}, Y ' _{I+1, j}, with the cancellation channel frequency response; Its two, the channel frequency response at different OFDM symbol same sub-carrier place is wanted approximately equal, i.e. H _{M, n}≈ H _{M+1, n}≈ H _nBut the structure of two synchronous heads maybe be inequality when practical application; In addition, the REN algorithm has two weak points: one of which, and along with the increase of channel maximum doppler frequency, it is big that the time selective fading of channel becomes, and the channel coefficients of different OFDM symbol same sub-carrier can change, and at this moment they can not approximately equal; Its two, if two synchronous heads other OFDM symbol at interval then still can exist between their channel coefficients than big-difference, also can influence the performance of REN algorithm.

We find through a large amount of experiments: the difference of two channel frequency responses of different OFDM mark space same sub-carrier is comparatively approaching, and the difference of the adjacent sub-carrier of adjacent OFDM symbol is minimum, i.e. (H _{M, n}-H _{M, n+1})-(H _{M+1, n}-H _{M+1, n+1}) ≈ 0, we just can offset the error amount of sub-carrier channels frequency response with such 4 points of adjacent sub-carrier same or the two adjacent OFDM symbol like this, and the error that it brings is littler.

Summary of the invention

In order to solve the above-mentioned technical problem that existing Boumard algorithm and REN algorithm exist, a kind of method that is used to estimate the noise power of ofdm signal that the present invention proposes is characterized in that comprising the steps:

Step 1: the initialization system parameter comprises: the selection decision according to the user adopts one of following four kinds of information to come estimated noise power: pilot tone, a plurality of synchronous head, single symmetrical synchronous head, single asymmetric synchronous head; Set left side boundary belt points N _B1, the right boundary belt points N _B2, circulating prefix-length N _Cp, synchronous head number N _p, frequency pilot sign number N _Pl, frame data of transmitting terminal are at the matrix that is expressed as of frequency domain $C=\left[\begin{array}{cccc}{\mathrm{<figure-callout\; id="0"\; label="c"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">c</figure-callout>}}_{\mathrm{0,0}}& \mathrm{<figure-callout\; id="0"\; label="L\; c"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">L</figure-callout>}& c_{}{}_{0,\mathrm{<figure-callout\; id="0"\; label="n\; L\; c"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">n</figure-callout>}}& L& c_{}{}_{0,N-1}\\ M& & M& & M\\ c_{m,0}& L& c_{m,n}& L& c_{m,N-1}\\ M& & M& & M\\ c_{M-\mathrm{1,0}}& L& c_{M-1,n}& L& c_{M-1,N-1}\end{array}\right],$ N is the number of ofdm system subcarrier, and M is the OFDM symbolic number that a frame contains;

Step 2: the information to the reception termination is received is carried out preliminary treatment, comprising:

Step 2-1: it is N that the OFDM symbol that receives is removed the sequence front end length _CpCyclic Prefix (CP), the time-domain information that obtains, following with matrix notation:

$X=\left[\begin{array}{ccccc}{x}_{\mathrm{0,0}}& L& x_{0,n}& L& x_{0,N-1}\\ M& & M& & M\\ x_{m,0}& L& x_{m,n}& L& x_{m,N-1}\\ M& & M& & M\\ x_{M-\mathrm{1,0}}& L& x_{M-1,n}& L& x_{M-1,N-1}\end{array}\right]$

Step 2-2: convert the time-domain signal X of step 2-1 into frequency-region signal: if adopt single symmetrical synchronous head to come estimated noise power, then extract first line data of the matrix X that step 2-1 obtains, it is divided into former and later two parts, that is: [x _0,0L x _{0, N/2-1}] and [x _{0, N/2}L x _{0, N-1}], respectively they are carried out the FFT of the energy normalized that N/2 orders, obtain matrix:

$Y=\left[\begin{array}{ccccc}{y}_{\mathrm{0,0}}& L& y_{0,j}& L& y_{0,N/2-1}\\ y_{\mathrm{1,0}}& L& y_{1,j}& L& y_{1,N/2-1}\end{array}\right]$

Otherwise, the energy normalized FFT that N is ordered is carried out in each provisional capital of matrix X that step 2-1 obtains, obtain matrix:

$Y=\left[\begin{array}{ccccc}{y}_{\mathrm{0,0}}& L& y_{0,n}& L& y_{0,N-1}\\ M& & M& & M\\ y_{m,0}& L& y_{m,n}& L& y_{m,N-1}\\ M& & M& & M\\ y_{M-\mathrm{1,0}}& L& y_{M-1,n}& L& y_{M-1,N-1}\end{array}\right]$

Step 3: extract useful data from receiving terminal:

(1), then according to the definition of this ofdm system standard, finds out from the structure of definition which OFDM symbol has pilot frequency information in a frame, and extract these symbols if adopt pilot tone to come estimated noise power; From them, confirm the position of pilot sub-carrier again, the reception signal data at this subcarrier place extracted, form a new pilot matrix:

$Z=\left[\begin{array}{ccccc}{\mathrm{<figure-callout\; id="0"\; label="z"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">z</figure-callout>}}_{0,0}& L& z_{0,j}& L& z_{0,J-1}\\ M& & M& & M\\ z_{i,0}& L& z_{i,j}& L& z_{i,J-1}\\ M& & M& & M\\ z_{N_{\mathrm{pl}}-\mathrm{1,0}}& L& z_{N_{\mathrm{pl}}-1,j}& L& z_{N_{\mathrm{pl}}-1,J-1}\end{array}\right]$

Wherein, N _PlBe the number of OFDM frequency pilot sign in the frame, J is counting of the interior pilot tone of a frequency pilot sign;

(2) if adopt a plurality of synchronous heads to come estimated noise power, each row at Np synchronous head sequence place among the matrix Y that then step 2 is obtained all removes the N of left side boundary belt _B1The N of individual point, the right boundary belt _B2It is the point on 0 the subcarrier that individual point, 0 frequency and corresponding sends data, obtains the matrix that useful data is formed:

$Z=\left[\begin{array}{ccccc}{\mathrm{<figure-callout\; id="0"\; label="y\; i"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">y</figure-callout>}}_{i_{}}& L& y_{i_0,j_k}& L& y_{i_0,j_{J-1}}\\ M& & M& & M\\ y_{i_m,j_0}& L& y_{i_m,j_k}& L& y_{i_m,j_{J-1}}\\ M& & M& & M\\ y_{i_{N_p-1},j_0}& L& y_{i_{N_p-1},j_k}& L& y_{i_{N_p-1},j_{J-1}}\end{array}\right]=\left[\begin{array}{cccc}{\mathrm{<figure-callout\; id="0"\; label="z"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">z</figure-callout>}}_{\mathrm{0,0}}& \mathrm{<figure-callout\; id="0"\; label="L\; z"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">L</figure-callout>}& z_{}{}_{0,\mathrm{<figure-callout\; id="0"\; label="j\; L\; z"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">j</figure-callout>}}& L& z_{}{}_{0,J-1}\\ M& & M& & M\\ z_{i,0}& L& z_{i,j}& L& z_{i,J-1}\\ M& & M& & M\\ z_{N_p-\mathrm{1,0}}& L& z_{N_p-1,j}& L& z_{N_p-1,J-1}\end{array}\right]$

Wherein J sends the number that data are not 0 point, j in the synchronous head symbol _kSpan at N _B1Between N/2-1, perhaps N/2+1 is to N-N _B2

(3) if adopt single asymmetric synchronous head to come estimated noise power, the synchronous head sequence of first row is removed left side boundary belt N among the matrix Y that then step 2 is obtained _B1Individual point, the right boundary belt N _B2It is the point on 0 the subcarrier that individual point, 0 frequency and corresponding sends data, obtains the useful data matrix

$Z=\left[\begin{array}{ccccc}{\mathrm{<figure-callout\; id="0"\; label="y"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">y</figure-callout>}}_{0,j_0}& \mathrm{<figure-callout\; id="0"\; label="L\; y"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">L</figure-callout>}& y_{}{}_{0,j_k}& \mathrm{<figure-callout\; id="0"\; label="L\; y"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">L</figure-callout>}& y_{}{}_{0,j_{J-1}}\end{array}\right]=\left[\begin{array}{cccc}{\mathrm{<figure-callout\; id="0"\; label="z"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">z</figure-callout>}}_{\mathrm{0,0}}& \mathrm{<figure-callout\; id="0"\; label="L\; z"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">L</figure-callout>}& z_{}{}_{0,\mathrm{<figure-callout\; id="0"\; label="j\; L\; z"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">j</figure-callout>}}& L& z_{}{}_{0,J-1}\end{array}\right]$

Wherein J sends the number that data are not 0 point, j in the synchronous head symbol _kSpan at N _B1Between N/2-1, perhaps N/2+1 is to N-N _B2

(4) if adopt single symmetrical synchronous head to come estimated noise power, the matrix formed of useful data then

$Z=\left[\begin{array}{ccccc}{y}_{\mathrm{0,0}}& L& y_{0,j}& L& y_{0,N/2-1}\\ y_{\mathrm{1,0}}& L& y_{1,j}& L& y_{1,N/2-1}\end{array}\right]=\left[\begin{array}{cccc}{\mathrm{<figure-callout\; id="0"\; label="z"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">z</figure-callout>}}_{\mathrm{0,0}}& \mathrm{<figure-callout\; id="0"\; label="L\; z"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">L</figure-callout>}& z_{}{}_{0,\mathrm{<figure-callout\; id="0"\; label="j\; L\; z"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">j</figure-callout>}}& L& z_{}{}_{0,J-1}\\ {\mathrm{<figure-callout\; id="1"\; label="z"\; filenames="HSA00000627483800024.png"\; state="\{\{state\}\}">z</figure-callout>}}_{1,0}& \mathrm{<figure-callout\; id="1"\; label="L\; z"\; filenames="HSA00000627483800024.png"\; state="\{\{state\}\}">L</figure-callout>}& z_{}{}_{1,\mathrm{<figure-callout\; id="1"\; label="j\; L\; z"\; filenames="HSA00000627483800024.png"\; state="\{\{state\}\}">j</figure-callout>}}& L& z_{}{}_{1,J-1}\end{array}\right]$

J＝N/2；

Step 4: extract useful data from transmitting terminal:

(1) if adopt pilot tone to come estimated noise power, then in transmitting terminal frequency-region signal Matrix C, confirm that earlier which OFDM symbol has pilot frequency information, extract corresponding row sequence among the C; From these row sequences, confirm the position of pilot sub-carrier again, the transmission information at these subcarrier places extracted composition-individual new pilot matrix:

$D=\left[\begin{array}{cccc}{\mathrm{<figure-callout\; id="0"\; label="d"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">d</figure-callout>}}_{\mathrm{0,0}}& \mathrm{<figure-callout\; id="0"\; label="L\; d"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">L</figure-callout>}& d_{}{}_{0,\mathrm{<figure-callout\; id="0"\; label="j\; L\; d"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">j</figure-callout>}}& L& d_{}{}_{0,J-1}\\ M& & M& & M\\ d_{i,0}& L& d_{i,j}& L& d_{i,J-1}\\ M& & M& & M\\ d_{N_{\mathrm{pl}}-\mathrm{1,0}}& L& d_{N_{\mathrm{pl}}-1,j}& L& d_{N_{\mathrm{pl}}-1,J-1}\end{array}\right]$

J is counting of the interior pilot tone of a symbol;

(2) if adopt a plurality of synchronous heads to come estimated noise power, then at the N of Matrix C _pDuring individual synchronous head sequence is expert at, remove left side boundary belt N _B1Individual point, the right boundary belt N _B2Individual point, 0 frequency and data are the point on 0 the subcarrier, obtain the matrix that useful data is formed:

$D=\left[\begin{array}{cccc}{\mathrm{<figure-callout\; id="0"\; label="d"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">d</figure-callout>}}_{\mathrm{0,0}}& \mathrm{<figure-callout\; id="0"\; label="L\; d"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">L</figure-callout>}& d_{}{}_{0,\mathrm{<figure-callout\; id="0"\; label="j\; L\; d"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">j</figure-callout>}}& L& d_{}{}_{0,J-1}\\ M& & M& & M\\ d_{i,0}& L& d_{i,j}& L& d_{i,J-1}\\ M& & M& & M\\ d_{N_p-\mathrm{1,0}}& L& d_{N_p-1,j}& L& d_{N_p-1,J-1}\end{array}\right]=\left[\begin{array}{ccccc}{\mathrm{<figure-callout\; id="0"\; label="c\; i"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">c</figure-callout>}}_{i_{}}& \mathrm{<figure-callout\; id="0"\; label="L\; c\; i"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">L</figure-callout>}& c_{i_{}}{{}_0,j_k}_{}& \mathrm{<figure-callout\; id="0"\; label="L\; c\; i"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">L</figure-callout>}& c_{i_{}}{{}_0,j_{J-1}}_{}\\ M& & M& & M\\ c_{i_m,j_0}& L& c_{i_m,j_k}& L& c_{i_m,j_{J-1}}\\ M& & M& & M\\ c_{i_{N_p-1},j_0}& L& c_{i_{N_p-1},j_k}& L& c_{i_{N_p-1},j_{J-1}}\end{array}\right]$

Wherein, J sends the number that data are not 0 point, j in the synchronous head symbol _kSpan at N _B1Between N/2-1, perhaps N/2+1 is to N-N _B2

(3) if adopt single asymmetric synchronous head to come estimated noise power,, remove left side boundary belt N then with first line data in the Matrix C _B1Individual point, the right boundary belt N _B2It is the point on 0 the subcarrier that individual point, 0 frequency and corresponding sends data, obtains the matrix that useful data is formed:

$D=\left[\begin{array}{cccc}{\mathrm{<figure-callout\; id="0"\; label="d"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">d</figure-callout>}}_{\mathrm{0,0}}& \mathrm{<figure-callout\; id="0"\; label="L\; d"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">L</figure-callout>}& d_{}{}_{0,\mathrm{<figure-callout\; id="0"\; label="j\; L\; d"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">j</figure-callout>}}& L& d_{}{}_{0,J-1}\end{array}\right]=\left[\begin{array}{cccc}{\mathrm{<figure-callout\; id="0"\; label="c"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">c</figure-callout>}}_{0,j_0}& \mathrm{<figure-callout\; id="0"\; label="L\; c"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">L</figure-callout>}& c_{}{}_{0,j_{\mathrm{<figure-callout\; id="0"\; label="k\; L\; c"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">k</figure-callout>}}}& L& c_{}{}_{0,j_{J-1}}\end{array}\right]$

J sends the number that data are not 0 point, j in the synchronous head symbol _kSpan at N _B1Between N/2-1, perhaps N/2+1 is to N-N _B2

(4) if adopt single symmetrical synchronous head to come estimated noise power, then:

$D=\left[\begin{array}{cccc}{\mathrm{<figure-callout\; id="0"\; label="d"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">d</figure-callout>}}_{\mathrm{0,0}}& \mathrm{<figure-callout\; id="0"\; label="L\; d"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">L</figure-callout>}& d_{}{}_{0,\mathrm{<figure-callout\; id="0"\; label="j\; L\; d"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">j</figure-callout>}}& L& d_{}{}_{0,J-1}\\ {\mathrm{<figure-callout\; id="1"\; label="d"\; filenames="HSA00000627483800024.png"\; state="\{\{state\}\}">d</figure-callout>}}_{\mathrm{1,0}}& L& d_{01,\mathrm{<figure-callout\; id="1"\; label="j\; L\; d"\; filenames="HSA00000627483800024.png"\; state="\{\{state\}\}">j</figure-callout>}}& L& d_{}{}_{1,J-1}\end{array}\right]=\left[\begin{array}{ccc}1& \mathrm{<figure-callout\; id="1"\; label="L"\; filenames="HSA00000627483800024.png"\; state="\{\{state\}\}">L</figure-callout>}& 1\\ 1& \mathrm{<figure-callout\; id="1"\; label="L"\; filenames="HSA00000627483800024.png"\; state="\{\{state\}\}">L</figure-callout>}& 1\end{array}\right]$

Wherein, J=N/2;

Step 5: the noise power of utilizing the useful sequence computing system of transmitting terminal that front two step obtains and receiving terminal: if adopt single asymmetric synchronous head to come estimated noise power, noise power then $\hat{W}=\frac{1}{4(J-3)}\underset{j=0}{\overset{J-4}{\mathrm{\&Sigma;}}}{|({\mathrm{<figure-callout\; id="0"\; label="z"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">z</figure-callout>}}_{0,\mathrm{<figure-callout\; id="0"\; label="j\; d"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">j</figure-callout>}}{d}_{}^{}{}_{0,j}^{*}-{\mathrm{<figure-callout\; id="0"\; label="z"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">z</figure-callout>}}_{0,j+2}{\mathrm{<figure-callout\; id="0"\; label="d"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">d</figure-callout>}}_{0,j+2}^{*})-({\mathrm{<figure-callout\; id="0"\; label="z"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">z</figure-callout>}}_{0,j+1}{\mathrm{<figure-callout\; id="0"\; label="d"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">d</figure-callout>}}_{0,j+1}^{*}-{\mathrm{<figure-callout\; id="0"\; label="z"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">z</figure-callout>}}_{0,j+3}{\mathrm{<figure-callout\; id="0"\; label="d"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">d</figure-callout>}}_{0,j+3}^{*})|}^2;$ Otherwise, from matrix Z and D, select any two row respectively and be used for calculating noise power $\hat{W}=\frac{1}{4(J-1)}\underset{j=0}{\overset{J-2}{\mathrm{\&Sigma;}}}{|(z_{i,j}{d}_{i,j}^{*}-z_{i,j+1}{d}_{i,j+1}^{*})-(z_{i^{\&prime;},j}{d}_{i^{\&prime;},j}^{*}-z_{i^{\&prime;},j+1}{d}_{i^{\&prime;},j+1}^{*})|}^2,$ Wherein, i, the row number of any two row of i ' representing matrix Z and D; d ^*Expression is got conjugation to d.

The FFT of said energy normalized, be meant under the expression formula of original FFT divided by invariant

wherein N be that FFT counts.

Any two row that said step 3 is selected from matrix Z and D respectively are adjacent two row in the matrix.

Can find out from top description; The technical scheme that the present invention proposes; Such four points of adjacent sub-carrier of one or two adjacent OFDM symbols of usefulness are offset the error amount of sub-carrier channels frequency response, and this just makes this noise power estimation method have the good anti-frequency selective fading and the characteristic of time selective fading.Simultaneously; This method both can adopt pilot tone to come estimated noise power, also can adopt synchronous head to come estimated noise power, when coming estimated noise power with synchronous head; Situation for a plurality of synchronous heads, single symmetrical synchronous head and single asymmetric synchronous head can both be suitable for, and has good adaptability.

Description of drawings

The overview flow chart of the method for a kind of noise power that is used to estimate ofdm signal that accompanying drawing 1 proposes for the present invention;

The pilot configuration figure that accompanying drawing 2 is used for the specific embodiment of the invention one;

The system frame structure figure that accompanying drawing 3 is used for the specific embodiment of the invention one;

The synchronous head structure figure that accompanying drawing 4 is used for the specific embodiment of the invention two;

The synchronous head structure figure that accompanying drawing 5 arrives for the specific embodiment of the invention three and embodiment four-function;

The useful sequential structure figure of synchronous head that accompanying drawing 6 arrives for specific embodiment of the invention four-function.

Embodiment

Specify the execution mode of technical scheme of the present invention below.

All embodiment select MBSFN (Multicast Broadcast Single Frequency Network) system, i.e. multicast broadcast single frequency network network for use.In the MBSFN system down link, it is identical that portable terminal receives each base station data sent, is similar to the user and receives the data of sending from a base station through multipath fading, and this channel is typical frequency selective fading channels.

Below the FFT of the energy normalized mentioned among each embodiment, be meant under the expression formula of original FFT divided by invariant

wherein N be that FFT counts.

Embodiment one: use the pilot frequency information estimated noise power

Present embodiment is an example with the pilot configuration of 3GPP standard, and this pilot configuration adopts block pilot tone, 3 OFDM data symbols of each OFDM pilot symbol interval, second OFDM frequency pilot sign and 1,31 seat carrier wave that staggers, and pilot tone adopts the QPSK modulation.This pilot configuration and system frame structure are as shown in Figures 2 and 3.In addition, present embodiment supposition OFDM sub-carrier number N is 512, circulating prefix-length N _CpBe N/4, the sub-carrier number of useful data is 300, left side boundary belt number N _B1=106, the right boundary belt number N _B2=105,0 frequency is 0, and frame data of transmitting terminal are at the matrix that is expressed as of frequency domain

$C=\left[\begin{array}{ccccc}{\mathrm{<figure-callout\; id="0"\; label="c"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">\; <figure-callout\; id="0"\; label="c"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">c</figure-callout>\; </figure-callout>}}_{\mathrm{0,0}}& \&CenterDot;\&CenterDot;\&CenterDot;& {\mathrm{<figure-callout\; id="0"\; label="c"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">c</figure-callout>}}_{0,n}& \&CenterDot;\&CenterDot;\&CenterDot;& {\mathrm{<figure-callout\; id="0"\; label="c"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">c</figure-callout>}}_{0,N-1}\\ \&CenterDot;& & \&CenterDot;& & \&CenterDot;\\ \&CenterDot;& & \&CenterDot;& & \&CenterDot;\\ \&CenterDot;& & \&CenterDot;& & \&CenterDot;\\ c_{m,0}& \&CenterDot;\&CenterDot;\&CenterDot;& c_{m,n}& \&CenterDot;\&CenterDot;\&CenterDot;& c_{m,N-1}\\ \&CenterDot;& & \&CenterDot;& & \&CenterDot;\\ \&CenterDot;& & \&CenterDot;& & \&CenterDot;\\ \&CenterDot;& & \&CenterDot;& & \&CenterDot;\\ c_{M-\mathrm{1,0}}& \&CenterDot;\&CenterDot;\&CenterDot;& c_{M-1,n}& \&CenterDot;\&CenterDot;\&CenterDot;& c_{M-1,N-1}\end{array}\right]$

Step 1: it is N that the OFDM symbol that receives is removed the sequence front end length _CpCyclic Prefix (CP), do the FFT computing of N point energy normalized then, convert frequency-region signal into;

Step 2: from receive data, take out pilot data: according to the system frame structure figure shown in pilot configuration figure shown in the accompanying drawing 2 and the accompanying drawing 3; The data that obtain for step 1; Remove left and right sides boundary belt and 0 frequency earlier; Take out the data on the pilot tone point again: for even numbered blocks, take out the 3rd OFDM symbol, in this symbol, take out the odd point data; For the odd number piece, only get the 1st OFDM symbol, in this symbol, take out the even number point data.Form new matrix:

$Z=\left[\begin{array}{ccccc}{\mathrm{<figure-callout\; id="0"\; label="z"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">\; <figure-callout\; id="0"\; label="z"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">z</figure-callout>\; </figure-callout>}}_{\mathrm{0,0}}& \&CenterDot;\&CenterDot;\&CenterDot;& {\mathrm{<figure-callout\; id="0"\; label="z"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">z</figure-callout>}}_{0,j}& \&CenterDot;\&CenterDot;\&CenterDot;& {\mathrm{<figure-callout\; id="0"\; label="z"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">z</figure-callout>}}_{0,J-1}\\ {\mathrm{<figure-callout\; id="1"\; label="z"\; filenames="HSA00000627483800024.png"\; state="\{\{state\}\}">\; <figure-callout\; id="0"\; label="z"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">z</figure-callout>\; </figure-callout>}}_{\mathrm{1,0}}& \&CenterDot;\&CenterDot;\&CenterDot;& z_{01,j}& \&CenterDot;\&CenterDot;\&CenterDot;& {\mathrm{<figure-callout\; id="1"\; label="z"\; filenames="HSA00000627483800024.png"\; state="\{\{state\}\}">z</figure-callout>}}_{1,J-1}\end{array}\right]$

Again in a manner described, from the data matrix C of transmitting terminal, take out the data of corresponding pilot tone point, form new matrix:

$D=\left[\begin{array}{ccccc}{\mathrm{<figure-callout\; id="0"\; label="d"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">\; <figure-callout\; id="0"\; label="d"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">d</figure-callout>\; </figure-callout>}}_{0,0}& \&CenterDot;\&CenterDot;\&CenterDot;& {\mathrm{<figure-callout\; id="0"\; label="d"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">d</figure-callout>}}_{0,j}& \&CenterDot;\&CenterDot;\&CenterDot;& {\mathrm{<figure-callout\; id="0"\; label="d"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">d</figure-callout>}}_{0,J-1}\\ {\mathrm{<figure-callout\; id="1"\; label="d"\; filenames="HSA00000627483800024.png"\; state="\{\{state\}\}">\; <figure-callout\; id="0"\; label="d"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">d</figure-callout>\; </figure-callout>}}_{\mathrm{1,0}}& \&CenterDot;\&CenterDot;\&CenterDot;& d_{01,j}& \&CenterDot;\&CenterDot;\&CenterDot;& {\mathrm{<figure-callout\; id="1"\; label="d"\; filenames="HSA00000627483800024.png"\; state="\{\{state\}\}">d</figure-callout>}}_{1,J-1}\end{array}\right]$

Wherein, J=150.

Step 3: utilize the transmitting terminal that step 2 obtains and the pilot tone dot matrix of receiving terminal, the computing system noise power: $\hat{W}=\frac{1}{4(J-1)}\underset{j=0}{\overset{J-2}{\mathrm{\&Sigma;}}}{|({\mathrm{<figure-callout\; id="0"\; label="z"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">z</figure-callout>}}_{0,\mathrm{<figure-callout\; id="0"\; label="j\; d"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">j</figure-callout>}}{d}_{}^{}{}_{0,j}^{*}-{\mathrm{<figure-callout\; id="0"\; label="z"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">z</figure-callout>}}_{0,j+1}{\mathrm{<figure-callout\; id="0"\; label="d"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">d</figure-callout>}}_{0,j+1}^{*})-({\mathrm{<figure-callout\; id="1"\; label="z"\; filenames="HSA00000627483800024.png"\; state="\{\{state\}\}">z</figure-callout>}}_{1,\mathrm{<figure-callout\; id="1"\; label="j\; d"\; filenames="HSA00000627483800024.png"\; state="\{\{state\}\}">j</figure-callout>}}{d}_{}^{}{}_{1,j}^{*}-{\mathrm{<figure-callout\; id="1"\; label="z"\; filenames="HSA00000627483800024.png"\; state="\{\{state\}\}">z</figure-callout>}}_{1,j+1}{\mathrm{<figure-callout\; id="1"\; label="d"\; filenames="HSA00000627483800024.png"\; state="\{\{state\}\}">d</figure-callout>}}_{1,j+1}^{*})|}^2,$ D wherein ^*Expression is got conjugation to d, J=150.

Need to prove that the pilot configuration of any ofdm system and form to those skilled in the art, are the information that can inquire through disclosed technical specification.Present embodiment is an example with the pilot configuration of 3GPP standard; The data that how to extract in the pilot tone are described, for other employing the ofdm system of pilot tone, during data in extracting pilot tone; As long as definition according to the corresponding techniques standard; Find out pilot data, and remove wherein boundary belt and 0 frequency (if any) data, just can form the matrix Z in the present embodiment.

Embodiment two: with two synchronous head data-evaluation noise powers.

In the present embodiment, adopt two identical synchronous heads, OFDM data symbol of synchronous head mark space, OFDM sub-carrier number N is 128, circulating prefix-length N _CpBe N/4, the sub-carrier number that data are arranged is 116, is bpsk signal and 0 form at interval, left side boundary belt number N _B1=6, the right boundary belt number N _B2=5,0 frequency is 0.The synchronous head overall structure is as shown in Figure 4.Suppose synchronous head left data among Fig. 4

A＝[1，0，1，0，-1，0，-1，0，1，0，1，0，-1，0，-1，0，1，0，1，0，1，0，1，0，-1，0，-1，0，-1，0，-1，0，1，0，-1，0，-1，0，1，0，-1，0，-1，0，-1，0，-10，-1，0，-1，0，-1，0，-1，0，-1，0]；

The synchronous head right data

B＝[0，-1，0，-1，0，-1，0，1，0，-1，0，1，0，-1，0，-1，0，-1，0，-1，0，-1，0，-1，0，-1，0，1，0，1，0，-1，0，1，0，1，0，-1，0，-1，0，1，0，-1，0，-1，0，-1，0，-1，0，-1，0，-1，0，-1，0，-1]。

Step 1: it is N that the OFDM symbol that receives is removed the sequence front end length _CpCyclic Prefix (CP), do N point energy normalized FFT computing then, convert the time-domain signal that receives into frequency-region signal;

Step 2: from receive data, take out the useful data in the synchronous head: according to synchronous head structure shown in Figure 4, from the data that step 1 obtains, get the sequence of first row and the third line earlier, these two is exactly the OFDM synchronous head; Remove left and right sides boundary belt and 0 frequency again, promptly take out in the receiving terminal data data of A and B position in the corresponding diagram 4; Afterwards, take out odd point and the point of the even number among the B among the A, promptly the corresponding receiving terminal frequency domain data of BPSK data loca is formed new matrix among A and the B:

$Z=\left[\begin{array}{ccccc}{\mathrm{<figure-callout\; id="0"\; label="z"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">\; <figure-callout\; id="0"\; label="z"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">z</figure-callout>\; </figure-callout>}}_{\mathrm{0,0}}& \&CenterDot;\&CenterDot;\&CenterDot;& {\mathrm{<figure-callout\; id="0"\; label="z"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">z</figure-callout>}}_{0,j}& \&CenterDot;\&CenterDot;\&CenterDot;& {\mathrm{<figure-callout\; id="0"\; label="z"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">z</figure-callout>}}_{0,J-1}\\ {\mathrm{<figure-callout\; id="1"\; label="z"\; filenames="HSA00000627483800024.png"\; state="\{\{state\}\}">\; <figure-callout\; id="0"\; label="z"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">z</figure-callout>\; </figure-callout>}}_{1,0}& \&CenterDot;\&CenterDot;\&CenterDot;& {\mathrm{<figure-callout\; id="1"\; label="z"\; filenames="HSA00000627483800024.png"\; state="\{\{state\}\}">z</figure-callout>}}_{1,j}& \&CenterDot;\&CenterDot;\&CenterDot;& {\mathrm{<figure-callout\; id="1"\; label="z"\; filenames="HSA00000627483800024.png"\; state="\{\{state\}\}">z</figure-callout>}}_{1,J-1}\end{array}\right]$

Wherein, J=58.

From the transmitting terminal data, take out useful data:

$D=\left[\begin{array}{ccccc}{\mathrm{<figure-callout\; id="0"\; label="d"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">\; <figure-callout\; id="0"\; label="d"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">d</figure-callout>\; </figure-callout>}}_{\mathrm{0,0}}& \&CenterDot;\&CenterDot;\&CenterDot;& {\mathrm{<figure-callout\; id="0"\; label="d"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">d</figure-callout>}}_{0,j}& \&CenterDot;\&CenterDot;\&CenterDot;& {\mathrm{<figure-callout\; id="0"\; label="d"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">d</figure-callout>}}_{0,J-1}\\ {\mathrm{<figure-callout\; id="1"\; label="d"\; filenames="HSA00000627483800024.png"\; state="\{\{state\}\}">\; <figure-callout\; id="0"\; label="d"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">d</figure-callout>\; </figure-callout>}}_{\mathrm{1,0}}& \&CenterDot;\&CenterDot;\&CenterDot;& d_{01,j}& \&CenterDot;\&CenterDot;\&CenterDot;& {\mathrm{<figure-callout\; id="1"\; label="d"\; filenames="HSA00000627483800024.png"\; state="\{\{state\}\}">d</figure-callout>}}_{1,J-1}\end{array}\right]=\left[\begin{array}{cc}{A}_1& B_1\\ A_1& {\mathrm{<figure-callout\; id="1"\; label="B"\; filenames="HSA00000627483800024.png"\; state="\{\{state\}\}">B</figure-callout>}}_1\end{array}\right]$

First row of matrix D is made up of BPSK data among BPSK data and the B among the A, and the second row sequence is identical with first row, is designated as:

A ₁＝[1，1，-1，-1，1，1，-1，-1，1，1，1，1，-1，-1，-1，-1，1，-1，-1，1，-1，-1，-1，-1-1，-1，-1，-1，-1]；

B ₁＝[1，-1，-1，1，-1，1，-1，-1，-1，-1，-1，-1，-1，1，1，-1，1，1，-1，-1，1，-1，-1，-1，-1，-1，-1，-1，-1]。

Step 3: utilize transmitting terminal that step 2 obtains and receiving terminal to the frequency matrix, the computing system noise power: $\hat{W}=\frac{1}{4(J-1)}\underset{j=0}{\overset{J-2}{\mathrm{\&Sigma;}}}{|({\mathrm{<figure-callout\; id="0"\; label="z"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">z</figure-callout>}}_{0,\mathrm{<figure-callout\; id="0"\; label="j\; d"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">j</figure-callout>}}{d}_{}^{}{}_{0,j}^{*}-{\mathrm{<figure-callout\; id="0"\; label="z"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">z</figure-callout>}}_{0,j+1}{\mathrm{<figure-callout\; id="0"\; label="d"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">d</figure-callout>}}_{0,j+1}^{*})-({\mathrm{<figure-callout\; id="1"\; label="z"\; filenames="HSA00000627483800024.png"\; state="\{\{state\}\}">z</figure-callout>}}_{1,\mathrm{<figure-callout\; id="1"\; label="j\; d"\; filenames="HSA00000627483800024.png"\; state="\{\{state\}\}">j</figure-callout>}}{d}_{}^{}{}_{1,j}^{*}-{\mathrm{<figure-callout\; id="1"\; label="z"\; filenames="HSA00000627483800024.png"\; state="\{\{state\}\}">z</figure-callout>}}_{1,j+1}{\mathrm{<figure-callout\; id="1"\; label="d"\; filenames="HSA00000627483800024.png"\; state="\{\{state\}\}">d</figure-callout>}}_{1,j+1}^{*})|}^2,$ D wherein ^*Expression is got conjugation to d, J=58.

Embodiment three: with single symmetrical synchronous head data-evaluation noise power.

Embodiment two used synchronous heads are a kind of synchronous head of symmetry, and three of embodiment adopt other conditions of synchronous head structure identical with embodiment two.Synchronous head structure is as shown in Figure 5.

OFDM sub-carrier number N is 128, circulating prefix-length N _CpBe N/4, the sub-carrier number that data are arranged is 116, is bpsk signal and 0 form at interval, left side boundary belt number N _B1=6, the right boundary belt number N _B2=5,0 frequency is 0.

Step 1: the signal that receives is removed Cyclic Prefix, obtain the sequence X that length is N, again it is divided into [x _0,0X _{0, N/2-1}] and [x _{0, N/2}X _{0, N-1}] two parts, all carrying out N/2 point energy normalized FFT computing, first row as the Z matrix is capable with second respectively, obtains matrix Y:

$Y=\left[\begin{array}{ccccc}{\mathrm{<figure-callout\; id="0"\; label="y"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">\; <figure-callout\; id="0"\; label="y"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">y</figure-callout>\; </figure-callout>}}_{\mathrm{0,0}}& \&CenterDot;\&CenterDot;\&CenterDot;& {\mathrm{<figure-callout\; id="0"\; label="y"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">y</figure-callout>}}_{0,j}& \&CenterDot;\&CenterDot;\&CenterDot;& {\mathrm{<figure-callout\; id="0"\; label="y"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">y</figure-callout>}}_{0,N/2-1}\\ {\mathrm{<figure-callout\; id="1"\; label="y"\; filenames="HSA00000627483800024.png"\; state="\{\{state\}\}">\; <figure-callout\; id="0"\; label="y"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">y</figure-callout>\; </figure-callout>}}_{\mathrm{1,0}}& \&CenterDot;\&CenterDot;\&CenterDot;& {\mathrm{<figure-callout\; id="1"\; label="y"\; filenames="HSA00000627483800024.png"\; state="\{\{state\}\}">y</figure-callout>}}_{1,j}& \&CenterDot;\&CenterDot;\&CenterDot;& {\mathrm{<figure-callout\; id="1"\; label="y"\; filenames="HSA00000627483800024.png"\; state="\{\{state\}\}">y</figure-callout>}}_{1,N/2-1}\end{array}\right]$

Step 2: directly with Y matrixing symbol, unification becomes the Z matrix, that is:

$Z=\left[\begin{array}{ccccc}{\mathrm{<figure-callout\; id="0"\; label="z"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">\; <figure-callout\; id="0"\; label="z"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">z</figure-callout>\; </figure-callout>}}_{\mathrm{0,0}}& \&CenterDot;\&CenterDot;\&CenterDot;& {\mathrm{<figure-callout\; id="0"\; label="z"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">z</figure-callout>}}_{0,j}& \&CenterDot;\&CenterDot;\&CenterDot;& {\mathrm{<figure-callout\; id="0"\; label="z"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">z</figure-callout>}}_{0,J-1}\\ {\mathrm{<figure-callout\; id="1"\; label="z"\; filenames="HSA00000627483800024.png"\; state="\{\{state\}\}">\; <figure-callout\; id="0"\; label="z"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">z</figure-callout>\; </figure-callout>}}_{\mathrm{1,0}}& \&CenterDot;\&CenterDot;\&CenterDot;& {\mathrm{<figure-callout\; id="1"\; label="z"\; filenames="HSA00000627483800024.png"\; state="\{\{state\}\}">z</figure-callout>}}_{1,j}& \&CenterDot;\&CenterDot;\&CenterDot;& {\mathrm{<figure-callout\; id="1"\; label="z"\; filenames="HSA00000627483800024.png"\; state="\{\{state\}\}">z</figure-callout>}}_{1,J-1}\end{array}\right]=\left[\begin{array}{ccccc}{\mathrm{<figure-callout\; id="0"\; label="y"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">\; <figure-callout\; id="0"\; label="y"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">y</figure-callout>\; </figure-callout>}}_{\mathrm{0,0}}& \&CenterDot;\&CenterDot;\&CenterDot;& {\mathrm{<figure-callout\; id="0"\; label="y"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">y</figure-callout>}}_{0,j}& \&CenterDot;\&CenterDot;\&CenterDot;& {\mathrm{<figure-callout\; id="0"\; label="y"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">y</figure-callout>}}_{0,N/2-1}\\ {\mathrm{<figure-callout\; id="1"\; label="y"\; filenames="HSA00000627483800024.png"\; state="\{\{state\}\}">\; <figure-callout\; id="0"\; label="y"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">y</figure-callout>\; </figure-callout>}}_{\mathrm{1,0}}& \&CenterDot;\&CenterDot;\&CenterDot;& {\mathrm{<figure-callout\; id="1"\; label="y"\; filenames="HSA00000627483800024.png"\; state="\{\{state\}\}">y</figure-callout>}}_{1,j}& \&CenterDot;\&CenterDot;\&CenterDot;& {\mathrm{<figure-callout\; id="1"\; label="y"\; filenames="HSA00000627483800024.png"\; state="\{\{state\}\}">y</figure-callout>}}_{1,N/2-1}\end{array}\right]$

And matrix D is complete 1 matrix

$D=\left[\begin{array}{ccccc}{\mathrm{<figure-callout\; id="0"\; label="d"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">\; <figure-callout\; id="0"\; label="d"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">d</figure-callout>\; </figure-callout>}}_{\mathrm{0,0}}& \&CenterDot;\&CenterDot;\&CenterDot;& {\mathrm{<figure-callout\; id="0"\; label="d"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">d</figure-callout>}}_{0,j}& \&CenterDot;\&CenterDot;\&CenterDot;& {\mathrm{<figure-callout\; id="0"\; label="d"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">d</figure-callout>}}_{0,J-1}\\ {\mathrm{<figure-callout\; id="1"\; label="d"\; filenames="HSA00000627483800024.png"\; state="\{\{state\}\}">\; <figure-callout\; id="0"\; label="d"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">d</figure-callout>\; </figure-callout>}}_{1,0}& \&CenterDot;\&CenterDot;\&CenterDot;& d_{01,j}& \&CenterDot;\&CenterDot;\&CenterDot;& {\mathrm{<figure-callout\; id="1"\; label="d"\; filenames="HSA00000627483800024.png"\; state="\{\{state\}\}">d</figure-callout>}}_{1,J-1}\end{array}\right]=\left[\begin{array}{ccc}1& \&CenterDot;\&CenterDot;\&CenterDot;& 1\\ 1& \&CenterDot;\&CenterDot;\&CenterDot;& 1\end{array}\right]$

Wherein, J=N/2=64.

Step 3: utilize the transmitting terminal that step 2 obtains and the matrix of receiving terminal, the computing system noise power: $\hat{W}=\frac{1}{4(J-1)}\underset{j=0}{\overset{J-2}{\mathrm{\&Sigma;}}}{|({\mathrm{<figure-callout\; id="0"\; label="z"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">z</figure-callout>}}_{0,\mathrm{<figure-callout\; id="0"\; label="j\; d"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">j</figure-callout>}}{d}_{}^{}{}_{0,j}^{*}-{\mathrm{<figure-callout\; id="0"\; label="z"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">z</figure-callout>}}_{0,j+1}{\mathrm{<figure-callout\; id="0"\; label="d"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">d</figure-callout>}}_{0,j+1}^{*})-({\mathrm{<figure-callout\; id="1"\; label="z"\; filenames="HSA00000627483800024.png"\; state="\{\{state\}\}">z</figure-callout>}}_{1,\mathrm{<figure-callout\; id="1"\; label="j\; d"\; filenames="HSA00000627483800024.png"\; state="\{\{state\}\}">j</figure-callout>}}{d}_{}^{}{}_{1,j}^{*}-{\mathrm{<figure-callout\; id="1"\; label="z"\; filenames="HSA00000627483800024.png"\; state="\{\{state\}\}">z</figure-callout>}}_{1,j+1}{\mathrm{<figure-callout\; id="1"\; label="d"\; filenames="HSA00000627483800024.png"\; state="\{\{state\}\}">d</figure-callout>}}_{1,j+1}^{*})|}^2,$ D wherein ^*Expression is got conjugation to d, J=64.

Embodiment four: with single asymmetric synchronous head data-evaluation noise power.

CAZAC is a kind of asymmetrical synchronous head, and present embodiment is that example is explained with the CAZAC sequence just.Framing format adopts Fig. 5 structure, and OFDM sub-carrier number N is 128, circulating prefix-length N _CpBe N/4, the sub-carrier number of useful data is 116, is bpsk signal and 0 form at interval, left side boundary belt number N _B1=6, the right boundary belt number N _B2=5,0 frequency is 0.

The useful data sequence be 0 with the form of the some space of CAZAC sequence, concrete form is as shown in Figure 6.The CAZAC sequence is produced by following formula:

CA(k)＝exp(-j2πk ²u/(2 _u))

Wherein, k=0:N _u-1, N _uBe CAZAC sequence length, N _u=58, the span of u is the same with k, is used to produce a certain specific sequence, u=1 in this example.Useful sequence is divided into isometric A, the B sequence, the CAZAC sequence of points is distributed in A, on the even number point of B sequence, puts into Fig. 6 again, generates synchronous head.

Step 1: the signal that receives is removed Cyclic Prefix, obtain the sequence X that length is N, it is carried out N point energy normalized FFT computing, obtain frequency domain sequence Y:

Y＝[y _0，0?…?y _0，j?…?y _0，N-1]

Step 2: from receive data, take out the useful data in the synchronous head: according to synchronous head structure shown in Figure 5, from the data that step 1 obtains, get the OFDM symbol sebolic addressing of first row earlier, Here it is OFDM synchronous head; Remove left and right sides boundary belt and 0 frequency again, promptly take out in the receiving terminal data data of A and B position in the corresponding diagram 4; Afterwards, take out even number point and the point of the even number among the B among the A, promptly the corresponding receiving terminal frequency domain data of CAZAC sequence loca is formed new matrix among A and the B:

Z＝[z _0，0?…?z _0，j?…?z _0，J-1]

Again in a manner described, from the data matrix C of transmitting terminal, take out the data of corresponding pilot tone point, form new matrix:

D＝[d _0，0?…?d _0，j?…?d _0，J-1]

Wherein, J=58.

Step 3: utilize the transmitting terminal that step 2 obtains and the matrix of receiving terminal, the computing system noise power: $\hat{W}=\frac{1}{4(J-3)}\underset{j=0}{\overset{J-4}{\mathrm{\&Sigma;}}}{|({\mathrm{<figure-callout\; id="0"\; label="z"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">z</figure-callout>}}_{0,\mathrm{<figure-callout\; id="0"\; label="j\; d"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">j</figure-callout>}}{d}_{}^{}{}_{0,j}^{*}-{\mathrm{<figure-callout\; id="0"\; label="z"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">z</figure-callout>}}_{0,j+2}{\mathrm{<figure-callout\; id="0"\; label="d"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">d</figure-callout>}}_{0,j+2}^{*})-({\mathrm{<figure-callout\; id="0"\; label="z"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">z</figure-callout>}}_{0,j+1}{\mathrm{<figure-callout\; id="0"\; label="d"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">d</figure-callout>}}_{0,j+1}^{*}-{\mathrm{<figure-callout\; id="0"\; label="z"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">z</figure-callout>}}_{0,j+3}{\mathrm{<figure-callout\; id="0"\; label="d"\; filenames="HSA00000627483800021.png,HSA00000627483800022.png"\; state="\{\{state\}\}">d</figure-callout>}}_{0,j+3}^{*})|}^2,$ D wherein ^*Expression is got conjugation to d, J=58.

Claims (3)

1. method that is used to estimate the noise power of ofdm signal; It is characterized in that comprising the steps: step 1: the initialization system parameter comprises: the selection decision according to the user adopts one of following four kinds of information to come estimated noise power: pilot tone, a plurality of synchronous head, single symmetrical synchronous head, single asymmetric synchronous head; Set left side boundary belt points N _B1, the right boundary belt points N _B2, circulating prefix-length N _Cp, synchronous head number N _p, frequency pilot sign number N _Pl, frame data of transmitting terminal are at the matrix that is expressed as of frequency domain $C=\left[\begin{array}{ccccc}{c}_{\mathrm{0,0}}& L& c_{0,n}& L& c_{0,N-1}\\ M& & M& & M\\ c_{m,0}& L& c_{m,n}& L& c_{m,N-1}\\ M& & M& & M\\ c_{M-\mathrm{1,0}}& L& c_{M-1,n}& L& c_{M-1,N-1}\end{array}\right]$ N is the number of ofdm system subcarrier, and M is the OFDM symbolic number that a frame contains;

Step 2: the information to the reception termination is received is carried out preliminary treatment, comprising:

Step 2-1: it is N that the OFDM symbol that receives is removed the sequence front end length _CpCyclic Prefix (CP), the time-domain information that obtains, following with matrix notation:

$X=\left[\begin{array}{ccccc}{x}_{\mathrm{0,0}}& L& x_{0,n}& L& x_{0,N-1}\\ M& & M& & M\\ x_{m,0}& L& x_{m,n}& L& x_{m,N-1}\\ M& & M& & M\\ x_{M-\mathrm{1,0}}& L& x_{M-1,n}& L& x_{M-1,N-1}\end{array}\right]$

Step 2-2: convert the time-domain signal X of step 2-1 into frequency-region signal: if adopt single symmetrical synchronous head to come estimated noise power, then extract first line data of the matrix X that step 2-1 obtains, it is divided into former and later two parts, that is: [x _0,0L x _{0, N/2-1}] and [x _{0, N/2}L x _{0, N-1}], respectively they are carried out the FFT of the energy normalized that N/2 orders, obtain matrix:

$Y=\left[\begin{array}{ccccc}{y}_{\mathrm{0,0}}& L& y_{0,j}& L& y_{0,N/2-1}\\ y_{\mathrm{1,0}}& L& y_{1,j}& L& y_{1,N/2-1}\end{array}\right]$

Otherwise, the energy normalized FFT that N is ordered is carried out in each provisional capital of matrix X that step 2-1 obtains, obtain matrix:

$Y=\left[\begin{array}{ccccc}{y}_{\mathrm{0,0}}& L& y_{0,n}& L& y_{0,N-1}\\ M& & M& & M\\ y_{m,0}& L& y_{m,n}& L& y_{m,N-1}\\ M& & M& & M\\ y_{M-\mathrm{1,0}}& L& y_{M-1,n}& L& y_{M-1,N-1}\end{array}\right]$

Step 3: extract useful data from receiving terminal:

(1), then according to the definition of this ofdm system standard, finds out from the structure of definition which OFDM symbol has pilot frequency information in a frame, and extract these symbols if adopt pilot tone to come estimated noise power; From them, confirm the position of pilot sub-carrier again, the reception signal data at this subcarrier place extracted, form a new pilot matrix:

$Z=\left[\begin{array}{ccccc}{z}_{0,0}& L& z_{0,j}& L& z_{0,J-1}\\ M& & M& & M\\ z_{i,0}& L& z_{i,j}& L& z_{i,J-1}\\ M& & M& & M\\ z_{N_{\mathrm{pl}}-\mathrm{1,0}}& L& z_{N_{\mathrm{pl}}-1,j}& L& z_{N_{\mathrm{pl}}-1,J-1}\end{array}\right]$

Wherein, N _PlBe the number of OFDM frequency pilot sign in the frame, J is counting of the interior pilot tone of a frequency pilot sign;

(2) if adopt a plurality of synchronous heads to come estimated noise power, the N among the matrix Y that then step 2 is obtained _pEach of individual synchronous head sequence place is gone, and all removes the N of left side boundary belt _B1The N of individual point, the right boundary belt _B2It is the point on 0 the subcarrier that individual point, 0 frequency and corresponding sends data, obtains the matrix that useful data is formed:

$Z=\left[\begin{array}{ccccc}{y}_{i_0,j_0}& L& y_{i_0,j_k}& L& y_{i_0,j_{J-1}}\\ M& & M& & M\\ y_{i_m,j_0}& L& y_{i_m,j_k}& L& y_{i_m,j_{J-1}}\\ M& & M& & M\\ y_{i_{N_p-1},j_0}& L& y_{i_{N_p-1},j_k}& L& y_{i_{N_p-1},j_{J-1}}\end{array}\right]=\left[\begin{array}{ccccc}{z}_{\mathrm{0,0}}& L& z_{0,j}& L& z_{0,J-1}\\ M& & M& & M\\ z_{i,0}& L& z_{i,j}& L& z_{i,J-1}\\ M& & M& & M\\ z_{N_p-\mathrm{1,0}}& L& z_{N_p-1,j}& L& z_{N_p-1,J-1}\end{array}\right]$

Wherein J sends the number that data are not 0 point, j in the synchronous head symbol _kSpan at N _B1Between N/2-1, perhaps N/2+1 is to N-N _B2

(3) if adopt single asymmetric synchronous head to come estimated noise power, the synchronous head sequence of first row is removed left side boundary belt N among the matrix Y that then step 2 is obtained _B1Individual point, the right boundary belt N _B2It is the point on 0 the subcarrier that individual point, 0 frequency and corresponding sends data, obtains the useful data matrix

$Z=\left[\begin{array}{ccccc}{y}_{0,j_0}& L& y_{0,j_k}& L& y_{0,j_{J-1}}\end{array}\right]=\left[\begin{array}{ccccc}{z}_{\mathrm{0,0}}& L& z_{0,j}& L& z_{0,J-1}\end{array}\right]$

Wherein J sends the number that data are not 0 point, j in the synchronous head symbol _kSpan at N _B1Between N/2-1, perhaps N/2+1 is to N-N _B2

(4) if adopt single symmetrical synchronous head to come estimated noise power, the matrix formed of useful data then

$Z=\left[\begin{array}{ccccc}{y}_{\mathrm{0,0}}& L& y_{0,j}& L& y_{0,N/2-1}\\ y_{\mathrm{1,0}}& L& y_{1,j}& L& y_{1,N/2-1}\end{array}\right]=\left[\begin{array}{ccccc}{z}_{\mathrm{0,0}}& L& z_{0,j}& L& z_{0,J-1}\\ z_{1,0}& L& z_{1,j}& L& z_{1,J-1}\end{array}\right]$

J＝N/2；

Step 4: extract useful data from transmitting terminal:

(1) if adopt pilot tone to come estimated noise power, then in transmitting terminal frequency-region signal Matrix C, confirm that earlier which OFDM symbol has pilot frequency information, extract corresponding row sequence among the C; From these row sequences, confirm the position of pilot sub-carrier again, the transmission information at these subcarrier places extracted, form a new pilot matrix:

$D=\left[\begin{array}{ccccc}{d}_{\mathrm{0,0}}& L& d_{0,j}& L& d_{0,J-1}\\ M& & M& & M\\ d_{i,0}& L& d_{i,j}& L& d_{i,J-1}\\ M& & M& & M\\ d_{N_{\mathrm{pl}}-\mathrm{1,0}}& L& d_{N_{\mathrm{pl}}-1,j}& L& d_{N_{\mathrm{pl}}-1,J-1}\end{array}\right]$

J is counting of the interior pilot tone of a symbol;

(2) if adopt a plurality of synchronous heads to come estimated noise power, then at the N of Matrix C _pDuring individual synchronous head sequence is expert at, remove left side boundary belt N _B1Individual point, the right boundary belt N _B2Individual point, 0 frequency and data are the point on 0 the subcarrier, obtain the matrix that useful data is formed:

$D=\left[\begin{array}{ccccc}{d}_{\mathrm{0,0}}& L& d_{0,j}& L& d_{0,J-1}\\ M& & M& & M\\ d_{i,0}& L& d_{i,j}& L& d_{i,J-1}\\ M& & M& & M\\ d_{N_p-\mathrm{1,0}}& L& d_{N_p-1,j}& L& d_{N_p-1,J-1}\end{array}\right]=\left[\begin{array}{ccccc}{c}_{i_0,j_0}& L& c_{i_0,j_k}& L& c_{i_0,j_{J-1}}\\ M& & M& & M\\ c_{i_m,j_0}& L& c_{i_m,j_k}& L& c_{i_m,j_{J-1}}\\ M& & M& & M\\ c_{i_{N_p-1},j_0}& L& c_{i_{N_p-1},j_k}& L& c_{i_{N_p-1},j_{J-1}}\end{array}\right]$

Wherein, J sends the number that data are not 0 point, j in the synchronous head symbol _kSpan at N _B1Between N/2-1, perhaps N/2+1 is to N-N _B2

(3) if adopt single asymmetric synchronous head to come estimated noise power,, remove left side boundary belt N then with first line data in the Matrix C _B1Individual point, the right boundary belt N _B2It is the point on 0 the subcarrier that individual point, 0 frequency and corresponding sends data, obtains the matrix that useful data is formed:

$D=\left[\begin{array}{ccccc}{d}_{\mathrm{0,0}}& L& d_{0,j}& L& d_{0,J-1}\end{array}\right]=\left[\begin{array}{ccccc}{c}_{0,j_0}& L& c_{0,j_k}& L& c_{0,j_{J-1}}\end{array}\right]$

J sends the number that data are not 0 point, j in the synchronous head symbol _kSpan at N _B1Between N/2-1, perhaps N/2+1 is to N-N _B2

(4) if adopt single symmetrical synchronous head to come estimated noise power, then:

$D=\left[\begin{array}{ccccc}{d}_{\mathrm{0,0}}& L& d_{0,j}& L& d_{0,J-1}\\ d_{\mathrm{1,0}}& L& d_{01,j}& L& d_{1,J-1}\end{array}\right]=\left[\begin{array}{ccc}1& L& 1\\ 1& L& 1\end{array}\right]$

Wherein, J=N/2;

Step 5: the noise power of utilizing the useful sequence computing system of transmitting terminal that front two step obtains and receiving terminal: if adopt single asymmetric synchronous head to come estimated noise power, noise power then $\hat{W}=\frac{1}{4(J-3)}\underset{j=0}{\overset{J-4}{\mathrm{\&Sigma;}}}{|(z_{0,j}{d}_{0,j}^{*}-z_{0,j+2}{d}_{0,j+2}^{*})-(z_{0,j+1}{d}_{0,j+1}^{*}-z_{0,j+3}{d}_{0,j+3}^{*})|}^2;$ Otherwise, from matrix Z and D, select any two row respectively and be used for calculating noise power $\hat{W}=\frac{1}{4(J-1)}\underset{j=0}{\overset{J-2}{\mathrm{\&Sigma;}}}{|(z_{i,j}{d}_{i,j}^{*}-z_{i,j+1}{d}_{i,j+1}^{*})-(z_{i^{\&prime;},j}{d}_{i^{\&prime;},j}^{*}-z_{i^{\&prime;},j+1}{d}_{i^{\&prime;},j+1}^{*})|}^2,$ Wherein, i, the row number of any two row of i ' representing matrix Z and D; d ^*Expression is got conjugation to d.

2. a kind of method that is used to estimate the noise power of ofdm signal according to claim 1; It is characterized in that: the FFT of said energy normalized, be meant under the expression formula of original FFT divided by invariant wherein N be that FFT counts.

3. a kind of method that is used to estimate the noise power of ofdm signal according to claim 1 is characterized in that: any two row that said step 3 is selected from matrix Z and D respectively are adjacent two row in the matrix.

Images