CN101800728A - OFDM (Orthogonal Frequency Division Multiplexing) related coefficient signal-to-noise ratio estimation algorithm - Google Patents

Abstract

The invention discloses an OFDM (Orthogonal Frequency Division Multiplexing) related coefficient signal-to-noise ratio estimation algorithm, which mainly solves the problems of needing of prior information and high complexity of the traditional OFDM related coefficient signal-to-noise ratio estimation algorithm. The OFDM related coefficient signal-to-noise ratio estimation algorithm comprises the steps of: 1, preprocessing receiving end data to obtain OFDM baseband frequency domain data; 2, obtaining frequency domain data of two adjacent OFDM pilot frequency symbols by an OFDM receiving end; 3, carrying out relevance coefficient computation on the frequency domain data of the two adjacent OFDM symbols, obtained by the OFDM receiving end; and 4, computing to obtain a signal-to-noise ratio value of the whole OFDM symbol by using the relevance coefficient obtained by the two adjacent OFDM pilot frequency symbols, obtained by the step 3. The invention can effectively estimate the signal-to-noise ratio parameter on the OFDM symbol, and can be used in an OFDM system in a multi-system communication environment.

Description

OFDM coefficient correlation signal-to-noise ratio estimation algorithm

Technical field

The invention belongs to communication technical field, be specifically related to ofdm signal and receive.Can be used under the situation of various communication systems ofdm signal being carried out the coefficient correlation signal-to-noise ratio (SNR) estimation.

Background technology

Be accompanied by the information age of current develop rapidly, the also former surprising speed that does not have of wireless communication technology is advancing.Particularly in the past ten years, the numeral of wireless mobile communications and radio circuit manufacturing technology aspect have obtained breakthrough.The appearance of technology such as a new generation's large scale integrated circuit makes that the volume of mobile device is littler, and price is more cheap, and function is more reliable, and these have all greatly promoted the development of mobile radio telecommunications.Along with the user to the increase of various real-time multimedia business demands and the fast development of Internet technology, can estimate that following wireless communication technology will be towards broadband, the direction of high speed develops.Following wireless network can provide peak rate up to 100Mbit/s to the user.In order to support the higher rate of information throughput and the user moving speed of Geng Gao, in next generation wireless communication, use spectrum efficiency higher, the novel transmission technology that ability of anti-multipath is stronger has become common recognition.OFDM (Orthogonal Frequency Division Multiplexing is an orthogonal frequency division multiplexi) has been subjected to widely paying close attention to as a kind of new multi-carrier modulation technology, and has become the B3G system and have one of technology of competitiveness.

The multi-carrier modulation technology that with the orthogonal frequency division multiplex OFDM is representative is widely used in Modern Communication System, for example based on the wlan system of 802.11a/g, digital video broadcasting DVB T system with based on WMAN system of 802.16 etc.Adaptive technique can change parameters such as adaptively modifying modulation system (number of constellation points), encoding rate, transmitted power, spreading gain and signalling bandwidth according to current channel circumstance, so that send information to greatest extent, thereby effectively improve spectrum efficiency, this point is even more important in radio communication.Adaptive technique has become a focus in the radio communication research in recent years, has been thought one of important means that effectively improves in the wireless communication system spectrum efficiency widely.For example UMTS HSDPA high speed downlink packet inserts and CDMA2000 1X EVDO/DV all with Adaptive Modulation as one of key technology.

Compare with traditional single-carrier system, multi-carrier OFDM systems combines with adaptive technique, except carrying out on traditional time-domain the self adaptation, can also utilize the frequency domain in the multicarrier system at an easy rate, promptly on different subcarriers, use different modulation systems and distribute corresponding transmitted power according to channel situation, thereby it has very high flexibility, and can obtain better system performance.Channel estimation technique also more and more is subjected to people's attention and research as a key of adaptive technique.And be a kind of signal-to-noise ratio estimation algorithm that does not need the channel prior information based on the signal-to-noise ratio estimation algorithm of coefficient correlation, do not knowing under the situation of channel condition information that we still can carry out signal-to-noise ratio (SNR) estimation.

Summary of the invention

The objective of the invention is to solve deficiency of the prior art, propose the signal-to-noise ratio estimation algorithm of a kind of OFDM based on coefficient correlation, with the signal-to-noise ratio (SNR) estimation of realization to the receiving terminal of ofdm signal, thus the adaptive technique of realization OFDM.

The technical scheme that realizes the object of the invention is in conjunction with OFDM receiving terminal data, utilizes the calculating of coefficient correlation to carry out the OFDM signal-to-noise ratio (SNR) estimation.

Ofdm signal intelligence method of reseptance of the present invention comprises the steps:

(1) carries out preliminary treatment in the receiving terminal data, obtain OFDM base band frequency domain data

(2), obtain the frequency domain data of two adjacent OFDM frequency pilot sign at the ofdm system receiving terminal.

(3) frequency domain data of the two adjacent OFDM frequency pilot sign that the ofdm system receiving terminal is obtained carries out relative coefficient and calculates;

(4) utilize the relative coefficient of the two adjacent OFDM frequency pilot sign that step (2) obtains, calculate the snr value of whole OFDM symbol.

Description of drawings

Fig. 1 is the ofdm system block diagram of application self-adapting technology of the present invention;

Fig. 2 is the pilot frequency locations schematic diagram in the OFDM symbol of the present invention;

Fig. 3 is the present invention under multipath channel, QPSK modulation, the comparison of signal-to-noise ratio estimation algorithm performance;

Fig. 4 is under multipath channel of the present invention, the QPSK modulation, the comparison of signal-to-noise ratio estimation algorithm performance NMSE;

Fig. 5 is the signal bandwidth estimator flow chart in the method for reseptance of the present invention;

Embodiment

Signal transmits in multidiameter fading channel, and channel impulse response (CIR) can be expressed as:

$h\left(n\right)=\underset{l=0}{\overset{L-1}{\mathrm{\Σ}}}{\mathrm{\α}}_l\mathrm{\δ}(h-{\mathrm{\τ}}_l),n=\mathrm{0,1},...,N-1$

α _l, τ _lBe respectively the channel gain and the propagation delay in l footpath, L is the channel multi-path number.Suppose that channel impulse response length is no more than circulating prefix-length.Channel frequency domain response (CFR) can be expressed as:

$H\left(k\right)=\underset{l=0}{\overset{L-1}{\mathrm{\&Sigma;}}}{\mathrm{\&alpha;}}_l{e}^{-\mathrm{<figure-callout\; id="2"\; label="j"\; filenames="HSA00000055921100021.png,HSA00000055921100022.png"\; state="\{\{state\}\}">j</figure-callout>}\frac{2\mathrm{\&pi;}}{N}k{\mathrm{\&tau;}}_l},k=\mathrm{0,1},...,N-1$

In slow fading channel, the transmission coefficient of channel can be expressed as:

H _i，j≈H _i+1，j???j＝0，1，…，N _P-1

And in the OFDM symbol: p _{I, j}=p _{I+1, j}

Ofdm signal intelligence method of reseptance of the present invention comprises the steps:

Step 1, the receiving terminal data are carried out preliminary treatment, obtain OFDM base band frequency domain data;

Suppose at ofdm system synchronously fully, total sub-carrier number is that (the pilot sub-carrier number is N to N in the OFDM symbol _P).In an OFDM symbol, the position of pilot sub-carrier can be expressed as so: 0,1,2 ..., N _P-1}, the pilot frequency locations schematic diagram as shown in Figure 1.

Can be expressed as at j normalization pilot sub-carrier of i OFDM symbol of receiving terminal:

Y _i，j＝p _i，jH _i，j+ _Wi，j??j＝0，1，…，N _P-1

P wherein _{I, j}Be transmitting terminal frequency pilot sign and | p _{I, j}| ²=1, H _{I, j}It is the Channel Transmission coefficient of j subchannel of i OFDM symbol.W _{I, j}Be to obey independent identically distributed white Gaussian noise, its average is 0, and variance is σ ²

Step 2 at the ofdm system receiving terminal, is obtained the frequency domain data of two adjacent OFDM frequency pilot sign.

The set of frequency pilot sign can be expressed as in i OFDM symbol of receiving terminal:

$U\left(i\right)=\{Y_{i,0},Y_{i,1},Y_{i,2},...,Y_{i,N_P-1}\}$

The set of frequency pilot sign can be expressed as in i+1 OFDM symbol of receiving terminal so:

$U(i+1)=\{Y_{i+\mathrm{1,0}},Y_{i+\mathrm{1,1}},Y_{i+\mathrm{1,2}},...,Y_{i+1,N_P-1}\}$

Step 3, the frequency domain data of the two adjacent OFDM symbol that the ofdm system receiving terminal is obtained carries out relative coefficient and calculates;

According to estimation theory, the cross-correlation coefficient r of U (i) and U (i+1) _{U (i) U (i+1)}(m) estimate partially in the nothing at m=0 place

(m) can be expressed as:

${\hat{r}}_{U\left(i\right)U(i+1)}\left(m\right){|}_{m=0}={\hat{r}}_{U\left(i\right)U(i+1)}\left(0\right)=\frac{1}{N_P}\underset{j=0}{\overset{N_P-1}{\mathrm{\&Sigma;}}}\left[Y_{i,j}{Y^{*}}_{i+1,j}\right]$

$=\frac{1}{N_P}\underset{j=0}{\overset{N_P-1}{\mathrm{\&Sigma;}}}\left\{\right[p_{i,j}{H}_{i,j}+W_{i,j}]\&CenterDot;{[p_{i+1,j}{H}_{i+1,j}+W_{i+1,j}]}^{*}\}$

$=\frac{1}{N_P}\underset{j=0}{\overset{N_P-1}{\mathrm{\&Sigma;}}}{\left|p_{i,j}\right|}^2{\left|H_{i,j}\right|}^2+\frac{1}{N_P}\underset{j=0}{\overset{N_P-1}{\mathrm{\&Sigma;}}}{p^{*}}_{i,j}{H^{*}}_{i,j}{W}_{i,j}$

$+\frac{1}{N_P}\underset{j=0}{\overset{N_P-1}{\mathrm{\&Sigma;}}}{p}_{i,j}{H}_{i,j}{W^{*}}_{i+1,j}+\frac{1}{N_P}\underset{j=0}{\overset{N_P-1}{\mathrm{\&Sigma;}}}{W}_{i,j}{W^{*}}_{i+1,j}$

$=\frac{1}{N_P}\underset{j=0}{\overset{N_P-1}{\mathrm{\&Sigma;}}}{\left|H_{i,j}\right|}^2+\frac{1}{N_P}\underset{j=0}{\overset{N_P-1}{\mathrm{\&Sigma;}}}{p^{*}}_{i,j}{H^{*}}_{i,j}{W}_{i,j}$

$+\frac{1}{N_P}\underset{j=0}{\overset{N_P-1}{\mathrm{\&Sigma;}}}{p}_{i,j}{H}_{i,j}{W^{*}}_{i+1,j}+\frac{1}{N_P}\underset{j=0}{\overset{N_P-1}{\mathrm{\&Sigma;}}}{W}_{i,j}{W^{*}}_{i+1,j}$

Do not have partially and estimate

(m) desired value can be expressed as:

$E\left[{\hat{r}}_{U\left(i\right)U(i+1)}\left(0\right)\right]$

$=E[\frac{1}{N_P}\underset{j=0}{\overset{N_P-1}{\mathrm{\&Sigma;}}}{\left|H_{i,j}\right|}^2+\frac{1}{N_P}\underset{j=0}{\overset{N_P-1}{\mathrm{\&Sigma;}}}{p^{*}}_{i,j}{H^{*}}_{i,j}{W}_{i,j}$

$+\frac{1}{N_P}\underset{j=0}{\overset{N_P-1}{\mathrm{\&Sigma;}}}{p}_{i,j}{H}_{i,j}{W^{*}}_{i+1,j}+\frac{1}{N_P}\underset{j=0}{\overset{N_P-1}{\mathrm{\&Sigma;}}}{W}_{i,j}{W^{*}}_{i+1,j}]$

$=\frac{1}{N_P}\underset{j=0}{\overset{N_P-1}{\mathrm{\&Sigma;}}}E\left[{\left|H_{i,j}\right|}^2\right]+\frac{1}{N_P}\underset{j=0}{\overset{N_P-1}{\mathrm{\&Sigma;}}}{p^{*}}_{i,j}{H^{*}}_{i,j}E\left[W_{i,j}\right]$

$+\frac{1}{N_P}\underset{j=1}{\overset{N_P-1}{\mathrm{\&Sigma;}}}{p}_{i,j}{H}_{i,j}E\left[{W^{*}}_{i+1,j}\right]+\frac{1}{N_P}\underset{j=0}{\overset{N_P-1}{\mathrm{\&Sigma;}}}E\left[W_{i,j}\right]E\left[{W^{*}}_{i+1,j}\right]]$

$=\frac{1}{N_P}\underset{j=0}{\overset{N_P-1}{\mathrm{\&Sigma;}}}E\left[{\left|H_{i,j}\right|}^2\right]=r_{U\left(i\right)U(i+1)}\left(0\right)$

The establishment of following formula is because white Gaussian noise itself is exactly incoherent, that is:

E[W _i，j]＝E[W* _i+1，j]＝0

Following formula cross-correlation coefficient r then _{U (i) U (i+1)}(m) can be expressed as at m=0:

$r_{U\left(i\right)U(i+1)}\left(m\right){|}_{m=0}\&ap;\frac{1}{N_P}\underset{j=0}{\overset{N_P-1}{\mathrm{\&Sigma;}}}{\left|H_{i,j}\right|}^2$

Order: $E_{U\left(i\right)}=\frac{1}{N_P}\underset{k=0}{\overset{N_P-1}{\mathrm{\&Sigma;}}}{\left|Y_{{i,\mathrm{kD}}_f}\right|}^2,E_{U(i+1)}=\frac{1}{N_P}\underset{k=0}{\overset{N_P-1}{\mathrm{\&Sigma;}}}{\left|Y_{i+1,{\mathrm{kD}}_f}\right|}^2$

Then have, $E\left[E_{U\left(i\right)}\right]\&cong;E\left[E_{U(i+1)}\right]$

The coefficient correlation of U (i) and these two sequences of U (i+1) can be expressed as:

${\mathrm{\&rho;}}_{U\left(i\right)U(i+1)}=\frac{\frac{1}{N_P}\underset{j=0}{\overset{N_P}{\mathrm{\&Sigma;}}}[Y_{i,j}\&CenterDot;Y_{i+1,j}]}{\sqrt{\frac{1}{N_P}\underset{j=0}{\overset{N_P-1}{\mathrm{\&Sigma;}}}{\left|Y_{i,j}\right|}^2}\sqrt{\frac{1}{N_P}\underset{k=0}{\overset{N_P-1}{\mathrm{\&Sigma;}}}{\left|Y_{i+1,j}\right|}^2}}$

$=\frac{r_{U\left(i\right)U(i+1)}\left(0\right)}{\sqrt{E_{U\left(i\right)}}\sqrt{E_{U(i+1)}}}=\frac{r_{U\left(i\right)U(i+1)}\left(0\right)}{E_{U\left(i\right)}}=\frac{\frac{1}{N_P}\underset{j=0}{\overset{N_P-1}{\mathrm{\&Sigma;}}}{\left|H_{i,j}\right|}^2}{E_{U\left(i\right)}}$

Step 4 is utilized the relative coefficient of the two adjacent OFDM symbol that step (3) obtains, and calculates the snr value of whole OFDM symbol.

The signal-to-noise ratio (SNR) estimation of being carried out for the set U (i) of frequency pilot sign in i OFDM symbol of receiving terminal can be expressed as:

${\mathrm{SNR}}_{U\left(i\right)}=10\log \left(\frac{\frac{1}{N_P}\underset{j=0}{\overset{N_P-1}{\mathrm{\&Sigma;}}}{\left|p_{i,j}{H}_{i,j}\right)|}^2}{\frac{1}{N_P}\underset{j=0}{\overset{N_P-1}{\mathrm{\&Sigma;}}}{\left|W_{i,j}\right|}^2}\right)=10\log \left(\frac{\frac{1}{N_P}\underset{j=0}{\overset{N_P-1}{\mathrm{\&Sigma;}}}{\left|H_{i,j}\right)|}^2}{\frac{1}{N_P}\underset{j=0}{\overset{N_P-1}{\mathrm{\&Sigma;}}}{\left|W_{i,j}\right|}^2}\right)$

$=10\log \left(\frac{\frac{1}{N_P}\underset{j=0}{\overset{N_P-1}{\mathrm{\&Sigma;}}}{\left|p_{i,j}{H}_{i,j}\right|}^2}{\frac{1}{N_P}\underset{j=0}{\overset{N_P}{\mathrm{\&Sigma;}}}{\left|Y_{i,j}\right|}^2-\frac{1}{N_P}\underset{j=0}{\overset{N_P}{\mathrm{\&Sigma;}}}{\left|p_{i,j}{H}_{i,j}\right)|}^2}\right)$

$=10\log \left(\frac{{\mathrm{\&rho;}}_{U\left(i\right)U(i+1)}\&CenterDot;E_{U\left(i\right)}}{E_{U\left(i\right)}-{\mathrm{\&rho;}}_{U\left(i\right)U(i+1)}\&CenterDot;E_{U\left(i\right)}}\right)=10\log \left(\frac{{\mathrm{\&rho;}}_{U\left(i\right)U(i+1)}}{1-{\mathrm{\&rho;}}_{U\left(i\right)U(i+1)}}\right)$

Be that we can carry out signal-to-noise ratio (SNR) estimation according to the correlation of signal, and do not need to know the channel statistical relevant information.Normalized root-mean-square error (Normalized Mean Square Error) can be expressed as:

$\mathrm{NMSE}=\frac{|{\mathrm{SNR}}_{U\left(i\right)}-\mathrm{SNR}|}{\left|\mathrm{SNR}\right|}$

The present invention compared with prior art has the following advantages:

1, method of reseptance of the present invention does not need prior information, can be used for the signal-to-noise ratio (SNR) estimation of ofdm signal.

2, the present invention has overcome other signal-to-noise ratio (SNR) estimation such as the high defective of MMSE signal-to-noise ratio estimation algorithm computation complexity owing to adopt coefficient correlation to carry out the OFDM signal-to-noise ratio (SNR) estimation;

3, it is less that signal-to-noise ratio (SNR) estimation of the present invention is subjected to the influence of frequency deviation, still can obtain good performance under the situation that has the frequency offset estimating error.

Advantage of the present invention can illustrate by following simulation performance:

1) simulated conditions: this partial simulation platform is with reference to the WLAN protocol emulation, and the concrete parameter of WLAN agreement is as shown in the table.Wherein: channel is the multipath Rayleigh channel; Comprise 48 of data subcarriers among the OFDM, 4 of pilot sub-carriers, 12 of gap carrier waves.The FFT cycle is 64, and modulation system is: BPSK, QPSK, 16QAM, 64QAM.

System parameters among the WLAN

Parameter	Value
		Length (the N of FFT FFT)	??64
Number of subcarriers (the N that uses used)	??48
		Protection is (T at interval g) and effective OFDM symbol lengths (T b) ratio	??1/4
The protection carrier number of frequency low side	??6
		The protection carrier number that frequency is high-end	??5
The numbering of protection subcarrier	??-32，…，-27，…，0，…，27，…，??31
		The pilot frequency locations subcarrier number	??-21，-7，7，21
Subcarrier modulation modes	??BPSK、QPSK、16QAM，64QAM

Emulation is under multipath channel, QPSK modulation, to XU, MMSE with based on the emulation of coefficient correlation signal-to-noise ratio estimation algorithm performance.Used two kinds of different pilot tones based on the coefficient correlation signal-to-noise ratio estimation algorithm in emulation, a kind of is the frame head training sequence of a frame, and a kind of is frequency pilot sign in the OFDM symbol.

2) simulation result

Fig. 3 finds out to the simulation result of Fig. 5: 1. under multipath channel, utilize the frame head training sequence best based on coefficient correlation signal-to-noise ratio estimation algorithm performance, next is the MMSE signal-to-noise ratio estimation algorithm that utilizes frequency pilot sign, then be utilize frequency pilot sign based on coefficient correlation signal-to-noise ratio estimation algorithm performance, be the XU algorithm at last.2. the signal to noise ratio of utilizing training sequence to estimate compares that to utilize frequency pilot sign to do the signal-to-noise ratio (SNR) estimation performance good, and this is that the amount of information that frequency pilot sign contained is fewer because have only 4 pilot tones in WLAN.3. exist under the situation of frequency deviation, signal-to-noise ratio estimation algorithm all can be affected, and relatively good based on the signal-to-noise ratio estimation algorithm performance of coefficient correlation comparatively speaking, the influence that is subjected to is smaller.

When system's average signal-to-noise ratio was 5dB-20dB, channel was under the situation of Gaussian channel, flat fading channel and multipath channel below, and the signal-to-noise ratio estimation algorithm based on coefficient correlation based on training sequence that adopts in this emulation platform is tested.Test data is as shown in the table:

The signal-to-noise ratio (SNR) estimation test data

??SNR(dB)	??2	??4	??6	??8	??10
						Gaussian channel	??1.99	??4.04	??6.01	??8.02	??10.00
Flat fading channel	??2.01	??4.08	??6.00	??8.08	??10.04
						Multipath channel	??2.07	??4.08	??6.04	??8.04	??10.06
??SNR(dB)	??12	??14	??16	??18	??20
						Gaussian channel	??12.04	??14.04	??16.06	??18.03	??20.06
Flat fading channel	??12.03	??14.03	??16.01	??18.04	??20.7
						Multipath channel	??12.06	??14.04	??16.04	??18.03	??19.98

By top test data as can be known, can under Gaussian channel, flat fading channel and multipath channel, obtain good estimation performance based on the signal-to-noise ratio estimation algorithm of coefficient correlation.

Claims (3)

1. OFDM signal-to-noise ratio estimation algorithm comprises following process:

At the ofdm system receiving terminal, obtain the frequency domain data of two adjacent OFDM frequency pilot sign.

The set of frequency pilot sign can be expressed as in i OFDM symbol of receiving terminal:

$U\left(i\right)=\{Y_{i,0},Y_{i,1},Y_{i,2},...,Y_{i,N_P-1}\}$

The set of frequency pilot sign can be expressed as in i+1 OFDM symbol of receiving terminal so:

$U(i+1)=\{Y_{i+\mathrm{1,0}},Y_{i+\mathrm{1,1}},Y_{i+\mathrm{1,2}},...,Y_{i+1,N_P-1}\}$

The coefficient correlation of U (i) and these two sequences of U (i+1) can be expressed as:

${\mathrm{\&rho;}}_{U\left(i\right)U(i+1)}=\frac{\frac{1}{N_P}\underset{j=0}{\overset{N_P-1}{\mathrm{\&Sigma;}}}[Y_{i,j}\&CenterDot;Y_{i+1,j}]}{\sqrt{\frac{1}{N_P}\underset{j=0}{\overset{N_P-1}{\mathrm{\&Sigma;}}}{\left|Y_{i,j}\right|}^2}\sqrt{\frac{1}{N_P}\underset{k=0}{\overset{N_P-1}{\mathrm{\&Sigma;}}}{\left|Y_{i+1,j}\right|}^2}}$

$=\frac{r_{U\left(i\right)U(i+1)}\left(0\right)}{\sqrt{E_{U\left(i\right)}}\sqrt{E_{U(i+1)}}}=\frac{r_{U\left(i\right)U(i+1)}\left(0\right)}{E_{U\left(i\right)}}=\frac{\frac{1}{N_P}\underset{j=0}{\overset{N_P-1}{\mathrm{\&Sigma;}}}{\left|H_{i,j}\right|}^2}{E_{U\left(i\right)}}$

The signal-to-noise ratio (SNR) estimation of being carried out for the set U (i) of frequency pilot sign in i OFDM symbol of receiving terminal can be expressed as:

${\mathrm{SNR}}_{U\left(i\right)}=10\log \left(\frac{\frac{1}{N_P}\underset{j=0}{\overset{N_P-1}{\mathrm{\&Sigma;}}}{\left|p_{i,j}{H}_{i,j}\right)|}^2}{\frac{1}{N_P}\underset{j=0}{\overset{N_P-1}{\mathrm{\&Sigma;}}}{\left|W_{i,j}\right|}^2}\right)=10\log \left(\frac{\frac{1}{N_P}\underset{j=0}{\overset{N_P-1}{\mathrm{\&Sigma;}}}{\left|H_{i,j}\right)|}^2}{\frac{1}{N_P}\underset{j=0}{\overset{N_P-1}{\mathrm{\&Sigma;}}}{\left|W_{i,j}\right|}^2}\right)$

$=10\log \left(\frac{\frac{1}{N_P}\underset{j=0}{\overset{N_P-1}{\mathrm{\&Sigma;}}}{\left|p_{i,j}{H}_{i,j}\right)|}^2}{\frac{1}{N_P}\underset{j=0}{\overset{N_P}{\mathrm{\&Sigma;}}}{\left|Y_{i,j}\right|}^2-\frac{1}{N_P}\underset{j=0}{\overset{N_P-1}{\mathrm{\&Sigma;}}}{\left|p_{i,j}{H}_{i,j}\right)|}^2}\right)$

$=10\log \left(\frac{{\mathrm{\&rho;}}_{U\left(i\right)U(i+1)}\&CenterDot;E_{U\left(i\right)}}{E_{U\left(i\right)}-{\mathrm{\&rho;}}_{U\left(i\right)U(i+1)}\&CenterDot;E_{U\left(i\right)}}\right)=10\log \left(\frac{{\mathrm{\&rho;}}_{U\left(i\right)U(i+1)}}{1-{\mathrm{\&rho;}}_{U\left(i\right)U(i+1)}}\right)$

2. OFDM signal-to-noise ratio estimation algorithm according to claim 1, ofdm system is synchronous fully, and in slow fading channel, the transmission coefficient of channel can be expressed as:

H _i，j≈H _i+1，j?????j＝0，1，…，N _P-1

3. OFDM signal-to-noise ratio estimation algorithm according to claim 1, ofdm system frequency offset estimating error can not be very big.The frequency offset estimating error is too big, can influence the signal-to-noise ratio (SNR) estimation value of ofdm system.

Images