CN101562590B - OFDM signal intelligent receiving system and receiving method - Google Patents

Abstract

The invention discloses an OFDM signal intelligent receiving system and a receiving method, which mainly solves the problem that the existing intelligent receiving technology is difficult to identify the OFDM signal and the high-order square QAM signal. The receiving process is: (1) preprocessing and down-converting the captured signal to obtain the baseband signal; (2) estimating the bandwidth of the baseband signal; (3) sampling the baseband signal at a frequency four times the signal bandwidth; ( 4) Use multi-layer wavelet decomposition to extract the eigenvector of the sampled signal, and compare it with the theoretical value to identify the OFDM signal; (5) In the case of identifying the OFDM signal, estimate the symbol length and cyclic prefix length of the OFDM signal, etc. Parameters to achieve symbol synchronization; and identify and demodulate subcarrier modulation modes for OFDM signals. The invention can effectively identify the OFDM signal, estimate the parameters of the OFDM signal, and can intelligently receive the OFDM signal in a multi-system communication environment.

Description

OFDM signal intelligent receiving system and receiving method

technical field

The invention belongs to the technical field of communication, and in particular relates to an OFDM signal receiving method. It can be used for identification and parameter estimation of OFDM signals in the case of unknown communication system.

Background technique

In recent years, with the development of digital communication technology and digital signal processing technology, there are more and more various communication modes, and the environment of wireless communication is becoming increasingly complex. In many applications, it is necessary to monitor the activity of the communication signal, distinguish the nature of the signal, and even intercept the content of the signal. The communication receiver should have stronger and stronger parameter identification capabilities, which are inseparable from the intelligent receiving technology of the communication signal.

In May 1992, Joseph Mitola of MILTRE company clearly proposed the concept of software radio for the first time. That is to construct an open, standardized, and modular general-purpose hardware platform, and use software to complete various functions, such as working frequency bands, modulation and demodulation types, data formats, encryption modes, and communication protocols, and make broadband A/D And the D/A converter is as close as possible to the antenna, in order to develop a new generation of wireless communication system with high flexibility and openness. Generally, it is based on a standardized and modular hardware platform with versatility, and realizes various functions of the radio station through software programming. Software radio emphasizes the openness and comprehensive programmability of the architecture, and changes the hardware configuration structure through software updates to achieve different functions.

Multi-carrier modulation technology represented by Orthogonal Frequency Division Multiplexing (OFDM) is widely used in modern communication systems, such as WLAN systems based on 802.11a/g, digital video broadcasting DVB_T systems and WMAN systems based on 802.16, etc. In order to identify the communication mode, the receiver first distinguishes the signal into single carrier signal and multi-carrier signal, then estimates the signal parameters, and identifies the communication mode according to the signal parameters. If the communication mode based on OFDM is unknown, the receiver needs to identify the parameters of the OFDM system to achieve subsequent monitoring, demodulation and other purposes.

The intelligent reception of signals includes signal acquisition, carrier estimation, baud rate estimation, signal identification and parameter estimation, etc. At present, there is no specific implementation process for the research on OFDM intelligent reception technology in the literature, and it is impossible to realize the intelligent reception of OFDM signals.

For the signal baud rate estimation unit in the signal intelligent receiving system. The existing baud rate estimation methods can be divided into time-domain baud rate estimation based on signal cyclostationary characteristics and frequency-domain baud rate estimation based on signal FFT transformation, mainly for single-carrier signal baud rate estimation. In the intelligent reception of signals, when it is necessary to estimate the baud rate of the signal, the type of the signal is unknown a priori. If an OFDM signal is received, the existing baud rate estimation algorithm will be invalid.

For signal identification of OFDM and single carrier in intelligent signal reception. In the document "Detection of multicarrier modulations using 4th-order cumulants", W.Akmouche first proposed the signal identification method of OFDM and single carrier using 4th-order cumulant in additive white Gaussian noise (AWGN) channel. In the document "Blind Identification of OFDM Signal in Rayleigh Channels", Bin Wang and Lindong Ge extended the method of using high-order cumulants to identify OFDM signals to Rayleigh fading channels, and achieved good identification results. However, using high-order cumulants to identify OFDM cannot distinguish OFDM signals from high-order square QAM signals in a single carrier.

Contents of the invention

The purpose of the present invention is to solve the deficiencies in the prior art, to propose an OFDM intelligent receiving system and its receiving method, to realize the reception and parameter estimation of OFDM signals, and to distinguish OFDM signals from those including high-order square QAM signals single carrier signal.

The technical solution for realizing the purpose of the present invention is to combine wavelet decomposition, reconstruction and Fourier transform to estimate baud rate, and utilize multi-layer wavelet decomposition to perform OFDM signal recognition. Its systems include:

a signal capturing unit, configured to capture and receive signals;

A preprocessing unit is used for filtering, amplifying and A/D conversion preprocessing on the received signal;

Carrier estimation and down-conversion unit, used for carrier frequency estimation and down-conversion to the preprocessed signal, to obtain baseband signal S(n);

A bandwidth estimation unit is used to estimate the bandwidth of the baseband signal S (n), and pass the estimation result W to the subsequent sampling unit, the OFDM parameter estimation unit and the OFDM signal demodulation unit;

The sampling unit is used to sample the baseband signal S(n) to obtain a 4 times oversampled baseband signal S'(n)

The OFDM signal identification unit is used to distinguish the received signal as an OFDM signal or a single carrier signal according to the 4 times oversampled baseband signal S'(n);

The OFDM signal parameter estimation unit is used to estimate the number of OFDM subcarriers N, cyclic prefix length N _G , and symbol start according to the baseband signal S(n) and the estimated signal bandwidth W under the condition that the received signal is judged to be an OFDM signal Position offset θ, obtain the OFDM signal S "(n) after symbol synchronization according to θ, and pass the estimated OFDM subcarrier number N and cyclic prefix length _N to the subsequent OFDM signal demodulation unit;

_The OFDM signal demodulation unit is used to judge the OFDM signal S "(n ) to identify and demodulate subcarrier modulation modes.

OFDM signal intelligent receiving method of the present invention, comprises the steps:

(1) Capture the signal in the monitoring frequency band, and perform filtering, amplification and A/D conversion preprocessing on the captured signal;

(2) Carrier estimation and down-conversion processing are performed on the preprocessed received signal to obtain the baseband signal S(n);

(3) Obtain the baseband signal spectrum by Fourier transform for the baseband signal S(n), use Haar wavelet decomposition and wavelet reconstruction to filter out the detailed components in the baseband signal spectrum to obtain a step-type baseband signal spectrum, and use statistical methods to judge the signal The edge position of the spectrum to get the signal bandwidth W;

(4) oversampling the baseband signal S(n) by 4 times the estimated signal bandwidth W, to obtain a 4 times oversampling baseband signal S'(n);

(5) Extract the eigenvector of S′(n) through multi-layer wavelet decomposition, calculate the Euclidean distance between the eigenvector and the theoretical value of the OFDM signal eigenvector, compare the distance with the threshold, and when the distance is less than the threshold, the discrimination reception The signal is an OFDM signal, and when the distance is greater than the threshold, it is judged that the received signal is a single-carrier signal;

(6) In the case that the discrimination signal is an OFDM signal, according to the baseband signal S(n) and the estimated signal bandwidth W, estimate the number N of OFDM subcarriers, the length of the cyclic prefix N _G and the offset θ of the symbol start position, Obtain the OFDM baseband signal S " (n) after symbol synchronization according to the estimation result of θ, and the signal bandwidth W of estimation is as the sample value rate of OFDM signal;

(7) In the case that the discrimination signal is an OFDM signal, according to the estimated number of OFDM subcarriers N, the cyclic prefix length N _G , the sampling rate W and the OFDM baseband signal S″(n) after symbol synchronization, the OFDM baseband The signal undergoes subcarrier modulation identification and demodulation.

Compared with the prior art, the present invention has the following advantages:

1. The present invention combines Fourier transform and Haar wavelet decomposition, and can estimate the baud rate of single-carrier signals and OFDM signals by estimating the signal bandwidth.

2. The present invention overcomes the shortcoming that high-order cumulants cannot distinguish OFDM signals and high-order square QAM signals due to the use of multi-layer wavelet decomposition for OFDM signal identification;

3. The receiving method of the present invention does not require prior information, and can be used for intelligent reception of OFDM signals.

Description of drawings

Fig. 1 is a system block diagram of the present invention;

Fig. 2 is a flow chart of the receiving method of the present invention;

Fig. 3 is a subflow chart of signal bandwidth estimation in the receiving method of the present invention;

Fig. 4 is a sub-flow chart of OFDM signal identification in the receiving method of the present invention.

Detailed ways

Referring to Fig. 1, the OFDM signal intelligent receiving system of the present invention includes: a signal acquisition unit, a signal preprocessing unit, a carrier estimation and a down-conversion unit, a signal bandwidth estimation unit, a sampling unit, an OFDM signal identification unit, an OFDM parameter estimation unit and an OFDM Signal demodulation unit. The signal capture unit detects the signal energy in the monitoring frequency band, and when the signal energy is greater than a certain threshold, captures and receives the signal, and sends the received signal to the signal preprocessing unit. The signal preprocessing unit performs filtering, amplification and A/D conversion preprocessing on the signal received by the signal acquisition unit, and sends the preprocessed signal to the carrier estimation and frequency down conversion unit. The carrier estimation and down-conversion unit performs carrier frequency estimation and down-conversion on the preprocessed signal to obtain the baseband signal S(n), and sends S(n) to the signal bandwidth estimation unit. The signal bandwidth estimation unit estimates the bandwidth of the baseband signal S(n), obtains the estimation result W, and transmits the baseband signal S(n) and the estimated signal bandwidth W to the subsequent sampling unit, OFDM parameter estimation unit and OFDM signal demodulation unit. The sampling unit samples the baseband signal S(n), the sampling frequency is 4 times the estimated signal bandwidth W, obtains the 4 times oversampled baseband signal S'(n), and passes S'(n) to the OFDM signal identification unit. The OFDM signal identification unit judges whether the signal S'(n) is an OFDM signal according to the 4 times oversampled baseband signal S'(n), and transmits the judgment result and the baseband signal S(n) to the OFDM signal synchronization and parameter estimation unit . The OFDM signal parameter estimation unit estimates the number N of OFDM subcarriers, the length of the cyclic prefix N _G and the start of the symbol according to the baseband signal S(n) and the estimated bandwidth W under the condition that the OFDM signal identification unit judges that the received signal is an OFDM signal The position offset θ, according to θ, the OFDM signal S″(n) after symbol synchronization is obtained, and the signal S″(n), the estimated number of OFDM subcarriers N, and the cyclic prefix length N _G are passed to the subsequent OFDM signal solution tune unit. OFDM signal demodulation unit is under the situation that OFDM signal identification unit distinguishes received signal as OFDM signal, according to OFDM subcarrier number N estimated by OFDM signal parameter estimation unit, cyclic prefix length N _G and the signal bandwidth W estimated by signal bandwidth estimation unit , performing subcarrier modulation identification and demodulation on the symbol-synchronized OFDM signal S″(n).

With reference to Fig. 2, OFDM signal intelligent receiving method of the present invention, comprises the steps:

Step 1, preprocessing the captured signal.

The signal energy in the monitoring frequency band is detected, and when the signal energy exceeds a certain threshold, the signal is captured and received, and the received signal is filtered, amplified and A/D converted for preprocessing.

Step 2, performing carrier estimation and down-conversion processing on the preprocessed received signal to obtain a baseband signal S(n).

Step 3, estimate the signal bandwidth W of the baseband signal S(n).

With reference to Fig. 3, the step of the signal bandwidth W of estimation baseband signal S (n) is as follows:

(3a) group the received baseband signal S(n), divide it into M groups, and divide it into N subgroups in each group;

(3b) performing Fourier transform on the baseband signal of the jth (1≤j≤N) subgroup in the ith (1≤i≤M) group to obtain the baseband signal spectrum S _{i, j} (f);

(3c) Perform n (n≥1) layers of Haar wavelet decomposition on the baseband signal spectrum S _i,j (f), and obtain the wavelet decomposition coefficient vector T.

(3d) Keep the approximate components in the coefficient vector T unchanged, and set the detail components to 0 to filter out the detail components in the baseband signal spectrum, and obtain the coefficient vector T' for wavelet reconstruction.

(3e) Perform Haar wavelet reconstruction on the signal spectrum according to the coefficient vector T′ to obtain the reconstructed step-type baseband signal spectrum. In steps (3c) and (3d), if the number of wavelet decomposition layers n is too low, the signal spectrum after wavelet reconstruction with high-frequency coefficients set to 0 is not smooth enough to perform edge detection; if the number of wavelet decomposition layers n is too high, The resolution of wavelet reconstruction after high-frequency coefficients are set to 0 is very low. By selecting an appropriate number of decomposition layers, high-frequency details in the spectrum can be eliminated and high resolution can be obtained. In this example, the number of wavelet decomposition layers n=6 is selected.

(3f) Statistically average the N step-type baseband signal spectra S' _{i, j} (f) (1≤j≤N) obtained by the i-th group of signals, and obtain the statistically averaged spectrum S' _i (f), The position where the difference between adjacent values is the largest in the frequency spectrum S' _i (f) is the steepest change position;

(3g) Find the two steepest changing positions of the S' _i (f), as the spectral edge positions of the i-th group of signals;

(3h) Count the spectrum edge positions of the M groups of signals, and use the bandwidth between the two positions with the highest statistical value as the estimated signal bandwidth W.

Step 4, sampling the baseband signal S(n).

Taking 4 times the signal bandwidth W as the sampling frequency, the baseband signal S(n) is sampled to obtain a 4 times oversampled baseband signal S'(n).

Step 5, judging whether the 4 times oversampled baseband signal S'(n) is an OFDM signal.

With reference to Fig. 4, the step of discriminating whether the baseband signal S' (n) of 4 times oversampling is OFDM signal is as follows:

(5a) Use multi-layer wavelet decomposition to construct the classification feature vector for the 4 times oversampled baseband signal S′(n):

$\mathrm{<span\; class="notranslate">\; x</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; C</span>}<span\; class="notranslate">\; \&Center\; Dot;</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{{\mathrm{<span\; class="notranslate">\; ap</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}}<span\; class="notranslate">\; [</span>{\mathrm{<span\; class="notranslate">\; ap</span>}}_{\mathrm{<span\; class="notranslate">\; 3</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; ap</span>}}_{\mathrm{<span\; class="notranslate">\; 4</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; ap</span>}}_{\mathrm{<span\; class="notranslate">\; 5</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; de</span>}}_{\mathrm{<span\; class="notranslate">\; 3</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; de</span>}}_{\mathrm{<span\; class="notranslate">\; 4</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; de</span>}}_{\mathrm{<span\; class="notranslate">\; 5</span>}}<span\; class="notranslate">\; ]</span>$

Among them, C is a parameter to adjust the magnitude of the eigenvector value, and this example selects C=100,

ap _i represents the approximate partial energy of the wavelet decomposition of the i-th layer of the signal, defined as follows:

${\mathrm{ap}}_i={\left|\right|a_i\left|\right|}_2^2=\underset{\mathrm{no}}{\mathrm{\&Sigma;}}{a}_{i,\mathrm{no}}^2,$ i=1,..., M, a _i represent the approximate partial number vector of the i-th layer wavelet decomposition, a _{i, n} represent the approximate partial coefficients of the i layer wavelet decomposition;

de _i represents the energy of the detail part of the wavelet decomposition of the i-th layer of the signal, defined as follows:

${\mathrm{de}}_i={\left|\right|d_i\left|\right|}_2^2=\underset{\mathrm{no}}{\mathrm{\&Sigma;}}{d}_{i,\mathrm{no}}^2,$ i=1,..., M, d _i represent the detail part coefficient vector of the i-th layer wavelet decomposition, d _{i, n} represents the detail part coefficient of the i-layer wavelet decomposition;

Since the energy of the OFDM signal is uniformly distributed in the frequency band, the theoretical value of its eigenvector x is $m=C\&times;[\frac{1}{2},\frac{1}{4},\frac{1}{8},\frac{1}{2},\frac{1}{4},\frac{1}{8}];$

(5b) Calculate the Euclidean distance D between the eigenvector x and the theoretical value M of the OFDM signal eigenvector:

$\mathrm{<span\; class="notranslate">\; D.</span>}<span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; l</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; 6</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; l</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; l</span>}}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; 2</span>}}$

Among them, x _l is the value of the lth dimension of the feature vector x, and m _l is the value of the lth dimension of the feature vector M;

(5c) Extract the eigenvector according to step (5a) for the single-carrier signal training sample, calculate the Euclidean distance D _SC between the eigenvector and the theoretical value M, and count the probability distribution p(D _SC ) of D _SC ; for the OFDM signal training sample, press Step (5a) extracts the feature vector, calculates the Euclidean distance D _OFDM between the feature vector and the theoretical value M, and counts the probability distribution p(D _OFDM ) of D _OFDM , according to p(D _SC ) and p(D _OFDM ) according to the maximum likelihood The detection criterion sets the judgment threshold, which can also be set based on experience, and the threshold is set to 33 in this example.

(5d) Comparing D with the decision threshold, when D is smaller than the decision threshold, it is judged as an OFDM signal, and when D is greater than the decision threshold, it is judged as a single-carrier signal.

Step 6, in the case that the received signal is judged as an OFDM signal, perform parameter estimation on the baseband signal S(n).

In the case that the received signal is identified as an OFDM signal, the parameter estimation of the baseband signal S(n) includes the number N of OFDM subcarriers, the length N _G of the cyclic prefix and the offset θ of the symbol starting position, and the steps are as follows:

(6a) The symbols of the OFDM signal have a cyclic prefix structure, and the cyclic prefix is a copy of the end of the valid data, so there is a correlation between them. If the autocorrelation of the received data is computed, the peak value is obtained when the cyclic prefix is correlated with its replicating source. Therefore, an autocorrelation algorithm with variable correlation length can be used to estimate the effective data length:

${\mathrm{<span\; class="notranslate">\; R</span>}}_{\mathrm{<span\; class="notranslate">\; S</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; \&Delta;</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span><span\; class="notranslate">\; |</span>\mathrm{<span\; class="notranslate">\; E.</span>}<span\; class="notranslate">\; [</span>\mathrm{<span\; class="notranslate">\; S</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; )</span>{\mathrm{<span\; class="notranslate">\; S</span>}}^{<span\; class="notranslate">\; *</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; \&Delta;</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; ]</span><span\; class="notranslate">\; |</span><span\; class="notranslate">\; =</span>\left\{\begin{array}{cc}{\mathrm{<span\; class="notranslate">\; \&sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; S</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; \&sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; w</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}& \mathrm{<span\; class="notranslate">\; \&Delta;</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0</span>}\\ {\mathrm{<span\; class="notranslate">\; \&sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; S</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}& \mathrm{<span\; class="notranslate">\; \&Delta;</span>}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; D.</span>}}<span\; class="notranslate">\; ;</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; \&Element;</span>\mathrm{<span\; class="notranslate">\; \&Omega;</span>}\\ \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; otherwise</span>}\end{array}\right.$

Among them, R(□) represents autocorrelation, Δ represents data position offset, E[□] represents mathematical expectation, S(n) represents baseband time domain signal, σ _s and σ _w represent signal energy and noise energy, respectively. N _D represents the length of the OFDM signal data domain represented by the number of sampling points, and Ω represents the collection of signal sampling points.

The estimate of _ND can be written as:

${\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; N</span>}}}_{\mathrm{<span\; class="notranslate">\; D.</span>}}<span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; \&Delta;</span>}}{\mathrm{<span\; class="notranslate">\; arg</span>}\mathrm{<span\; class="notranslate">\; max</span>}}<span\; class="notranslate">\; \{</span><span\; class="notranslate">\; |</span>{\mathrm{<span\; class="notranslate">\; R</span>}}_{\mathrm{<span\; class="notranslate">\; S</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; \&Delta;</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; |</span><span\; class="notranslate">\; \}</span><span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; \&Delta;</span>}<span\; class="notranslate">\; ></span>\mathrm{<span\; class="notranslate">\; 0</span>}$

是以采样点数表示的OFDM信号有效数据区长度；子载波个数 $\hat{N}=\frac{{\hat{N}}_D}{f_s/W},$ 其中W为步骤3估计的信号带宽，f_s为步骤1中进行A/D变换的采样频率；estimated result

The effective data area length of the OFDM signal represented by the number of sampling points; the number of subcarriers $\hat{N}=\frac{{\hat{N}}_{\mathrm{D.}}}{f_{\mathrm{the\; s}}/W},$ Wherein W is the signal bandwidth estimated in step 3, and f _s is the sampling frequency for performing A/D conversion in step 1;

修正为 ${\hat{N}}^{\&prime;}=\underset{m\&Element;S}{\arg \min }|\hat{N}-m|,$

修正为 ${\hat{N}}_D^{\&prime;}={\hat{N}}^{\&prime;}\&times;(f_s/W);$ (6b) The number of OFDM signal subcarriers in the actual system is generally taken in the set S={2 ⁿ |n is a positive integer}, and the estimated result of step (6a)

amended to ${\hat{N}}^{\&prime;}=\underset{m\&Element;S}{\arg \min }|\hat{N}-m|,$

amended to ${\hat{N}}_{\mathrm{D.}}^{\&prime;}={\hat{N}}^{\&prime;}\&times;(f_{\mathrm{the\; s}}/W);$

(6c) In order to demodulate the received OFDM signal, it is necessary to estimate the symbol start position offset to complete OFDM symbol synchronization. In some systems, the cyclic prefix length of the OFDM signal is optional, and the cyclic prefix length of the OFDM signal needs to be estimated. The present invention uses a maximum likelihood estimation algorithm to estimate the cyclic prefix length N _G and the symbol starting position offset θ, and the likelihood function is as follows:

$\mathrm{<span\; class="notranslate">\; \&Lambda;</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; S</span>}<span\; class="notranslate">\; ;</span>{\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; D.</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; G</span>}}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; \&theta;</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; log</span>}<span\; class="notranslate">\; (</span>\underset{\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; \&Element;</span>\mathrm{<span\; class="notranslate">\; \&Omega;</span>}}{\mathrm{<span\; class="notranslate">\; \&Pi;</span>}}\mathrm{<span\; class="notranslate">\; p</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; S</span>}<span\; class="notranslate">\; [</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; ]</span><span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; S</span>}<span\; class="notranslate">\; [</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; D.</span>}}<span\; class="notranslate">\; ]</span><span\; class="notranslate">\; )</span><span\; class="notranslate">\; \&Center\; Dot;</span>\underset{\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; \&NotElement;</span>\mathrm{<span\; class="notranslate">\; \&Omega;</span>}<span\; class="notranslate">\; \&cup;</span>{\mathrm{<span\; class="notranslate">\; \&Omega;</span>}}^{<span\; class="notranslate">\; \&prime;</span>}}{\mathrm{<span\; class="notranslate">\; \&Pi;</span>}}\mathrm{<span\; class="notranslate">\; p</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; S</span>}<span\; class="notranslate">\; [</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; ]</span><span\; class="notranslate">\; )</span><span\; class="notranslate">\; )</span>$

$<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; log</span>}<span\; class="notranslate">\; (</span>\underset{\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; \&Element;</span>\mathrm{<span\; class="notranslate">\; \&Omega;</span>}}{\mathrm{<span\; class="notranslate">\; \&Pi;</span>}}\frac{\mathrm{<span\; class="notranslate">\; p</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; S</span>}<span\; class="notranslate">\; [</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; ]</span><span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; S</span>}<span\; class="notranslate">\; [</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; D.</span>}}<span\; class="notranslate">\; ]</span><span\; class="notranslate">\; )</span>}{\mathrm{<span\; class="notranslate">\; p</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; S</span>}<span\; class="notranslate">\; [</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; ]</span><span\; class="notranslate">\; )</span><span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; p</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; S</span>}<span\; class="notranslate">\; [</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; D.</span>}}<span\; class="notranslate">\; ]</span><span\; class="notranslate">\; )</span>}\underset{\mathrm{<span\; class="notranslate">\; no</span>}}{\mathrm{<span\; class="notranslate">\; \&Pi;</span>}}\mathrm{<span\; class="notranslate">\; p</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; S</span>}<span\; class="notranslate">\; [</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; ]</span><span\; class="notranslate">\; )</span><span\; class="notranslate">\; )</span>$

In the formula, S represents the baseband signal vector, S[□] represents the baseband signal, the set Ω is the data set of the cyclic prefix area in the OFDM signal, and the set Ω′ is the effective data set copied to the cyclic prefix area in the OFDM signal. Assuming that S has a joint Gaussian distribution, the final result of simplifying the likelihood function is as follows:

代表向下取整，^*代表取复共轭。In the formula, S represents the baseband signal vector, S[□] represents the baseband signal,

Represents rounding down, ^* represents complex conjugate.

According to the simplified likelihood function and the revised estimate in step (6b) , the cyclic prefix length N _G and the symbol start position offset θ are estimated as follows:

$<span\; class="notranslate">\; (</span>{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; N</span>}}}_{\mathrm{<span\; class="notranslate">\; G</span>}}<span\; class="notranslate">\; ,</span>\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; \&theta;</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\underset{{\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; G</span>}}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; \&theta;</span>}}{\mathrm{<span\; class="notranslate">\; arg</span>}\mathrm{<span\; class="notranslate">\; max</span>}}<span\; class="notranslate">\; \{</span>\stackrel{<span\; class="notranslate">\; ~</span>}{\mathrm{<span\; class="notranslate">\; \&Lambda;</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; S</span>}<span\; class="notranslate">\; ;</span>{{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; N</span>}}}^{<span\; class="notranslate">\; \&prime;</span>}}_{\mathrm{<span\; class="notranslate">\; D.</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; G</span>}}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; \&theta;</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; \}</span>$

代表向下取整，^*代表取复共轭。In the formula, S represents the baseband signal vector, S[□] represents the baseband signal,

Represents rounding down, ^* represents complex conjugate.

的取值可以限定在

的1/2，1/4，1/8，1/16，1/32这些倍数范围内。根据θ的估计结果得到符号同步后的OFDM基带信号S”(n)。Typically, the cyclic prefix length _NG is some multiple of _ND . For example, in the 802.16 system, the available lengths are 1/4, 1/8, 1/16, and 1/32 of _ND . Therefore, in the above formula

The value of can be limited to

1/2, 1/4, 1/8, 1/16, 1/32 of these multiples range. According to the estimation result of θ, the OFDM baseband signal S"(n) after symbol synchronization is obtained.

Step 7: When the received signal is judged as an OFDM signal, perform subcarrier modulation identification and demodulation on the symbol-synchronized OFDM signal S″(n).

The sampling frequency f _s of the baseband signal S(n) is known, and the signal bandwidth W estimated in step 3 is taken as the sample rate of the OFDM signal, and according to the number N of OFDM subcarriers estimated in step 6, and the cyclic prefix length N _G , Carry out Fourier transform to the OFDM signal S″(n) after symbol synchronization to obtain the subcarrier modulation data. According to the high-order cumulant of the subcarrier modulation data, the eigenvector is constructed to identify the subcarrier modulation method, and according to the identified subcarrier modulation method, the The modulated data is demodulated.

The identification steps of OFDM signal subcarrier modulation mode are as follows:

(7a) In this example, the OFDM signal subcarrier modulation mode is selected in the set {BPSK, QPSK, 16QAM}, and energy normalization is performed on the OFDM subcarrier modulation data to obtain the signal x(n);

(7b) For the energy-normalized OFDM subcarrier signal x(n), construct the eigenvector F _r according to the following formula:

${\mathrm{<span\; class="notranslate">\; f</span>}}_{\mathrm{<span\; class="notranslate">\; r</span>}}<span\; class="notranslate">\; =</span><span\; class="notranslate">\; [</span><span\; class="notranslate">\; |</span>\frac{\mathrm{<span\; class="notranslate">\; mean</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; x</span>}<span\; class="notranslate">\; )</span>}{\sqrt[\mathrm{<span\; class="notranslate">\; 4</span>}]{<span\; class="notranslate">\; |</span>{\mathrm{<span\; class="notranslate">\; C</span>}}_{\mathrm{<span\; class="notranslate">\; x</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; 42</span>}}<span\; class="notranslate">\; |</span>}}<span\; class="notranslate">\; |</span><span\; class="notranslate">\; ,</span><span\; class="notranslate">\; |</span>\frac{{\mathrm{<span\; class="notranslate">\; C</span>}}_{\mathrm{<span\; class="notranslate">\; x</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; 40</span>}}}{{\mathrm{<span\; class="notranslate">\; C</span>}}_{\mathrm{<span\; class="notranslate">\; x</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; 42</span>}}}<span\; class="notranslate">\; |</span><span\; class="notranslate">\; ,</span><span\; class="notranslate">\; |</span>\frac{{\mathrm{<span\; class="notranslate">\; C</span>}}_{\mathrm{<span\; class="notranslate">\; x</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; 41</span>}}}{{\mathrm{<span\; class="notranslate">\; C</span>}}_{\mathrm{<span\; class="notranslate">\; x</span>}<span\; class="notranslate">\; .</span><span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; 42</span>}}}<span\; class="notranslate">\; |</span><span\; class="notranslate">\; ]</span>$

Let M _pq ＝E[x(n) ^pq (x(n) ^* ) _q ], E[□] represents the mathematical expectation, in the above formula:

Mean(□) represents mean value;

${\mathrm{<span\; class="notranslate">\; C</span>}}_{\mathrm{<span\; class="notranslate">\; x</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; 40</span>}}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; 40</span>}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 3</span>}{\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; 20</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}$

C _x,41 =M ₄₁ -3M ₂₁ M ₂₀

${\mathrm{<span\; class="notranslate">\; C</span>}}_{\mathrm{<span\; class="notranslate">\; x</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; 42</span>}}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; 42</span>}}<span\; class="notranslate">\; -</span>{<span\; class="notranslate">\; |</span>{\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; 20</span>}}<span\; class="notranslate">\; |</span>}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; twenty\; one</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; ;</span>$

(7c) the eigenvector F _r of the OFDM subcarrier modulation data that step (7b) obtains is judged according to the following formula:

${\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; m</span>}}}_{\mathrm{<span\; class="notranslate">\; k</span>}}<span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1,2,3</span>}}{\mathrm{<span\; class="notranslate">\; arg</span>}\mathrm{<span\; class="notranslate">\; min</span>}}<span\; class="notranslate">\; (</span><span\; class="notranslate">\; |</span>{\mathrm{<span\; class="notranslate">\; f</span>}}_{\mathrm{<span\; class="notranslate">\; r</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; f</span>}}_{{\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; k</span>}}}<span\; class="notranslate">\; |</span><span\; class="notranslate">\; )</span>$

代表调制方式为M_k的信号对应的特征向量理论值，

代表估计的信号调制方式，M₁，M₂，M₃三种调制方式对应的特征向量理论值如下表1：In the formula, M _k represents the modulation mode of the signal, M ₁ , M ₂ , and M ₃ represent the three modulation modes of BPSK, QPSK, and 16QAM respectively.

Represents the theoretical value of the eigenvector corresponding to the signal whose modulation mode is M _k ,

Represents the estimated signal modulation mode, and the theoretical values of the eigenvectors corresponding to the three modulation modes of M ₁ , M ₂ , and M ₃ are shown in Table 1:

Table 1 Theoretical values of eigenvectors corresponding to the three modulation modes of _{M 1} , M ₂ , and M ₃

The advantages of the present invention can be illustrated by the following simulation performance:

1. Simulation of the bandwidth estimation method described in step 3

(1a) Simulation conditions and content

The signal types used in the simulation: OFDM, BPSK, QPSK, 16QAM, 64QAM, in which the number of subcarriers of the OFDM signal is 64, the length of the cyclic prefix is 16, and the single carrier signal uses a raised cosine roll-off filter with a roll-off coefficient of 0.5 for pulse shaping , the signals are all 8 times oversampled baseband signals.

Estimate the signal bandwidth W according to the method described in step 3. For a single carrier signal, W is equal to the signal baud rate, and for an OFDM signal, W is equal to the signal sample rate. Under AWGN and multipath channel, estimation accuracy is used to measure estimation performance.

(1b) Simulation results

Table 2 and Table 3 show the simulation results of estimation accuracy under AWGN channel and multipath channel.

Table 2 Accuracy of signal bandwidth estimation under AWGN channel

Table 3 Accuracy of signal bandwidth estimation under multipath channel

It can be seen from the simulation results in Table 2 and Table 3 that for OFDM signals, under AWGN channel and multipath channel, the estimation of the signal sample rate has a high accuracy rate when the signal-to-noise ratio is greater than 0dB. For single-carrier signals, under the AWGN channel, the accuracy rate of baud rate estimation is above 93% when the signal-to-noise ratio is greater than 4dB; More than 91%.

2. Simulation of the OFDM signal identification method described in step 5

(2a) Simulation conditions and content

The signal types used in the simulation: OFDM, BPSK, QPSK, square 16QAM, square 64QAM. Among them, the number of subcarriers of the OFDM signal is 64, and the length of the cyclic prefix is 16. The single-carrier signal uses a raised cosine roll-off filter with a roll-off coefficient of 0.5 for pulse shaping, and the signals are all 4 times oversampled baseband signals.

According to the method described in step 5, the signal feature vector is extracted using multi-layer wavelet decomposition. Taking C=100, the theoretical value of the OFDM signal eigenvector is M=[50, 25, 12.5, 50, 25, 12.5]. The Euclidean distance D is calculated by comparing the received signal eigenvector with the theoretical value of the OFDM signal eigenvector. Comparing D with the decision threshold R, when D<R, it is judged as an OFDM signal, and when D>R, it is judged as a single carrier signal. Adjust the threshold R to get the best recognition effect. It is regarded as a correct judgment if the judgment result of the OFDM signal is an OFDM signal, and it is regarded as a correct judgment if the judgment result of a single carrier signal is a non-OFDM signal.

(2b) Simulation results

The simulation results of recognition accuracy obtained under the AWGN channel and multipath channel are shown in Table 4 and Table 5:

Table 4 Correct rate of OFDM signal recognition under AWGN channel

Table 5 Correct rate of OFDM signal recognition under multipath channel

According to the simulation results in Table 4 and Table 5, it can be seen that under the AWGN channel, when the SNR is higher than 0dB, the correct recognition rate is above 94%; under the multipath channel, when the SNR is above 4dB, the correct recognition rate The probability is above 96%.

Claims (4)

1. ofdm signal intelligence receiving system comprises:

The signal capture unit is used for signal is caught reception;

Pretreatment unit is used for carrying out to the received signal filtering, amplification and A/D preconditioning;

Carrier estimation, down-converter unit are used for obtaining baseband signal S (n) to carrying out estimating carrier frequencies and down-conversion through pretreated signal;

The signal bandwidth estimation unit is used for baseband signal S (n) is carried out following bandwidth estimation, and estimated result W is passed to follow-up sampling unit and ofdm signal parameter estimation unit;

(a) the baseband signal S (n) that receives is divided into groups, it is divided into M group, be further divided into N group in every group;

(b) to i, j in 1≤i≤M group, 1≤j≤N group baseband signal is carried out Fourier transform and is obtained base-band signal spectrum S _{I, j}(f);

(c) to base-band signal spectrum S _{I, j}(f) carry out n, n 〉=1 layer Haar wavelet decomposition obtains the coefficient of wavelet decomposition vector T;

(d) the approximate component in the retention coefficient vector T is constant, and the details component is set to 0, and with the details component in the elimination base-band signal spectrum, obtains the coefficient vector T ' for wavelet reconstruction;

(e) according to described coefficient vector T ' signal spectrum is carried out the Haar wavelet reconstruction, obtain the step change type base-band signal spectrum after the reconstruct;

(f) N the step change type base-band signal spectrum S ' that i group signal is obtained _{I, j}(f), 1≤j≤N carries out statistical average, obtains the frequency spectrum after the statistical average

This frequency spectrum

The position that middle consecutive value differs maximum is the steepest change location;

(g) search described

Two steepest change locations, as the position, frequency spectrum edge of i group signal;

(h) position, frequency spectrum edge of statistics M group signal is with the signal bandwidth W of the bandwidth between two the highest positions of statistic as estimation;

Sampling unit is used for baseband signal S (n) is sampled, and the signal S ' after obtaining sampling (n);

The ofdm signal recognition unit is used for (n) carrying out following differentiation for ofdm signal or single-carrier signal to received signal according to the signal S ' after the sampling:

A) (n) adopt multi-level Wavelet Transform to decompose the structural classification characteristic vector to the baseband signal S ' of 4 times of over-samplings:

$x=C\&CenterDot;\frac{1}{{\mathrm{ap}}_2}[{\mathrm{ap}}_3,{\mathrm{ap}}_4,{\mathrm{ap}}_5,{\mathrm{de}}_3,{\mathrm{de}}_4,{\mathrm{de}}_5]$

Wherein, C is the parameter that the amplitude of characteristic vector value is adjusted,

Ap _iThe approximate portion of energy of representation signal i layer wavelet decomposition is defined as follows:

${\mathrm{ap}}_i={\left|\right|a_i\left|\right|}_2^2=\underset{n}{\mathrm{\&Sigma;}}{a}_{i,n}^2,$ I=1 ..., M, a _iRepresent the approximate of i layer wavelet decomposition

The part coefficient vector, a _{I, n}The approximate part coefficient of expression i layer wavelet decomposition;

De _iThe detail section energy of representation signal i layer wavelet decomposition is defined as follows:

${\mathrm{de}}_i={\left|\right|d_i\left|\right|}_2^2=\underset{n}{\mathrm{\&Sigma;}}{d}_{i,n}^2,$ I=1 ..., M, d _iRepresent the details of i layer wavelet decomposition

The part coefficient vector, d _{I, n}The detail section coefficient of expression i layer wavelet decomposition;

B) the Euclidean distance D of calculated characteristics vector x and ofdm signal characteristic vector theoretical value M:

$D=\underset{l=1}{\overset{6}{\mathrm{\&Sigma;}}}{(x_l-m_l)}^2$

X wherein _lBe characteristic vector x l dimension value, m _lBe characteristic vector M l dimension value;

C) with D and thresholding relatively, as D during less than thresholding, differentiate and be ofdm signal, as D during greater than thresholding, differentiate and be single-carrier signal;

The ofdm signal parameter estimation unit, being used for differentiating the reception signal is in the situation of ofdm signal, according to the bandwidth W of baseband signal S (n) and estimation, estimating OFDM subcarrier number N, circulating prefix-length N _GWith symbol original position side-play amount θ, obtain ofdm signal S after the sign synchronization according to θ " (n), and with the OFDM subcarrier number N, the circulating prefix-length N that estimate _GPass to follow-up ofdm signal demodulating unit;

The ofdm signal demodulating unit, being used for differentiating the reception signal is in the situation of ofdm signal, according to the OFDM subcarrier number N, the circulating prefix-length N that estimate _G" (n) carry out subcarrier modulation modes identification and demodulation with signal bandwidth W, to the ofdm signal S after the sign synchronization.

2. an ofdm signal intelligence method of reseptance comprises the steps:

(1) catches the interior signal of monitoring frequency range, and the signal of catching is carried out filtering, amplification and A/D preconditioning;

(2) pretreated reception signal is carried out carrier estimation and down-converted, obtain baseband signal S (n);

(3) baseband signal S (n) is obtained base-band signal spectrum by Fourier transform, use Haar wavelet decomposition and wavelet reconstruction, details component in the elimination base-band signal spectrum obtains the step change type base-band signal spectrum, the step change type base-band signal spectrum is carried out statistical average, obtain the frequency spectrum after the statistical average, search two steepest change locations of the frequency spectrum after this statistical average, as position, frequency spectrum edge, statistics position, frequency spectrum edge is with the signal bandwidth W of the bandwidth between two the highest positions of statistic as estimation;

(4) baseband signal S (n) is carried out over-sampling by 4 times of estimated signal bandwidth W, obtain 4 times of over-sampling baseband signal S ' (n);

(5) signal S ' (n) is decomposed the extraction characteristic vector by multi-level Wavelet Transform, calculate the Euclidean distance of this characteristic vector and ofdm signal characteristic vector theoretical value, this distance is compared with thresholding, when distance during less than thresholding, differentiating the reception signal is ofdm signal, when distance during greater than thresholding, differentiate that to receive signal be single-carrier signal

Described S ' (n) decompose is extracted characteristic vector by multi-level Wavelet Transform, carries out as follows:

(5a) (n) adopt multi-level Wavelet Transform to decompose the structural classification characteristic vector to the baseband signal S ' of 4 times of over-samplings:

$x=C\&CenterDot;\frac{1}{{\mathrm{ap}}_2}[{\mathrm{ap}}_3,{\mathrm{ap}}_4,{\mathrm{ap}}_5,{\mathrm{de}}_3,{\mathrm{de}}_4,{\mathrm{de}}_5]$

Wherein, C is the parameter that the amplitude of characteristic vector value is adjusted,

Ap _iThe approximate portion of energy of representation signal i layer wavelet decomposition is defined as follows:

${\mathrm{ap}}_i={\left|\right|a_i\left|\right|}_2^2=\underset{n}{\mathrm{\&Sigma;}}{a}_{i,n}^2,$ I=1 ..., M, a _iRepresent the approximate of i layer wavelet decomposition

The part coefficient vector, a _{I, n}The approximate part coefficient of expression i layer wavelet decomposition;

De _iThe detail section energy of representation signal i layer wavelet decomposition is defined as follows:

${\mathrm{de}}_i={\left|\right|d_i\left|\right|}_2^2=\underset{n}{\mathrm{\&Sigma;}}{d}_{i,n}^2,$ I=1 ..., M, d _iRepresent the detail section coefficient vector of i layer wavelet decomposition, d _{I, n}The detail section coefficient of expression i layer wavelet decomposition;

(5b) the Euclidean distance D of calculated characteristics vector x and ofdm signal characteristic vector theoretical value M:

$D=\underset{l=1}{\overset{6}{\mathrm{\&Sigma;}}}{(x_l-m_l)}^2$

X wherein _lBe characteristic vector x l dimension value, m _lBe characteristic vector M l dimension value;

(5c) with D and thresholding relatively, as D during less than thresholding, differentiate and be ofdm signal, as D during greater than thresholding, differentiate and be single-carrier signal;

(6) be in the situation of ofdm signal in judgment signal, according to the signal bandwidth W of baseband signal S (n) and estimation, estimating OFDM subcarrier number N, circulating prefix-length N _GWith symbol original position side-play amount θ, obtain OFDM baseband signal S after the sign synchronization according to the estimated result of θ " (n), and with the signal bandwidth W that the estimates sample value speed as ofdm signal;

(7) be in the situation of ofdm signal in judgment signal, according to the OFDM subcarrier number N, the circulating prefix-length N that estimate _G, the OFDM baseband signal S after sample value speed W and the sign synchronization " (n), carries out subcarrier-modulated identification and demodulation to the OFDM baseband signal.

3. intelligent method of reseptance according to claim 2, wherein step (3) described use Haar wavelet decomposition and wavelet reconstruction, the details component in the elimination base-band signal spectrum obtains the base-band signal spectrum of step change type, carries out as follows:

(3a) base-band signal spectrum is carried out n, n 〉=1 layer Haar wavelet decomposition obtains the coefficient of wavelet decomposition vector T;

(3b) the approximate component in the retention coefficient vector T is constant, and the details component is set to 0, and with the details component in the elimination base-band signal spectrum, obtains the coefficient vector T ' for wavelet reconstruction;

(3c) according to described coefficient vector T ' signal spectrum is carried out the Haar wavelet reconstruction, obtain the step change type signal spectrum after the reconstruct.

4. intelligent method of reseptance according to claim 2, wherein step (3) is described carries out statistical average to the step change type base-band signal spectrum, carries out as follows:

3a) the baseband signal data that receive are divided into groups, it is divided into M group, be further divided into N group in every group;

3b) j group baseband signal in the i group is carried out Fourier transform, obtain signal spectrum S _{I, j}(f), 1≤i≤M, 1≤j≤N;

3c) to signal spectrum S _{I, j}(f) use Haar wavelet decomposition and wavelet reconstruction, elimination details component obtains step change type base-band signal spectrum S ' _{I, j}(f);

N the step change type base-band signal spectrum S ' that 3d) i group signal is obtained _{I, j}(f), 1≤j≤N carries out statistical average, obtains the frequency spectrum after the statistical average

This frequency spectrum

The position that middle consecutive value differs maximum is the steepest change location.

Images