CN107454024A - A kind of underwater sound OFDM MFSK channel equalization methods based on Virtual time reversal mirror - Google Patents

Abstract

The invention discloses an underwater acoustic OFDM-MFSK channel equalization method based on a virtual time reversal mirror, and belongs to the technical field of underwater acoustic communication. The invention adds linear frequency modulation signals at the beginning and end of a frame of data through the transmitting end; the receiving end detects the preamble signal of the frame, completes data interception and Doppler factor estimation, and performs Doppler compensation on the received signal; uses Doppler The compensated frame preamble signal is used for high-precision channel impulse response estimation; the virtual time-reversal channel equalization is performed on the data according to the estimated channel impulse response result; finally, the equalized data is demodulated. The present invention utilizes the chirp signals at the beginning and end of the data to perform Doppler estimation, avoiding the problems of single-frequency signals; utilizing the orthogonal matching tracking algorithm to realize high-precision estimation of the underwater acoustic channel, avoiding the traditional passive time-reversal mirror due to The performance loss caused by the detection signal effectively improves the performance of the OFDM‑MFSK underwater acoustic communication system under severe multi-channel extended channels.

Description

An Underwater Acoustic OFDM-MFSK Channel Equalization Method Based on Virtual Time Reversal Mirror

technical field

The invention belongs to the technical field of underwater acoustic communication, and in particular relates to an underwater acoustic OFDM-MFSK channel equalization method based on a virtual time reversal mirror.

Background technique

OFDM-MFSK is a modulation technology combining multi-carrier technology and MFSK. During the implementation process, the transmitting end of OFDM-MFSK divides all subcarriers into groups of M elements, uses MFSK to map information, and the receiving end follows MFSK signal detection is performed in a non-coherent manner without channel estimation and equalization processes. OFDM-MFSK is compatible with the high communication rate of multi-carrier modulation technology, and retains the robust performance of MFSK modulation. Compared with Orthogonal Frequency Division Multiplexing (OFDM) modulation, it can better balance the transmission rate and robust performance.

The underwater acoustic channel is the biggest obstacle for the underwater acoustic communication technology to lag far behind the terrestrial wireless communication technology. The underwater acoustic channel is by far one of the most complex wireless channels, with severe multi-path extension and Doppler extension, and limited available bandwidth . Due to the slow speed of sound wave transmission in water, when there is relative motion between the two parties in communication, the received signal will be compressed and expanded, resulting in Doppler frequency offset, and multi-channel expansion will cause inter-symbol interference when coherent multi-channel signals reach the receiving end (ISI), in a multi-carrier system, a large multi-path spread will cause the system to be affected by severe inter-symbol interference (ISI). In order to eliminate the ISI, the system needs to add a longer cyclic prefix for elimination, which reduces the frequency band utilization of the system.

In order to effectively reduce the impact of ISI, some scholars applied the passive time reversal mirror technology to the multi-carrier OFDM-MFSK system, using the good time compression and space focusing characteristics of the time reversal mirror technology to shorten the channel. The results show that the time Compared with the untimely anti-processing system, the anti-processing system can effectively reduce the system bit error rate and improve the system performance, but this method does not consider the influence of the detection signal autocorrelation on the processing results. Based on this, some scholars proposed to apply the Virtual Time Reversal Mirror (VTRM) technology to the OFDM system. This method first uses the detection signal and uses the matching pursuit algorithm to accurately estimate the channel impulse response, and then the estimated The results are time reversed and convolved with the signal to complete the channel equalization. The results show that compared with the passive time reversed mirror technology, the virtual time reversed mirror technology avoids the influence of the detection signal autocorrelation due to the accurate estimation of the channel. And it has better system bit error performance.

To sum up, the time-reversal mirror technology can effectively reduce the performance loss caused by the multi-channel expansion of the channel. However, the matching pursuit algorithm is used in this method to estimate the channel. Although it has advantages over the traditional matched filtering method, However, the selected atoms are not orthogonal, and there is a certain deviation between the estimation effect and the true value, and the single-frequency signal is used for Doppler estimation, which leads to a significant decrease in the estimation stability and accuracy under the channel with severe multi-channel expansion, and sacrifices greater system power.

The present invention utilizes the chirp signals at the beginning and end of the data to perform Doppler estimation, effectively avoiding the problems existing in the use of single-frequency signals; at the same time, the orthogonal matching and tracking algorithm is used to further improve the channel estimation accuracy; and the virtual time reversal mirror is applied In the OFDM-MFSK underwater acoustic communication system.

Contents of the invention

The purpose of the present invention is to provide an underwater acoustic OFDM-MFSK channel equalization method based on a virtual time-reversal mirror that can effectively solve the performance loss caused by channel multi-channel expansion to the underwater acoustic OFDM-MFSK communication system.

The purpose of the present invention is achieved like this:

The invention discloses an underwater acoustic OFDM-MFSK channel equalization method based on a virtual time reversal mirror, and its specific implementation steps include:

(1) The transmitter adds a chirp signal at the beginning and end of a frame of data;

(2) The receiving end detects the frame preamble signal, completes data interception and Doppler factor estimation, and performs Doppler compensation on the received signal;

(3) Using Doppler-compensated frame preamble signals to perform high-precision channel impulse response estimation;

(4) Perform virtual time-reversal channel equalization on the data according to the estimated channel impulse response result;

(5) Demodulate the equalized data.

For a kind of underwater acoustic OFDM-MFSK channel equalization method based on virtual time-reversal mirror, the bandwidth of chirp signal described in step (1) is the system operating bandwidth, and the time product of the bandwidth of chirp signal and chirp signal is greater than 100, and the length of the guard interval between the data and the two chirps is greater than the maximum multipath extension length of the channel.

For a kind of underwater acoustic OFDM-MFSK channel equalization method based on virtual time reversal mirror, the specific implementation steps of described step (2) include:

(2.1) Use the chirp signals at the beginning and end of the data to perform matched filtering, and perform correlation calculations on the received data with the synchronous signal as a reference signal, find the peak coordinate points of the pre-frame and post-frame synchronous signals, and calculate the time interval between the two signals;

(2.2) Calculate the average Doppler factor of the entire frame of data, and use the method of linear interpolation to complete the Doppler compensation of the received signal according to the average Doppler factor.

For a kind of underwater acoustic OFDM-MFSK channel equalization method based on virtual time reversal mirror, the specific implementation steps of described step (3) include:

(3.1) The frame preamble signal after Doppler compensation is used as the detection signal, and the orthogonal matching pursuit algorithm is used to estimate the impulse response of the underwater multi-path channel, and the time delay and amplitude information of each multi-path path are obtained.

Preferably, the specific implementation steps of the orthogonal matching pursuit algorithm described in step (3.1) include:

(3.1.1) Constructing a linearized model for sparse channels

y=Φx+v

Among them, y∈R ^M is the channel observation vector, x∈R ^N is the sparse channel to be estimated, v∈R ^M is the channel noise vector, Φ∈R ^M×N is the measurement matrix (or atomic library), and can be expressed as is the atom, and the observation matrix is the Toeplitz matrix formed by the detection signal;

(3.1.2) Initialize residual r ₀ =y, index set iteration index i=1,

(3.1.3) Find the atom in the measurement matrix Φ that best matches the residual:

(3.1.4) Increase the support set Λ _i ＝ Λ _i-1 ∪{λ _i }, the augmented matrix

(3.1.5) Approximate the observation vector using the existing atoms in the index set:

(3.1.6) Update residuals:

(3.1.7) If i<k, add 1 to the iteration index and return to step (3.1.3); if i=k, get the final channel estimation result

For an underwater acoustic OFDM-MFSK channel equalization method based on a virtual time-reversal mirror, the specific implementation of the step (4) is: after the estimated channel impulse response is time-reversed, it is compared with the received data Convolution, to complete the virtual time-reversed channel equalization of the data.

For a kind of underwater acoustic OFDM-MFSK channel equalization method based on virtual time reversal mirror, the specific implementation steps of described step (5) include:

(5.1) Carry out parallel-serial conversion to data, remove cyclic prefix and cyclic suffix;

(5.2) DFT is operated to obtain all subcarrier frequency domain data;

(5.3) Take the modulus of the data and divide it into a group according to M elements, calculate the number of bits represented by the energy maximum value of each group of data, and realize the demodulation of the data.

The beneficial effect of the present invention is that: the underwater acoustic OFDM-MFSK channel equalization method based on the virtual time reversal mirror disclosed by the present invention adopts the linear frequency modulation signal at the beginning and end of a frame of data to complete the estimation of data Doppler, compared with single-frequency Signal Doppler estimation has better stability and accuracy in fading underwater acoustic channels; linear interpolation is used to achieve Doppler compensation to effectively reduce the amount of calculation in the system;

At the same time, the present invention uses the Doppler compensated frame preamble signal for channel estimation, avoids the use of detection signals, and improves the power utilization rate of the system; and uses the orthogonal matching tracking algorithm to realize high-precision estimation of the underwater acoustic channel , the channel equalization of the OFDM-MFSK system is realized through the convolution operation with the received signal, which effectively improves the system signal-to-noise ratio and reduces the impact of channel multi-channel expansion on system performance.

Description of drawings

Fig. 1 is a schematic diagram of the transmission frame structure of the OFDM-MFSK underwater acoustic communication system in the present invention;

Fig. 2 is a schematic diagram of signal processing for virtual time reversal at the receiving end of the OFDM-MFSK underwater acoustic communication system in the present invention.

detailed description

The present invention will be further described below in conjunction with the accompanying drawings.

The invention discloses an underwater acoustic OFDM-MFSK channel equalization method based on a virtual time reversal mirror, and its specific implementation steps include:

(1) The transmitter adds a chirp signal at the beginning and end of a frame of data;

(2) The receiving end detects the frame preamble signal, completes data interception and Doppler factor estimation, and performs Doppler compensation on the received signal;

(3) Using Doppler-compensated frame preamble signals to perform high-precision channel impulse response estimation;

(4) Perform virtual time-reversal channel equalization on the data according to the estimated channel impulse response result;

(5) Demodulate the equalized data.

(1) The transmitter adds a chirp signal at the beginning and end of a frame of data;

Referring to FIG. 1 , according to the structure of the transmission frame, the chirp signal before the data is called a frame preamble, and the chirp signal after the data is called a frame postamble. The parameters of the two chirp signals are consistent, and the length of the guard interval between the data and the two chirp signals is greater than the maximum multipath extension length of the channel. The OFDM-MFSK data block is composed of time-domain OFDM-MFSK symbols adding a cyclic prefix and a cyclic suffix. The cyclic prefix is to copy the data after the OFDM-MFSK symbol to the front of the symbol, and the cyclic suffix is to copy the data in front of the OFDM-MFSK symbol to the symbol.

(2) The receiving end detects the frame preamble signal, completes data interception and Doppler factor estimation, and performs Doppler compensation on the received signal;

Perform correlation calculation on the received data with the synchronization signal as the reference signal, find the peak coordinate points of the synchronization signal before and after the frame, calculate the time interval T _r between the two signals, and calculate the Doppler factor of the entire data block according to the following formula

Wherein, T _t is the time interval between two chirp signals in the transmitted frame signal.

Assuming that the original sampling point vector is x, according to the calculated Doppler factor, the position of the new sampling point after Doppler compensation is The amplitude of the signal after Doppler compensation is completed by linear interpolation.

Let the function f(x) be f(x ₀ ) and f(x ₁ ) at the two endpoints of the interval [x ₀ ,x ₁ ], and use the linear function L(x)=ax+b to approximate the interval [x ₀ ,x ₁ ], by selecting parameters a and b, so that L(x ₀ )=f(x ₀ ), L(x ₁ )=f(x ₁ ), the two points of the straight line equation can be The expression to obtain L(x) is

Combining formula (2) and the new sampling point position is The Doppler compensation of the received signal can be completed.

(3) Using Doppler-compensated frame preamble signals to perform high-precision channel impulse response estimation;

The present invention uses the frame preamble signal after Doppler compensation as the detection signal, and uses an Orthogonal Matching Pursuit (OMP) algorithm to complete the high-precision estimation of the underwater acoustic channel. Compared with the matching pursuit (Matching Pursuit, OMP) algorithm, the OMP algorithm adds an orthogonalization process to the candidate atoms, which can reduce the residual error and improve the channel estimation accuracy. The specific steps of the OMP algorithm are given below:

Now consider the linearized model of the sparse channel

y=Φx+v (3)

Among them, y∈R ^M is the channel observation vector, x∈R ^N is the sparse channel to be estimated, v∈R ^M is the channel noise vector, Φ∈R ^M×N is the measurement matrix (or atomic library), and can be expressed as is an atom, and the observation matrix in the present invention is a Toeplitz matrix formed by detection signals.

(3.1) Initialize residual r ₀ =y, index set iteration index i=1,

(3.2) Find the atom in the measurement matrix Φ that best matches the residual:

(3.3) Increase the support set Λ _i ＝ Λ _i-1 ∪{λ _i }, the augmented matrix

(3.4) Approximate the observation vector using the existing atoms in the index set:

(3.5) Update residuals:

(3.6) If i<k, add 1 to the iteration index and return to step (3.2); if i=k, get the final channel estimation result

(4) Perform virtual time-reversal channel equalization on the data according to the estimated channel impulse response result;

Combined with Figure 2, according to the signal processing block diagram of the virtual time reversal at the receiving end of the OFDM-MFSK system, it is assumed that the system does not have the influence of Doppler after the Doppler estimation and compensation processing in the previous stage.

Suppose the transmitted data is s[n], and the impulse response of the underwater acoustic channel is h[n], then the received data r[n] is expressed as follows:

in, Represents the convolution operation, and z[n] represents the channel noise.

Suppose the estimated channel impulse response is h[n], then the received data can be expressed as

In the formula, is the Q function of virtual time inverse processing, and z′[n] is the noise component after time inverse processing.

The Q function is actually the equivalent total channel processed by the virtual time-reversal mirror of the system. If the channel estimation is correct, the result is the autocorrelation function of the channel impulse response. When the channel is more complex, the Q function is approximated as a Sinc function, which is manifested as the main peak with concentrated energy and lower side lobes. Therefore, the virtual time-reversal mirror technology realizes channel compression and shortening, and completes channel equalization.

(5) Demodulate the equalized data;

By performing parallel-to-serial conversion on the data, removing the cyclic prefix and cyclic suffix, and performing DFT operations to obtain all subcarrier frequency domain data, the data is modulo-divided into groups of M elements, and the maximum energy value of each group of data is calculated Represents the number of bits to achieve data demodulation.

The above descriptions are only preferred embodiments of the present invention, and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and changes. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included within the protection scope of the present invention.

Claims (7)

1. a kind of underwater acoustic OFDM-MFSK channel equalization method based on virtual time reversal mirror, it is characterized in that, concrete realization step comprises: (1) The transmitter adds a chirp signal at the beginning and end of a frame of data; (2) The receiving end detects the frame preamble signal, completes data interception and Doppler factor estimation, and performs Doppler compensation on the received signal; (3) Using Doppler-compensated frame preamble signals to perform high-precision channel impulse response estimation; (4) Perform virtual time-reversal channel equalization on the data according to the estimated channel impulse response result; (5) Demodulate the equalized data. 2. a kind of underwater acoustic OFDM-MFSK channel equalization method based on virtual time reversal mirror according to claim 1, it is characterized in that: the bandwidth of chirp signal described in step (1) is system operating bandwidth, linear The time product of the bandwidth of the FM signal and the chirp signal is greater than 100, and the length of the guard interval between the data and the two chirp signals is greater than the maximum multipath extension length of the channel. 3. a kind of underwater acoustic OFDM-MFSK channel equalization method based on virtual time-reversal mirror according to claim 1, is characterized in that, the specific implementation steps of described step (2) comprises: (2.1) Use the chirp signals at the beginning and end of the data to perform matched filtering, and perform correlation calculations on the received data with the synchronous signal as a reference signal, find the peak coordinate points of the pre-frame and post-frame synchronous signals, and calculate the time interval between the two signals; (2.2) Calculate the average Doppler factor of the entire frame of data, and use the method of linear interpolation to complete the Doppler compensation of the received signal according to the average Doppler factor. 4. a kind of underwater acoustic OFDM-MFSK channel equalization method based on virtual time-reversal mirror according to claim 1, is characterized in that, the specific implementation steps of described step (3) comprises: (3.1) The frame preamble signal after Doppler compensation is used as the detection signal, and the orthogonal matching pursuit algorithm is used to estimate the impulse response of the underwater multi-path channel, and the time delay and amplitude information of each multi-path path are obtained. 5. a kind of underwater acoustic OFDM-MFSK channel equalization method based on virtual time-reversal mirror according to claim 1, is characterized in that, the specific implementation mode of described step (4) is: the channel that will estimate After the excitation response is time-reversed, it is convoluted with the received data to complete the virtual time-reversed channel equalization of the data. 6. a kind of underwater acoustic OFDM-MFSK channel equalization method based on virtual time-reversal mirror according to claim 1, is characterized in that, the concrete realization step of described step (5) comprises: (5.1) Carry out parallel-serial conversion to data, remove cyclic prefix and cyclic suffix; (5.2) DFT is operated to obtain all subcarrier frequency domain data; (5.3) Take the modulus of the data and divide it into a group according to M elements, calculate the number of bits represented by the energy maximum value of each group of data, and realize the demodulation of the data. 7. a kind of underwater acoustic OFDM-MFSK channel equalization method based on virtual time-reversal mirror according to claim 4, is characterized in that, the specific implementation steps of the orthogonal matching pursuit algorithm described in step (3.1) comprises: (3.1.1) Constructing a linearized model for sparse channels y=Φx+v Among them, y∈R ^M is the channel observation vector, x∈R ^N is the sparse channel to be estimated, v∈R ^M is the channel noise vector, Φ∈R ^{M ×N} is the measurement matrix (or atomic library), and can be expressed as is the atom, and the observation matrix is the Toeplitz matrix formed by the detection signal; (3.1.2) Initialize residual r ₀ =y, index set iteration index i=1, (3.1.3) Find the atom in the measurement matrix Φ that best matches the residual: (3.1.4) Increase the support set Λ _i ＝ Λ _i-1 ∪{λ _i }, the augmented matrix (3.1.5) Approximate the observation vector using the existing atoms in the index set: <mrow><msub><mover><mi>h</mi><mo>^</mo></mover><mrow><mi>&amp;Lambda;</mi><mi>i</mi></mrow></msub><mo>=</mo><mi>arg</mi><munder><mi>min</mi><msub><mi>h</mi><mrow><mi>&amp;Lambda;</mi><mi>i</mi></mrow></msub></munder><mo>|</mo><mo>|</mo><mi>y</mi><mo>-</mo><msub><mover><mi>&amp;Psi;</mi><mo>~</mo></mover><mi>i</mi></msub><msub><mi>h</mi><mrow><mi>&amp;Lambda;</mi><mi>i</mi></mrow></msub><mo>|</mo><mo>|</mo></mrow> (3.1.6) Update residuals: <mrow><msub><mi>r</mi><mi>i</mi></msub><mo>=</mo><mi>y</mi><mo>-</mo><msub><mover><mi>&amp;Psi;</mi><mo>~</mo></mover><mi>i</mi></msub><msub><mover><mi>h</mi><mo>^</mo></mover><mrow><mi>&amp;Lambda;</mi><mi>i</mi></mrow></msub></mrow> (3.1.7) If i<k, add 1 to the iteration index and return to step (3.1.3); if i=k, get the final channel estimation result