Method for High-Speed Uncoordinated Concurrent Access in Underwater Acoustic Internet of Things

Abstract

The present invention discloses a method for high-speed non-coordinated concurrent access in an underwater acoustic Internet of Things, which realizes a time-varying channel against Doppler and strong multipath, and improves the throughput of a time-varying underwater acoustic system while solving conflicts. The method includes: 1. Allocate two ZC sequences with different root indexes to different users, which are used as auxiliary sequences and modulation sequences respectively; 2. Use a sliding correlator to perform cross-correlation operations on the received baseband signal. The preamble consists of three repeated ZC base sequences; when detecting continuously equidistant signals exceeding a threshold, retain the potential peaks and their corresponding root indexes, and determine the earliest peak as the starting position of the preamble, and then determine the starting position of data demodulation for different users according to the preamble length; 3. Demodulate the template signal based on the auxiliary sequence, demodulate the signal to be demodulated based on the modulation sequence, introduce the Pearson correlation coefficient with an adaptive weight to calculate the correlation coefficient between the template signal and the signal to be demodulated, and demodulate the data of the specified user from the superimposed data symbols.

Description

Technical Field

[0001] The present invention belongs to the field of communication technologies, and relates to a method for processing received data symbols in a spread-spectrum communication transmission scheme, to a system design for symbol processing in underwater acoustic communication for combating multipath, Doppler, and time-varying channels, and also to multi-user access in the Internet of Things, and particularly refers to a method for high-speed non-coordinated concurrent access in an underwater acoustic Internet of Things. Background Art

[0002] Underwater Internet of Things (IoUT) is an emerging technology that extends the functions of the Internet of Things (IoT) to the underwater environment, and its data collection and analysis promote the exploration of the world's oceans. Different from traditional wireless IoT, underwater IoT usually uses acoustic waves for communication because radio waves have the characteristic of high attenuation and have an extremely limited propagation distance underwater.

[0003] High-speed underwater acoustic communication is a challenging technology: on the one hand, due to the complex underwater environment (such as reflection and refraction of the seabed and water surface), the underwater channel has severe multipath effects; on the other hand, the Doppler effect and the dynamic environment caused by mobile devices or moving water make the UWA channel time-varying.

[0004] In addition, the design of an efficient underwater acoustic network MAC protocol is also a challenge. The propagation delay underwater cannot be ignored because the propagation speed of acoustic waves in water is about 1500 meters per second, which is much lower than the radio propagation speed in the air, resulting in very unreliable acoustic wave communication connections. If underwater IoT devices want to access the shared acoustic medium in an orthogonal manner, it requires a high coordination overhead. Non-coordinated random access protocols (such as ALOHA) are particularly suitable for underwater IoT data collection applications because of their simplicity. However, when users transmit without coordination, the performance of these protocols will be severely limited by collisions, which also affects the data transmission rate.

[0005] For high-speed underwater acoustic communication (UAC), existing technologies have proposed new modulation methods such as multi-carrier chirp spread spectrum (CSS) and multi-carrier orthogonal frequency division multiplexing (OFDM). Current underwater CSS and OFDM systems are all designed for point-to-point communication. When there are multiple users in concurrency, their demodulation methods cannot effectively handle collisions, resulting in limited networking performance.

[0006] The currently developed underwater CDMA (Code Division Multiple Access) system can support non-coordinated concurrent access and can reasonably handle user collisions. However, each symbol can only carry 1 bit, which affects the data rate. In addition, although the Rake receiver has relatively good performance in anti-multipath, its anti-Doppler time-varying effect is limited; although the nCSK system can carry multiple bits per symbol and also supports multi-user concurrent transmission, it also uses the Rake receiver for equalization and is difficult to cope with the time-varying channels of Doppler and strong multipath. Summary of the Invention

[0007] The object of the present invention is to provide a method for high-speed non-coordinated concurrent access in an underwater Internet of Things, to solve the problems existing in the prior art. By using high-rate ZC modulation and leveraging the orthogonality of different ZC sequences to enable a low-overhead non-coordinated concurrent access protocol, and at the same time adopting a modulation and demodulation method based on an auxiliary sequence to achieve resistance to the time-varying channels of Doppler and strong multipath, the throughput of the time-varying underwater acoustic system is improved while resolving conflicts.

[0008] To achieve the above object, the solution of the present invention is:

[0009] A method for high-speed non-coordinated concurrent access in an underwater Internet of Things, including:

[0010] Step 1, symbol modulation

[0011] For the data of different users, ZC sequences with different roots are used; each user is assigned two ZC sequences with different root indices and the root index is not 1. One is used as the auxiliary sequence for generating the template signal, and the other is used as the modulation sequence for modulating the data.

[0012] If different users have data to send, they can directly access the channel asynchronously to send data signals without coordination. The data signals include a preamble and data symbols. The preamble consists of three repeated ZC base sequences, and the data symbols are modulated by the above two ZC sequences with different root indices and the root index is not 1. The length of the data symbols is determined by the data length.

[0013] Step 2, frame synchronization

[0014] A sliding correlator is used to perform a cross-correlation operation on the received baseband signal; when detecting continuous equidistant signals exceeding the threshold, the potential peaks and their corresponding root indices are retained, and the earliest peak is determined as the starting position of the preamble. Then, according to the preamble length, the starting position of the data symbol demodulation of different users is determined.

[0015] Step 3, symbol demodulation

[0016] Using the operations of DFT-IDFT, the template signal is demodulated based on the auxiliary sequence assigned to each user, and the signal to be demodulated is demodulated based on the modulation sequence assigned to each user. The Pearson correlation coefficient with an adaptive weight is introduced to calculate the correlation coefficient between the template signal and the signal to be demodulated, and the data of the specified user is demodulated from the superimposed data symbols.

[0017] Preferably, the specific steps of symbol modulation in step 1 include:

[0018] 1.1. Define each ZC sequence as , where represents the natural base, represents the imaginary unit, represents the root index, represents the length of the symbol; adopt the coding mapping method based on the ZC sequence, assign ZC sequences with different root indexes to different users respectively, and each user is assigned two ZC sequences with different root indexes, one as the auxiliary sequence for generating the template signal and the other as the modulation sequence for modulating the data;

[0019] 1.2. After the ZC sequence is circularly shifted, it is expressed as , where represents the number of times of sequence circular shift, represents the modulo operation; each symbol of a user is modulated based on the circular shift sequences of two root indexes to obtain a symbol with bits, where .

[0020] Preferably, the specific steps of symbol demodulation in step 3 include:

[0021] 3.1. Spectrum demodulation

[0022] According to the starting position of data symbol demodulation of different users in the received signal obtained in step 2, the received symbol with a length of is correlated with sequences of possible cases, and the one with the largest correlation value is taken as the transmitted sequence; specifically, in step 3.1, demodulation is performed through DFT-IDFT. First, the symbol is multiplied by the conjugate of the ZC base sequence with different root indexes in the frequency domain, and then converted into the time domain to obtain the sequence , and its calculation formula is:

[0023] ;

[0024] where represents the th symbol sequence after spectrum demodulation, represents the inverse discrete Fourier transform, represents the inverse discrete Fourier transform, represents a ZC base sequence, with superscript represents the conjugate operation; then find the starting point with the largest peak value as the position of the transmitted data symbol, that is , get the sending sequence;

[0025] 3.2. Matching demodulation based on auxiliary sequence

[0026] Perform spectrum demodulation on each received data symbol and two root indices, and use the auxiliary sequence to find the template signal to obtain the first The template signal of , which is used to reflect the channel response of the data packet, including multipath, Doppler effect and time offset caused by frame synchronization error. Its calculation formula is:

[0027] ;

[0028] 3.3. Calculation of Pearson correlation coefficient of adaptive weights

[0029] First, calculate the correlation between the template signal and the data symbol in the received spectrum demodulation sequence, and give the samples with amplitude greater than the preset threshold Set a higher weight; then the decoded data is:

[0030] ;

[0031] in , ;

[0032] 3.4. According to Get the data packet bits corresponding to the user.

[0033] After adopting the above technical solution, the present invention has the following technical effects:

[0034] (1) The symbol modulation method proposed in the present invention can essentially solve the problem of data symbol collision of different users during uncoordinated concurrent transmission of underwater IoT. By allocating ZC sequences with different root indexes to different users, it can better reduce interference between users and has better performance than traditional CSS modulation and OFDM modulation.

[0035] (2) The present invention improves the data rate in multi-user concurrent access scenarios. Through the definition of the ZC sequence, each data symbol can carry n bits, which is n times the data rate improvement compared to the traditional CDMA concurrent system.

[0036] (3) The present invention achieves high-performance demodulation and high system throughput in a strongly time-varying channel. Specifically, by defining an auxiliary sequence to generate a template signal that reflects the channel characteristics of data symbols, it can track the rapidly changing underwater acoustic channel. And by introducing an adaptive weight in the demodulation process to adjust the calculation of the Pearson correlation coefficient, the influence of noise samples in a multi-user scenario on decoding is reduced, improving the system's robustness to noise, thereby increasing the system throughput in a concurrent scenario. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 It is a schematic diagram of the system framework and scenario applied in the present invention;

[0038] Figure 2 It is a flowchart of the present invention;

[0039] Figure 3 It is a schematic diagram of the implementation process of the present invention;

[0040] Figure 4 It is a schematic diagram of the channel response impulse during experimental testing;

[0041] Figure 5 It is a schematic diagram of the Doppler spectral power density during experimental testing;

[0042] Figure 6 It is the correlation spectrum of the received signals of two users during experimental testing;

[0043] Figure 7 It is the superimposed correlation spectrum of the received signals during experimental testing;

[0044] Figure 8 It is the signal-to-noise ratio-bit error rate curve during experimental testing;

[0045] Figure 9 It is a schematic diagram of the physical layer throughput of the present invention at different duty cycles during experimental testing. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0046] To further explain the technical solution of the present invention, the present invention will be elaborated in detail through specific embodiments below.

[0047] Referring to Figures 1-3 shown, Figure 1 a schematic diagram of the system framework and scenario applied in the present invention is given. Static and mobile underwater Internet of Things devices transmit sensed data to a gateway near the sea surface through the modulation technology of the present invention, and then the gateway calls the method proposed by the present invention for high-performance demodulation and decoding; Figure 2 a schematic flowchart of the method of the present invention is given. The steps of symbol modulation, frame synchronization, and symbol demodulation are executed in sequence, and its specific implementation process is as Figure 3 .

[0048] Accordingly, the present invention discloses a method for high-speed uncoordinated concurrent access in an underwater acoustic Internet of Things, including:

[0049] Step 1, symbol modulation

[0050] For the data of different users, Zadoff-Chu (ZC) sequences with different roots are used; each user is assigned two ZC sequences with different root indices and the root indices are not 1. One is used as an auxiliary sequence for generating a template signal (representing the channel response), and the other is used as a modulation sequence for modulating the data.

[0051] If different users have data to send, without coordination, they directly access the channel asynchronously to send data signals. The data signals include a preamble and data symbols. The preamble is composed of three repeated ZC base sequences (root index is 1), and the data symbols are modulated by the above two ZC sequences with different root indices and the root indices are not 1. The length of the data symbols is determined by the data length.

[0052] Step 2, frame synchronization

[0053] A sliding correlator is used to perform a cross-correlation operation on the received baseband signal; the frame synchronization module of the receiver continuously performs cross-correlation and detects based on the correlation value. When detecting consecutive equally spaced signals exceeding the threshold, the potential peaks and their corresponding root indices are retained, and the earliest peak is determined as the starting position of the preamble. Then, the starting position for demodulating the data symbols of different users is determined according to the preamble length.

[0054] Step 3, symbol demodulation

[0055] Using the operations of DFT-IDFT (Discrete Fourier Transform - Inverse Discrete Fourier Transform), based on the auxiliary sequence assigned to each user, the template signal is demodulated, and based on the modulation sequence assigned to each user, the signal to be demodulated is demodulated. The Pearson correlation coefficient with an adaptive weight is introduced to calculate the correlation coefficient between the template signal and the signal to be demodulated, and the data of the specified user is demodulated from the superimposed data symbols.

[0056] Through the above solution, the symbol modulation method proposed by the present invention can essentially solve the problem of data symbol collision of different users during non-coordinated concurrent transmission in the underwater Internet of Things. By allocating ZC sequences with different root indexes to different users, it is possible to better reduce interference between users and has better performance than traditional CSS modulation and OFDM modulation. The present invention realizes the improvement of data rate in the scenario of multi-user concurrent access. By defining the ZC sequence, each data symbol can carry n bits, which is n times the data improvement compared with the traditional CDMA concurrent system. The present invention realizes high-performance demodulation and high system throughput in a strongly time-varying channel. Specifically, by defining an auxiliary sequence to generate a template signal reflecting the channel characteristics of the data symbol, it can track the rapidly changing underwater acoustic channel, and by introducing an adaptive weight in the demodulation process to adjust the calculation of the Pearson correlation coefficient, the influence of noise samples under multiple users on decoding is reduced, improving the robustness of the system to noise, thereby improving the system throughput in the concurrent scenario.

[0057] The following shows specific embodiments of the present invention.

[0058] In the above step 1, the specific steps of symbol modulation include:

[0059] 1.1. Define each ZC sequence as , where represents the natural base, represents the imaginary unit, represents the root index, represents the length of the symbol; adopt a coding mapping method based on the ZC sequence, and assign ZC sequences with different root indexes to different users respectively. Each user is assigned two ZC sequences with different root indexes, one as the auxiliary sequence for generating the template signal, and the other as the modulation sequence for modulating the data;

[0060] 1.2. After cyclic shifting the ZC sequence, it is expressed as , where represents the number of times of sequence cyclic shift, represents the modulo operator; each symbol of a user is modulated based on the cyclic shift sequences of two root indexes to obtain a symbol with bits, where .

[0061] In the above step 3, the specific steps of symbol demodulation include:

[0062] 3.1. Spectrum demodulation

[0063] According to the starting position of data symbol demodulation of different users in the received signal obtained in step 2, the symbol with a length of received is compared with the possible Correlate the sequences of all cases, and select the one with the largest correlation value as the transmitted sequence.

[0064] Further, in step 3.1 above, to reduce complexity, demodulation is performed through DFT-IDFT. First, multiply the symbol by the conjugate of the ZC base sequence with different root indices in the frequency domain, and then convert it to the time domain to obtain the sequence , and its calculation formula is:

[0065] ;

[0066] where represents the symbol sequence after the -th spectrum demodulation, represents the inverse discrete Fourier transform, represents the inverse discrete Fourier transform, represents the ZC base sequence, and the superscript represents the conjugate operation; then find the starting point with the largest peak as the position of the transmitted data symbol, that is, , to obtain the transmitted sequence.

[0067] Secondly, in step 3 above, the specific steps of symbol demodulation further include:

[0068] 3.2. Matched demodulation based on the auxiliary sequence

[0069] Perform spectrum demodulation operations on each received data symbol with two root indices. The auxiliary sequence is used to find the template signal to obtain the template signal of the -th symbol, which is used to reflect the channel response of the data packet, including multipath, Doppler effect, and time offset caused by frame synchronization error. Its calculation formula is:

[0070] ;

[0071] 3.3. Calculation of the Pearson correlation coefficient with adaptive weights

[0072] First, calculate the correlation degree between the template signal and the data symbol in the received sequence after spectrum demodulation, and at the same time, give a higher weight to the samples whose amplitudes are greater than a preset threshold (set artificially) (because these may be data symbols to suppress noise); then the decoded data is:

[0073] ;

[0074] where , ;

[0075] 3.4. According to the data packet bits corresponding to the user are obtained.

[0076] The technical effects of the present invention are verified through experimental tests as follows.

[0077] An experiment is conducted using a transducer, a hydrophone, and two NI acquisition boards for experimental evaluation. Among them, the transducer, power amplifier, and one NI acquisition board serve as the transmitter, and the hydrophone, STM32 MCU, and the other NI acquisition board serve as the receiver. NI LabVIEW is used in conjunction with the NI acquisition board to achieve the transmission and reception of passband sampling. At the same time, the waveform and demodulation algorithms are implemented in MATLAB.

[0078] The experiment is carried out in a pool, and artificial movement is applied to generate the Doppler effect. The channel response is shown in Figure 4 and 5 : Figure 4 It shows that there is a certain multipath effect, and the time delay at the peak energy position at each moment is shifted; Figure 5 It shows that there is a relatively serious Doppler spread phenomenon in this channel.

[0079] In the experiment, the frequency range of the hardware filter is 20 - 30 kHz, the center frequency of all signals is 25 kHz, and the bandwidth is 6 kHz. Each ZCMod contains the preamble and data symbols of three ZC sequences with repeated root indexes. The data symbols have a zero-padding guard interval with a length of 1 / 6 of the symbol duration. The bit error rate at different signal-to-noise ratios is simulated by adding Gaussian white noise. To evaluate the network layer throughput, a random access system with 8 users and duty cycles (20%, 40%, 60%, 80%, 100%) is further considered, where the value is defined as the ratio of the packet length to the frame length. The transmitter selects 8 positions to simulate real devices with different channels. For each position, 240 data packets are collected. Specifically, random delays for each device are generated within the frame length, and its signal is added with the delay to generate a superimposed signal at the same transmission power (signal-to-noise ratio = 10 dB in the experiment) to simulate the multi-user concurrent scenario. The superimposed signal is used for demodulation.

[0080] Figure 6 and 7 It shows the actually measured cyclic correlation spectrum affected by multipath interference. The transmitted signal is 1, but the position of its peak is severely interfered, which will lead to decision errors in the traditional peak demodulation method.

[0081] Figure 8 The signal-to-noise ratio-bit error rate curve for point-to-point transmission is evaluated. Compared with the modulation method of CSS, the performance gain of the present invention is large, and the error floor is eliminated.

[0082] Figure 9 It shows the physical layer throughput of the present invention under different duty cycles in the experiment, which has a significant improvement compared with the prior art.

[0083] The above embodiments and diagrams do not limit the product form and style of the present invention. Any appropriate changes or modifications made by those of ordinary skill in the art shall be considered as not departing from the patent scope of the present invention.

Claims

1. A method for high-speed non-coordinated concurrent access of underwater acoustic Internet of Things, characterized in that include: Step 1: Symbol modulation For data of different users, use ZC sequences with different roots; Each user is assigned two ZC sequences with different root indices and root indices other than 1, one of which is used as an auxiliary sequence to generate a template signal representing the channel response, and the other is used as a modulation sequence to modulate the data; after the ZC sequence is cyclically shifted, each symbol of a user is modulated based on the cyclic shift sequences of the two root indices to obtain a The symbol of the bit; If different users have data to send, they can directly access the channel and send data signals asynchronously without coordination. The data signal includes a preamble and a data symbol. The preamble is composed of three repeated ZC base sequences. The data symbol is modulated by the two ZC sequences with different root indexes and the root index is not 1. The length of the data symbol is determined by the data length. Step 2: Frame synchronization A sliding correlator is used to perform cross-correlation operation on the received baseband signal; when continuous equidistant signals exceeding the threshold are detected, potential peaks and their corresponding root indexes are retained, and the earliest peak is determined as the starting position of the preamble code, and then the demodulation starting position of the data symbols of different users is determined according to the preamble code length; Step 3: Symbol demodulation Using the DFT-IDFT operation, the template signal is demodulated based on the auxiliary sequence assigned to each user, and the demodulated signal is demodulated based on the modulation sequence assigned to each user. The Pearson correlation coefficient with adaptive weights is introduced to calculate the correlation coefficient between the template signal and the signal to be demodulated, and the specified user data is demodulated from the superimposed data symbols.

2. The method for high-speed non-coordinated concurrent access of the underwater acoustic Internet of Things as claimed in claim 1, characterized in that The specific steps of symbol modulation in step 1 include: 1.

1. Define each ZC sequence as ,in represents the natural base, represents the imaginary unit, Represents the root index, Indicates the length of the symbol; adopts a coding mapping method based on a ZC sequence, assigns ZC sequences with different root indices to different users, and allocates two ZC sequences with different root indices to each user, one of which is used as an auxiliary sequence for generating a template signal, and the other is used as a modulation sequence for modulating data; 1.

2. After cyclic shift, the ZC sequence is expressed as ,in, represents the number of cyclic shifts of the sequence, Represents a modulo operation; each symbol of a user is modulated based on a cyclic shift sequence of two root indices to obtain a The symbol of the bit, where .

3. The method for high-speed non-coordinated concurrent access of underwater acoustic Internet of Things according to claim 2, characterized in that The specific steps of symbol demodulation in step 3 include: 3.1 Spectrum Demodulation According to step 2, the demodulation start position of the data symbols of different users in the received signal is obtained, and the received length is Symbol With possible Correlate the sequences of the two situations and find the one with the largest correlation value as the sending sequence.

4. The method for high-speed non-coordinated concurrent access of underwater acoustic Internet of Things as claimed in claim 3, characterized in that: In step 3.1, demodulation is performed by DFT-IDFT. First, the symbol The conjugate of the ZC basis sequence with different root indices is multiplied in the frequency domain and then converted to the time domain to obtain the sequence , and its calculation formula is: ； in Indicates The symbol sequence after spectrum demodulation is represents the inverse discrete Fourier transform, represents the discrete Fourier transform, represents a ZC base sequence, with superscript represents the conjugate operation; then find the starting point with the largest peak value as the position of the transmitted data symbol, that is , get the sending sequence.

5. The method for high-speed non-coordinated concurrent access of the underwater acoustic Internet of Things as claimed in claim 4, characterized in that The specific steps of symbol demodulation in step 3 also include: 3.

2. Matching demodulation based on auxiliary sequence Perform spectrum demodulation on each received data symbol and two root indices, and use the auxiliary sequence to find the template signal to obtain the first The template signal of , which is used to reflect the channel response of the data packet, including multipath, Doppler effect and time offset caused by frame synchronization error. Its calculation formula is: ； 3.

3. Calculation of Pearson correlation coefficient of adaptive weights First, calculate the correlation between the template signal and the data symbol in the received spectrum demodulation sequence, and give the samples with amplitude greater than the preset threshold Set a higher weight; then the decoded data is: ； in , ; 3.

4. According to Get the data packet bits corresponding to the user.

Images