Environment-aware underwater acoustic communication system adaptive switching method

Abstract

The invention discloses an environment-aware underwater acoustic communication system adaptive switching method, and belongs to the technical field of underwater acoustic communication. Aiming at a complex and changeable ocean channel environment and the problem of low switching efficiency of a traditional communication system, the invention optimizes environment parameter perception and a system decision algorithm. At a receiving end, weights of Doppler frequency shift, a signal-to-noise ratio and a channel structure recognition algorithm are obtained based on a preset threshold value, an optimal communication mode is decided through fusion, switching is completed, the efficiency bottleneck caused by multiple channel detection and handshake operations in traditional communication is avoided, and robust self-adaptive communication in a complex marine environment is achieved. The method does not need to frequently use detection signals for handshaking, the communication system switching efficiency is remarkably improved, and the adaptive capacity of the underwater acoustic communication system in different environments is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to an environment-aware underwater acoustic communication system adaptive switching method, belonging to the technical field of underwater acoustic communication. The present application relates to the technical field of underwater acoustic communication. The method focuses on reliable communication in complex ocean acoustic environment, dynamically decides and switches to the optimal communication system by real-time sensing of channel parameters (such as Doppler shift, signal-to-noise ratio, multipath structure), to overcome the performance limitation of traditional underwater acoustic communication system caused by fixed system. The core is the intelligent matching of environmental characteristics and communication mode, which can be widely used in underwater sensor network, autonomous underwater vehicle (AUV) communication, ocean environment monitoring and other fields, to improve the communication reliability and transmission efficiency in dynamic scenarios. BACKGROUND

[0002] The underwater acoustic channel has strong space-time-frequency variation characteristics, and there are problems such as significant Doppler shift, multipath effect, environmental noise and propagation attenuation. The traditional fixed communication system cannot meet the performance requirements in different scenarios. For example, the orthogonal frequency division multiplexing (OFDM) technology is easily affected by inter-carrier interference in high Doppler scenarios, and the single-carrier modulation has a high bit error rate in long time delay spread multipath channels, and the spread spectrum is the most robust but has a low rate. The existing solutions mostly rely on multiple handshakes for adaptive rate of a single communication system, lack of joint optimization and dynamic adaptability of environmental parameters under multi-communication system integration, resulting in switching lag or wrong decision. In addition, most methods need to add complex channel sounding signals, which occupy additional bandwidth resources and have high actual deployment cost. Therefore, there is an urgent need for a low-overhead and high-robustness adaptive communication system switching method to realize real-time sensing of the channel and dynamic adjustment of the communication strategy. SUMMARY

[0003] The present application proposes an environment-aware underwater acoustic communication system adaptive switching method, which realizes dynamic optimization and selection of communication system by using communication frame synchronization header to complete efficient estimation of environmental parameters at the receiving end, and fusing multi-dimensional parameters for decision-making. Specifically, at the receiving end, a linear frequency modulation or a pseudo-random sequence is used as the synchronization header of the communication frame, and the signal-to-noise ratio, Doppler shift and multipath time delay spread parameters are estimated in real time at the receiving end; based on historical data, threshold values are set to divide, and the matching score is calculated for the pre-defined single-carrier BPSK, OFDM, spread spectrum, single-carrier QPSK and other communication systems, and finally the communication mode with the highest score is selected. The method uses periodic parameter re-estimation and feedback mechanism to ensure that the system continuously adapts to environmental changes, significantly improving the communication robustness and resource utilization in complex underwater acoustic channels. Compared with the traditional scheme, the present application avoids additional detection overhead by multiplexing the synchronization header, and the dynamic optimization mechanism of weights and thresholds can flexibly adapt to diversified ocean scenarios, with low complexity and high engineering practicability.

[0004] To achieve the above object, the application provides an environment-aware underwater acoustic communication system adaptive switching method, which is characterized in that, at the receiving end, the current environment parameters are perceived by using a synchronization header, weights are obtained based on preset thresholds of Doppler frequency shift, signal-to-noise ratio and channel structure identification algorithm, and the optimal communication mode is decided by fusion and switching is completed.

[0005] Specifically, the current environment parameters are perceived by using a synchronization header, including signal-to-noise ratio, Doppler frequency shift and multipath structure, including the following steps:

[0006] Step 1: Linear frequency modulation signal (LFM) or pseudo-random sequence (such as m sequence) is used as a synchronization header, and the receiving end synchronization header design and parameter estimation are adopted. The synchronization header is embedded in the communication frame header, repeated once per frame, and used for real-time environment perception.

[0007] Step 2: The signal-to-noise ratio SNR is calculated by the ratio of the synchronization header signal power P sync to the noise power P noise , and the specific formula is

[0008]

[0009] Step 3: The frequency shift is calculated by using the time delay offset of the synchronization header autocorrelation peak, and the formula is: where △τ is the time delay difference, υ is the sound speed, and λ is the wavelength. The Doppler frequency shift f d is estimated.

[0010] Step 4: The multipath time delay spread τ max and the path number N path are extracted by matching filtering or channel impulse response (CIR) analysis, and the multipath structure identification is completed.

[0011] Based on preset thresholds of Doppler frequency shift, signal-to-noise ratio and channel structure identification algorithm, weights are obtained, and the optimal communication mode is decided by fusion and switching is completed, and the specific steps are as follows:

[0012] Step 1: Environment parameter threshold and weight distribution. The preset initial threshold is: Doppler frequency shift threshold f SNR threshold SNR th , and multipath time delay spread threshold τ

[0013] Step 2: The weight coefficient calculation adopts weight distribution. A judgment vector matrix is constructed, and the weight coefficient represents the importance of each parameter (such as SNR> Doppler> multipath). After normalization, the weight vector is obtained.

[0014] Step 3: Communication mode decision and switching. Candidate communication mode library, predefine 4 communication modes, each mode has different application scenarios. For example: mode 1: single carrier BPSK (high Doppler, low SNR applicable), mode 2: OFDM (high speed, low Doppler required), mode 3: spread spectrum communication (anti-multipath, high Doppler, low speed), mode 4: single carrier QPSK (high SNR, high Doppler applicable).

[0015] Step 4: Calculate the environment matching score for each communication mode: select the mode with the highest score as the optimal mode, k * = argmax (S k ).

[0016] Step 5: Dynamic feedback and adaptive update online learning mechanism. Record the CRC check after each switch as a feedback signal. If the CRC check is wrong for 2 consecutive times, trigger the threshold and weight re-optimization (return to step 1). And perform periodic parameter re-estimation, when SNR mutation is detected (such as change > 6dB), re-execute the above steps to update the communication mode.

[0017] Optimize the weight coefficient, which can dynamically adjust the current communication mode, and update the communication mode used by the system according to the next parameter estimation result.

[0018] The main advantages of the present application include:

[0019] 1. By real-time sensing of multi-dimensional parameters such as Doppler shift, signal-to-noise ratio and multipath structure, the system can match different ocean environments (such as shallow water strong multipath, deep water long distance, mobile high noise, etc.), and realize autonomous switching of communication mode.

[0020] 2. Low overhead and high compatibility: reuse communication frame synchronization header for parameter estimation, no need to send additional probe signal, save bandwidth resources.

[0021] 3. Resource utilization optimization: by dynamically selecting communication mode, in low SNR scenario, prefer to enable spread spectrum anti-noise, in high Doppler scenario, switch to single carrier anti-frequency offset, in stable channel, use high speed communication system (OFDM and single carrier) to improve throughput, thereby maximize the utilization of channel capacity, avoid the waste of resources caused by fixed mode.

[0022] 5. Strong engineering practicability: the decision algorithm has low complexity, can be embedded in low power embedded chip, meets the strict requirements of underwater equipment on power consumption and real-time performance. In addition, the initialization of weight and threshold can be adjusted and set through historical data, which reduces the risk of mis-switching in cold start stage, and is suitable for large-scale underwater network rapid deployment. BRIEF DESCRIPTION OF DRAWINGS

[0023] Figure 1is the adaptive switching process of the underwater acoustic communication system in the application;

[0024] Figure 2 is the signal frame structure and data packet design in the application. DETAILED DESCRIPTION

[0025] The application will be described in more detail below with examples in conjunction with the accompanying drawings.

[0026] An environment-aware adaptive switching method of an underwater acoustic communication system, as shown in Figure 1 , includes the following steps:

[0027] Step 1: In the communication frame structure design stage, use a linear frequency modulation or an m sequence as a synchronization detection signal of the communication frame, insert the same synchronization signal in each frame, and append a CRC check at the tail of the data for dynamic feedback, as shown in Figure 2 . The preset initial threshold values are: a Doppler frequency shift threshold , an SNR threshold SNR th = 15 dB, and a multipath time delay spread threshold Each system is allocated a score Sc i (1-4 points, 4 being the best) in each performance dimension. Four communication modes are predefined, mode 1: single-carrier BPSK (high Doppler, lower SNR applicable), mode 2: OFDM (high speed, low Doppler required), mode 3: spread spectrum communication (anti-multipath, high Doppler, low speed), and mode 4: single-carrier QPSK (high SNR, high Doppler applicable).

[0028] Step 2: At the receiving end, use copy correlation to estimate the signal-to-noise ratio of the synchronization signal of the communication frame. After the synchronization algorithm detects the signal frame, save the frame waveform, intercept the noise end and the synchronization signal segment of the waveform, calculate the signal and noise power, and calculate the signal-to-noise ratio SNR through the ratio of the synchronization head signal power P sync to the noise power P noise , with the specific formula being

[0029]

[0030] Step 3: Use the front and rear synchronization heads of the communication frame to obtain the time delay offset of the front and rear autocorrelation peaks to calculate the Doppler frequency shift , where △τ is the time delay difference, υ is the sound speed, and λ is the wavelength. The Doppler frequency shift f d is estimated.

[0031] Step 4: Use copy correlation to obtain a rough channel impulse response, set the multipath threshold to 0.2, detect the multipath time delay spread τ max , and complete the multipath structure identification.

[0032] Step 5: The environmental parameter needs a threshold to determine whether the current environment is challenging (condition indication), when greater than the parameter threshold, the calculation of the parameter is triggered, that is, C i = 1, otherwise C i = 0.

[0033] Step 5: Build a judgment vector, and the weight coefficient represents the importance of each parameter (such as SNR > Doppler > multipath). After normalization, the weight vector is obtained (for example: 0.7, 0.1, 0.2).

[0034] Step 6: Calculate the environmental matching score for each communication mode: Select the mode with the highest score as the optimal mode, k * = argmax(S k ).

[0035] Step 7: Record the CRC check after each switch as a feedback signal. If the CRC check is wrong for 2 times in a row, trigger the threshold and weight optimization (return to step 1). Periodic parameter re-estimation is also performed, and when a SNR mutation (such as a change of > 6dB) is detected, the above steps are re-executed to update the communication mode.

[0036] In specific embodiments, the test data processing of the application is as follows:

[0037] (1) Test conditions and parameters:

[0038] In November 2024, a communication experiment was conducted on the Songhua River in China. Two underwater acoustic communication machines were used for testing, numbered 4# and 5#. The communication system was set up to include four communication systems: single-carrier BPSK, multi-order spread spectrum, single-carrier QPSK, and direct sequence spread spectrum. According to the steps, different communication systems were used for transmission at different signal-to-noise ratios. The communication distance was 2.5 km. Anti-low signal-to-noise ratio performance (for low signal-to-noise ratio environment): direct sequence spread spectrum > M-order spread spectrum > BPSK > QPSK. Anti-Doppler performance (for high Doppler shift environment): direct sequence spread spectrum > BPSK >> M-order spread spectrum > QPSK (">>" means that BPSK is significantly better than M-order spread spectrum). Anti-multipath performance (for large multipath spread environment): M-order spread spectrum > BPSK > direct sequence spread spectrum > QPSK. To quantify the performance, each system is assigned a score (1-4 points, 4 being the best) on each performance dimension, as shown in Table 1.

[0039] Table 1: Each system is assigned a score on each performance dimension

[0040]

[0041] The preset initial threshold is: Doppler shift threshold SNR threshold SNRth = 15 dB, multipath time delay threshold The performance score of the signal-to-noise ratio parameter is increased by a one-level threshold division SNR th = 5 dB, when the SNR is greater than 5 dB, the anti-low signal-to-noise performance score 1 is used, and when the SNR is less than 5 dB, the anti-low signal-to-noise performance score 2 is used.

[0042] (2) Test results

[0043] Both devices are set to receive signals, and the system is determined according to the current environment, and the test data is replied according to the system.

[0044] Test 1, first use 4# communication machine to send system 1 test data to trigger the system, as shown in table 2, 5# receives system 1 and uses algorithm for adaptive switching, the current signal-to-noise ratio is calculated to be about 25 dB, according to the above method, the environment matching degree evaluation S = [0, 0, 0, 0], according to the rate agreement, the system 3 rate is the highest, so the current best communication system is system 3. In the following several rounds, both sides use system 3 for communication, the channel condition is good, and the best communication efficiency is realized.

[0045] Table 2 Songhua River environment perception adaptive system switching test case 1

[0046]

[0047] Test 2, in order to simulate the change of environment, the communication transmission power is artificially reduced, as shown in table 3, 4# transmits system 3, at this time 5# communication machine receives signal-to-noise ratio calculation for 12 dB, the communication system switching is system 1.

[0048] Table 3 Songhua River environment perception adaptive system switching test case 2

[0049]

[0050] The initial receiving calculation obtains the current signal-to-noise ratio of 12 dB, according to the above method, the environment matching degree evaluation S = [1.4, 0, 1.05, 0], the current best communication system is system 1. In the following several rounds, the received signal-to-noise ratio is about 25 dB, and the communication system 3 is used for communication, and the best communication efficiency is realized.

Claims

1. A method for adaptively switching an underwater acoustic communication system based on environmental awareness, characterized in that: At the receiving end, the synchronization head is used to perceive the current environmental parameters, and the Doppler frequency shift, signal-to-noise ratio, and channel structure recognition algorithm are weighted based on the preset threshold. The optimal communication mode is determined through fusion and the switching is completed.

2. The method for adaptively switching an underwater acoustic communication system for environmental perception according to claim 1, characterized in that: The synchronization head is used to sense the current environmental parameters, including signal-to-noise ratio, Doppler shift, and multipath structure. The specific steps are as follows: Step 1: Use a linear frequency modulation signal (LFM) or a pseudo-random sequence as a synchronization header, and design and estimate the parameters of the synchronization header at the receiving end. The synchronization header is embedded in the communication frame header and repeated once per frame for real-time environmental perception. Step 2: Use the synchronization head signal power P sync and noise power P noise The signal-to-noise ratio SNR is calculated by the ratio of Step 3: Calculate the frequency shift using the time delay offset of the synchronization header autocorrelation peak. Formula: Among them, △τ is the time delay difference, υ is the speed of sound, and λ is the wavelength. The Doppler frequency shift f is estimated. d ; Step 4: Extract the multipath delay spread τ by matched filtering or channel impulse response (CIR) analysis. max , completing multi-path structure recognition.

3. The method for adaptively switching an underwater acoustic communication system for environmental awareness according to claim 1, characterized in that: The Doppler frequency shift, signal-to-noise ratio, and channel structure recognition algorithm are weighted based on a preset threshold, and the optimal communication mode is determined through fusion and switching is completed. The specific steps are: Step 1: Environmental parameter thresholds and weight assignment; preset initial threshold: Doppler shift threshold SNR threshold SNR th , multipath delay spread threshold Step 2: Weight coefficient calculation uses weight distribution; construct a judgment vector matrix, and the weight coefficient represents the weight vector obtained after normalization of the importance of each parameter Step 3: Communication mode decision and switching; The candidate communication standard library predefines four communication modes, each of which is applicable to different scenarios, for example: Mode 1: single-carrier BPSK, Mode 2: OFDM, Mode 3: spread spectrum communication, Mode 4: single-carrier QPSK; Step 4: Calculate the matching degree of each communication mode: select the mode with the highest score as the optimal mode, k * =argmax(S k ); Step 5: Online learning mechanism for dynamic feedback and adaptive update; record the CRC check after each switch as a feedback signal; If the CRC check fails twice in a row, the trigger threshold and weight are optimized again and the process returns to step 1. Periodic parameter re-estimation is performed. When a sudden change in SNR is detected, the above steps are re-executed to update the communication mode.

Images