Frequency hopping communication method suitable for deep sea RAP channel

Abstract

The invention discloses a frequency hopping communication method suitable for a deep sea RAP channel, and the method comprises the steps: calculating a channel boundary of a sparse multipath structure under the RAP channel according to the prior information of geography, depth and hydrology, and designing a frequency hopping communication signal structure suitable for the RAP channel; estimating channel parameters by using a matched filtering method, and measuring and calculating multipath time delay, phase difference and in-band energy; performing time domain equalization by using the estimated sparse multipath channel, and eliminating inter-symbol interference of communication; and performing frequency response phase shift equalization on each sub-band, and compensating frequency selective attenuation and phase deflection of the deep sea channel. The method is robust in algorithm, is suitable for robust communication in a deep sea RAP channel which is low in signal-to-noise ratio and dynamically fluctuates, and has reference value for communication system design in various typical deep sea sound channels.

Description

Technical Field

[0001] The present invention relates to the field of communication under the RAP channel, which is a key technology for deep-sea communication networks, and particularly relates to a frequency-hopping communication design method that uses the deep-sea RAP three-dimensional channel model as the prior information for deep-sea communication networks. Background Art

[0002] With the in-depth understanding of ocean information by people, distributed underwater acoustic networks in deep-sea scenarios have increasingly important application prospects. Reasonably distributed network nodes can more efficiently utilize a large range of deep-sea areas to achieve reliable communication under deep-sea channels. However, compared with terrestrial wireless networks that use electromagnetic waves and light waves as carriers of communication information, underwater acoustic channels use low-speed, unstable, narrow-bandwidth, and short-distance acoustic channels as the transmission channels for communication signals. The deep-sea acoustic channel contains the superposition of multiple direct waves and reflected acoustic rays in a large range of airspace, and its propagation structure is relatively complex. Among them, the RAP channel in the non-complete sound channel is the most typical. The deep-sea communication network uses the RAP with a large depth span to achieve long horizontal distance communication between near-surface nodes and near-bottom nodes.

[0003] The underwater acoustic frequency-hopping communication technology is relatively mature. However, specifically for the communication method under deep-sea channels, due to the long channel delay spread of multipath and the harsh conditions of low signal-to-noise ratio in long-distance transmission, the blind equalization filtering method based on dynamic error adjustment has poor effects, and the equalization method based on prior pilot estimation of channel characteristics has not been effectively designed for the channel characteristics of RAP. Since the multipath structure of the RAP deep-sea channel can be simplified into three types of constant long-delay multipath structures: direct wave, first sea surface reflection, and first sea bottom reflection, and can be pre-localized by the inversion of sound field information, pilot signals can be reasonably designed, and channel parameters can be accurately extracted at the receiving end, combined with the design of equalizers to improve communication effects. Summary of the Invention

[0004] The purpose of the present invention is to propose a frequency-hopping communication method suitable for the deep-sea RAP channel by combining the characteristics of the deep-sea RAP channel to address the three problems of long channel delay, frequency-selective fading, and phase change.

[0005] The technical solution for achieving the purpose of the present invention is as follows: A frequency-hopping communication method suitable for the deep-sea RAP channel, the steps are as follows:

[0006] Step 1: Calculate the spatial eigen - acoustic ray propagation trajectories between the source and receiver in the depth and distance dimensions related to frequency, as well as the propagation loss, interface reflection loss, and phase deflection of each ray in the acoustic ray cluster, generate the impulse response under the deep - sea RAP channel characterizing the attenuation of the communication signal. Based on the three - type multipath structure assumptions of the deep - sea RAP channel with an incomplete sound channel, calculate the channel boundaries of the sea - surface reflection multipath cluster and the sea - floor reflection multipath cluster relative to the direct wave, and construct an M - element frequency - hopping communication signal using frequency - modulated pulse signals with different frequencies;

[0007] Step 2: Perform cross - correlation transformation on the received signal, use the peak - energy detection method to detect the moment with the strongest correlation - peak energy between the entire signal and the time series of the transmitted copy signal as the time index of the signal synchronization moment at the receiving end, and calculate the average spectral - level energy and sub - band amplitude estimation in the sub - bands equally divided in the frequency domain for the main path;

[0008] Step 3: According to the multipath frequency - domain sub - band amplitude estimation and multipath delay, construct the received - channel transfer function and multipath time - reversal filter for each sub - band, reconstruct the received communication signal segment by sub - band from the received signal segment, perform time - domain equalization with the multipath time - reversal filter, and superimpose the outputs of each sub - band after time - domain equalization in the time domain to achieve time - domain multipath equalization;

[0009] Step 4: Perform frequency - response phase - shift equalization on each sub - band, compensate for the frequency - selective attenuation and phase deflection of the deep - sea channel, perform cross - correlation transformation with the M - element frequency - hopping sequence copy signal, and demodulate the communication signal using the peak - detection method to output the symbol sequence.

[0010] Furthermore, in Step 1: Calculate the spatial eigen - acoustic ray propagation trajectories between the source and receiver in the depth and distance dimensions related to frequency, as well as the propagation loss, interface reflection loss, and phase deflection of each ray in the acoustic ray cluster, generate the impulse response under the deep - sea RAP channel characterizing the attenuation of the communication signal. Based on the three - type multipath structure assumptions of the deep - sea RAP channel with an incomplete sound channel, calculate the channel boundaries of the sea - surface reflection multipath cluster and the sea - floor reflection multipath cluster relative to the direct wave, and construct an M - element frequency - hopping communication signal using frequency - modulated pulse signals with different frequencies. The specific method is as follows:

[0011] (1 - 1) Define the calculation parameters of the deep - sea sound - field environment, the spatial transceiver position range in the RAP path, and the communication - bandwidth parameters, and calculate the spatial eigen - acoustic ray propagation trajectories between the source and receiver in the depth and distance dimensions related to frequency;

[0012] Set the calculation - parameter configuration item ξ(f, D bottom , ssp(c, d), d bottom , A bottom ) of the deep - sea RAP channel, including: the center frequency f of the communication band = (f l +fh ) / 2, seabed depth D bottom , equal-depth sound speed profile ssp(c, d), seabed medium density d bottom , seabed sound attenuation coefficient A bottom ; Set the deep-sea airspace transceiver parameter configuration item ψ(z s , z r , r), including: transmitting sound source depth z s , receiver depth z r and the horizontal transceiver distance r;

[0013] Calculate the spatial propagation trajectory of the eigen-ray cluster of the RAP channel from the transmitting sound source to the receiver;

[0014] where n beam represents the number of rays in the eigen-sound ray cluster, represents the i-th beam sound ray in the cluster at a distance depth angle variation relationship;

[0015] (1-2) According to the spatial eigen-ray propagation trajectory, calculate the propagation loss, interface reflection loss and phase deflection of each ray in the sound ray cluster, and generate the impulse response under the deep-sea RAP channel characterizing the attenuation of the communication signal;

[0016] The reflection coefficient calculation formula for the sound energy loss caused by the sound ray passing through the interface is:

[0017]

[0018] where Z1 and Z2 respectively represent the characteristic impedances Z of the incident medium and the outgoing medium i = c i ·d i ;

[0019] The propagation loss of the sound ray is:

[0020]

[0021] where, l ray represents the length of the sound ray trajectory;

[0022] Set the transmitting sound source level SPL, and calculate the sound energy level of the i-th beam sound ray reaching the receiver by combining the interface reflection loss and the propagation loss;

[0023]

[0024] where, n refl represents the i-thbeam The number of times the root sound ray is reflected on the interface represents the reflection loss;

[0025] The i beam time delay of the root sound ray reaching the receiver is:

[0026]

[0027] The i beam arrival phase of the root sound ray is expressed as:

[0028]

[0029] where dem(·) represents taking the remainder, represents the reflection phase;

[0030] Combining the sound energy level and time delay of each root sound ray, the channel time-domain impulse response corresponding to the i-th spatial domain parameter configuration item ψ i is obtained;

[0031]

[0032] where e i =[z s i ,z r i ,r i represents the i-th spatial domain configuration item including the depth z of the transmitting sound source s i , the depth z of the receiver r i and the horizontal distance r between the transmitter and the receiver i ;

[0033] (1-3) Based on the three types of multipath structure assumptions of the deep-sea RAP channel with an imperfect sound channel, the channel boundaries of the sea surface reflection multipath cluster and the seabed reflection multipath cluster relative to the direct wave are calculated: the propagation loss difference boundary, the phase deflection amount boundary, and the propagation time delay difference boundary;

[0034] The channel boundary matrix considering RAP is expressed as

[0035] The RAP with an imperfect sound channel can simplify the multipath structure into the direct wave D, the sea surface reflection sound ray SR, and the seabed reflection sound ray BR, three groups of ray clusters beam]D,SR,BR];

[0036] Define the propagation time delay difference boundary of RAP:

[0037] Θ τ = max(Δτ D-BR ,Δτ D-SR)

[0038] Propagation loss difference boundary:

[0039]

[0040] Phase deflection amount boundary:

[0041]

[0042] (1-4) Construct an M-ary frequency-hopping communication signal using frequency-modulated pulse signals of different frequencies;

[0043]

[0044] where T code represents the length of the frequency-hopping communication signal, f i represents the center frequency of the FSK carrier corresponding to the i-th symbol, and B represents the constant bandwidth of any symbol, where the bandwidths of each symbol carrier and cover the entire communication frequency band;

[0045] (1-4) Construct the carrier symbol unit of the M-ary frequency-hopping communication signal using frequency-modulated pulse signals of different frequencies;

[0046]

[0047] where T code represents the length of the frequency-hopping communication signal, f i represents the center frequency of the M-ary frequency-hopping communication signal carrier corresponding to the i-th symbol, and B represents the constant bandwidth of any symbol, where the bandwidths of each symbol carrier and cover the entire communication frequency band.

[0048] Generate a symbol sequence of M-ary with the number of symbols being c num according to the actual application requirements and generate the M-ary frequency-hopping communication signal s com (t) according to formula (1).

[0049]

[0050] where f s represents the sampling frequency, c len represents the symbol length, c num represents the number of symbols in the symbol sequence, T feq represents the time domain length of the frequency domain energy compensation signal s(t feq ),

[0051]

[0052] Set the pilot signal It is expressed as a frequency division multiplexing LFM structure with M equal divisions of the communication frequency band;

[0053] According to the time delay boundary Θ τ and the pilot signal length Set the zero-padding length between the pilot signal and the communication signal It is expressed as having the same length as the propagation time delay difference boundary.

[0054] Set the transmitted signal as the time-domain superposition form s of the pilot signal, zero-padding sequence, and communication signal send_sig = [s pilot , o zp , s com (t)].

[0055] Furthermore, in step 2, perform cross-correlation transformation on the received signal. Use the peak energy detection method to detect the moment with the strongest correlation peak energy between the entire signal and the time series of the transmitted copy signal, and use it as the time index of the signal synchronization moment at the receiving end. Then calculate the average spectral level energy and sub-band amplitude estimation in the sub-bands with M equal divisions in the frequency domain of the main path. The specific method is as follows:

[0056] (2-1) Assume the signal at the receiving end is Perform cross-correlation operation with the pilot signal s pilot to achieve cross-correlation transformation of the received signal;

[0057]

[0058] Use the peak energy detection method to detect the moment with the strongest correlation peak energy between the entire signal and the time series of the transmitted copy signal Use it as the time index of the signal synchronization moment at the receiving end, satisfying:

[0059]

[0060] The received pilot signal, copy signal, and communication signal are respectively expressed as:

[0061]

[0062] (2-2) Define the received signal segment extended from the time index to the same time length as the pilot signal at the receiving end as Define the spectrum transformation formula fft(·), and calculate the average spectral level energy in the sub-bands with M equal divisions in the frequency domain of the main path;

[0063]

[0064] (2-3) From the time index to In the cross-correlation output signal segment with the time domain length, the coherent detection method is used to find the maximum value in the copied signal at the receiving end as the multipath delay

[0065]

[0066] Among them, It is expressed as local maximum search, that is, in the domain of a ∈ S, search for all that satisfy the relationship of [b(a) > mean(b)] ∩ [b(a) > 2·max(b(a - Δa), b(a + Δa))], where Δa is the amplitude attenuation coefficient of the cross-correlation of the matched filter for the unit impulse;

[0067] Calculate the moments of each maximum value The phase deflection Δψ corresponding to the received signal segment beam ;

[0068]

[0069] (2 - 4) According to the prior phase deflection boundary and the calculated phase deflection matching, classify into sea surface reflection multipath clusters, seabed reflection multipath clusters, and direct waves;

[0070]

[0071] Divide each multipath into M - element sub - bands and reconstruct Perform cross - correlation on the signals of each sub - band, where the cross - correlation output for i ∈ [1, …, M] is:

[0072]

[0073] Then the channel amplitudes in different M - element equal - sub - bands of each multipath are:

[0074]

[0075] Furthermore, in step 3, according to the multipath frequency - domain sub - band amplitude estimation and multipath delay, construct the received channel transfer function and multipath time - reversal filter for each sub - band, reconstruct the received communication signal segment by dividing the received signal segment into sub - bands, perform time - domain equalization with the multipath time - reversal filter, and superimpose the outputs of each sub - band after time - domain equalization in the time domain to achieve time - domain multipath equalization. The specific method is as follows:

[0076] (3 - 1) According to the multipath frequency - domain sub - band amplitude estimation And the multipath delay Construct the received channel transfer function for each sub - band;

[0077]

[0078] Among them, (beama , beam b ) ∈ (D, SR, BR);

[0079] Construct a multi-path time-reversal equalizer for the molecular band Among them, the filter coefficients for i ∈ [1, …, M] are:

[0080]

[0081] Among them, Indicates the length of the time-reversal filter equivalent to the channel transfer function;

[0082] (3-2) Reconstruct the received communication signal segment for the molecular band of the received signal segment, perform time-domain equalization with the multi-path time-reversal equalizer in (3-1) respectively, and superimpose the outputs after time-domain equalization of each sub-band in the time domain to achieve time-domain multi-path equalization;

[0083] Reconstruct the received communication signal segment for the molecular band of the received signal segment;

[0084]

[0085] Pass the reconstructed received communication signal end through the multi-path time-reversal filter Perform time-domain equalization;

[0086]

[0087] And superimpose the signals of each sub-band in the time domain reconstruction to obtain the signal after time-domain multi-path equalization;

[0088]

[0089] Furthermore, in step 4, perform frequency response phase shift equalization on each sub-band, compensate for the frequency selectivity attenuation and phase deflection of the deep sea channel, perform cross-correlation transformation with the M-element frequency hopping sequence copy signal, and use the peak detection method to demodulate the communication signal and output the symbol sequence. The specific method is:

[0090] (4-1) Perform spectrum transformation on the communication signal after multi-path time-domain equalization to obtain the spectral level of the communication signal;

[0091]

[0092] (4-2) According to the average spectral level energy of the main path in the M-element sub-band, perform linear frequency domain equalization on the output of (4-1);

[0093] F feq (f) = F com · F pilot / |F com |

[0094] Perform time-domain phase shift equalization;

[0095] F fpeq (f) = F feq ·e iΔψ

[0096] Perform inverse Fourier transform and map back to the time-domain signal;

[0097] s eq (t) = ifft(F fpeq (f))

[0098] (4 - 3) Perform cross-correlation transformation on the output of (4 - 2) and the M-ary frequency-hopping sequence copy signal, and use the peak detection method to demodulate the communication signal and output the symbol sequence;

[0099] Use the M-ary frequency-hopping sequence copy signal as the copy signal, and perform matched filtering on the received signal through cross-correlation operation to achieve;

[0100]

[0101] Perform peak detection on the c -th integer multiple chip position in the cross-correlation output of the matched filtering and output the symbol sequence; num

[0102]

[0103] A frequency-hopping communication system applicable to the deep-sea RAP channel, implementing the frequency-hopping communication method applicable to the deep-sea RAP channel to achieve frequency-hopping communication applicable to the deep-sea RAP channel, which is divided into four modules and respectively executes steps 1 to 4.

[0104] A computer device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, it implements the frequency-hopping communication method applicable to the deep-sea RAP channel to achieve frequency-hopping communication applicable to the deep-sea RAP channel.

[0105] A computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it implements the frequency-hopping communication method applicable to the deep-sea RAP channel to achieve frequency-hopping communication applicable to the deep-sea RAP channel.

[0106] ​Compared with the prior art, the significant advantages of the present invention are as follows: 1) The pilot signal and FH-MFSK communication signal are designed and constructed based on the time delay, phase, and propagation loss threshold of RAP multipath components, and the design of the transmitted signal is more suitable for the real deep-sea RAP scenario; 2) By propagating through the deep-sea RAP to the receiving end, extracting channel characteristics, and performing equalization and demodulation based on the channel characteristics, the communication effectiveness in this scenario is more prominent. Description of the Drawings

[0107] Figure 1 It is a schematic flowchart of an embodiment of the present invention;

[0108] Figure 2 It is the spatial propagation trajectory of the eigen sound ray of the deep-sea RAP channel;

[0109] Figure 3 It is the multipath channel impulse response containing time delay and amplitude information;

[0110] Figure 4 It is the time-frequency diagram of the transmitted signal, received signal, and output signal after equalization in the communication signal section;

[0111] Figure 5 It is the bit error rate performance of this method under different signal-to-noise ratio conditions; Detailed Embodiment

[0112] In order to make the objectives, technical solutions, and advantages of this application clearer, the following further elaborates on this application in combination with the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not used to limit this application.

[0113] A frequency hopping communication method applicable to the deep-sea RAP channel according to the present invention designs in detail the composition of the transmitted frequency hopping signal and the receiving and processing method through the mechanism combined with the modeling method of the deep-sea RAP channel. Different from the previous methods for evaluating communication performance, which are limited to: only combining the empirical distribution of the bit error rate of a certain type of communication signal at different signal-to-noise ratios with the channel propagation loss, and calculating whether the detection threshold of the communication signal can be reached through the sonar equation. The specific method is as follows:

[0114] Step 1: Calculate the channel boundary of the sparse multipath structure under the RAP channel according to the prior information of geography, depth, and hydrology, and design the M-ary frequency hopping communication signal and the pilot signal according to the propagation time delay difference boundary;

[0115] (1-1) Calculate the spatial eigen sound ray propagation trajectory between the source and receiver in the depth and distance dimensions related to frequency according to the defined deep-sea sound field environment calculation parameters, the spatial transceiver position range in the RAP path, and the communication bandwidth parameters;

[0116] Set the calculation parameter configuration item ξ(f, D bottom , ssp(c, d), d bottom , A bottom ) for the sound field environment in the deep - sea RAP channel, including: the center frequency f of the communication band = (f l + f h ) / 2, the seabed depth D bottom , the constant - depth sound - speed profile ssp(c, d), the seabed medium density d bottom , the seabed sound - attenuation coefficient A bottom ; Set the transceiver parameter configuration item ψ(z s , z r , r), including: the depth z s of the transmitting sound source, the depth z r of the receiver, and the horizontal distance r between the transmitter and the receiver.

[0117] Use the ray - calculation model to calculate the spatial propagation trajectory of the eigen - ray cluster of the RAP channel from the transmitting sound source to the receiver where n beam represents the number of rays in the eigen - sound - ray cluster, represents the variation relationship of the i beam -th sound ray in the cluster at a distance depth angle .

[0118] (1 - 2) According to the spatial eigen - ray propagation trajectory, calculate the propagation loss, interface reflection loss, and phase deflection of each ray in the sound - ray cluster, and generate the impulse response under the deep - sea RAP channel characterizing the attenuation of the communication signal;

[0119] The reflection - coefficient calculation formula for the sound - energy loss caused by the sound ray passing through the interface is

[0120]

[0121] where Z1 and Z2 respectively represent the characteristic impedances Z i = c i ·d i ;

[0122] The propagation loss of the sound ray is

[0123]

[0124] where l ray represents the length of the sound - ray trajectory.

[0125] Set the transmitting - sound - source level SPL, and calculate the i beamThe sound energy level of the i-th sound ray reaching the receiver;

[0126]

[0127] where n refl represents the number of reflections of the i-th sound ray on the interface, beam and represents the reflection loss.

[0128] The time delay of the i-th sound ray reaching the receiver is beam

[0129]

[0130] The arrival phase of the i-th sound ray is expressed as beam

[0131]

[0132] where dem(·) represents taking the remainder, and represents the reflection phase.

[0133] Combining the sound energy level and time delay of each sound ray, the channel time-domain impulse response corresponding to the i-th spatial domain parameter configuration item ψ i is obtained;

[0134]

[0135] where l i = [z s i , z r i , r i represents the i-th spatial domain configuration item including the depth z s i of the transmitting sound source, the depth z r i of the receiver, and the horizontal distance r i between the transmitter and the receiver;

[0136] (1-3) Based on the three types of multipath structure assumptions of the deep-sea RAP channel with an incomplete sound channel, the channel boundaries of the sea surface reflection multipath cluster (SR) and the seabed reflection multipath cluster (BR) relative to the direct wave (D) are calculated: the propagation loss difference boundary, the phase deflection amount boundary, and the propagation time delay difference boundary;

[0137] The channel boundary matrix considering RAP is expressed as

[0138] The RAP with an incomplete sound channel can simplify the multipath structure into the direct wave D, the sea surface reflection sound ray SR, and the seabed reflection sound ray BR, and the three groups of ray clusters beam ∈ [D, SR, BR].​

[0139] Define the propagation delay difference boundary of RAP

[0140] Θ τ = max(Δτ D-BR , Δτ D-SR )

[0141] Propagation loss difference boundary

[0142]

[0143] Phase deflection amount boundary

[0144]

[0145] (1 - 4) Construct the carrier symbol unit of the M - element frequency - hopping communication signal using frequency - modulated pulse signals of different frequencies;

[0146]

[0147] where T code represents the length of the frequency - hopping communication signal, f i represents the center frequency of the FSK carrier corresponding to the i - th symbol, and B represents the constant bandwidth of any symbol, where the bandwidth of each symbol carrier covers the entire communication frequency band.

[0148] Generate a symbol sequence of M - ary with the number of symbols c num according to the actual application requirements and generate the FSK transmitted communication signal s (t) according to the transmitted symbol sequence FSK .

[0149]

[0150] where f s represents the sampling frequency, c len represents the symbol length, c num represents the number of symbols in the symbol sequence, and T feq represents the time - domain length of the frequency - domain energy compensation signal s(t feq );

[0151]

[0152] Set the pilot signal which is expressed as a frequency - division multiplexing LFM structure that equally divides the communication frequency band into M elements.

[0153] Set the zero - padding length between the pilot signal and the communication signal according to the delay boundary Θ τ and the length of the pilot signal ​ It means that the length is equal to the boundary of the propagation time delay difference.

[0154] Set the transmitted signal as the time-domain superposition form s of a pilot signal, a zero-padding sequence, and a communication signal send_sig =[s pilot ,o zp ,s com (t)].

[0155] Step 2: Use the matched filtering method to estimate the channel parameters, and measure the multipath time delay, phase deflection amount, and in-band energy;

[0156] (2-1) Assume that the received signal at the receiving end is Perform a cross-correlation operation on it with the pilot signal s in formula (1) pilot Cross-correlation operation Realize the cross-correlation transformation of the received signal.

[0157]

[0158] Use the peak energy detection method to detect the moment when the correlation peak energy of the entire signal and the transmitted copy signal time series is the strongest Use it as the time index of the receiving end signal synchronization moment, satisfying

[0159]

[0160] The received pilot signal, copy signal, and communication signal are respectively expressed as

[0161]

[0162] (2-2) Define the received signal segment at the receiving end extended from the time index to the same time length as the pilot signal in step 1-4 as Define the spectrum transformation formula fft(·), and calculate the average spectral energy in the sub-bands equally divided into M elements in the frequency domain of the main path;

[0163]

[0164] (2-3) From the time index To In the cross-correlation output signal segment with the time domain length, use the coherent detection method to find the maximum value in the copy signal at the receiving end

[0165]

[0166] Among them, It is expressed as local maximum search, that is, in the a∈S domain, search for all that satisfy

[0167] related to the relationship of [[b(a)>mean(b)]] ∩ [[b(a)>2·max(b(a-Δa),b(a+Δa))]], where Δa is the amplitude attenuation coefficient of the matched filter for the unit impulse cross-correlation.

[0168] Calculate the moments of each maximum value The phase deflection Δψ corresponding to the received signal segment beam ;

[0169]

[0170] (2-4) Classify D, SR, and BR according to the prior phase deflection boundary and the calculated phase deflection matching.

[0171]

[0172] Divide each multipath into M-element subbands and reconstruct Cross-correlate the signals of each subband, where the cross-correlation output for i ∈ [1,…,M] is:

[0173]

[0174] Then the channel amplitude within different M-element equal subbands of each multipath;

[0175]

[0176] Step 3, perform time-domain equalization using the estimated multipath channel to eliminate inter-symbol interference in communication. The specific method is:

[0177] (3-1) According to the multipath frequency-domain subband amplitude estimation in (2-4) And the multipath delay Construct the received channel transfer function of each subband;

[0178]

[0179] Among them, (beam a , beam b ) ∈ (D, SR, BR).

[0180] Construct the time-reversal filters for the subbands respectively The filter coefficients for i ∈ [1,…,M] are:

[0181]

[0182] Among them, Represents the length of the time-reversal filter equivalent to the channel transfer function.

[0183] (3-2) Reconstruct the received signal segment into sub-band received communication signal segments, perform time-domain equalization with the multi-path time-reversal equalizer in step 3-1 respectively, and superimpose the outputs of the time-domain equalization of each sub-band in the time domain to achieve time-domain multi-path equalization.

[0184] Reconstruct the received signal segment into sub-band received communication signal segments;

[0185]

[0186] Pass the reconstructed received communication signal end through the multi-path time-reversal filter Perform time-domain equalization

[0187]

[0188] And superimpose the signals of each sub-band in the time domain reconstruction to obtain the signal after time-domain multi-path equalization;

[0189]

[0190] Step 4, perform frequency response phase shift equalization on each sub-band, compensate for the frequency selectivity attenuation and phase deflection of the deep sea channel, and achieve communication decoding. The specific method is as follows:

[0191] (4-1) Perform spectrum transformation on the communication signal after multi-path time-domain equalization in (3-2) to obtain the spectral level of the communication signal;

[0192]

[0193] (4-2) According to the average spectral level energy of the main path calculated in (2-2) in the M-element sub-band, perform linear frequency domain equalization on the output of (4-1);

[0194] F feq (f) = F com ·F pilot / |F com |

[0195] Perform phase shift equalization on the time domain;

[0196] F fpeq (f) = F feq ·e iΔψ

[0197] Perform inverse Fourier transform on the output and map it back to the time domain signal;

[0198] s eq (t) = ifft(F fpeq (f))

[0199] (4-3) performing cross-correlation transformation on the output of (4-2) and the M-ary frequency hopping sequence copy signal, demodulating the communication signal using a peak detection method, and outputting a codeword sequence;

[0200] The signal is copied using the M-ary frequency hopping sequence in formula (1) As a copy signal, through the cross-correlation operation Realize matched filtering of the received signal.

[0201]

[0202] Cross-correlation output of matched filter c num Peak detection is performed at integer multiple code chip positions and a code symbol sequence is output.

[0203]

[0204] The present invention also proposes a frequency hopping communication system applicable to a deep-sea RAP channel, implements the frequency hopping communication method applicable to a deep-sea RAP channel, and realizes frequency hopping communication applicable to a deep-sea RAP channel.

[0205] A computer device comprises a memory, a processor and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the frequency hopping communication method applicable to a deep-sea RAP channel is implemented to realize the frequency hopping communication applicable to a deep-sea RAP channel.

[0206] A computer-readable storage medium stores a computer program. When the computer program is executed by a processor, the frequency hopping communication method applicable to a deep-sea RAP channel is implemented to realize the frequency hopping communication applicable to a deep-sea RAP channel.

[0207] Example

[0208] In order to verify the effectiveness of the solution of the present invention, the following experiment was conducted.

[0209] (1) Using the ray calculation model, according to the configuration parameters of the deep-sea RAP channel: the center frequency of the communication band, the seabed depth, the equal-sea-depth sound velocity profile, the seabed medium density, the seabed sound attenuation coefficient, the transmitting sound source depth, the receiver depth and the horizontal distance between transmission and reception, the spatial propagation trajectory of the intrinsic ray cluster of the RAP channel from the transmitting sound source to the receiver is calculated, and the multipath reflection loss, phase deflection and multipath delay are calculated.

[0210] Figure 3 The figure shows the propagation trajectory of the airspace acoustic ray under the RAP channel parameter configuration. The simulation parameters are: the center frequency of the communication band is 7.5kHz, the seabed depth is 5000m, and the density of the seabed medium is 1.8g / cm3 , the underwater sound attenuation coefficient is 0.8 dB / Hz, and the deep-sea airspace transceiver parameter configuration item ψ i (z s ,z r ,r) is shown in Table 1. The transmitting sound source depth is 300 m, the receiver depth is 4500 m, and the horizontal distance between the transmitter and the receiver is 15 km.

[0211] According to the propagation trajectory of the eigenrays in the airspace, calculate the propagation loss, interface reflection loss, and phase deflection of each ray in the acoustic ray cluster, and generate the impulse response under the deep-sea RAP channel characterizing the attenuation of the communication signal;

[0212] Figure 4 The impulse response of the multipath time-domain channel is shown as follows. The simulation parameters are: the center frequency of the communication band is 7.5 kHz, the seabed depth is 5000 m, the density of the seabed medium is 1.8 g / cm 3 , the underwater sound attenuation coefficient is 0.8 dB / Hz, and the deep-sea airspace transceiver parameter configuration item ψ i (z s ,z r ,r) is shown in Table 1. The transmitting sound source depth is 300 m, the receiver depth is 4500 m, and the horizontal distance between the transmitter and the receiver is 15 km.

[0213] Based on the three types of multipath structure assumptions of the deep-sea RAP channel with an incomplete sound channel, calculate the propagation loss difference boundary, phase deflection boundary, and propagation time delay difference boundary of the sea surface reflection multipath cluster (SR) and the seabed reflection multipath cluster (BR) relative to the direct wave (D); design the minimum zero-padding length between the pilot signal and the M-element frequency-hopping coding signal according to the channel boundary, where the pilot signal is set to a frequency-division multiplexing LFM structure that equally divides the communication band into M elements.

[0214] (2) Perform copy correlation on the pilot signal at the receiving end, and use the time-domain peak position as the synchronization time and amplitude of the receiving-end signal; perform spectral transformation on the synchronized received pilot signal and calculate the spectral level; find each maximum time in the copy correlation output as the multipath time delay, and intercept the time-domain signals of the same length as the pilot signal corresponding to each multipath, and calculate the phase deflection based on the pilot signal; classify D, SR, and BR according to the calculated prior boundary and the phase deflection of each maximum value, equally divide the M elements of each multipath pilot signal, and calculate the amplitude of each sub-band after reconstruction;

[0215] (3) According to the channel estimation output, construct a time-reversal multipath time-reversal equalizer for each sub-band; reconstruct the received communication signal segment for each sub-band, equalize it with the sub-band of the multipath time-reversal equalizer in (3-1), and superimpose it in the time domain to achieve time-domain multipath equalization.

[0216] (4) Perform spectral transformation on the communication signal after (3-2) multipath time-domain equalization; according to the energy difference of the M subbands, perform linear frequency-domain equalization on the output of the multipath delay equalization, and perform inverse Fourier transform on the output to map it back to the time-domain signal; correlate the output with the M-element frequency-hopping sequence copy signal respectively, and use the peak detection method to demodulate the communication signal and output the symbol sequence;

[0217] Figure 5 The shown is the LOFAR time-frequency spectrogram of the communication signal segments of the transmitted signal, received signal and equalized output signal of FH-MFSK. The simulation parameters are: the sampling frequency is 50 kHz, the length of the time-domain discrete sequence of the symbol is 500, the upper and lower limits of the FH-MFSK communication frequency band are 5 kHz and 10 kHz respectively, the number of carriers is 8 elements, the number of symbols is 50, and the signal-to-noise ratio in the band is 25 dB.

[0218] Figure 5 The shown is the output of the demodulation bit error rate of 1000 Monte Carlo times with the signal-to-noise ratio in the range of -10 to 15 dB and the number of carriers being 4, 8, and 16 elements respectively. The simulation parameters are: the sampling frequency is 50 kHz, the length of the time-domain discrete sequence of the symbol is 500, the upper and lower limits of the FH-MFSK communication frequency band are 5 kHz and 10 kHz respectively, and the number of symbols is 100.

[0219] From the experimental data processing results, it can be concluded that this method can have good communication performance under certain signal-to-noise ratio conditions in the RAP channel.

[0220] The technical features of the above embodiments can be combined arbitrarily. For the sake of brief description, not all possible combinations of the technical features in the above embodiments are described. However, as long as the combinations of these technical features do not conflict, they should all be considered as the scope described in this specification.

Claims

1. A frequency hopping communication method suitable for deep sea RAP channel, characterized in that: Here are the steps: Step 1, calculate the frequency-related depth, distance dimension spatial intrinsic sound ray propagation trajectory between the sound source and the receiver, as well as the propagation loss of each ray in the sound ray cluster, the interface reflection loss and the phase deflection, generate the impulse response under the deep-sea RAP channel that characterizes the attenuation of the communication signal, and based on the three types of multipath structure assumptions of the deep-sea RAP channel of the incomplete sound channel, calculate the channel boundaries of the sea surface reflection multipath cluster and the seabed reflection multipath cluster relative to the direct wave, and use frequency-modulated pulse signals of different frequencies to construct an M-ary frequency-hopping communication signal; Step 2, perform cross-correlation transformation on the received signal, use the peak energy detection method to detect the moment when the correlation peak energy of the entire signal and the transmitted copy signal time series is the strongest, as the time index of the receiving end signal synchronization moment, and calculate the average spectral level energy and sub-band amplitude estimation of the main path in the M-element equally divided sub-band in the frequency domain; Step 3: Based on the multipath frequency domain subband amplitude estimation and multipath delay, construct the receiving channel transfer function and multipath time inversion filter of each subband, reconstruct the received communication signal segment by using the received signal segment subband, perform time domain equalization with the multipath time inversion filter, and superimpose the outputs of each subband after time domain equalization in the time domain to achieve time domain multipath equalization; Step 4: Perform frequency response phase shift equalization on each sub-band to compensate for the frequency selective attenuation and phase deviation of the deep-sea channel, perform cross-correlation transformation with the M-ary frequency hopping sequence copy signal, use the peak detection method to demodulate the communication signal, and output the codeword sequence.

2. The frequency hopping communication method applicable to deep sea RAP channel according to claim 1, characterized in that: Step 1: Calculate the frequency-related depth, distance-dimensional spatial intrinsic sound ray propagation trajectory between the sound source and the receiver, as well as the propagation loss of each ray in the sound ray cluster, the interface reflection loss and the phase deflection, generate the impulse response under the deep-sea RAP channel that characterizes the attenuation of the communication signal, and calculate the channel boundaries of the sea surface reflection multipath cluster and the seabed reflection multipath cluster relative to the direct wave based on the three-type multipath structure assumption of the deep-sea RAP channel of the incomplete sound channel, and use frequency-modulated pulse signals of different frequencies to construct an M-ary frequency-hopping communication signal. The specific method is as follows: (1-1) Define the deep-sea acoustic field environment calculation parameters, the spatial transmission and reception position range in the RAP path, and the communication bandwidth parameters, and calculate the frequency-related depth and distance-dimensional spatial intrinsic sound ray propagation trajectory between the sound source and the receiver; Set the sound field environment calculation parameter configuration item ξ(f,D bottom ,ssp(c,d),d bottom ,A bottom ), including: the center frequency of the communication frequency band f = (f l +f h ) / 2, seafloor depth D bottom , constant sea depth sound velocity profile ssp(c,d), seabed medium density d bottom , Seabed sound attenuation coefficient A bottom ; Set the deep sea airspace transmission and reception parameter configuration item ψ(z s ,z r ,r), including: the depth of the transmitting sound source z s , receiver depth z r and the horizontal distance r between sending and receiving; Calculate the spatial propagation trajectory of the eigenray cluster of the RAP channel from the transmitting sound source to the receiver; where n beam represents the number of rays in the intrinsic acoustic ray cluster, represents the i-th beam The sound ray is at a distance depth angle The changing relationship on (1-2) According to the propagation trajectory of the spatial eigenray, the propagation loss of each ray in the acoustic ray cluster, the interface reflection loss and the phase deflection are calculated to generate the impulse response under the deep-sea RAP channel that represents the attenuation of the communication signal; The calculation formula of the reflection coefficient of the sound energy loss caused by the sound ray passing through the interface is: Where Z1 and Z2 represent the characteristic impedance Z of the incident medium and the output medium respectively. i =c i ·d i ; The propagation loss of sound rays is: Among them, l ray represents the length of the sound ray trajectory; Set the emission sound source level SPL, and calculate the i-th beam The sound energy level of the sound ray reaching the receiver; Among them, n refl Indicates the i beam The number of times a sound ray is reflected on the interface, Represents reflection loss; i-th beam The time delay of an acoustic ray reaching the receiver is: i-th beam The arrival phase of each sound ray is expressed as: Among them, dem(·) means remainder, represents the reflection phase; Combining the sound energy level and time delay of each sound ray, we can get the parameter configuration item ψ in the i-th spatial domain: i The corresponding channel time domain impulse response; in, Indicates the depth z of the ith type of emitted sound source s i , receiver depth z r i And the horizontal distance between sending and receiving r i Airspace configuration items; (1-3) Based on the assumption of three types of multipath structures of the deep-sea RAP channel with incomplete sound channels, the channel boundaries of the sea surface reflection multipath cluster and the seabed reflection multipath cluster relative to the direct wave are calculated: the propagation loss difference boundary, the phase deflection boundary, and the propagation delay difference boundary; Considering RAP, the channel boundary matrix is ​​expressed as The RAP of the incomplete sound channel can simplify the multipath structure into the direct wave D, the sea surface reflected sound ray SR, and the seabed reflected sound ray BR, three groups of ray clusters beam∈[D,SR,BR]; Define the propagation delay difference boundary of RAP: I τ =max(Δτ D-BR ,Dt D-SR ) Propagation loss difference boundary: Phase deflection amount boundary: (1-4) Constructing an M-ary frequency hopping communication signal using frequency modulated pulse signals of different frequencies; Among them, T code Indicates the length of the frequency hopping communication signal, f i represents the center frequency of the FSK carrier corresponding to the i-th code symbol, and B represents the constant bandwidth of any code symbol, where the bandwidth of each symbol carrier and covers the complete communication frequency band; (1-4) constructing a carrier symbol unit of an M-ary frequency hopping communication signal using frequency modulated pulse signals of different frequencies; Among them, T code Indicates the length of the frequency hopping communication signal, f i represents the center frequency of the M-element frequency hopping communication signal carrier corresponding to the i-th code symbol, and B represents the constant bandwidth of any code symbol, wherein the bandwidth of each symbol carrier and covers the complete communication frequency band; The number of generated code elements is c according to the actual application requirements. num The M-ary codeword sequence i∈[1,…,M], and generate an M-ary frequency hopping communication signal s according to formula (1) com (t); Among them, f s represents the sampling frequency, c len Indicates the code length, c num represents the number of codes in the codeword sequence, T feq represents the frequency domain energy compensation signal s(t feq )’s time domain length; Setting the pilot signal It is represented as a frequency division multiplexing LFM structure that divides the communication frequency band into M equal parts; According to the delay boundary Θ τ and pilot signal length Set the zero padding length between the pilot signal and the communication signal It indicates that the length is equal to the propagation delay difference boundary. Set the transmission signal to be the pilot signal, zero-filling sequence, and communication signal in the time domain superposition form. send_sig =[s pilot ,o zp ,s com (t)].

3. The frequency hopping communication method applicable to deep sea RAP channel according to claim 2, characterized in that: Step 2, perform cross-correlation transformation on the received signal, use the peak energy detection method to detect the moment when the correlation peak energy of the entire signal and the transmitted copy signal time series is the strongest, as the time index of the receiving end signal synchronization moment, and calculate the average spectral level energy and sub-band amplitude estimation of the main path in the M-element equally divided sub-band in the frequency domain. The specific method is: (2-1) Assume that the signal at the receiving end is Combine it with the pilot signal s pilot Cross-correlation operation Implementing cross-correlation transformation of received signals; Use the peak energy detection method to detect the moment when the correlation peak energy of the entire signal and the transmitted copy signal time series is the strongest Use it as the time index of the receiving end signal synchronization moment, satisfying: The received pilot signal, copy signal, and communication signal are respectively expressed as: (2-2) Define the receiving end's received signal segment extending from the time index to the same time length as the pilot signal as Define the spectrum transformation formula fft(·) to calculate the average spectrum level energy of the main path in the M-element equally divided subband in the frequency domain; (2-3) From the time index to In the cross-correlation output signal segment of the time domain length, the coherent detection method is used to find the maximum value in the copy signal at the receiving end as the multipath delay in, It is expressed as a local maximum search, that is, in the domain a∈S, searching for all the values ​​that satisfy the relationship [b(a)>mean(b)]∩[b(a)>2·max(b(a-Δa),b(a+Δa))], where Δa is the amplitude attenuation coefficient of the matched filter for the unit impulse cross-correlation; Calculate each maximum moment The phase shift Δψ corresponding to the received signal segment beam ; (2-4) According to the matching of the prior phase deflection amount boundary and the calculated phase deflection amount, the multipaths are classified into sea surface reflection multipath clusters, seabed reflection multipath clusters and direct waves; Each multipath is divided into M sub-bands and reconstructed The signal of each sub-band is cross-correlated, where i∈[1,…,M], and the cross-correlation output is: Then the channel amplitude in each multipath with different M elements is:

4. The frequency hopping communication method applicable to deep sea RAP channel according to claim 3, characterized in that: Step 3: Based on the multipath frequency domain subband amplitude estimation and multipath delay, construct the receiving channel transfer function and multipath time inversion filter of each subband, reconstruct the received communication signal segment by using the received signal segment subband, perform time domain equalization with the multipath time inversion filter, and superimpose the outputs of each subband after time domain equalization in the time domain to achieve time domain multipath equalization. The specific method is as follows: (3-1) Based on the multipath frequency domain sub-band amplitude estimation and multipath delay Construct the receiving channel transfer function of each sub-band; in, (beam a ,beam b )∈(D,SR,BR); Construct a multipath time-reversal filter for each subband The filter coefficients for i∈[1,…,M] are: in, represents the length of the time-inverse filter equivalent to the channel transfer function; (3-2) Reconstruct the received communication signal segment by dividing the received signal segment into sub-bands, perform time domain equalization with the multipath time inversion equalizer of (3-1) respectively, and superimpose the outputs of each sub-band after time domain equalization in the time domain to achieve time domain multipath equalization; Reconstructing received communication signal segments by dividing received signal segments into bands; The reconstructed received communication signal is passed through a multipath time inversion filter Perform time domain equalization; The signals of each sub-band are reconstructed and superimposed in the time domain to obtain a signal after time domain multipath equalization; 5. The frequency hopping communication method applicable to deep sea RAP channel according to claim 4, characterized in that: Step 4: perform frequency response phase shift equalization on each sub-band to compensate for the frequency selective attenuation and phase deflection of the deep-sea channel, and perform cross-correlation transformation with the M-ary frequency hopping sequence copy signal, use the peak detection method to demodulate the communication signal, and output the code element sequence. The specific method is as follows: (4-1) performing spectrum transformation on the communication signal after multipath time domain equalization to obtain the spectrum level of the communication signal; (4-2) Perform linear frequency domain equalization on the output of (4-1) according to the average spectral level energy of the main path in the M-element subband; F feq (f)=F com ·F pilot / |F com | Perform time domain phase shift equalization; F fpeq (f)=F feq ·e iΔψ Perform inverse Fourier transform and map back to time domain signal; s eq (t)=ifft(F fpeq (f)) (4-3) performing cross-correlation transformation on the output of (4-2) and the M-ary frequency hopping sequence copy signal, demodulating the communication signal using a peak detection method, and outputting a codeword sequence; Copy the signal with M-ary frequency hopping sequence As a copy signal, through the cross-correlation operation Realize matched filtering of received signals; Cross-correlation output of matched filter c num Perform peak detection on integer multiple chip positions and output a symbol sequence; 6. A frequency hopping communication system suitable for deep sea RAP channels, characterized in that: Implement the frequency hopping communication method applicable to deep-sea RAP channels as described in any one of claims 1-5 to realize frequency hopping communication applicable to deep-sea RAP channels, which is divided into four modules and executes steps 1 to 4 respectively.

7. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the frequency hopping communication method applicable to a deep-sea RAP channel as described in any one of claims 1 to 5 is implemented to realize frequency hopping communication applicable to a deep-sea RAP channel.

8. A computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the frequency hopping communication method applicable to a deep-sea RAP channel as described in any one of claims 1 to 5 is implemented to realize frequency hopping communication applicable to a deep-sea RAP channel.

Images