A diversity gain and decoding mode switching method for a low frequency wireless communication system

Abstract

The application discloses a diversity gain and decoding mode switching method of a low-frequency wireless communication system, which comprises the following steps: collecting two pieces of data in a time period T1 before code sending and a time period T2 during code sending and carrying out band-pass filtering to obtain N3, N4, X3 and X4; calculating a noise covariance matrix C1 of N3 and N4, a sample covariance matrix C2 of a known sequence section in X3 and X4 and a cross covariance matrix C3; expanding C1 to the dimension of C2 to obtain C4; multiplying the sum of C4, C2 and a regularization coefficient and a unit matrix to obtain two filters after inverting, and then multiplying the two filters with C3 to obtain X7 after filtering X3 and X4; accumulating the total energy of the carriers of each frequency of the known sequence section of X7 to obtain a detection coefficient; comparing the relative signal-to-noise ratio of each detection coefficient with the size of a deep fading threshold to determine the applicable decoding mode of X7. The application can improve the signal-to-noise ratio and adaptively switch the decoding mode based on the carrier detection state.

Description

Technical Field

[0001] This invention relates to the field of wireless communication and signal processing, and in particular to a method for switching diversity gain and decoding mode in a low-frequency wireless communication system. Background Technology

[0002] Submarine optical cables are a core infrastructure of global information and communication networks. Besides optical fibers for data transmission, they typically integrate copper core conductors for power supply. Utilizing this structural characteristic, modulated current signals are applied to the copper core conductors from shore-based stations, generating an alternating magnetic field through the conductive medium of seawater. This establishes a communication link between the submarine optical cable and a submersible. This has become a crucial short-range communication technology solution supporting submersibles in conducting inspections, maintenance, and data exchange near the optical cable. Its communication reliability directly affects the safety and efficiency of submarine optical cable operation and maintenance. However, in practical marine engineering applications, this magnetic induction communication technology has long been limited by several key technological bottlenecks, hindering further performance improvements and its widespread engineering application.

[0003] The inherent frequency-selective fading of the channel is a core issue affecting the integrity of magnetic field signal transmission. The copper core conductor, armor structure, and seawater medium of submarine optical cables together constitute a complex magnetic field transmission channel. During propagation, the alternating magnetic field generated by the modulation current experiences multipath propagation due to scattering by the optical cable structure, the inhomogeneity of the seawater medium, and reflection from the marine environment. Specifically, when the magnetic field signal reaches the submersible receiver, it experiences depth attenuation at a specific frequency point, and this attenuation point dynamically shifts with the relative position of the submersible and the optical cable. The direct consequence is that when the carrier frequency used by the communication system coincides with the instantaneous attenuation point, the carrier signal will be significantly weakened or even completely obliterated, leading to communication link interruption. Furthermore, physical constraints in engineering applications further increase the difficulty of signal reception. Due to the stringent size and weight limitations of the submersible platform, the installation distance between the two magnetic sensors is typically extremely small, much smaller than the wavelength of the magnetic field signal in seawater. This results in a high correlation between the signal envelopes received by the two sensors. This strong spatial correlation fails to meet the fundamental requirement of uncorrelated sensor signals in classical spatial diversity reception theory, severely limiting the gain potential of spatial diversity technology.

[0004] To address the aforementioned technical issues, existing solutions still have significant shortcomings. Currently, magnetic induction technology based on submarine optical cables is limited to cable detection and positioning, relying solely on DC signals or fixed-frequency power frequency signals to determine cable location. It lacks effective acquisition and demodulation techniques for modulated signals carrying complex information, failing to meet practical data communication needs. Specifically, existing technologies largely focus on interference sources (such as brushless DC motors), reducing magnetic interference through shielding or structural optimization. However, research on adaptive noise reduction techniques suitable for real-time communication systems is scarce. Furthermore, existing multi-carrier modulation magnetic communication systems often employ fixed decoding strategies. When any carrier experiences deep fading, the system lacks an adaptive switching mechanism to a more robust modulation scheme, significantly reducing communication reliability. In addition, most existing algorithms are based on the ideal assumption that noise is white noise and that noise between sensors is uncorrelated, failing to consider the colored noise properties of marine magnetic interference and the correlation characteristics of actual sensor noise. This leads to severe performance degradation in real, complex marine noise environments. Furthermore, existing research on magnetic communication is mostly based on laboratory simulation environments or relies on computer platforms, which makes it difficult to directly meet the strict limitations of submersible platforms on size, weight, and power consumption, and thus makes it difficult to achieve engineering applications.

[0005] In summary, under the current technological background, there is an urgent need for an innovative solution that can comprehensively address multiple technical challenges such as frequency-selective fading of magnetic field channels, strongly correlated marine magnetic interference, and limited spatial diversity of submersible platforms. Summary of the Invention

[0006] To address the problems of high bit error rate and frequent communication interruptions in existing low-frequency wireless communication systems under frequency-selective fading environments due to carrier depth attenuation, sensor spatial correlation, and the inability to adaptively switch decoding modes, this invention proposes a diversity gain and decoding mode switching method for low-frequency wireless communication systems. This scheme has adaptive adjustment capabilities, can make coordinated use of limited spatial diversity and frequency diversity gains, and can intelligently switch decoding modes when channel quality deterioration is detected. It also adapts to the constraints of submersible platforms on size, weight, and power consumption, ultimately ensuring the stable and reliable operation of communication links in harsh marine environments.

[0007] The specific technical solution is as follows:

[0008] A method for switching diversity gain and decoding mode in a low-frequency wireless communication system includes the following steps:

[0009] S1: When the interference source is running stably, during the time period T1 before the signal generator transmits the code, collect the raw data N1 at position P1 and the raw data N2 at position P2; during the time period T2 when the code is being transmitted, collect the raw data X1 at position P1 and the raw data X2 at position P2.

[0010] S2: Bandpass filtering is performed on N1, N2, X1, and X2 in sequence to obtain bandpass data N3 and N4 before code transmission and bandpass data X3 and X4 during code transmission;

[0011] S3: Calculate the noise covariance matrix C1 based on N3 and N4; obtain the observed signals of the known sequence segments in X3 and X4 based on the maximum cross-correlation values ​​between X3 and X4 and the local known sequence S, and complete signal synchronization; calculate the sample covariance matrix C2 based on the observed signals of the known sequence segments; extend C1 to the same dimension as C2 to obtain matrix C4; obtain the cross covariance matrix C3 based on the observed signals of the known sequence segments and the local known sequence S; invert the sum of C4, C2, and the product of the regularization coefficient E and the identity matrix I, and multiply the resulting inverse matrix with C3 to obtain filters L1 and L2 applied to X3 and X4; apply L1 filtering to X3 and L2 filtering to X4, and combine the filtering results to obtain X7;

[0012] S4: Based on the known sequence portion in X7, calculate the total energy of the carrier at each frequency of the known sequence portion as the detection coefficient; calculate the relative signal-to-noise ratio of each detection coefficient and determine its relationship with the set deep fading threshold γ to determine the appropriate decoding method for signal X7.

[0013] Furthermore, in S1, the pre-code transmission period T1 and the mid-code transmission period T2 are adjacent, and the time of transition between the two periods is the time when the signal generator starts transmitting codes; the code transmission data of the signal generator includes known sequences and random sequences.

[0014] Furthermore, in S1, the original data N1 and N2 include noise, and the original data X1 and X2 include received signals and noise; the received signals include signals with known sequences and signals with unknown sequences, and the noise includes white noise and interference noise.

[0015] Furthermore, the carrier frequency of the received signal satisfies the following: within each symbol period, different symbols are orthogonal; the known sequence is any one of an m-sequence, a Zadoff-Chu sequence, or other pseudo-random sequences.

[0016] Furthermore, in S2, the bandpass filtering is performed based on the system bandwidth B and the operating frequency f0 of the interference source; the system bandwidth B includes the effective bandwidth of the signal determined by the carrier frequency and symbol rate of the received signal; the system bandwidth B avoids the operating frequency f0 of the interference source while covering the effective bandwidth of the signal.

[0017] Furthermore, the design method for the bandpass filter includes the window function method, the equiripple method, or the least squares method; the low cutoff frequency of the bandpass filter is less than the lower limit frequency of the system bandwidth B, and the high cutoff frequency is greater than the upper limit frequency of the system bandwidth B.

[0018] Furthermore, in S3, C1 is extended to the same dimension as C2 according to the Kronecker product extension to obtain matrix C4.

[0019] Furthermore, in S4, the total energy of the carrier at each frequency of the known sequence portion is calculated cumulatively, specifically through the following operations:

[0020] Extract a synchronization signal containing a known sequence Q from signal X7, generate a reference signal with the same length as the known sequence Q; calculate the cross-correlation function between the synchronization signal and the reference signal, and take the position of its maximum value as the starting position of the known sequence Q to achieve symbol synchronization;

[0021] In signal X7, starting from the beginning position of the known sequence Q, extract a signal of the same length as the known sequence Q. Calculate the signals respectively The energy of each frequency carrier on each symbol is calculated by summing the energy of each symbol to obtain the total energy of the frequency carrier.

[0022] Furthermore, in S4, the decoding method applicable to signal X7 is: a combination decoding method of all carriers or part of the carriers corresponding to the coding method of the signal generator;

[0023] When the signal generator's encoding mode is FSK, the decoding mode for all carriers corresponding to the signal generator's encoding mode is FSK decoding mode, and the decoding mode for some carriers corresponding to the signal generator's encoding mode is OOK decoding mode.

[0024] Furthermore, for any two detection coefficients Y1 and Y2, the relative signal-to-noise ratio of Y1 to Y2 is: 10 multiplied by the logarithm to the base 10, where the argument of the logarithm is the ratio of Y1 to Y2.

[0025] For a modulation scheme with two carrier frequencies, the total energy of the low-frequency carrier is Y. L The total energy of the high-frequency carrier is Y. HAt this point, the deep fading detection method is as follows: when the signal-to-noise ratio of the high frequency relative to the low frequency is greater than γ, it is determined that the low frequency carrier is deeply fading, and the high frequency carrier is subsequently decoded in OOK mode; when the signal-to-noise ratio of the low frequency relative to the high frequency is greater than γ, it is determined that the high frequency carrier is deeply fading, and the low frequency carrier is subsequently decoded in OOK mode; when neither of these conditions is met, it is determined that there is no carrier with deep fading, and the 2FSK mode is subsequently used for decoding.

[0026] For a modulation scheme with two carrier frequencies, the total energy of the low-frequency, intermediate-frequency, and high-frequency carriers is Y, respectively. L Y M Y H At this point, the deep fading detection method is as follows: When the signal-to-noise ratio (SNR) of the intermediate frequency (IF) relative to the low frequency (LFM) is greater than γ, and the SNR of the high frequency relative to the LFM is greater than γ, the LFM carrier is determined to be severely fading. Subsequently, OOK mode decoding is performed on the IF carrier and the high frequency carrier respectively. If the symbols of both carriers are 0, the symbol of the original LFM carrier is determined to be 1. When the SNR of the LFM relative to the IF is greater than γ, and the SNR of the high frequency relative to the IF is greater than γ, the IF carrier is determined to be severely fading. Subsequently, OOK mode decoding is performed on the LFM carrier and the high frequency carrier respectively. If the symbols of both carriers are 0, the symbol of the original IF carrier is determined to be 1. When the SNR of the LFM relative to the high frequency (LFM) is greater than γ, and the SNR of the IF relative to the high frequency (LFM) is greater than γ, the high frequency carrier is determined to be severely fading. Subsequently, OOK mode decoding is performed on the LFM carrier and the IF carrier respectively. If the symbols of both carriers are 0, the symbol of the original high frequency carrier is determined to be 1. When none of the above conditions are met, it is determined that there is no carrier with deep fading, and 3FSK mode decoding is subsequently used.

[0027] The beneficial effects of this invention are:

[0028] (1) The present invention can still achieve signal-to-noise ratio gain when the sensor spacing is limited by noise covariance estimation and merging filtering.

[0029] (2) The adaptive switching decoding method based on carrier detection state of the present invention can avoid the adverse effects of deep fading carrier on decoding.

[0030] (3) The present invention has low computational complexity, requires no modification to existing hardware facilities, and can be directly deployed on FPGA, PC or on-chip system in the field. Attached Figure Description

[0031] Figure 1 This is a flowchart of the diversity gain and decoding mode switching method of a low-frequency wireless communication system in an embodiment of the present invention.

[0032] Figure 2 This is a schematic diagram of the original signal (signal 1) received by sensor 1 in an embodiment of the present invention.

[0033] Figure 3 This is a schematic diagram of the original signal (signal 2) received by sensor 2 in an embodiment of the present invention.

[0034] Figure 4 This is a flowchart illustrating the design of the merging filter in an embodiment of the present invention.

[0035] Figure 5 This is a schematic diagram of the output signal (signal 3) after the received signal 1 is equalized by MMSE in an embodiment of the present invention.

[0036] Figure 6 This is a schematic diagram of the output signal (signal 4) after the received signal 2 is equalized by MMSE in an embodiment of the present invention.

[0037] Figure 7 This is a schematic diagram of the output signal (signal 5) after signal 3 and signal 4 are combined by MMSE in an embodiment of the present invention. Detailed Implementation

[0038] The present invention will be described in detail below with reference to the accompanying drawings and preferred embodiments. The objectives and effects of the present invention will become clearer as a result. The present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention. To more fully illustrate the principle of the present invention, 2FSK modulation is used as an example for explanation. It can be understood that this method can be applied by analogy to modulation schemes with more carriers.

[0039] This invention addresses the frequency-selective fading and strong interference noise problems faced by low-frequency wireless communication telemetry systems. It proposes a diversity gain and decoding mode switching method for low-frequency wireless communication systems, such as... Figure 1 As shown, the method specifically includes the following steps:

[0040] S1: Signal acquisition before and during code transmission.

[0041] Two sensors are positioned at locations P1 and P2 of the low-frequency wireless communication telemetry system to collect data. When the interference source is running stably, during the period T1 before the signal generator transmits codes, raw data N1 is collected at location P1 and raw data N2 is collected at location P2; during this period, there is no signal transmission, only noise (including interference noise and thermal noise) exists.

[0042] During the period T2 after T1 ends and the signal generator begins coding, raw data X1 is collected at position P1 and raw data X2 is collected at position P2. Preferably, the pre-coding period T1 and the coding period T2 are adjacent, and the transition time between the two periods is the start time of the signal generator's coding. This design helps to eliminate noise, as it ensures the stable operation of the interference source. Both X1 and X2 contain noise and the received signal. The noise includes white noise, interference noise, and other noise. The received signal includes signals with known sequences and signals with unknown sequences. The known sequence can be any one of an m-sequence, a Zadoff-Chu sequence, or other pseudo-random sequences. The carrier frequency of the received signal satisfies the following condition: within each symbol period, different symbols are orthogonal.

[0043] like Figure 2 As shown, this is the raw signal (i.e., signal 1) received by sensor 1 at position P1, where the signal before 5 seconds on the time axis belongs to the signal corresponding to N1, and the signal after 5 seconds belongs to the signal corresponding to X1. Figure 3 As shown, the original signal (i.e. signal 2) received by sensor 2 at position P2 is the signal corresponding to N2 before 5s on the time axis, and the signal corresponding to X2 after 5s.

[0044] In this embodiment, the signal generator's coding method is as follows: low-frequency carrier frequency f L The high-frequency carrier frequency is 5Hz, f. H The modulation scheme is 2FSK, with a frequency of 10Hz and a symbol rate of 5 bits per second. The transmitted data consists of an m-sequence and a random sequence, totaling 200 symbols; the m-sequence is generated using a 5-bit linear shift register and has a length of N. p =31 symbols, the corresponding waveforms of which are the waveforms S of a signal with a known sequence.

[0045] S2: Data preprocessing.

[0046] Bandpass filtering is performed on N1 and N2 based on the system bandwidth B and the operating frequency f0 of the interference source to obtain the pre-code bandpass data N3 and N4. The vector formed by N3 and N4 is denoted as... Bandpass filtering is performed on X1 and X2 based on the system bandwidth B and the operating frequency f0 of the interference source to obtain the bandpass data X3 and X4 in the transmitted code. The vector formed by X3 and X4 is denoted as... Bandpass filtering is used to preserve the complete code information of the signal generator and to eliminate the DC component and major interference noise components of the received signal.

[0047] The system bandwidth B includes the effective bandwidth of the signal, which is determined by the carrier frequency and symbol rate of the received signal; the system bandwidth B avoids the operating frequency f0 of the interference source while covering the effective bandwidth of the signal.

[0048] Bandpass filter design methods include the window function method, the equiripple method, or the least squares method. The low cutoff frequency of the bandpass filter is lower than the lower limit frequency of the system bandwidth B, and the high cutoff frequency is higher than the upper limit frequency of the system bandwidth B. In this embodiment, the low cutoff frequency of the bandpass filter is 2Hz, and the high cutoff frequency is 13Hz.

[0049] S3: Design of merging filters. For example... Figure 4 As shown, S3 is implemented through the following sub-steps:

[0050] (3.1) Calculate the covariance matrix of data vectors N3 and N4 as the noise covariance matrix C1 to avoid performance deviation caused by the assumption of white noise. Specifically, estimate the true noise covariance matrix C1 using the bandpass data before code transmission, as shown in the following expression:

[0051]

[0052] In the formula, N n This represents the number of noise samples. In practical applications, 5000 sampling points can be used, corresponding to a sampling rate f. s =Noise data for 5 seconds at 1000Hz; It is a 2×1 dimensional noise sample vector, corresponding to the sampled values ​​of the pre-code pass data N3 and N4 from the two sensors at position j.

[0053] This estimation process can capture the correlation between sensors and the colored noise characteristics of interference noise, thereby improving the numerical stability of the covariance matrix estimation.

[0054] (3.2) Signal synchronization: Calculate the position of the maximum cross-correlation value between X3 and the waveform S of the local known sequence to obtain the starting position of the known sequence in X3. Calculate the position of the maximum cross-correlation value between X4 and the waveform S of the local known sequence to obtain the starting position of the known sequence in X4, and complete the signal synchronization.

[0055] (3.3) Stack the known sequence segment observation signals of the two sensors (i.e., the signals in X3 and X4 corresponding to their known sequences) into a 2×N stack. p 3D observation matrix X p Calculate the sample covariance matrix C2 of the received signal:

[0056]

[0057] The noise covariance matrix C1 is extended to the same dimension as the sample covariance C2 using the Kronecker product extension, resulting in the extended noise covariance matrix C4. .

[0058] (3.4) The waveform S of the locally known sequence is copied and stacked as 2×N.p The matrix S of the waveform of a known sequence p Calculate the cross-covariance matrix C3 between the received signal and the waveform S of the known sequence, as shown in the following expression:

[0059]

[0060] (3.5) Based on the minimum mean square error (MMSE) criterion, the optimal linear combination of the two sensors is achieved; the length of the linear combination vector is set to I. L Its length is the same as the number of sampling points in the known sequence, and needs to be greater than the number of sampling points corresponding to the empirical value of the channel length. Specifically, the sum of the products of the extended noise covariance matrix C4, the sample covariance matrix C2, and the regularization coefficient E and the identity matrix I is inverted. The resulting inverse matrix is ​​multiplied by the cross covariance matrix C3 to obtain filter L1, which can be applied to X3, and filter L2, which can be applied to X4, respectively. The expression for the combined filter is as follows:

[0061]

[0062] In this embodiment, the regularization coefficient E = 10 -6 This is used to ensure the numerical stability of matrix inversion, and the resulting matrix L is I. L A 2x2 matrix, with the first column taken as the I value of the sensor at position P1. L The filter L1 has one sampling point, and the second column serves as the I of the sensor at position P2. L Filter L2 with one sampling point.

[0063] (3.6) Applying L1 filtering to X3 yields X5, as follows: Figure 5 As shown (after 5 seconds on the timeline), applying L2 filtering to X4 yields X6, as follows. Figure 6 As shown (after 5 seconds on the time axis), the obtained signals X5 and X6 are added and combined to obtain the diversity gain result X7, as shown. Figure 7 As shown (after 5 seconds on the timeline), this is used as the basis for subsequent signal processing. The expression is as follows:

[0064]

[0065] In the formula, This represents the convolution operation.

[0066] Subsequent signal processing, including carrier depth fading detection and subsequent signal processing procedures such as noise reduction and equalization decoding involved in the embodiments of the present invention, are all based on the merged signal X7.

[0067] S4. Carrier Deep Fading Detection: Based on the known sequence portion of signal X7, the total energy of the carriers at each frequency of the known sequence portion is calculated cumulatively and used as detection coefficients. The number of detection coefficients is the same as the number of carriers (divided according to frequency). The relative signal-to-noise ratio (SNR) of each detection coefficient is calculated. Based on the relationship between the relative SNR and the set deep fading threshold γ, the appropriate decoding method for signal X7 is determined to avoid interference from noise in the frequency band where the deep fading carrier is located on the original multi-carrier decoding method under a strong frequency-selective channel. S4 is specifically implemented through the following sub-steps:

[0068] (4.1) Perform FSK symbol synchronization on signal X7: Extract a synchronization signal from signal X7 that includes a known sequence Q (the channel-modified version of S). Generate a reference signal (of the same length) corresponding to the known sequence Q; calculate Cross-correlation function with reference signal:

[0069]

[0070] In the formula, N is the synchronization signal. The number of sampling points.

[0071] The location where the maximum value of the cross-correlation function is obtained. This is the starting position of the known sequence Q, which enables symbol synchronization.

[0072] (4.2) For signal X7, extract a signal of the same length as the known sequence Q, starting from the beginning position of the known sequence Q, and use it as the signal. Calculate the signal respectively The energy y of the low-frequency carrier on each symbol L [k] and the energy y of the high-frequency carrier H [k], the expression is as follows:

[0073]

[0074]

[0075] In the formula, f s T is the sampling rate. s For the symbol period, f L f is the low-frequency carrier frequency. H It is a high-frequency carrier frequency; For the nth sampling point of the kth symbol, , It is the number of sampling points for each symbol.

[0076] (4.3) The energy of each symbol is summed to obtain the total low-frequency carrier energy Y of the known sequence portion of signal X7. LThe total energy Y of the high-frequency carrier H These are taken as detection coefficients, and the expressions are as follows:

[0077]

[0078]

[0079] (4.4) By performing deep fading hypothesis verification on low-frequency and high-frequency carriers respectively (i.e., based on the relationship between the relative signal-to-noise ratio of each detection coefficient and the set deep fading threshold γ), the fading situation of low-frequency and high-frequency carriers is determined, and the appropriate decoding method for signal X7 is determined. The appropriate decoding method for signal X7 is a combination decoding method of all carriers or some carriers corresponding to the transmission method; in particular, when the transmission method is FSK, the corresponding decoding methods include FSK (all carriers corresponding to the transmission method) decoding method and OOK (some carriers corresponding to the transmission method) decoding method. The following is a detailed explanation with reference to the embodiments.

[0080] As an embodiment of the present invention, for a modulation scheme with two carrier frequencies (2FSK coding scheme), the deep fading threshold γ is calculated based on the energy ratio (in dB) between the low-frequency and high-frequency carriers. The deep fading detection method in this case is: when the signal-to-noise ratio of the high-frequency carrier relative to the low-frequency carrier... When the signal-to-noise ratio is greater than γ, it is determined to be deep fading of the low-frequency carrier, and subsequent OOK mode decoding is performed on the high-frequency carrier; when the signal-to-noise ratio of the low frequency relative to the high frequency is... When the value is greater than γ, it is determined to be deep fading of the high-frequency carrier, and the low-frequency carrier is subsequently decoded in OOK mode; when neither condition is met, the fading degree of either the low-frequency or high-frequency carrier is significantly greater than that of the other, and the 2FSK mode is subsequently used for decoding.

[0081] As another embodiment of the present invention, for a modulation scheme with three carrier frequencies, such as 3FSK, the low-frequency, intermediate-frequency, and high-frequency carrier frequencies are fi, fj, ... L f M f H The total energy of the low-frequency, mid-frequency, and high-frequency carriers is Y, respectively. L Y M Y H And used as the detection coefficient; under this modulation method, the deep fading detection method is: when the signal-to-noise ratio of the intermediate frequency relative to the low frequency and the high frequency relative to the low frequency satisfies and When the low-frequency carrier is severely fading, OOK mode decoding is subsequently performed on the intermediate frequency (IF) and high-frequency (HF) carriers respectively. For carriers where both symbols are 0, the original low-frequency carrier is identified as having a symbol of 1. When the signal-to-noise ratio (SNR) of the low-frequency carrier relative to the IF and the high-frequency carrier relative to the IF satisfy... and When the intermediate frequency (IF) carrier is severely fading, OOK mode decoding is subsequently performed on the low-frequency and high-frequency carriers respectively. For carriers where both symbols are 0, the original IF carrier is determined to have a symbol of 1. When the signal-to-noise ratio (SNR) of the low-frequency carrier relative to the high-frequency carrier and the IF carrier relative to the high-frequency carrier satisfies the following condition... and If the high-frequency carrier is severely fading, the low-frequency carrier and the intermediate-frequency carrier are then decoded in OOK mode. If the symbols of both carriers are 0, the symbol of the original high-frequency carrier is determined to be 1. If none of the above conditions are met, it is determined that neither carrier is severely fading, and the 3FSK mode decoding is continued.

[0082] It will be understood by those skilled in the art that the above descriptions are merely preferred examples of the invention and are not intended to limit the invention. Although the invention has been described in detail with reference to the foregoing examples, those skilled in the art can still modify the technical solutions described in the foregoing examples or make equivalent substitutions for some of the technical features. All modifications and equivalent substitutions made within the spirit and principles of the invention should be included within the scope of protection of the invention.

Claims

1. A method for switching diversity gain and decoding mode in a low-frequency wireless communication system, characterized in that, Includes the following steps: S1: When the interference source is running stably, during the time period T1 before the signal generator transmits the code, collect the raw data N1 at position P1 and the raw data N2 at position P2; during the time period T2 when the code is being transmitted, collect the raw data X1 at position P1 and the raw data X2 at position P2. S2: Bandpass filtering is performed on N1, N2, X1, and X2 in sequence to obtain bandpass data N3 and N4 before code transmission and bandpass data X3 and X4 during code transmission; S3: Calculate the noise covariance matrix C1 based on N3 and N4; obtain the observed signals of the known sequence segments in X3 and X4 based on the maximum cross-correlation values ​​between X3 and X4 and the local known sequence S, and complete signal synchronization; calculate the sample covariance matrix C2 based on the observed signals of the known sequence segments; extend C1 to the same dimension as C2 to obtain matrix C4; obtain the cross covariance matrix C3 based on the observed signals of the known sequence segments and the local known sequence S; invert the sum of C4, C2, and the product of the regularization coefficient E and the identity matrix I, and multiply the resulting inverse matrix with C3 to obtain filters L1 and L2 applied to X3 and X4; apply L1 filtering to X3 and L2 filtering to X4, and combine the filtering results to obtain X7; S4: Based on the known sequence portion in X7, calculate the total energy of the carrier at each frequency of the known sequence portion as the detection coefficient; calculate the relative signal-to-noise ratio of each detection coefficient and determine its relationship with the set deep fading threshold γ to determine the appropriate decoding method for signal X7.

2. The method for switching diversity gain and decoding mode in a low-frequency wireless communication system according to claim 1, characterized in that, In S1, the pre-code transmission period T1 and the mid-code transmission period T2 are adjacent, and the time of transition between the two periods is the time when the signal generator starts transmitting codes; the code transmission data of the signal generator includes known sequences and random sequences.

3. The method for switching diversity gain and decoding mode in a low-frequency wireless communication system according to claim 1, characterized in that, In S1, the original data N1 and N2 include noise, and the original data X1 and X2 include received signals and noise; the received signals include signals with known sequences and signals with unknown sequences, and the noise includes white noise and interference noise.

4. The diversity gain and decoding mode switching method for a low-frequency wireless communication system according to claim 3, characterized in that, The carrier frequency of the received signal satisfies the following: within each symbol period, different symbols are orthogonal; the known sequence is any one of an m-sequence, a Zadoff-Chu sequence, or other pseudo-random sequences.

5. The method for switching diversity gain and decoding mode in a low-frequency wireless communication system according to claim 1, characterized in that, In S2, bandpass filtering is performed based on the system bandwidth B and the operating frequency f0 of the interference source; the system bandwidth B includes the effective bandwidth of the signal determined by the carrier frequency and symbol rate of the received signal; the system bandwidth B avoids the operating frequency f0 of the interference source while covering the effective bandwidth of the signal.

6. The method for switching diversity gain and decoding mode in a low-frequency wireless communication system according to claim 5, characterized in that, The design methods for the bandpass filter include the window function method, the equiripple method, or the least squares method; the low cutoff frequency of the bandpass filter is less than the lower limit frequency of the system bandwidth B, and the high cutoff frequency is greater than the upper limit frequency of the system bandwidth B.

7. The method for switching diversity gain and decoding mode in a low-frequency wireless communication system according to claim 1, characterized in that, In S3, C1 is extended to the same dimension as C2 according to the Kronecker product extension to obtain matrix C4.

8. The method for switching diversity gain and decoding mode in a low-frequency wireless communication system according to claim 1, characterized in that, In step S4, the total energy of the carrier at each frequency of the known sequence portion is calculated cumulatively, which is achieved through the following operations: Extract a synchronization signal containing a known sequence Q from signal X7, generate a reference signal with the same length as the known sequence Q; calculate the cross-correlation function between the synchronization signal and the reference signal, and take the position of its maximum value as the starting position of the known sequence Q to achieve symbol synchronization; In signal X7, starting from the beginning position of the known sequence Q, extract a signal of the same length as the known sequence Q. Calculate the signals respectively The energy of each frequency carrier on each symbol is calculated by summing the energy of each symbol to obtain the total energy of the frequency carrier.

9. The method for switching diversity gain and decoding mode in a low-frequency wireless communication system according to claim 1, characterized in that, In S4, the decoding method applicable to signal X7 is: a combination decoding method of all carriers or some carriers corresponding to the coding method of the signal generator; When the signal generator's encoding mode is FSK, the decoding mode for all carriers corresponding to the signal generator's encoding mode is FSK decoding mode, and the decoding mode for some carriers corresponding to the signal generator's encoding mode is OOK decoding mode.

10. The method for switching diversity gain and decoding mode in a low-frequency wireless communication system according to claim 9, characterized in that, For any two detection coefficients Y1 and Y2, the relative signal-to-noise ratio of Y1 to Y2 is: 10 multiplied by the logarithm to the base 10, where the argument of the logarithm is the ratio of Y1 to Y2. For a modulation scheme with two carrier frequencies, the total energy of the low-frequency carrier is Y. L The total energy of the high-frequency carrier is Y. H At this point, the deep fading detection method is as follows: when the signal-to-noise ratio of the high frequency relative to the low frequency is greater than γ, it is determined that the low frequency carrier is deeply fading, and the high frequency carrier is subsequently decoded in OOK mode; when the signal-to-noise ratio of the low frequency relative to the high frequency is greater than γ, it is determined that the high frequency carrier is deeply fading, and the low frequency carrier is subsequently decoded in OOK mode; when neither of these conditions is met, it is determined that there is no carrier with deep fading, and the 2FSK mode is subsequently used for decoding. For a modulation scheme with two carrier frequencies, the total energy of the low-frequency, intermediate-frequency, and high-frequency carriers is Y, respectively. L Y M Y H At this point, the deep fading detection method is as follows: When the signal-to-noise ratio (SNR) of the intermediate frequency (IF) relative to the low frequency (LFM) is greater than γ, and the SNR of the high frequency relative to the LFM is greater than γ, the LFM carrier is determined to be severely fading. Subsequently, OOK mode decoding is performed on the IF carrier and the high frequency carrier respectively. If the symbols of both carriers are 0, the symbol of the original LFM carrier is determined to be 1. When the SNR of the LFM relative to the IF is greater than γ, and the SNR of the high frequency relative to the IF is greater than γ, the IF carrier is determined to be severely fading. Subsequently, OOK mode decoding is performed on the LFM carrier and the high frequency carrier respectively. If the symbols of both carriers are 0, the symbol of the original IF carrier is determined to be 1. When the SNR of the LFM relative to the high frequency (LFM) is greater than γ, and the SNR of the IF relative to the high frequency (LFM) is greater than γ, the high frequency carrier is determined to be severely fading. Subsequently, OOK mode decoding is performed on the LFM carrier and the IF carrier respectively. If the symbols of both carriers are 0, the symbol of the original high frequency carrier is determined to be 1. When none of the above conditions are met, it is determined that there is no carrier with deep fading, and 3FSK mode decoding is subsequently used.

Images