OFDM low pilot channel estimation method based on coherence of underwater acoustic slow time-varying channel

Abstract

This paper proposes a low-pilot channel estimation method for OFDM based on the coherence of slow time-varying underwater acoustic channels. In the field of underwater acoustic communication technology, the previous method required allocating pilot signals for each symbol within a communication frame, wasting already limited underwater acoustic bandwidth resources. This negatively impacted the accuracy of the initial channel estimate. The proposed solution addresses this issue by using the channel estimate updated from the previous symbol for the initial channel estimate in OFDM symbols after the third symbol. Since only the first two OFDM symbols in a frame of underwater acoustic communication signals containing multiple OFDM symbols have bandwidth resources allocated for channel estimation, the third and subsequent OFDM symbols do not require bandwidth allocation. Therefore, compared to communication systems that allocate pilot resources for every OFDM symbol, the proposed solution significantly improves the utilization of underwater acoustic bandwidth resources, ensuring sufficient resources for channel estimation and thus improving the accuracy of the initial channel estimate.

Description

Technical Field

[0001] This application relates to the field of underwater acoustic communication technology, specifically to an OFDM low pilot channel estimation method based on the coherence of underwater acoustic slow time-varying channels. Background Technology

[0002] Increased data throughput and improved data recovery quality depend on the accuracy of channel estimation. Due to the complexity of underwater channels, pilot-assisted channel estimation algorithms remain the mainstream approach. However, pilot allocation requires bandwidth resources, thus the accuracy of channel estimation algorithms and the improvement of communication rates are mutually constrained. Typically, channel estimation algorithms assume that the underwater acoustic channels experienced by communication symbols are independent, therefore the receiver processes the communication signal symbol by symbol. Symbol-by-symbol processing requires allocating pilots to each symbol within the communication frame, wasting already limited underwater acoustic bandwidth resources. This results in the communication system not being able to allocate sufficient resources for channel estimation, thereby affecting the accuracy of the initial channel estimate. Summary of the Invention

[0003] The purpose of this invention is to address the problem that symbol-by-symbol processing requires allocating pilot signals for each symbol within a communication frame, which wastes the already limited underwater acoustic bandwidth resources and consequently affects the accuracy of the initial channel estimate. This invention provides an OFDM low-pilot channel estimation method based on the coherence of slow time-varying underwater acoustic channels.

[0004] The technical solution adopted by the present invention to solve the above-mentioned technical problems is as follows:

[0005] OFDM low-pilot channel estimation methods based on the coherence of underwater acoustic slow time-varying channels include:

[0006] Step 1: Acquire each frame of signal transmitted by the underwater acoustic communication system;

[0007] Step 2: For the first OFDM symbol in each frame of signal, i.e. =1, allocate pilot signals to obtain the fuzzy channel. and utilize Signal detection is performed to obtain data symbols. Then utilize Feedback updates yield accurate channel information. Finally, set the channel phase. =0, and set the channel gain to 0. = ;

[0008] Step 3: For the second OFDM symbol in each frame of signal, i.e. =2, repeat step 2 to obtain the accurate channel. ;

[0009] Step 4: Calculate the channel phase shift change based on the precise channel of the current symbol and the precise channel of the previous symbol. ;

[0010] Step 5: Based on the phase shift changes of adjacent symbol channels Calculate the residual Doppler factor ;

[0011] Step 6: Based on residual Doppler factors Update the channel phase of the current symbol using the channel phase of the previous symbol. ;

[0012] Step 7: Channel phase based on the current symbol And the precise channel, and the channel gain of the previous symbol, update the channel gain of the current symbol. ;

[0013] Step 8: Determine if the current symbol is the last symbol. If yes, end the process and output the precise channel, channel gain, and channel phase of the current symbol. Otherwise, proceed to step 9.

[0014] Step 9: Predict the channel gain estimate and channel phase estimate of the next symbol, and obtain the fuzzy channel of the next symbol based on the channel gain estimate and channel phase estimate, and perform signal detection to obtain the accurate channel of the next symbol. Finally, use the accurate channel as the accurate channel of the current symbol in step 4, and repeat steps 4 to 8.

[0015] Furthermore, the channel phase shift change Represented as:

[0016] ,

[0017] in, Indicates frequency, This represents the residual Doppler factor after Doppler compensation. Indicates the symbol length of CP-OFDM.

[0018] Furthermore, the residual Doppler factor after Doppler compensation Represented as:

[0019] ,

[0020] in, The Doppler factor representing the path. This represents the channel Doppler factor estimated by the receiver.

[0021] Furthermore, the residual Doppler factor Represented as:

[0022] ,

[0023] in, Indicates a pair of elements Phase operation, Indicates the total number of OFDM subcarriers. It represents the imaginary unit.

[0024] Furthermore, the channel phase Represented as:

[0025] ,

[0026] in, The channel phase corresponding to the current OFDM symbol. This is the channel phase corresponding to the previous OFDM symbol.

[0027] Furthermore, the channel gain Represented as:

[0028] ,

[0029] in, This represents the set proportionality coefficient. Indicates the channel corresponding to the current symbol.

[0030] Furthermore, the channel gain estimate for the next OFDM symbol in step 9 is expressed as follows:

[0031] ,

[0032] in, This represents the estimated channel gain for the next OFDM symbol.

[0033] Furthermore, the channel phase estimate for the next OFDM symbol in step 9 is expressed as follows:

[0034] ,

[0035] in, This represents the channel phase estimate for the next OFDM symbol.

[0036] Furthermore, the fuzzy channel representation of the next OFDM symbol in step 9 is as follows:

[0037] .

[0038] Furthermore, the signal transmitted by the underwater acoustic communication system in step 1 is a compensated signal. Represented as:

[0039] ,

[0040] in, This represents the channel Doppler factor estimated by the receiver. This represents the OFDM passband time domain signal arriving at the receiving hydrophone. This represents a continuous time-domain time variable.

[0041] The beneficial effects of this invention are:

[0042] The technical solution of this application uses an initial channel estimate in the third and subsequent OFDM symbols based on the channel estimate updated in the previous symbol. Since only the first two OFDM symbols in a frame of underwater acoustic communication signal containing multiple OFDM symbols are allocated bandwidth resources for channel estimation, the third and subsequent OFDM symbols do not require bandwidth allocation for channel estimation. Therefore, compared to communication systems where pilot resources are allocated to each OFDM symbol, the technical solution of this application significantly improves the utilization rate of underwater acoustic bandwidth resources, ensuring the communication system has sufficient resources for channel estimation and thus improving the accuracy of the initial channel estimate. Attached Figure Description

[0043] Figure 1 This is a comparison chart of the bit error rate performance of this application and conventional channel estimation algorithms;

[0044] Figure 2 This application compares the constellation diagram performance of constellation diagrams with conventional channel estimation algorithms. Figure 1 ;

[0045] Figure 3 This application compares the constellation diagram performance of constellation diagrams with conventional channel estimation algorithms. Figure 2 ;

[0046] Figure 4 This is the overall flowchart of this application. Detailed Implementation

[0047] It should be noted that, where there is no conflict, the various embodiments disclosed in this application can be combined with each other.

[0048] Specific Implementation Method 1: The OFDM low pilot channel estimation method based on underwater acoustic slow time-varying channel coherence described in this implementation method:

[0049] Slow time-varying underwater acoustic channel modeling

[0050] The underwater acoustic channel can be modeled as having Discrete multipath model of path, the first path The channel gain and time delay corresponding to each path are expressed as follows: and Since the duration of a single OFDM symbol is typically shorter than the coherence time of an underwater acoustic channel, it can be assumed that the path gain remains constant within a single communication symbol. Furthermore, the path delay change over one symbol time can be approximated by a first-order polynomial:

[0051] Obtain OFDM symbols and model the underwater acoustic channel experienced by the OFDM symbols as having The discrete multipath model of the path is obtained, and based on the discrete multipath model, the first path is... The channel gain and time delay corresponding to each path are expressed as follows: and ,

[0052] (1)

[0053] in Indicates the initial path delay during the OFDM duration. This represents the Doppler factor for each path.

[0054] When the main Doppler shift of a communication system is caused by the relative motion between the transmitter and receiver, the Doppler factor of all paths can be considered consistent, i.e. Even if the actual situation does not meet this assumption, the resulting error can still be considered as additional noise. Under calm ocean conditions, the underwater acoustic channel exhibits slow time-varying behavior, meaning that parameters such as path delay, path gain, and Doppler factor are assumed to remain constant within one OFDM symbol period, while channel parameters are compressed or expanded by Doppler effects between adjacent OFDM blocks. Therefore, the first... The path delay experienced by an OFDM symbol can be modeled as follows:

[0055] (2)

[0056] in The symbol length for CP-OFDM.

[0057] In practical communication systems, the receiver first performs Doppler estimation and compensation processing on the received signal to mitigate the impact of inter-carrier interference (ICI) caused by the time-varying underwater acoustic channel on communication demodulation. Assume the channel Doppler factor estimated by the receiver is... The compensated passband signal is then expressed as:

[0058] (3)

[0059] in This is the OFDM passband time domain signal that reaches the receiving hydrophone.

[0060] Doppler factor compensation is applied to the OFDM symbol, and the compensated passband signal is: .

[0061] definition The remaining Doppler factor after Doppler compensation affects the variation of the frequency domain channel response (CFR) between adjacent symbols. The Doppler-compensated... The first OFDM symbol The underwater acoustic channel transfer function experienced on each subcarrier can be expressed as the equivalent channel complex gain. and equivalent phase change Combinations:

[0062] (4)

[0063] (5)

[0064] (6)

[0065] in , For the first The frequencies corresponding to each OFDM subcarrier. Due to channel gain The frequency domain channel response changes slowly between adjacent OFDM symbols. Mainly affected It changes due to the influence of [the environment / the environment].

[0066] Compare two adjacent OFDM symbols The CFR relationship corresponding to each OFDM subcarrier is obtained as follows:

[0067] (7)

[0068] in Indicates the frequency of two adjacent symbols The phase change at CFR is written as:

[0069] (8)

[0070] Therefore, the residual Doppler factor of the channel can be estimated from the phase change of the channel according to formulas (7) and (8). Furthermore, the channel frequency response can be predicted by the Doppler factor according to the formula, thus realizing adaptive channel estimation.

[0071] Based on the above, the adaptive channel estimation proposed in this application consists of four parts: prior information acquisition, channel tracking, channel prediction, and channel update.

[0072] Prior information acquisition

[0073] Define each frame of signal transmitted by the underwater acoustic communication system as being composed of The signal consists of several OFDM symbols. The first two OFDM symbols of each frame are assigned pilot signals. The frequency domain channel response is obtained by applying the pilot-assisted orthogonal matched pursuit channel estimation method. and Apply to symbol 1 and symbol 2 and Signal detection is performed, and then the detected data symbols are... and Treating the pilot symbols as known, they are used together with the original pilots for data feedback channel estimation to obtain an updated channel. and Due to the initial phase shift value The initial value of the channel gain is assumed to be... .

[0074] Update the currently processed symbol index to .

[0075] Channel tracking

[0076] Calculate the current OFDM symbol block according to formula (7). and the previous symbol block At carrier frequency The channel frequency response ratio at that point, i.e.:

[0077] (9)

[0078] According to formula (8), the change in channel phase angle is affected by the residual Doppler factor. Due to the influence of this, the estimated value of the Doppler factor of the current symbol block can be obtained from the phase angle of the ratio of the estimated values ​​of adjacent symbol channels in formula (9), that is:

[0079] (10)

[0080] in Indicates a pair of elements Phase operation, This represents the total number of OFDM subcarriers. It should be noted that there is a potential issue with the estimation of the phase difference in equation (10). The problem is fuzzy, therefore the range of phase difference that ensures the algorithm can accurately estimate is [range missing]. .

[0081] Based on the assumption of linear change in channel frequency response phase, according to formula (6), the current symbol block The channel phase shift can be determined by the previous symbol block. Channel phase The current symbol block Doppler factor estimate obtained by formula (10) Perform tracking and prediction to obtain symbol blocks. Channel phase:

[0082] (11)

[0083] After that, due to Given the channel phase prediction value obtained according to formula (11), and channels By inversely calculating the equivalent channel complex gain, we obtain:

[0084] (12)

[0085] In slow time-varying channel models, channel gain The change is slow, and the result obtained from formula (12) can be compared with the equivalent channel gain of the previous symbol block. Combined, update the channel gain of the current symbol block to obtain:

[0086] (13)

[0087] in The set proportional coefficient, In practical applications, Mostly The selection is made within the specified range, taking into account the signal-to-noise ratio and channel conditions of the actual underwater acoustic environment.

[0088] The parameters obtained in this step , and Used for channel prediction of the next symbol block.

[0089] Channel prediction

[0090] After obtaining the relevant channel parameters according to formulas (10), (11), and (13), the channel characteristics are combined to determine the next symbol block. The channel information is used to predict the channel gain and phase, which are:

[0091] (14)

[0092] (15)

[0093] The predicted channel for the next symbol block is:

[0094] (16)

[0095] Channel update

[0096] Update the currently processed symbol index to ( ).

[0097] because OFDM symbol blocks are not allocated pilot resources. The channel response predicted by formula (16) is used to perform OFDM signal detection in the current symbol block to obtain the transmitted bit stream. Then, the detected bit stream data is used as known pilots to perform channel estimation based on data feedback and update the channel response of the current symbol. .

[0098] when At this time, no pilot is assigned, and the channel frequency response is achieved through... The channel prediction steps for each OFDM symbol block are obtained.

[0099] Finally, the channel tracking, channel prediction, and channel update steps are repeated until all symbols in the frame have been processed.

[0100] Compared to conventional symbol-independent channel estimation methods that configure pilots in each OFDM symbol, the channel estimation method proposed in this application achieves comparable reception processing performance, as shown in the following figures. Figure 1 As shown. By Figure 1 As can be seen, the channel estimation method proposed in this application has slightly better performance than conventional channel estimation methods when the signal-to-noise ratio is 3-9 dB. This is because in symbol-independent data feedback channel estimation algorithms, the initial channel estimation for each symbol uses the channel value based on the original pilot, while the initial channel estimation value of the algorithm in this application for OFDM symbols after the third symbol is based on the channel estimation value updated by the previous symbol. Therefore, the initial channel estimation value of the algorithm in this application is more accurate. Figure 2 For the corresponding Figure 1 When the signal-to-noise ratio is 6dB, the data subcarrier constellation diagram after initial channel equalization shows that the constellation points corresponding to the method proposed in this application are more concentrated and closer to the standard constellation points. This proves that the algorithm of this application can provide more accurate channel feedback results, thereby obtaining more accurate channel update results and ultimately achieving a lower bit error rate.

[0101] Figure 2 and Figure 3 For the corresponding Figure 1 The data subcarrier constellation diagram after initial channel equalization when the signal-to-noise ratio is 6dB. Figure 2 and Figure 3 As can be seen, the constellation point distribution corresponding to the method proposed in this application is more concentrated and closer to the standard constellation points. This result shows that the algorithm proposed in this application can obtain more accurate data detection results, and on this basis, form more reliable data feedback channel estimation results, thereby further improving the accuracy of channel update results and ultimately achieving a lower bit error rate.

[0102] On the other hand, the algorithm proposed in this application can achieve a higher communication rate. Definition The signal length of the underwater acoustic OFDM system is [value], and the system communication bandwidth is [value]. =6kHz, each symbol contains =1024 subcarriers, the number of pilot subcarriers per symbol is =256, data subcarriers use For QPSK modulation with a value of 4, the channel coding code rate is... =0.5, the formula for calculating the communication rate is:

[0103] (17)

[0104] Based on the above parameters and formulas, the algorithm proposed in this application has a lower pilot ratio and a higher communication rate compared with conventional symbol-independent channel estimation algorithms, as shown in Table 1:

[0105] Table 1. Parameters related to communication rate of MIMO system

[0106]

[0107] It should be noted that the specific embodiments are merely explanations and illustrations of the technical solution of the present invention and should not be used to limit the scope of protection. Any modifications made in accordance with the claims and specification of the present invention that are only partial should still fall within the protection scope of the present invention.

Claims

1. An OFDM low-pilot channel estimation method based on the coherence of underwater acoustic slow time-varying channels, characterized in that, include: Step 1: Acquire each frame of signal transmitted by the underwater acoustic communication system; Step 2: For the first OFDM symbol in each frame of signal, i.e. =1, allocate pilot signals to obtain the fuzzy channel. and utilize Perform signal detection to obtain data symbols Then utilize Feedback updates yield accurate channel information. Finally, set the channel phase. =0, and set the channel gain to 0. = ; Step 3: For the second OFDM symbol in each frame of signal, i.e. =2, repeat step 2 to obtain the accurate channel. ; Step 4: Calculate the channel phase shift change based on the precise channel of the current symbol and the precise channel of the previous symbol. ; Step 5: Based on the phase shift changes of adjacent symbol channels Calculate the residual Doppler factor ; Step 6: Based on residual Doppler factors Update the channel phase of the current symbol using the channel phase of the previous symbol. ; Step 7: Channel phase based on the current symbol And the precise channel, and the channel gain of the previous symbol, update the channel gain of the current symbol. ; Step 8: Determine if the current symbol is the last symbol. If yes, end the process and output the precise channel, channel gain, and channel phase of the current symbol. Otherwise, proceed to step 9. Step 9: Predict the channel gain estimate and channel phase estimate of the next symbol, and obtain the fuzzy channel of the next symbol based on the channel gain estimate and channel phase estimate, and perform signal detection to obtain the accurate channel of the next symbol. Finally, use the accurate channel as the accurate channel of the current symbol in step 4, and repeat steps 4 to 8.

2. The OFDM low pilot channel estimation method based on the coherence of underwater acoustic slow time-varying channels according to claim 1, characterized in that, The channel phase shift change Represented as: ， in, Indicates frequency, This represents the residual Doppler factor after Doppler compensation. Indicates the symbol length of CP-OFDM.

3. The OFDM low pilot channel estimation method based on the coherence of underwater acoustic slow time-varying channels according to claim 2, characterized in that, The residual Doppler factor after Doppler compensation Represented as: ， in, The Doppler factor representing the path. This represents the channel Doppler factor estimated by the receiver.

4. The OFDM low pilot channel estimation method based on the coherence of underwater acoustic slow time-varying channels according to claim 3, characterized in that, The residual Doppler factor Represented as: ， in, Indicates a pair of elements Phase operation, Indicates the total number of OFDM subcarriers. It represents the imaginary unit.

5. The OFDM low pilot channel estimation method based on the coherence of underwater acoustic slow time-varying channels according to claim 4, characterized in that, The channel phase Represented as: ， in, The channel phase corresponding to the current OFDM symbol. This is the channel phase corresponding to the previous OFDM symbol.

6. The OFDM low pilot channel estimation method based on the coherence of underwater acoustic slow time-varying channels according to claim 5, characterized in that, The channel gain Represented as: ， in, This represents the set proportionality coefficient. Indicates the channel corresponding to the current symbol.

7. The OFDM low pilot channel estimation method based on the coherence of underwater acoustic slow time-varying channels according to claim 6, characterized in that, The channel gain estimate for the next OFDM symbol in step 9 is expressed as follows: ， in, This represents the estimated channel gain for the next OFDM symbol.

8. The OFDM low pilot channel estimation method based on the coherence of underwater acoustic slow time-varying channels according to claim 7, characterized in that, The channel phase estimate for the next OFDM symbol in step 9 is expressed as follows: ， in, This represents the channel phase estimate for the next OFDM symbol.

9. The OFDM low pilot channel estimation method based on the coherence of underwater acoustic slow time-varying channels according to claim 8, characterized in that, The fuzzy channel representation of the next OFDM symbol in step 9 is as follows: 。 10. The OFDM low pilot channel estimation method based on the coherence of underwater acoustic slow time-varying channels according to claim 9, characterized in that, The signal frame transmitted by the underwater acoustic communication system obtained in step 1 is the compensated signal. Represented as: ， in, This represents the channel Doppler factor estimated by the receiver. This represents the OFDM passband time domain signal arriving at the receiving hydrophone. This represents a continuous time-domain time variable.

Images