Self-adaptive modulation and coding method for fast convergence of ad hoc network of unmanned ship

Abstract

The invention provides a self-adaptive modulation and coding method for rapid convergence of an ad hoc network of an unmanned ship, which comprises the following steps of: S1, measuring channel quality, acquiring a channel condition, and acquiring HARQ / CRC (Hybrid Automatic Repeat reQuest / Cyclic Redundancy Check) feedback according to a current MCS (Modulation and Coding Scheme); s2, performing an inner loop link adaptive algorithm by combining a dynamic linear filtering mechanism based on a greedy algorithm according to the channel condition to obtain an illaSINR value; and S3, carrying out an outer loop link adaptive algorithm based on HARQ / CRC feedback and the target BLER, and obtaining an ollaSINR value. And S4, integrating the illaSINR value and the ollaSINR value, obtaining a decision MCS (Modulation and Coding Scheme), and carrying out adaptive modulation coding based on the decision MCS. According to the invention, rapid convergence and stable operation of the unmanned ship ad hoc network communication system are realized.

Description

Technical Field

[0001] The present invention relates to the technical field of wireless communication technology, and in particular to an adaptive modulation and coding method for rapid convergence of an unmanned ship self-organizing network. Background Art

[0002] During the unmanned ship self-organizing network mission, the complexity of the marine environment (such as water surface fluctuations, signal interference, and obstacle obstruction) causes the wireless channel conditions to change continuously and dynamically. The existing adaptive modulation and coding (AMC) technology is difficult to quickly and accurately adjust the modulation and coding strategy when facing rapidly changing channel conditions, resulting in problems such as low communication efficiency and link instability. Therefore, there is a need for an adaptive modulation and coding method that can converge quickly and is stable and reliable.

[0003] With the development of wireless communication technology, especially the promotion of 5G and future 6G networks, higher requirements are placed on the throughput, latency, coverage and spectrum efficiency of wireless communication systems. Adaptive modulation and coding (AMC), as a key technology to improve system transmission efficiency, can effectively optimize spectrum utilization and system performance by dynamically selecting appropriate modulation and coding methods according to channel quality. However, existing AMC systems face many challenges, especially in high mobility, large-scale MIMO systems and multi-user environments. How to efficiently process channel state information (CSI) and make decisions in real time is still a difficult problem.

[0004] Traditional AMC methods rely on the inner loop algorithm (ILLA) to dynamically adjust the modulation and coding scheme according to channel feedback information. However, since the inner loop algorithm is usually based on the fast response of local channel information, its optimization goal is limited to improving local performance, and the optimization of global performance is often ignored. In addition, the outer loop algorithm (OLLA) is usually responsible for global optimization, and its adjustment cycle is long, resulting in large feedback delays and an inability to quickly respond to channel changes. Therefore, how to achieve effective collaborative optimization between the two to simultaneously improve local and global performance has become an important research topic in the field of wireless communications.

[0005] The invention patent with the publication number CN114222328A discloses a wireless link adaptation method, including the following steps: setting of adjustment step group: according to different target block error rates BLER (Block Error Rate), setting a group of outer loop link adaptive OLLA adjustment step corresponding to different modulation and coding strategy MCS adjustment speeds; system step initialization: selecting an adjustment step as the default step and initializing the system step; actual block error rate statistics: statistical actual block error rate BLER within the time window; step adjustment: when the actual block error rate BLER exceeds the adjustment threshold, the system step is changed to the outer loop link adaptive OLLA adjustment step corresponding to the actual block error rate BLER. This invention can avoid problems such as rate drop and / or BLER jitter. However, this method cannot improve local and global performance at the same time. Summary of the invention

[0006] In view of the technical problem that the existing technology cannot improve local and global performance at the same time, the present invention proposes an adaptive modulation and coding method for rapid convergence of unmanned ship self-organizing network. The adaptive modulation and coding algorithm (AMC) of the present invention contains two core components: the inner loop link adaptation (ILLA) algorithm and the outer loop link adaptation (OLLA) algorithm, which work together to achieve rapid convergence and optimize the modulation and coding strategy. Among them, the inner loop link adaptation (ILLA) adopts a dynamic filtering mechanism based on a greedy strategy, and adjusts the filter factor in real time to adapt to channel changes to combat large-scale fading, effectively solving the performance problem of the traditional fixed filter coefficient solution in complex channel environments; the outer loop link adaptation (OLLA) algorithm adjusts the demodulation performance of the uplink and downlink at different layers based on HARQ (Hybrid Automatic Repeat Request) feedback and CRC feedback to combat small-scale fading. This design makes the modulation and coding adjustment both fast and accurate, so as to better cope with channel changes in complex marine environments.

[0007] In order to achieve the above object, the technical solution of the present invention is achieved as follows:

[0008] An adaptive modulation and coding method for rapid convergence of unmanned ship self-organizing network, comprising the steps of:

[0009] S1: Perform channel quality measurement, obtain channel status, and obtain HARQ / CRC feedback based on the current MCS;

[0010] S2: According to the channel conditions, the inner loop link adaptation algorithm is combined with the dynamic linear filtering mechanism based on the greedy algorithm to obtain the illaSINR value;

[0011] S3: Perform outer loop link adaptation algorithm based on HARQ / CRC feedback and target BLER to obtain ollaSINR value;

[0012] S4: The illaSINR value and the ollaSINR value are integrated to obtain a decision MCS, and adaptive modulation and coding is performed based on the decision MCS.

[0013] The channel quality measurement includes uplink channel measurement and downlink channel measurement; the uplink channel measurement measures the sounding reference signal SRS through the physical layer in the base station, calculates the uplink channel SINR value, and the physical layer transmits the uplink channel SINR value to the AMC algorithm module of the MAC layer; the downlink channel measurement measures the reference signal RS through the neighboring node of the user node to obtain the SINR value, maps the SINR value into a channel quality indication CQI through the uplink control information UCI in the uplink feedback channel PUSCH and feeds it back to the physical layer in the base station, and the physical layer transmits the downlink channel quality indication CQI to the AMC algorithm module of the MAC layer.

[0014] The specific process of the inner loop link adaptation algorithm is as follows: mapping the downlink channel quality indication CQI obtained from the physical layer to a downlink channel SINR value through a CQI-SINR mapping table, and obtaining an illaSINR value from the downlink channel SINR value or the uplink channel SINR value through a dynamic linear filtering mechanism based on a greedy algorithm;

[0015] The specific process of the outer loop link adaptation algorithm is: combining the CRC feedback of the uplink and the BLER target of different services or the HARQ feedback of the downlink and the target BLER of different services, dynamically adjusting the offset ollaSINR value relative to the inner loop illaSINR value based on the step size parameter and the target BLER dynamic matching algorithm.

[0016] The implementation method of the dynamic linear filtering mechanism based on the greedy algorithm is:

[0017] Step 1: Obtain channel measurement data, including downlink channel SINR value and uplink channel SINR value;

[0018] Step 2: Check the preheat counter; set the threshold value of the statistical number, use the preheat counter to count the number of times the inner loop link adaptive algorithm is performed; determine whether to perform filtering based on the threshold value of the statistical number; if the preheat counter statistical number is less than the threshold value of the statistical number, skip filtering, otherwise proceed to Step 3;

[0019] Step 3: Adaptive dynamic filter factor adjustment based on greedy algorithm and linear filtering algorithm;

[0020] Step 4: Perform a linear filtering algorithm based on the filter factor adjusted in step 3 to obtain an updated illaSINR value;

[0021] Step 5: Inner loop decision information update: record the updated illaSINR value.

[0022] The implementation method based on the greedy algorithm and the linear filtering algorithm is:

[0023] Step 3.1: Definition of performance index: SINR value is used as performance index and SINR value is optimized as target;

[0024] Step 3.2: Local optimal selection: Select the optimal filter factor based on the current SINR value and the historical illaSINR value;

[0025] Step 3.3: Update the filter factor: store the optimal filter factor through the wegtFltrIlla function.

[0026] The method for selecting the optimal filter factor described in step 3.2 is:

[0027] Step 3.2.1: For each iteration period t, calculate the current SINR trend change ΔS based on the current SINR value and the historical illaSINR value;

[0028] Step 3.2.2: Determine the search direction based on the current SINR trend change ΔS;

[0029] Step 3.2.3: Select the optimal filter factor x in the search space S * :

[0030] The method for performing linear filtering described in step 4 is: combining the historical illaSINR value, the current SINR value, and the optimal filtering factor to update the illaSINR value, and the update formula is:

[0031] illaSINR(t)=(1-wegtFltrIlla(t))*memIllaSinrDb+wegtFltrIlla(t)*SINR(t)

[0032] Among them, illaSINR(t) is the updated illaSINR value, memIllaSinrDb represents the historical illaSINR value, SINR(t) is the current SINR value, and wegtFltrIlla(t) represents the optimal filter factor x * .

[0033] The steps of the dynamic matching algorithm based on step size parameter and target BLER are as follows:

[0034] ST1: Initialization parameters: define and initialize the parameters required by the OLLA algorithm, including increasing step size and decreasing step size;

[0035] ST2: Receive HARQ / CRC feedback: Determine the current MCS transmission effect based on the uplink CRC feedback or downlink HARQ feedback, and decide whether to adjust the ollaSINR value;

[0036] ST3: Adjust the step size up and down based on the HARQ / CRC feedback obtained from the historical MCS and the current MCS;

[0037] ST4: Perform demodulation performance feedback processing based on the adjusted boost step size and reduction step size to obtain the final ollaSINR value.

[0038] The method for adjusting the step size increase and the step size decrease is as follows: according to different target BLERs, a multi-level step size parameter strategy is obtained using the relationship between the target BLER and the step size increase and the step size decrease; the step size parameter strategy includes obtaining the step size increase ollaSinrUpStep and the step size decrease ollaSinrDnStep that meet the target BLER; adjusting the step size decrease ollaSinrDnStep that meets the target BLER based on the NACK number limit convergence algorithm to obtain the step size decrease ollaSinrDnStep (X), and finally obtaining the step size increase ollaSinrUpStep and the step size decrease ollaSinrDnStep (X), where X represents the step size multiples corresponding to different NACK numbers.

[0039] The method for performing demodulation performance feedback processing is:

[0040] When ACK feedback in the uplink CRC feedback or ACK feedback in the downlink HARQ feedback is received, ACK feedback processing is performed: the ollaSINR value is increased by increasing the step size ollaSinrUpStep; when NACK feedback in the uplink CRC feedback or NACK feedback in the downlink HARQ feedback is received, NACK feedback processing is performed: the ollaSINR value is reduced by reducing the step size ollaSinrDnStep(X); wherein when multiple consecutive uplink NACK feedbacks or downlink NACK feedbacks are received, the ollaSINR value is adjusted through the fast decrease and slow increase mechanism when performing demodulation performance feedback processing.

[0041] The relationship between the target BLER and the step size increase and decrease is:

[0042]

[0043] The NACK number limit convergence algorithm is: when multiple consecutive NACK feedbacks in uplink CRC feedbacks or NACK feedbacks in downlink HARQ feedbacks are received, the convergence is accelerated by increasing the reduction step size ollaSinrDnStep, and the formula is:

[0044] ollaSinrDnStep(X)=ollaSinrDnStep*X

[0045] The formula for the ACK feedback processing is:

[0046] ollaSINR(t)=min(ollaSINR(t-1)+ollaSinrUpStep(bler),OllaUpperThrd);

[0047] The formula for NACK feedback processing is:

[0048] ollaSINR(t)=max(ollaSINR(t-1)-ollaSinrDnStep(bler),OllaLowerThrd);

[0049] Among them, OllaUpperThrd is the maximum value of the step length, and OllaLowerThrd is the minimum value of the step length.

[0050] Beneficial effects of the present invention:

[0051] Significantly improved channel adaptability: The dynamic filtering mechanism optimized by the greedy algorithm improves the response speed to channel changes; the multi-level step adjustment strategy of fast decrease and slow increase ensures the optimal convergence performance under different channel conditions;

[0052] More accurate service quality assurance: Differentiated BLER target settings meet different service requirements; NACK fast convergence mechanism improves system reliability in poor channels;

[0053] Comprehensive optimization of system performance: improved communication efficiency: quick response to channel changes; enhanced link stability: avoid rate jitter problems; improved resource utilization: accurate MCS selection strategy;

[0054] In summary, the present invention innovatively introduces the greedy algorithm into the inner loop dynamic filtering mechanism and designs a multi-level step adjustment strategy to optimize the outer loop adaptability, thereby achieving rapid convergence and stable operation of the unmanned ship self-organizing network communication system. The present invention provides an effective technical solution for the communication of unmanned ship self-organizing networks in complex marine environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0056] Figure 1 It is a schematic diagram of the process of the present invention.

[0057] Figure 2 This is the flow chart of the dynamic linear filtering mechanism based on the greedy algorithm.

[0058] Figure 3 FIG. 4 is a schematic diagram of adjusting the ollaSINR value of the present invention.

[0059] Figure 4 This is a flowchart of the inner loop link adaptation process of the present invention.

[0060] Figure 5 This is a flowchart of the outer ring link adaptation process of the present invention. DETAILED DESCRIPTION

[0061] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0062] Example 1

[0063] An adaptive modulation and coding method for fast convergence of unmanned ship self-organizing network, such as Figure 1 As shown, the steps include:

[0064] S1: Perform channel quality measurement, obtain channel status, and obtain HARQ / CRC feedback based on the current MCS.

[0065] The quality measurement includes uplink channel measurement and downlink channel measurement.

[0066] The method for uplink channel measurement is as follows: Sounding Reference Signals (SRS) are periodically sent by user equipment (such as unmanned ships) to the base station to estimate the uplink channel quality; the physical layer (PHY) in the base station measures the sounding reference signal SRS and calculates the uplink channel signal-to-noise ratio (SINR), and the physical layer transmits the uplink channel SINR value to the adaptive modulation and coding (AMC) algorithm module of the control layer (MAC).

[0067] The method for downlink channel measurement is as follows: the reference signal (RS) is sent by the base station to estimate the downlink channel quality; the neighboring node of the user node measures the reference signal RS to obtain the SINR value, and maps the SINR value into a channel quality indication (CQI) through the uplink control information (UCI) in the uplink feedback channel PUSCH and feeds it back to the physical layer in the base station, and the physical layer transmits the downlink channel quality indication (CQI) to the AMC algorithm module of the MAC layer.

[0068] The SINR value obtained by channel quality measurement only reflects the interference of the wireless channel, and cannot fully reflect the demodulation performance of the receiving end in the real environment. In this way, the AMC algorithm needs to determine the basic MCS according to the current channel conditions of the uplink and downlink channels, and then adjust the MCS accordingly based on the demodulation BLER on the physical channel. Therefore, the ACK / NACK feedback of HARQ is required as another important input of the AMC algorithm module.

[0069] The MCS obtained in the previous adaptive modulation and coding process is equivalent to the current MCS before the current adaptive modulation and coding process is completed. The HARQ / CRC feedback is obtained according to the current MCS:

[0070] For the downlink, the AMC algorithm selects an MCS for data transmission in the previous adaptive modulation and coding process, and sends the MCS to the neighbor nodes of the user node through the downlink control information (DCI) in the downlink control channel PDSCH; the base station sends data packets to the neighbor nodes of the user node based on the selected MCS, and the neighbor nodes of the user node decode the received data packets and calculate the cyclic redundancy check (CRC) to verify the data integrity and perform HARQ feedback. If the CRC check succeeds, the neighbor nodes of the user node send ACK feedback to the physical layer of the base station, indicating that the data has been successfully received; if the CRC check fails, the neighbor nodes of the user node send NACK feedback to the physical layer of the base station, requesting to resend the data packet.

[0071] For the uplink, the neighboring nodes of the user node send data packets to the base station according to the MCS. The physical layer of the base station decodes the received data packets and calculates the cyclic redundancy check (CRC) to verify the data integrity, generates CRC feedback, and feeds back CRC correctness or CRC error to the MAC layer. The MAC layer generates ACK feedback or NACK feedback.

[0072] S2: According to the channel conditions, the inner loop link adaptation (ILLA) algorithm is performed in combination with a dynamic linear filtering mechanism based on a greedy algorithm to obtain an illaSINR value.

[0073] like Figure 4 As shown, the specific process of the inner loop link adaptation algorithm is: the downlink channel quality indication (CQI) obtained from the physical layer is mapped to a downlink channel SINR value through a CQI-SINR mapping table, and the downlink channel SINR value or the uplink channel SINR value is obtained by a dynamic linear filtering mechanism based on a greedy algorithm to obtain a stable SINR, which is recorded as illaSINR, which is used to reflect the basic interference situation of the downlink channel or the uplink channel.

[0074] The CQI-SINR mapping is obtained by link simulation, specifically: defining the target BLER; selecting MCS; determining the frequency point and channel model related to the mapping relationship, and the channel model can adopt Gaussian, indoor, dense urban area, urban area, and rural model; simulating data transmission conditions under different SINR levels through simulation tools; based on the simulation results, creating a CQI-SINR mapping table for the selected MCS.

[0075] The illaSINR value output by the ILLA algorithm should reflect the basic interference situation of the current channel, which is mostly the change of large-scale fading, so the illaSINR value should be relatively stable. Therefore, it is necessary to filter the downlink channel SINR value or the uplink channel SINR value.

[0076] The present invention adopts a dynamic linear filtering mechanism based on a greedy algorithm, such as Figure 2 As shown, the specific method is:

[0077] Step 1: Obtain channel measurement data, including downlink channel SINR value and uplink channel SINR value.

[0078] Step 2: Check the warm-up counter. Use the warm-up counter to count the number of times the inner-loop link adaptation algorithm is performed to control the initial state of the algorithm and ensure that no over-adjustment is performed before the algorithm stabilizes.

[0079] The specific steps are as follows: setting the threshold value SCH_AMC_ILLA_WARMUP_NUM for the statistical times; if the statistical times IllaWarmUpCntr of the warm-up counter is less than the threshold value SCH_AMC_ILLA_WARMUP_NUM for the statistical times, skipping the filtering adjustment step to avoid overreaction in the algorithm initialization phase, that is, the algorithm adjusts its parameters too early or too aggressively without fully understanding the current channel characteristics, resulting in inaccurate or unnecessary adjustments; if the statistical times IllaWarmUpCntr of the warm-up counter is greater than or equal to the threshold value SCH_AMC_ILLA_WARMUP_NUM for the statistical times, proceeding to step 3.

[0080] Step 3: Adaptive dynamic filter factor adjustment based on greedy algorithm and linear filtering algorithm.

[0081] According to the trend of the historical illaSINR value and the current SINR value, the filter factor is dynamically adjusted - the filter factor to be enlarged or reduced is selected, that is, a filter factor that can maximize or minimize the SINR value is selected in each iteration. In order to achieve a more accurate filtering effect, the current optimal decision is selected at each step to achieve a global optimization effect. Each dynamic linear filtering based on the greedy algorithm is an iterative process.

[0082] The specific steps are:

[0083] Step 3.1: Definition of performance index: SINR value is used as performance index and SINR value is the optimization target

[0084] The performance indicators are defined as:

[0085]

[0086] Among them, F(x) represents the performance index, x is the value of the filtering factor, and output represents the performance index value after filtering.

[0087] Step 3.2: Local optimal selection: In each iteration, a filter factor that can maximize the performance index is selected based on the current SINR value and the historical SINR value;

[0088] The specific process is:

[0089] Step 3.2.1: For each iteration period t, calculate the current SINR trend change based on the current SINR value and the historical illaSINR value:

[0090] ΔS=SINR(t)-memIllaSinrDb

[0091] Wherein, SINR(t) is the current SINR value, memIllaSinrDb is the historical illaSINR value;

[0092] Step 3.2.2: Determine the search direction based on the current SINR trend change ΔS;

[0093] Step 3.2.3: Select the optimal filter factor in the search space S:

[0094] x * = argmax(F(x)),x∈S

[0095] Among them, x * is the optimal filtering factor, argmax() represents the input parameter that makes the function reach the maximum value;

[0096] In order to avoid falling into local extreme values, restrictions are added to the local optimal selection of the greedy algorithm. By setting the upper and lower limits of the filter factor, it is ensured that the selected filter factor does not deviate too much from the reasonable range of the historical value or the current channel environment.

[0097] Step 3.3: Update the filter factor: If the performance is improved, store the optimal filter factor through the wegtFltrIlla function; otherwise, keep the filter factor obtained in the previous iteration for use in the next iteration; the formula is:

[0098]

[0099] Step 4: Linear filtering algorithm: Combine the historical illaSINR value memIllaSinrDb, the current SINR value SINR(t), and the selected filtering factor wegtFltrIlla(t) to calculate and update the illaSINR value;

[0100] The update formula is:

[0101] illaSINR(t)=(1-wegtFltrIlla(t))*memIllaSinrDb+wegtFltrIlla(t)*SINR(t)

[0102] Wherein, illaSINR(t) is the updated inner loop illaSINR value.

[0103] Step 5: Inner loop decision information update: record the illaSINR(t) value and store the illaSINR(t) value in memIllaSinrDb as the historical illaSINR value for the next SINR update and modulation coding decision.

[0104] S3: Perform an outer loop link adaptation algorithm based on HARQ / CRC feedback and target BLER to obtain the ollaSINR value.

[0105] like Figure 5 As shown, the specific process of the outer loop link adaptation algorithm is: combining the CRC feedback of the uplink and the BLER target of different services or the HARQ feedback of the downlink and the target BLER of different services, dynamically adjusting the offset ollaSINR relative to the inner loop illaSINR based on the step size parameter and the target BLER dynamic matching algorithm.

[0106] The steps of the dynamic matching algorithm based on step size parameters and target BLER are:

[0107] ST1: Initialization parameters: Define and initialize the parameters required by the OLLA algorithm, including the step size increase and step size decrease, which are used to adjust the OLLA SINR according to the target BLER;

[0108] ST2: Receive HARQ / CRC feedback: Determine the current MCS transmission effect based on the uplink CRC feedback or downlink HARQ feedback, and decide whether to adjust the ollaSINR.

[0109] The AMC algorithm counts HARQ / CRC feedback information in each time window. When the number of statistical windows is greater than or equal to the set threshold, the current MCS transmission effect is judged. When the CRC is correct, the step size adjustment of step ST3 is performed.

[0110] ST3: Based on the historical MCS and the HARQ / CRC feedback obtained based on the current MCS, predict the channel deterioration and recovery, adjust the step size in advance, and speed up the convergence.

[0111] When adjusting the step size increase and decrease, different step size adjustment strategies are set according to different target BLERs. The setting of target BLER is adjusted according to different business requirements. For example, real-time communication (such as voice and video) requires a lower BLER (such as 10%) to ensure reliability; while for services with higher fault tolerance (such as file transfer), a higher BLER (such as 20%) can be set to improve throughput.

[0112] The method for adjusting the step size increase and step size decrease is: according to different target BLERs, a multi-level step size parameter strategy is obtained by using the relationship between the target BLER and the step size increase and step size decrease; the step size parameter strategy includes obtaining the step size increase ollaSinrUpStep and the step size decrease ollaSinrDnStep that meet the target BLER.

[0113] The relationship between the target BLER and the step size of increase and decrease is:

[0114]

[0115] Based on the NACK number limit convergence algorithm, the reduction step size ollaSinrUpStep that meets the target BLER is adjusted to obtain the reduction step size ollaSinrDnStep(X), and finally the increase step size ollaSinrUpStep and the reduction step size ollaSinrDnStep(X) are obtained.

[0116] The NACK number limit convergence algorithm is: when multiple consecutive uplink NACK feedbacks or downlink NACK feedbacks are received, the convergence is accelerated by increasing the ollaSinrDnStep; the formula is:

[0117] ollaSinrDnStep(X)=ollaSinrDnStep*X

[0118] Where X is the step size multiple corresponding to different NACK times, X = [2, 4, 6, 16]. The NACK number limit convergence algorithm can accelerate convergence under harsh channel conditions. When NACK is 1, the NACK number limit convergence algorithm is not used. In order to facilitate the calculation and representation described later, when NACK is 1, the step size is reduced to ollaSinrDnStep(X), where X = 1.

[0119] ST4: Obtain the adjusted increase step size ollaSinrUpStep and decrease step size ollaSinrDnStep(X) that meet the target BLER; perform demodulation performance feedback processing to obtain the final ollaSINR value:

[0120] The method for performing demodulation performance feedback processing is as follows: when receiving ACK feedback in the uplink CRC feedback or ACK feedback in the downlink HARQ feedback, it indicates that the data can be correctly received according to the current MCS transmission data, so the ollaSINR is tried to be increased, and ACK feedback processing is performed: the ollaSINR is increased by increasing the step size ollaSinrUpStep; when receiving NACK feedback in the uplink CRC feedback or NACK feedback in the downlink HARQ feedback, it indicates that the data cannot be correctly received according to the current MCS transmission data, so the ollaSINR is tried to be reduced, and NACK feedback processing is performed: according to the historical MCS, it is judged that the ollaSINR is reduced by reducing the step size ollaSinrDnStep(X). When multiple consecutive uplink NACK feedbacks or downlink NACK feedbacks are received, the ollaSinrDnStep(X) value increases due to the NACK number limit convergence algorithm. When performing demodulation performance feedback processing, the ollaSINR value is adjusted through a fast reduction and slow increase mechanism, such as Figure 3 shown.

[0121] ACK feedback processing: If the transmission is successful (ACK), the ollaSINR is increased with a step size of ollaSinrUpStep; the formula is:

[0122] ollaSINR(t)=min(ollaSINR(t-1)+ollaSinrUpStep,OllaUpperThrd);

[0123] NACK feedback processing: If the transmission fails (NACK), the ollaSINR is reduced with a step size of ollaSinrDnStep(X):

[0124] ollaSINR(t)=max(ollaSINR(t-1)-ollaSinrDnStep(X),OllaLowerThrd);

[0125] Among them, OllaUpperThrd is the maximum value of ollaSINR, and OllaLowerThrd is the minimum value of ollaSINR.

[0126] S4: The illaSINR value and the ollaSINR value are integrated to obtain a decision MCS, and adaptive modulation and coding is performed based on the decision MCS.

[0127] Calculate the adaptive modulation coding SINR value AMCSINR=illaSINR+ollaSINR, map the AMCSINR value to the recommended MCS according to the SINR-MCS mapping table, determine the level of the recommended MCS and the system configuration SchMCS, the system configuration SchMCS includes the uplink SchULMcs and the downlink SchDLMcs, select the lower-level MCS as the decision MCS, and pass the MSC information to the physical layer through the transmission control indication TCI, and perform downlink encoding and sending data and uplink decoding and receiving data through the physical layer.

[0128] Example 2

[0129] An adaptive modulation and coding method for rapid convergence of unmanned ship self-organizing network, comprising the steps of:

[0130] S1: Perform channel quality measurement, obtain channel status, and obtain HARQ / CRC feedback based on the current MCS.

[0131] Channel quality measurement requires initialization and parameter configuration; after the user (unmanned ship) accesses the unmanned ship self-organizing network, first configure the system default parameters for the user node, and then configure the inner ring link adaptation parameters and outer ring link adaptation parameters for the user;

[0132] As shown in Table 1, the system default parameters include the default uplink modulation and coding scheme (MCS) level (SchULMcs) supported by the system, the default downlink MCS level (SchDLMcs) supported by the system; target block error rate (TarBler, TargetBLER); Olla adjustment control switch (AMCOllaSwitch), Illa adjustment control switch (AMCIllaSwitch); uplink modulation and demodulation mapping table (UlModulation), downlink modulation and demodulation mapping table (DlModulation).

[0133] Table 1. System default parameter configuration

[0134]

[0135] As shown in Table 2, the inner loop link adaptation parameters include the Illa filter coefficient (IllaAlpha) and the Illa conservative SINR (IllaMinSinr).

[0136] Table 2. Inner ring link adaptation configuration parameters

[0137]

[0138] As shown in Table 3, the outer loop link adaptation parameters include Olla ACK adjustment step (ollaSinrUpStep), OllaNACK adjustment step (ollaSinrDownStep); Olla maximum upper limit value (OllaUpperThrd), Olla minimum lower limit value (OllaLowerThrd).

[0139] Table 3. Outer ring link adaptation configuration parameters

[0140]

[0141]

[0142] S2: According to the channel conditions, the inner loop link adaptation (ILLA) algorithm is performed in combination with a dynamic linear filtering mechanism based on a greedy algorithm to obtain an illaSINR value.

[0143] The specific process of the inner loop link adaptation algorithm is as follows: the downlink channel quality indication (CQI) obtained from the physical layer is mapped to a downlink channel SINR value through a CQI-SINR mapping table, and the downlink channel SINR value or the uplink channel SINR value is obtained by a dynamic linear filtering mechanism based on a greedy algorithm to obtain a stable SINR, which is recorded as illaSINR, to reflect the basic interference situation of the downlink channel or the uplink channel.

[0144] The present invention adopts a dynamic linear filtering mechanism based on a greedy algorithm, and the specific method is as follows:

[0145] Step 1: Obtain channel measurement data, including downlink channel SINR value and uplink channel SINR value.

[0146] Step 2: Perform a warm-up counter check; use the warm-up counter to control the initial state of the algorithm to ensure that no over-adjustment is performed before the algorithm stabilizes. The threshold value for the number of statistics is 16.

[0147] Step 3: Adaptive dynamic filter factor adjustment based on greedy algorithm and linear filtering algorithm.

[0148] Step 4: Linear filtering algorithm.

[0149] Step 5: Update the inner loop decision information.

[0150] S3: Perform an outer loop link adaptation algorithm based on HARQ / CRC feedback and target BLER to obtain the ollaSINR value.

[0151] The initial up-stepping step size ollaSinrUpStep is 1 / 128dB, and the initial down-stepping step size ollaSinrDnStep is {(1 / 128)*(1-0.1) / 0.1}dB;

[0152] S4: The illaSINR value and the ollaSINR value are integrated to obtain a decision MCS, and adaptive modulation and coding is performed based on the decision MCS.

[0153] The other methods are the same as those in Example 1.

[0154] Example 3

[0155] An adaptive modulation and coding method for rapid convergence of unmanned ship self-organizing network, comprising the steps of:

[0156] S1: Perform channel quality measurement, obtain channel status, and obtain HARQ / CRC feedback based on the current MCS.

[0157] S2: According to the channel conditions, the inner loop link adaptation (ILLA) algorithm is performed in combination with a dynamic linear filtering mechanism based on a greedy algorithm to obtain an illaSINR value.

[0158] S3: Perform an outer loop link adaptation algorithm based on HARQ / CRC feedback and target BLER to obtain the ollaSINR value.

[0159] The following multi-step parameter strategy example is given:

[0160] Table 4. Step size parameter strategy table

[0161]

[0162]

[0163] Among them, MeasSinrDb is the current SINR value, MeasSinrThrd1, MeasSinrThrd2, MeasSinrThrd2+5, MeasSinrThrd1, and MeasSinrThrd2 are all Illa conservative SINRs, indicating the minimum SINR value for each adjustment.

[0164] As shown in Table 4. The adjustment step size is related to the demodulation SINR convergence speed: the larger the step size, the faster the demodulation SINR convergence speed. For example, if the target BLER is also 10%, configuration 1 is [ollaSinrUpStep, ollaSinrDnStep] = [X / 1024, 9X / 1024] dB, configuration 2 is [ollaSinrUpStep, ollaSinrDnStep] = [2X / 1024, 18X / 1024] dB, and configuration 3 is [ollaSinrUpStep, ollaSinrDnStep] = [4X / 1024, 36X / 1024] dB. The convergence speed of configuration 2 is twice that of configuration 1, and the convergence speed of configuration 3 is 4 times that of configuration 1. It is not the case that the larger the configuration step size, the better. When the user is in a good wireless environment, a larger step size will cause its MCS fluctuation range to become larger and the BLER will not be stable enough. When the UE is in a radio environment that changes suddenly from bad to good, a large step size will accelerate the convergence to the target BLER.

[0165] S4: The illaSINR value and the ollaSINR value are integrated to obtain a decision MCS, and adaptive modulation and coding is performed based on the decision MCS.

[0166] The other methods are the same as those in Example 2.

[0167] Dual-loop framework adjustment strategy: The present invention adopts an inner-outer-loop adjustment strategy. The inner loop is adjusted based on channel quality (SINR / CQI) to combat large-scale fading, and the outer loop is adjusted based on bit error rate (BLER) to combat small-scale fading. The present invention adopts a dual-loop framework of inner-loop link adaptation (ILLA) and outer-loop link adaptation (OLLA) to achieve fast and accurate modulation and coding adjustment.

[0168] Dynamic filtering mechanism: Different from the design of fixed forgetting factor of traditional smoothing filter, the present invention innovatively designs an adaptive dynamic filter factor adjustment mechanism based on the idea of ​​smoothing filter algorithm and greedy algorithm. While smoothing the channel quality measurement value, filtering the mutation signal to reduce the impact caused by instantaneous channel changes, and obtaining accurate filtering effect, a filter factor that can maximize or minimize a certain performance indicator (such as SINR value) is selected in each iteration. In order to achieve a more accurate filtering effect, the current optimal decision is selected at each step to achieve a global optimization effect.

[0169] Dynamic matching of step size parameters and target BLER: The present invention adopts a step size adjustment strategy based on target BLER. By adjusting the relationship between the step size parameters and the target BLER, precise control of the MCS adjustment speed is achieved. According to HARQ feedback and the target BLER requirements of different services, the step size of SINR offset is dynamically adjusted to quickly respond to channel changes and optimize MCS selection. By analyzing historical data and current HARQ feedback, channel deterioration and recovery are predicted, and the AMC OLLA adjustment step size is adjusted in advance to speed up convergence so that MCS selection can keep up with channel changes. Avoid rate drops and / or BLER jitter problems.

[0170] Adapting to the transmission quality requirements of different business needs: The present invention can set different target BLERs according to the different requirements of different business types for transmission quality. For services with high real-time requirements, a low BLER target is set to ensure high reliability of transmission; and for services with high tolerance, a high BLER target is set to optimize throughput and spectrum efficiency. It supports three BLER mode selections: voice, video, and file.

[0171] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. An adaptive modulation and coding method for rapid convergence of unmanned ship self-organizing network, characterized in that: Includes steps: S1: Perform channel quality measurement, obtain channel status, and obtain HARQ / CRC feedback based on the current MCS; S2: According to the channel conditions, the inner loop link adaptation algorithm is combined with the dynamic linear filtering mechanism based on the greedy algorithm to obtain the illaSINR value; S3: Perform outer loop link adaptation algorithm based on HARQ / CRC feedback and target BLER to obtain ollaSINR value; S4: The illaSINR value and the ollaSINR value are integrated to obtain a decision MCS, and adaptive modulation and coding is performed based on the decision MCS.

2. The adaptive modulation and coding method for rapid convergence of unmanned ship ad hoc network according to claim 1 is characterized in that: The channel quality measurement includes uplink channel measurement and downlink channel measurement; the uplink channel measurement measures the sounding reference signal SRS through the physical layer in the base station, calculates the uplink channel SINR value, and the physical layer transmits the uplink channel SINR value to the AMC algorithm module of the MAC layer; the downlink channel measurement measures the reference signal RS through the neighboring node of the user node to obtain the SINR value, maps the SINR value into a channel quality indication CQI through the uplink control information UCI in the uplink feedback channel PUSCH and feeds it back to the physical layer in the base station, and the physical layer transmits the downlink channel quality indication CQI to the AMC algorithm module of the MAC layer.

3. The adaptive modulation and coding method for rapid convergence of unmanned ship ad hoc network according to claim 2 is characterized in that: The specific process of the inner loop link adaptation algorithm is as follows: mapping the downlink channel quality indication CQI obtained from the physical layer to a downlink channel SINR value through a CQI-SINR mapping table, and obtaining an illaSINR value from the downlink channel SINR value or the uplink channel SINR value through a dynamic linear filtering mechanism based on a greedy algorithm; The specific process of the outer loop link adaptation algorithm is: combining the CRC feedback of the uplink and the BLER target of different services or the HARQ feedback of the downlink and the target BLER of different services, dynamically adjusting the offset ollaSINR value relative to the inner loop illaSINR value based on the step size parameter and the target BLER dynamic matching algorithm.

4. The adaptive modulation and coding method for rapid convergence of unmanned ship ad hoc network according to claim 3 is characterized in that: The implementation method of the dynamic linear filtering mechanism based on the greedy algorithm is: Step 1: Obtain channel measurement data, including downlink channel SINR value and uplink channel SINR value; Step 2: Check the preheat counter; set the threshold value of the statistical number, use the preheat counter to count the number of times the inner loop link adaptive algorithm is performed; determine whether to perform filtering based on the threshold value of the statistical number; if the preheat counter statistical number is less than the threshold value of the statistical number, skip filtering, otherwise proceed to Step 3; Step 3: Adaptive dynamic filter factor adjustment based on greedy algorithm and linear filtering algorithm; Step 4: Perform a linear filtering algorithm based on the filter factor adjusted in step 3 to obtain an updated illaSINR value; Step 5: Inner loop decision information update: record the updated illaSINR value.

5. The adaptive modulation and coding method for rapid convergence of unmanned ship ad hoc network according to claim 4 is characterized in that: The implementation method based on the greedy algorithm and the linear filtering algorithm is: Step 3.1: Definition of performance index: SINR value is used as performance index and SINR value is optimized as target; Step 3.2: Local optimal selection: Select the optimal filter factor based on the current SINR value and the historical illaSINR value; Step 3.3: Update the filter factor: store the optimal filter factor through the wegtFltrIlla function.

6. The adaptive modulation and coding method for rapid convergence of unmanned ship ad hoc network according to claim 5 is characterized in that: The method for selecting the optimal filter factor described in step 3.2 is: Step 3.2.1: For each iteration period t, calculate the current SINR trend change ΔS based on the current SINR value and the historical illaSINR value; Step 3.2.2: Determine the search direction based on the current SINR trend change ΔS; Step 3.2.3: Select the optimal filter factor x in the search space S * : The method for performing linear filtering described in step 4 is: combining the historical illaSINR value, the current SINR value, and the optimal filtering factor to update the illaSINR value, and the update formula is: illaSINR(t)=(1-wegtFltrIlla(t))*memIllaSinrDb+wegtFltrIlla(t)*SINR(t) Among them, illaSINR(t) is the updated illaSINR value, memIllaSinrDb represents the historical illaSINR value, SINR(t) is the current SINR value, and wegtFltrIlla(t) represents the optimal filter factor x * .

7. The adaptive modulation and coding method for rapid convergence of unmanned ship self-organizing network according to any one of claims 3 to 6, characterized in that: The steps of the dynamic matching algorithm based on step size parameter and target BLER are as follows: ST1: Initialization parameters: define and initialize the parameters required by the OLLA algorithm, including increasing step size and decreasing step size; ST2: Receive HARQ / CRC feedback: Determine the current MCS transmission effect based on the uplink CRC feedback or downlink HARQ feedback, and decide whether to adjust the ollaSINR value; ST3: Adjust the step size up and down based on the HARQ / CRC feedback obtained from the historical MCS and the current MCS; ST4: Perform demodulation performance feedback processing based on the adjusted boost step size and reduction step size to obtain the final ollaSINR value.

8. The adaptive modulation and coding method for rapid convergence of unmanned ship ad hoc network according to claim 7 is characterized in that: The method for adjusting the step size increase and the step size decrease is as follows: according to different target BLERs, a multi-level step size parameter strategy is obtained using the relationship between the target BLER and the step size increase and the step size decrease; the step size parameter strategy includes obtaining the step size increase ollaSinrUpStep and the step size decrease ollaSinrDnStep that meet the target BLER; adjusting the step size decrease ollaSinrDnStep that meets the target BLER based on the NACK number limit convergence algorithm to obtain the step size decrease ollaSinrDnStep (X), and finally obtaining the step size increase ollaSinrUpStep and the step size decrease ollaSinrDnStep (X), where X represents the step size multiples corresponding to different NACK numbers.

9. The adaptive modulation and coding method for rapid convergence of unmanned ship ad hoc network according to claim 8, characterized in that: The method for performing demodulation performance feedback processing is: When ACK feedback in the uplink CRC feedback or ACK feedback in the downlink HARQ feedback is received, ACK feedback processing is performed: the ollaSINR value is increased by increasing the step size ollaSinrUpStep; when NACK feedback in the uplink CRC feedback or NACK feedback in the downlink HARQ feedback is received, NACK feedback processing is performed: the ollaSINR value is reduced by reducing the step size ollaSinrDnStep(X); wherein when multiple consecutive uplink NACK feedbacks or downlink NACK feedbacks are received, the ollaSINR value is adjusted through the fast decrease and slow increase mechanism when performing demodulation performance feedback processing.

10. The adaptive modulation and coding method for rapid convergence of unmanned ship ad hoc network according to claim 9, characterized in that: The relationship between the target BLER and the step size increase and decrease is: The NACK number limit convergence algorithm is: when multiple consecutive NACK feedbacks in uplink CRC feedbacks or NACK feedbacks in downlink HARQ feedbacks are received, the convergence is accelerated by increasing the reduction step size ollaSinrDnStep, and the formula is: ollaSinrDnStep(X)=ollaSinrDnStep*X The formula for the ACK feedback processing is: ollaSINR(t)=min(ollaSINR(t-1)+ollaSinrUpStep(bler),OllaUpperThrd); The formula for NACK feedback processing is: ollaSINR(t)=max(ollaSINR(t-1)-ollaSinrDnStep(bler),OllaLowerThrd); Among them, OllaUpperThrd is the maximum value of the step length, and OllaLowerThrd is the minimum value of the step length.

Images