Adaptive adjustment method for data transmission link of micro-nano unmanned aerial vehicle in complex environment

Abstract

The invention relates to the technical field of uplink transmission, and discloses an adaptive adjustment method for a data transmission link of a micro-nano unmanned aerial vehicle in a complex environment, which comprises the following steps: calculating a second-order statistical variance of a downlink to represent environmental fluctuation, and determining the segmentation granularity of an uplink transmission block; on the premise of keeping the total power constant, the data amplitude is reduced and the pilot frequency amplitude is improved along with the increase of the segmentation granularity, and asymmetric power distribution is established; dividing the transmission block into a plurality of independent verification micro-frames for sending; according to the method, a physical layer structure and energy distribution reconstruction are directly driven through the environmental volatility, the problems of link self-adaption failure and channel estimation bottleneck caused by feedback lag in a fast fading environment are solved, and through micro-framing spatial isolation and power zero-sum gaming, the self-adaption performance of a channel is improved. And the uplink stability and the effective throughput in a complex multipath environment are improved.

Description

TECHNICAL FIELD

[0001] The application relates to a micro-nano unmanned aerial vehicle data transmission link adaptive adjustment method in a complex environment, and belongs to the technical field of uplink transmission. BACKGROUND

[0002] Currently, the existing micro-nano unmanned aerial vehicle data transmission link is based on a cellular mobile communication protocol, and adopts a measurement-feedback-execution closed-loop link adaptive mechanism. In a relatively stable channel state scenario, the receiving end measures the uplink channel quality to calculate the optimal modulation and coding strategy, and adjusts the transmission parameters of the transmitting end through downlink feedback to achieve high spectral efficiency. However, when the micro-nano unmanned aerial vehicle performs ultra-low altitude inspection, city canyon traversal or tree canopy detection tasks, the environment has serious multipath effect and fast time-varying characteristics, and the channel coherence time is less than the round-trip feedback delay of the communication protocol. In a fast fading environment, the existing closed-loop feedback mechanism faces inherent lag. The receiving end generates scheduling instructions based on the channel quality at a historical time, and by the time the scheduling instructions are executed at the transmitting end, the physical channel conditions have already changed, the channel quality has deteriorated sharply, and the scheduling instructions maintain high-order modulation, resulting in demodulation failure of the transmission block. The existing hybrid automatic repeat request mechanism is based on transmission block-level verification, and is vulnerable to sudden local interference. Only a small number of symbols are damaged, and the entire transmission block needs to be retransmitted. In a channel fluctuating environment, the whole frame retransmission strategy causes a sharp decline in uplink throughput and a sharp increase in invalid energy consumption, which restricts the endurance and task efficiency of the micro-nano unmanned aerial vehicle with limited battery capacity.

[0003] In addition to the feedback lag defect of the protocol mechanism, the existing technology focuses on macroscopic path selection or parameter estimation based on long-time statistical models to cope with complex environments. However, it lacks underlying self-healing capabilities for single physical link transient deterioration. For example, the Chinese invention patent with the authorization announcement number CN120582685B discloses a method for constructing a data transmission link for an unmanned aerial vehicle for ecological monitoring. The method obtains the sequence of unmanned aerial vehicle operating parameters and spatial distribution information, calculates the disturbance characteristic coefficient, and calculates the fading loss value based on the Nakagami-m model. The transmission link is constructed by selecting the best from multiple unmanned aerial vehicles. This method mainly relies on the spatial diversity advantage of multiple nodes and the analysis of statistical characteristics over a long period of time. It is a network layer routing optimization strategy. In the scenario of high dynamic task execution by a single micro-nano unmanned aerial vehicle, there is a lack of alternative links and the channel changes very quickly. The calculation of the disturbance coefficient relies on historical message data, which has high computational overhead and is difficult to complete real-time adjustment of the physical layer waveform structure within a millisecond channel coherence time. It cannot solve the problem of communication blockage caused by single-link deep fading. The size, weight and power consumption constraints of the micro-nano unmanned aerial vehicle result in a very low transmission power budget. In a deep fading channel environment, the conventional data and pilot power or fixed proportion allocation strategy is used. The limited total transmission power causes the signal-to-noise ratio of the demodulation reference signal to drop below the demodulation threshold, resulting in the failure of the pilot signal demodulation, which is used as the channel estimation reference. High-intensity error correction coding cannot correctly detect the data part, resulting in link interruption.

[0004] Therefore, the technical problem to be solved by this invention is to construct a robust transmission mechanism for fast fading channels that does not rely on instantaneous feedback accuracy, adaptively adjusts the transmission structure and resource allocation using local statistical characteristics, and achieves power-constrained transmission. Summary of the Invention

[0005] To address the problems mentioned in the background art, the technical solution of the present invention is as follows: A method for adaptive adjustment of data transmission links in complex environments for micro / nano unmanned aerial vehicles, comprising the following steps: The communication module calculates the second-order statistical variance of the signal quality index within the time window of receiving downlink signals, defines the second-order statistical variance as the channel fluctuation characteristic value, and uses the channel fluctuation characteristic value to characterize the long-term statistical uncertainty of the current channel environment. Query the preset fluctuation granularity mapping table to determine the uplink transport block segmentation granularity parameters that are positively correlated with the channel fluctuation characteristic value; Under the physical constraint of keeping the total transmit power constant within the transmission time interval, a power reconfiguration operation directly related to the segment granularity parameters is performed: the power offset value of the demodulation reference signal relative to the data symbol is determined according to the segment granularity parameters, the transmit amplitude of the data symbol is reduced and the transmit amplitude of the demodulation reference signal is increased. The power offset value increases monotonically with the increase of the segment granularity parameters, and an asymmetric energy allocation is established between the channel estimation reference and the data payload. Based on the segmentation granularity parameters, the uplink transmission block to be transmitted is divided into multiple microframes at the physical layer. An independent cyclic redundancy check sequence is added to each microframe to generate a composite transmission block. The composite transmission block is transmitted using the transmit amplitude determined by the power reconfiguration operation. Based on the verification result indicated by the state bitmap, the receiving party parses the state bitmap and extracts the specific micro-frames that indicate the verification failed, and performs independent encoding and retransmission on the specific micro-frames.

[0006] Preferably, the step of determining the uplink transport block segmentation granularity parameter that is positively correlated with the channel fluctuation characteristic value includes: performing multi-level comparisons between the channel fluctuation characteristic value and a preset set of stability thresholds; when the channel fluctuation characteristic value is lower than the lowest stability threshold, locking the segmentation granularity parameter to one; when the channel fluctuation characteristic value is higher than the stability threshold, increasing the segmentation granularity parameter according to a preset nonlinear ladder rule until the minimum microframe length limit allowed by the physical channel protocol is reached, so as to construct a data discretization structure that matches the degree of environmental fluctuation in the time domain dimension.

[0007] Preferably, in the step of performing the power reconfiguration operation directly related to the segmented granularity parameters, the power bias value is determined according to the following relationship: ,in, The power bias value of the demodulation reference signal relative to the data symbols, For segmented granularity parameters, These are weighting coefficients preset based on the system's allowed channel estimation error tolerance. The basic bias constant; the communication module is based on the calculated... The power ratio of the demodulation reference signal and data symbols is dynamically adjusted to ensure that the demodulation signal-to-noise ratio of the demodulation reference signal is prioritized when the channel estimation difficulty increases due to the increase in segmentation granularity parameters.

[0008] Preferably, the steps of performing independent encoding and retransmission on a specific microframe include: calculating the maximum run length of consecutive bits indicating check failure in the state bitmap, calculating the ratio of the maximum run length to the segmentation granularity parameter, and defining the ratio as the error clustering degree; comparing the error clustering degree with a preset morphological threshold; if the error clustering degree is less than the morphological threshold, determining that the current channel impairment is discrete random interference, and encoding the specific microframe using the same redundant version and modulation order as the initial transmission; if the error clustering degree is greater than or equal to the morphological threshold, determining that the current channel impairment is clustered physical obstruction, encoding the specific microframe using a redundant version containing incremental check information, and reducing the modulation order of the specific microframe.

[0009] Preferably, before the step of transmitting the composite transmission block using the transmit amplitude determined by the power reconfiguration operation, the method further includes: obtaining the recommended modulation and coding strategy issued by the receiving feedback party, and determining the first modulation order corresponding to the recommended modulation and coding strategy; querying a preset granularity-modulation constraint table, determining the maximum allowable modulation order corresponding to the current segment granularity parameter, and defining the maximum allowable modulation order as the second modulation order; comparing the first modulation order with the second modulation order; if the first modulation order is higher than the second modulation order, configuring the modulation method of the current composite transmission block to the second modulation order, and using the remaining time-frequency resources generated by the reduction in modulation order to load error correction redundancy information.

[0010] Preferably, the step of transmitting the composite transport block using the transmit amplitude determined by the power reconfiguration operation further includes: determining whether the segmentation granularity parameter exceeds a preset frequency domain discretization threshold; if it does not exceed the frequency domain discretization threshold, mapping the data symbols in the composite transport block to continuously distributed physical resource blocks in the frequency domain; if it exceeds the frequency domain discretization threshold, discretizing the data symbols in the composite transport block to non-contiguous physical resource blocks within the system bandwidth according to a preset frequency domain spacing rule, and ensuring that the data symbols within the same microframe are not adjacent in the frequency domain.

[0011] Preferably, in the step of receiving the verification result based on the state bitmap indication, the generation rules of the state bitmap include: the receiving feedback party independently performs cyclic redundancy check on each microframe in the received composite transport block according to the number corresponding to the segmentation granularity parameter; generates a binary bit sequence with a length equal to the segmentation granularity parameter, establishes a mapping relationship between each bit in the binary bit sequence and the verification result of each microframe; encapsulates the binary bit sequence into a state bitmap, and feeds it back to the communication module through the downlink control channel.

[0012] Preferably, the method further includes: when the communication module transmits the composite transport block, it does not send additional signaling to indicate the power offset value; the receiving feedback party obtains the segmentation granularity parameters by blind detection or reading downlink scheduling instructions, and derives the power offset value based on the same correlation as the communication module; the receiving feedback party performs amplitude normalization processing and channel estimation on the received signal based on the derived power offset value.

[0013] Preferably, in the step of calculating the second-order statistical variance of the signal quality index: the communication module adopts a sliding window mechanism to collect a sample sequence of downlink reference signal received power or downlink signal-to-noise ratio within a preset time length; calculates the variance value of the sample sequence, and performs a differential operation between the variance value and the preset reference noise variance measured in the system idle time slot, and uses the operation result as the channel fluctuation characteristic value to eliminate the interference of the system's inherent thermal noise on the environmental fluctuation assessment.

[0014] Preferably, the method further includes: after the communication module performs the transmission step, starting a retransmission timer; if no status bitmap is received or feedback indicating a full frame error is received before the retransmission timer expires, it is determined that the current link is in a deep fading blocking state; the communication module adjusts the segmentation granularity parameter to the maximum value, and adopts the lowest order modulation and coding strategy and the maximum power offset value to initiate a probing consistency retransmission, where consistency retransmission refers to sending a transmission block with the same content and structure as the previous transmission.

[0015] Compared with the prior art, the beneficial effects of the present invention are: 1. In the adaptive adjustment of data transmission links in complex environments of micro / nano UAVs, the downlink statistical variance is calculated to characterize environmental fluctuations. Based on this, the physical layer segmentation granularity of the uplink transmission block is dynamically adjusted. An open-loop adaptive mechanism independent of instantaneous channel state feedback is established. By utilizing the long-term stability of channel statistical characteristics, the problem of outdated channel state information caused by feedback delay in traditional protocols is avoided. In environments with fast fading or multipath effects, as the environmental fluctuation variance increases, the number of transmission block segments is automatically increased and an independent check sequence is configured for each microframe. The bit errors caused by sudden interference are limited to a specific microframe range. The physical layer structured isolation allows the receiver to initiate retransmission only for damaged small data segments, maintaining the link connection without the need for full frame retransmission, and reducing invalid data throughput and transmission delay in non-stationary channel environments.

[0016] 2. A positive correlation mapping relationship is established between segmentation granularity parameters and pilot power offset values. Under the physical constraint of maintaining a constant total uplink transmit power, symbol-level dynamic energy reconfiguration is achieved. When high environmental fluctuations trigger high-granularity segmentation, the transmitter automatically reduces the data symbol transmission amplitude and adds the saved power budget to the demodulation reference signal. By sacrificing the signal-to-noise ratio margin of the data payload under high fluctuation conditions, the channel estimation accuracy of the demodulation reference signal in deep fading channels is improved, solving the problem of full-frame demodulation failure caused by channel estimation errors. The resource game strategy ensures that the receiver establishes an accurate channel equalization benchmark under harsh conditions with extremely short channel coherence time, providing a physical layer prerequisite for the correct demodulation of micro-frame data.

[0017] 3. Using the ratio of the run length of the erroneous bits to the total number of segments in the feedback bitmap as the basis for error morphology diagnosis, the system identifies whether the current channel impairment is discrete random noise interference or clustered physical obstruction fading. For different topology errors, the system automatically switches the retransmission redundancy version and modulation order at the physical layer: for discrete errors, the original parameters are maintained to obtain energy combining gain; for clustered errors, the modulation order is downgraded and incremental redundancy mode information reconstruction is adopted. The simple feedback index information is transformed into inference of the physical properties of the channel, avoiding blind energy combining in deep fading scenarios that could lead to retransmission deadlock. This achieves accurate matching between the hybrid automatic retransmission request strategy and the channel impairment mechanism. Attached Figure Description

[0018] Fig. 1 This is a schematic diagram illustrating the entire process of adaptive adjustment of the data link in complex environments for the micro / nano-type UAV of the present invention. Fig. 2 This is a schematic diagram comparing the effective throughput performance of the system under different moving speeds according to the present invention; Fig. 3 This is a schematic diagram of the logical architecture and module interaction of the transceiver system of the present invention, which enables collaborative operation between the airborne and ground terminals. Detailed Implementation

[0019] To enable those skilled in the art to better understand the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0020] This invention discloses an adaptive adjustment method for data transmission links in complex environments for micro / nano unmanned aerial vehicles (UAVs). The method operates between the onboard communication module of the UAV and the ground receiving station, and is used to solve the link instability problem caused by feedback lag in fast-fading channels. The main execution component of this method is the baseband processing unit integrated on the UAV. Within the time window for receiving downlink signals, the communication module initiates a process with a start-up length of [missing information]. A sliding sampling window, preferably with a length of 20ms, is used. The communication module acquires the received power sample sequence S of the downlink demodulation reference signal at a preset sampling frequency, such as 1kHz. The processing unit calculates the variance of the sample sequence. The variance value was then compared with the baseline variance of thermal noise measured during the quiescent period. Perform difference operations to obtain the net environmental fluctuation characteristic value. ,Right now , this eigenvalue This feature value is used to characterize the degree of environmental fluctuation in the current channel; the processing unit utilizes this feature value. Using the index key, a pre-defined fluctuation-granularity mapping table is queried. This mapping table defines the fluctuation characteristic value range and the uplink transport block segmentation granularity parameters. The hierarchical correspondence, in specific engineering calibration, when Less than the first threshold At that time, the system is locked. If it is 1, maintain the standard transport block structure; when In and In between, set It is 4; when Exceed At that time, set The value is 16. Through this step, the system determines the degree of data discretization in the physical layer based on statistical grading; and determines the segmentation granularity parameters. Subsequently, the system performs a power reconfiguration operation based on a constant power constraint, with the linear operating region of the micro / nano UAV's RF front-end limiting the upper limit of the total transmit power. The processing unit is based on the formula Calculate the power bias value of the demodulation reference signal relative to the data symbol. ,in The weighting coefficient is preset based on the channel estimation error tolerance, and its value is, for example, 0.5; This is the base bias constant, with a value such as 0dB. The processor uses the calculated value... In the digital domain, amplitude scaling is performed on the generated waveform data; the processor downscales the amplitude factor of the data symbols. And adjust the amplitude factor of the modulation reference signal accordingly. This ensures that the total power is within the integration time of the symbol period. Strictly equal to This operation reduces the signal-to-noise ratio of the data payload while increasing the transmit power of the demodulation reference signal.

[0021] The baseband processor performs physical layer framing and resource mapping, and the processor bases the data on the physical layer. The uplink transport block to be sent is divided into bits. Each microframe is an independent microframe, and a length of [length missing] is calculated and added to each microframe separately. An independent cyclic redundancy check sequence, preferably 8 bits in length, is used to form a composite transport block. During the physical resource mapping phase, the system determines the current... Does it exceed the preset frequency domain discretization threshold? The threshold is preferably set to 8. Less than or equal to The processor maps composite transport blocks to frequency-contiguous physical resource blocks, if Greater than The processor initiates the comb-like discrete mapping logic, according to the system bandwidth. and The ratio determines the frequency domain spacing. Data symbols belonging to the same microframe are discretely loaded into frequency indices K. On subcarriers; after signal transmission, the communication module processes the feedback information from the receiving end and executes retransmission control, while the ground receiver processes the received signal. Each microframe is independently verified, and a feedback length of [length missing] is returned. The binary status bitmap is used, where 0 bits represent verification failure and 1 bits represent success. After receiving the bitmap, the UAV performs run-length analysis to calculate the maximum length of consecutive 0s in the bitmap. System calculation With total length The ratio of these values ​​is defined as the error clustering degree. ,Right now The system will With preset shape threshold For comparison, the threshold is preferably 0.3. Less than The system determines the interference to be discrete random interference and executes the Chase combining and retransmission strategy for microframes that fail verification. This involves sending a redundant version RV0 identical to the initial transmission while maintaining the modulation order. Greater than or equal to The system determines that the sudden fading is caused by physical obstruction and forcibly switches to the incremental redundancy retransmission strategy, selecting the redundant version RV3 that includes a parity bit, and reverting the modulation scheme of the microframe from the current order, such as 16QAM, to the lowest order, such as QPSK. In addition, the transmitter performs modulation and coding strategy arbitration before framing, and the processor reads the first modulation order from the scheduling instruction issued by the ground station. Simultaneously, it queries the locally stored granularity and modulation constraint table to obtain the current... Maximum allowed modulation order System comparison and ,like Higher than The system configures the modulation scheme of the current transport block as follows: And use the remaining time-frequency resources generated by the order reduction to fill in additional error correction coding redundancy information.

[0022] Example 1: This example is applied to a scenario where a micro / nano-sized UAV performs low-altitude inspections in densely built-up urban areas. Under such conditions, the physical channel exhibits severe frequency-selective fading, and the channel coherence time is shortened to the millisecond level, which is less than the round-trip delay of the measurement and feedback loop in the communication protocol. When the channel state and the ground scheduling command are severely mismatched—that is, when the ground receiving station issues a high-order scheduling command using 64QAM modulation based on the excellent channel state measured in historical time, but the UAV has entered the shadow area of ​​buildings at the time of execution, causing the actual channel quality to deteriorate dramatically—the adaptive adjustment mechanism of this invention is triggered. The baseband processing unit integrated into the UAV calculates the net environmental fluctuation characteristic value within the sliding window of the received downlink signal. This value currently exceeds the system's preset second threshold. The processing unit queries the fluctuation and granularity mapping table to determine the segmentation granularity parameters of the uplink transmission block. Locked at 16, power reconfiguration based on segmented granularity is performed according to the formula. The positive power bias value is calculated. The weighting coefficients Set to 0.5, the basic bias constant. Setting it to 0dB, the baseband processing unit reduces the transmission amplitude of the data symbols in the digital domain, while simultaneously transferring the saved power budget to the demodulation reference signal, thereby increasing the transmission power of the demodulation reference signal relative to the data symbols. Decibels, while the MCS arbitration mechanism compares the 64QAM instructions issued from the ground with the current The corresponding 16QAM upper limit will revert the modulation scheme to 16QAM and use the released time and frequency resources to fill additional error correction redundancy.

[0023] The composite transport block is transmitted to the ground receiving station via a comb-like discrete mapping. Deep fading caused by building obstruction results in subcarrier damage at specific locations in the frequency domain. The ground receiving station independently verifies the received 16 microframes. The generated feedback state bitmap shows that verification failed for microframes 4 through 7. After receiving this bitmap, the UAV performs topology feature analysis to calculate the error clustering degree. Exceeding the preset morphological threshold Based on this, the system determines that the error is caused by physical obstruction. For the 4th to 7th microframes, an incremental redundancy retransmission strategy is initiated. The redundant version RV3 with parity bit is selected, and the modulation method of the above specific microframes is further downgraded to QPSK for independent retransmission. The ground receiving station uses the demodulation reference signal with high signal-to-noise ratio to establish a channel equalization reference and recovers the damaged microframes by merging the retransmitted parity bit information.

[0024] Example 2: This example constructs a hardware-in-the-loop simulation verification platform to objectively quantify the physical layer performance of the adaptive adjustment method proposed in this invention in a simulated real urban canyon environment. Specifically, it empirically verifies link stability and effective throughput in multipath fast fading scenarios caused by high-speed movement. The test platform consists of a vector signal generator, a wireless channel simulator, and a terminal under test (DUT). The channel simulator is configured to load the TDL-C (Urban Macro) delay tap-type channel model defined by the 3GPP TR38.901 standard, which has non-line-of-sight (NLOS) transmission characteristics and rich multipath components, used to simulate complex urban densely built-up areas. To reproduce electromagnetic interference in real engineering environments, additive white Gaussian noise (AWGN) is actively injected into the signal link, clamping the receiver's signal-to-noise ratio (SNR) benchmark at a low level of 5.0 dB, serving as a stress test boundary for system stability. During the setting of test parameters, the downlink signal sampling window length... The determination follows an engineering trade-off logic based on Doppler frequency shift. Too short a sample size leads to insufficient statistical variance. Severely affected by transient noise disturbances; if If the time is too long, it will be impossible to track the non-stationary changes in the channel in a timely manner. Based on the moving speed range of the tested terminal from 0 to 120 km / h (approximately 33.3 m / s), and considering that the maximum Doppler frequency shift corresponding to a carrier frequency of 2.4 GHz is approximately 266 Hz and the coherence time is approximately 1.6 ms, this experiment will... The value is set to 20ms, which is approximately 12 times the coherence time. This ensures both the convergence stability of the second-order statistics and the sensitivity to environmental fluctuations. The update frequency (50Hz) is sufficient to cover the maneuvering dynamics of the UAV. The experiment set up two control groups: the control group adopted the standard fixed transport block structure and the adaptive modulation and coding strategy based on instantaneous CSI feedback; the experimental group adopted the dynamic micro-frame and power reconfiguration strategy based on statistical variance of the present invention. During the experiment, the moving speed of the terminal under test increased linearly from 5m / s to 35m / s to simulate the process of the UAV accelerating from hovering to high-speed cruising.

[0025] The collection and analysis of experimental data revealed the characteristic values ​​of environmental fluctuations. The inherent correlation between physical layer parameters and final link performance means that as mobile speed increases, the time-selective fading of the channel intensifies, leading to... It exhibits a nonlinear growth trend. Table 1 extracts key intermediate parameters and final performance data at different speed nodes during the experiment. Under quasi-static conditions at low speed (5.0 m / s), The experimental group maintained a low level of 0.02. The long-frame structure had a packet error rate (PER) of 0.5%, the same as the control group, and the system did not incur additional verification overhead. However, when the speed was increased to 20.0 m / s, The value climbed to 0.15, triggering the segmented granularity parameter of the experimental group. Adjusting it to 4 will result in a power bias of 1.5 dB. Despite the reduction in the transmit power of the data symbols, the channel estimation mean square error (NMSE) remained at an excellent level of -25dB due to the 1.5dB power enhancement of the demodulation reference signal, resulting in a slight increase in the PER of the experimental group to 2.1%. In contrast, the control group suffered from a sharp deterioration in PER to 18.4% due to insufficient pilot signal-to-noise ratio leading to a deterioration in channel estimation.

[0026] Table 1: Comparative Test Data of Link Performance under Variable Doppler Frequency Shift Environment

[0027] Further data analysis reveals the performance inflection point characteristics of this invention in the high dynamic range. When the speed reaches the extreme condition of 30.0 m / s, The number surged to 0.420, and the system will... Locked to 16 and subjected to a 2.8dB deep power offset, the PER of the control group collapsed to 85.7%, essentially losing its communication capability. This indicates that the link adaptation mechanism based on instantaneous feedback completely fails under fast fading channels. The experimental group, by sacrificing the signal-to-noise ratio of the data portion (reduced data power leading to an increase in the original bit error rate), obtained an absolutely stable channel estimation benchmark (enhanced pilot power). Combined with the spatial isolation effect of microframes, the final PER was controlled within the usable range of 5.4%. It is worth noting that although the effective throughput of the experimental group dropped to 6.2Mbps at 30.0m / s, far lower than in the low-speed state... However, compared to the near-zero throughput of the control group, this represents a qualitative leap from no communication to reliable low-speed communication. Furthermore, when the speed is further increased to 35.0 m / s, the PER of the experimental group also rises to 12.8%, indicating that the Doppler frequency shift of the physical channel is close to the subcarrier spacing tolerance of the Orthogonal Frequency Division Multiplexing (OFDM) system. This establishes the effective working boundary of this method under the current physical layer standard. This invention, through statistical feature-driven physical layer structure reconstruction and asymmetric energy allocation, avoids the risks of feedback lag and channel estimation failure in complex environments with both fast fading and strong interference, and expands the stable working range of the micro-nano UAV data link.

[0028] Example 3: This example combines Figs. 1 to 3 This document describes an adaptive adjustment method for data transmission links in complex environments for micro / nano unmanned aerial vehicles (UAVs). Fig. 1 As shown, the process begins with the channel fluctuation characteristic value calculation step. The communication module receives downlink reference signal received power or signal-to-noise ratio samples, calculates long-term statistical uncertainty based on second-order statistical variance and removes thermal noise interference, and enters the uplink transport block segment granularity determination step. By querying the fluctuation granularity mapping table, a data discretization structure matching the environment is constructed, and the process enters the parallel processing stage: on the one hand, asymmetric power allocation is performed, and the amplitude of the demodulated reference signal is increased according to the segment granularity while keeping the total power constant; on the other hand, modulation and coding strategy arbitration is performed, the ground scheduling instruction is received to recommend the modulation and coding strategy and the instruction is compared with the local constraints. If necessary, the modulation order is backed up and redundancy is filled, and the composite transport block generation and transmission step is entered. The original data payload of the uplink transport block to be transmitted is added with micro-frame independent verification sequences, and power reconfiguration and frequency domain discrete mapping are applied. It is transmitted to the complex multipath environment via uplink radio frequency signal. After the receiving end feeds back the status bitmap through the downlink control channel, the system enters the error morphology diagnosis and retransmission strategy decision step. Based on the error clustering degree, discrete interference or clustered physical blockage is identified, the incremental redundancy mode is switched and the feedback is looped back to the composite transport block generation stage.

[0029] like Fig. 2As shown, the horizontal axis represents the moving speed in m / s, and the vertical axis represents the effective throughput in Mbps. The solid line in the graph represents the throughput of the traditional method, and the dashed line represents the throughput of the method of this invention. Observing the curve trend, it can be seen that in the low-speed range of 5 m / s to 15 m / s, the two curves basically overlap. As the speed increases above 15 m / s, the effective throughput of the traditional method shows a sharp downward trend, dropping to near 0 Mbps at 30 m / s. In contrast, the effective throughput of the method of this invention decreases more gradually, maintaining an effective throughput of 4 Mbps even at a high speed of 35 m / s. Fig. 3 As shown, the core component of the airborne terminal is the baseband processing unit, which integrates an environmental fluctuation sensing module for performing downlink signal sliding sampling, second-order statistical variance calculation, and thermal noise benchmark removal. The adaptive strategy decision module is responsible for fluctuation and segmented granularity mapping and modulation and coding strategy arbitration. The power and resource reconstruction module is responsible for performing asymmetric energy allocation and pilot power bias enhancement. The physical layer framing engine is responsible for microframe segmentation and independent verification, frequency domain discrete resource mapping, and connects to the RF transceiver front end to send composite transmission blocks or discrete microframes through the uplink data link. The core component of the ground terminal is the signal processing and scheduling center, which connects to a high-gain receiving antenna array. It includes an enhanced channel estimation module for channel equalization and amplitude normalization based on high-power pilots. The multi-granularity verification module is responsible for microframe-level independent cyclic redundancy check and error location. The error morphology analysis logic is responsible for error run length calculation and interference type (discrete / clustered) determination. The intelligent feedback generation module is used to generate a state bitmap and assemble retransmission scheduling instructions. The state bitmap and reference signal are fed back to the airborne terminal through the downlink control and feedback link to complete closed-loop control.

[0030] Example 4: This example provides a fully reproducible procedure for channel fluctuation feature extraction and dynamic threshold determination, aiming to transform abstract statistical analysis logic into concrete engineering operation steps. The execution entity in this example is the baseband processing unit of a micro / nano-type UAV. The core task is to extract quantitative indicators that can accurately characterize environmental instability from discrete downlink signal samples, generate segmented granularity adjustment instructions, and the system executes the downlink reference signal received power sample sequence. Preprocessing, considering the high-frequency motor noise interference that may be introduced by the micro-nano UAV platform, the processing unit applies a length of [missing information] to the original sample sequence S. The moving average filter, the filter window length The settings follow the denoising principles in signal processing: smoothing out random jitter caused by non-channel factors while retaining the trend characteristics reflecting the true changes in the channel. In this embodiment, based on the relationship between the UAV motor speed frequency range (usually between 100Hz and 500Hz) and the downlink signal sampling frequency (1kHz), the settings are determined through analysis. The sample value is taken from 5 sampling points, and after smoothing, a denoised sample sequence is obtained. Each element It is the arithmetic mean of 5 consecutive sampling points within the corresponding window in the original sequence.

[0031] Next, based on the denoised sample sequences Calculate the second-order statistical variance The specific calculation logic is as follows: calculate the sequence mean , where N is the sequence length; then calculate the sum of squares of the differences between each sample point and the mean, and divide by the total number of samples, i.e. This step strictly follows the statistical definition of variance to ensure the mathematical rigor of the index calculation. To eliminate the interference of inherent thermal noise in the system on the environmental fluctuation assessment, the processing unit retrieves the reference noise variance stored in non-volatile memory. This benchmark value is obtained by collecting received power samples under no-load channels during the system's factory calibration phase or periodic quiet periods (i.e., periods without downlink signal transmission) and calculating the variance using the same algorithm. Finally, it is obtained through differential operations. Obtain net environmental fluctuation characteristic value This value purely reflects the multipath and fading variations of the external channel environment; finally, the execution is based on... The segmented granularity decision-making, with a black box for threshold setting, is disclosed in this embodiment. and The adaptive calibration logic sets the threshold not as a fixed empirical value, but dynamically correlated with the current modulation and coding scheme (MCS) level. The system uses a pre-defined lookup table to define the variance limits corresponding to the maximum Doppler frequency shift that different MCS levels, such as QPSK, 16QAM, and 64QAM, can tolerate. For example, when the current MCS is 16QAM, based on theoretical simulations and measured data, the variance critical point where the demodulation bit error rate increases sharply is marked as... The system sets the first threshold to Set the second threshold to When calculated in real time At that time, the channel is determined to be stable. Keep it at 1; when At that time, it was determined to be a moderate fluctuation. Adjusted to 4; when At that time, it was determined to be a violent fluctuation. Adjusted to 16.

[0032] Example 5: In a controlled environment of a microwave anechoic chamber, an offline calibration procedure for constructing a wave-granularity mapping table was executed. The channel simulator generated test signals with gradient-varying Doppler frequency shift and multipath delay spread to simulate environmental wave characteristics of different levels. For each discrete The system iterates through all allowed segmentation granularity parameters for each node. It also monitors the block error rate of the demodulated reference signal, and the criterion for establishing the mapping relationship is to select the current... The minimum value that can maintain the demodulation reference signal block error rate below a preset safety threshold The values ​​are used to fill the state switching boundaries in the lookup table with experimental data, ensuring that the segmentation strategy is based on objective channel physical limits rather than empirical estimations. During the RF parameter solidification phase of the communication module, the power bias formula is executed. Medium coefficient and Hardware adaptation and calibration are performed by monitoring the output signal of the RF front-end using a spectrum analyzer. The system sends test sequences with different segmentation granularities, and the coefficients... The calibration is based on the linearity curve of the RF power amplifier. Under the premise of ensuring that the adjacent channel leakage ratio does not exceed the standard, the linearity is determined accordingly. The maximum allowable slope for pilot power increase when the pilot power is increased, a constant. The calibration is then performed based on the fundamental insertion loss difference of the system link to ensure the calculated results. It can accurately compensate for fixed deviations introduced by the hardware.

[0033] Example 6: This example provides a set of reference parameter self-calibration and environmental boundary adaptation procedures for the system power-on initialization phase. It aims to solve the parameter reference deviation problem caused by hardware component differences and temperature drift in actual engineering deployments. Before the UAV communication module initiates the uplink establishment process, the baseband processing unit executes a dynamic calibration program for the thermal noise reference. The processing unit controls the transmit / receive switch of the RF front-end to switch the antenna port to the internal matched load. A thermal noise power sample sequence of a preset length is collected in an environment isolated from external electromagnetic signals. The processing unit calculates the variance of this sequence and writes it into the reference noise variance. The register is used to update the subtrahend term in the environmental fluctuation characteristic value calculation formula in real time, thereby avoiding the cumulative error of the environmental fluctuation assessment accuracy caused by the receiver noise floor drift due to device temperature changes or aging.

[0034] The system executes a frequency domain correlation adaptation procedure for the error clustering determination logic. Considering the differences in channel coherence bandwidth under different multipath environments, a fixed morphological threshold may lead to misjudgment of error type. The processing unit estimates the channel coherence bandwidth based on the frequency domain response of the current downlink demodulation reference signal. The processing unit reads the currently configured system transmission bandwidth. And based on physical formulas Calculate and update the morphological threshold Where k is a weighting constant preset based on the error rate tolerance, ensuring the accuracy of error clustering. The physical judgment criteria always maintain a strict linear correspondence with the frequency selective fading scale of the current channel, preventing normal frequency selective fading from being misjudged as physical blockage in strongly correlated channels or physical blockage events from being missed in weakly correlated channels.

[0035] Example 7: This example details the offline engineering calibration procedure for key system control parameters, aiming to establish the fluctuation granularity mapping table and the physical source of the power bias coefficient. The fluctuation granularity mapping table is constructed based on offline traversal testing to maximize throughput. In a controlled channel simulation environment, the test equipment loads a standardized multipath channel model and progressively increases the Doppler frequency shift parameters with a fixed frequency step size. For each discrete frequency shift test point, the system traverses all protocol-allowed segmented granularity parameters. The system provides real-time statistics on uplink block error rate and throughput data, and tests how to maximize throughput under the current frequency shift. The option defines the corresponding channel fluctuation characteristic value as the threshold for triggering segmentation granularity determination. The calibration data is compiled into a static lookup table and burned into the non-volatile storage space of the baseband processing unit as a runtime adaptive adjustment logic benchmark; the power bias formula weighting coefficients The values ​​were determined through nonlinear distortion boundary tests of the RF transmit channel. With the transmit output power set to its maximum nominal value, the test instrument gradually increased the power ratio of the demodulated reference signal relative to the data symbols. The spectrum analyzer simultaneously monitored the transmit signal error vector amplitude (EVM) and adjacent channel leakage ratio (ACLR). When any indicator deteriorated to the limits specified in the communication protocol (e.g., EVM exceeding the demodulation threshold of the corresponding modulation order or ACLR exceeding 45 dBc), the power ratio was recorded as the maximum allowable offset limit by the hardware. Based on the maximum bias limit and the linear slope relationship between the maximum segment granularity supported by the system, the power reconfiguration is limited to the linear dynamic range of the RF front end.

[0036] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0037] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for adaptive adjustment of data transmission links in complex environments for micro / nano unmanned aerial vehicles (UAVs), characterized in that, Includes the following steps: The communication module calculates the second-order statistical variance of the signal quality index within the time window of receiving downlink signals, defines the second-order statistical variance as the channel fluctuation characteristic value, and uses the channel fluctuation characteristic value to characterize the long-term statistical uncertainty of the current channel environment. Query the preset fluctuation granularity mapping table to determine the uplink transport block segmentation granularity parameters that are positively correlated with the channel fluctuation characteristic value; Under the physical constraint of keeping the total transmit power constant within the transmission time interval, a power reconfiguration operation directly related to the segment granularity parameters is performed: the power offset value of the demodulation reference signal relative to the data symbol is determined according to the segment granularity parameters, the transmit amplitude of the data symbol is reduced and the transmit amplitude of the demodulation reference signal is increased. The power offset value increases monotonically with the increase of the segment granularity parameters, and an asymmetric energy allocation is established between the channel estimation reference and the data payload. Based on the segmentation granularity parameters, the uplink transmission block to be transmitted is divided into multiple microframes at the physical layer. An independent cyclic redundancy check sequence is added to each microframe to generate a composite transmission block. The composite transmission block is transmitted using the transmit amplitude determined by the power reconfiguration operation. Based on the verification result indicated by the state bitmap, the receiving party parses the state bitmap and extracts the specific micro-frames that indicate the verification failed, and performs independent encoding and retransmission on the specific micro-frames.

2. The adaptive adjustment method for data transmission links in complex environments for micro / nano unmanned aerial vehicles according to claim 1, characterized in that, The steps for determining the uplink transport block segmentation granularity parameters that are positively correlated with the channel fluctuation characteristic value include: comparing the channel fluctuation characteristic value with a preset set of stable thresholds at multiple levels; when the channel fluctuation characteristic value is lower than the lowest stable threshold, locking the segmentation granularity parameter to one; when the channel fluctuation characteristic value is higher than the stable threshold, increasing the segmentation granularity parameter according to a preset nonlinear ladder rule until the minimum microframe length limit allowed by the physical channel protocol is reached, so as to construct a data discretization structure that matches the degree of environmental fluctuation in the time domain dimension.

3. The method for adaptive adjustment of data transmission links in complex environments for micro / nano-type unmanned aerial vehicles according to claim 1, characterized in that, In the steps of performing power reconfiguration operations directly related to segmented granularity parameters, the power bias value is determined based on the following relationship: ,in, The power bias value of the demodulation reference signal relative to the data symbols, For segmented granularity parameters, These are weighting coefficients preset based on the system's allowed channel estimation error tolerance. The basic bias constant; the communication module is based on the calculated... The power ratio of the demodulation reference signal and data symbols is dynamically adjusted to ensure that the demodulation signal-to-noise ratio of the demodulation reference signal is prioritized when the channel estimation difficulty increases due to the increase in segmentation granularity parameters.

4. The adaptive adjustment method for data transmission links in complex environments for micro / nano unmanned aerial vehicles according to claim 1, characterized in that, The steps for performing independent encoding and retransmission on a specific microframe include: calculating the maximum run length of consecutive bits indicating check failure in the state bitmap; calculating the ratio of the maximum run length to the segmentation granularity parameter; defining the ratio as the error clustering degree; comparing the error clustering degree with a preset morphological threshold; if the error clustering degree is less than the morphological threshold, determining that the current channel impairment is discrete random interference, and encoding the specific microframe using the same redundant version and modulation order as the initial transmission; if the error clustering degree is greater than or equal to the morphological threshold, determining that the current channel impairment is clustered physical obstruction, encoding the specific microframe using a redundant version containing incremental check information, and reducing the modulation order of the specific microframe.

5. The adaptive adjustment method for data transmission links in complex environments for micro / nano unmanned aerial vehicles according to claim 1, characterized in that, Before the step of transmitting the composite transport block using the transmit amplitude determined by the power reconfiguration operation, the method further includes: obtaining the recommended modulation and coding strategy issued by the receiving feedback party, determining the first modulation order corresponding to the recommended modulation and coding strategy; querying the preset granularity-modulation constraint table, determining the maximum allowable modulation order corresponding to the current segment granularity parameter, and defining the maximum allowable modulation order as the second modulation order; comparing the first modulation order with the second modulation order; if the first modulation order is higher than the second modulation order, configuring the modulation mode of the current composite transport block to the second modulation order, and using the remaining time-frequency resources generated by the reduction in modulation order to load error correction redundancy information.

6. The adaptive adjustment method for data transmission links in complex environments for micro / nano unmanned aerial vehicles according to claim 1, characterized in that, The step of transmitting the composite transport block using the transmit amplitude determined by the power reconfiguration operation further includes: determining whether the segmentation granularity parameter exceeds a preset frequency domain discretization threshold; if it does not exceed the frequency domain discretization threshold, mapping the data symbols in the composite transport block to continuously distributed physical resource blocks in the frequency domain; if it exceeds the frequency domain discretization threshold, discretizing the data symbols in the composite transport block to non-contiguous physical resource blocks within the system bandwidth according to a preset frequency domain spacing rule, and ensuring that the data symbols within the same microframe are not adjacent in the frequency domain.

7. The adaptive adjustment method for data transmission links in complex environments for micro / nano unmanned aerial vehicles according to claim 1, characterized in that, In the step of receiving the verification result based on the state bitmap indication, the generation rules of the state bitmap include: the receiving feedback party independently performs cyclic redundancy check on each microframe in the received composite transport block according to the number corresponding to the segmentation granularity parameter; generates a binary bit sequence with a length equal to the segmentation granularity parameter, establishes a mapping relationship between each bit in the binary bit sequence and the verification result of each microframe; encapsulates the binary bit sequence into a state bitmap, and feeds it back to the communication module through the downlink control channel.

8. The adaptive adjustment method for data transmission links in complex environments for micro / nano unmanned aerial vehicles according to claim 3, characterized in that, The method also includes: when the communication module transmits the composite transport block, it does not send additional signaling to indicate the power offset value; the receiving feedback party obtains the segmentation granularity parameters by blind detection or reading downlink scheduling instructions, and derives the power offset value based on the same correlation as the communication module; the receiving feedback party performs amplitude normalization processing and channel estimation on the received signal based on the derived power offset value.

9. The adaptive adjustment method for data transmission links in complex environments for micro / nano unmanned aerial vehicles according to claim 1, characterized in that, In the steps of calculating the second-order statistical variance of signal quality indicators: the communication module adopts a sliding window mechanism to collect downlink reference signal received power or downlink signal-to-noise ratio sample sequences within a preset time length; The variance of the sample sequence is calculated, and the variance is differentially calculated with the preset reference noise variance obtained by measuring in the idle time slot of the system. The result is used as the channel fluctuation characteristic value to eliminate the interference of the inherent thermal noise of the system on the environmental fluctuation assessment.

10. The method for adaptive adjustment of data transmission links in complex environments for micro / nano-type unmanned aerial vehicles according to claim 1, characterized in that, The method also includes: after the communication module performs the transmission step, starting a retransmission timer; if no status bitmap is received or feedback indicating a full frame error is received before the retransmission timer expires, the current link is determined to be in a deep fading blocking state; the communication module adjusts the segmentation granularity parameter to the maximum value, and adopts the lowest order modulation and coding strategy and the maximum power offset value to initiate a probing consistency retransmission, which means sending a transport block with the same content and structure as the previous transmission.

Images