Morse code encoding and decoding method and system based on artificial intelligence

Abstract

The application relates to the technical field of digital information transmission, and discloses a Morse code encoding and decoding method and system based on artificial intelligence, which comprises the following steps: converting text information into an initial dot-dash symbol sequence according to a standard Morse code mapping table, and collecting channel characteristic parameters to input a light decision tree model to determine a time sequence characteristic weight vector; optimizing the initial dot-dash symbol sequence to generate an optimized dot-dash symbol sequence; calculating an absolute sending time length, combining a multipath fading coefficient of a channel to dynamically adjust sending power and waveform envelope parameters, and generating an encoding signal; demodulating a received signal to obtain a time sequence characteristic weight vector used by a sending end, performing envelope detection on a baseband signal, and obtaining a dot-dash symbol sequence according to the time sequence characteristic weight vector, and converting to obtain decoded text information according to a standard Morse code mapping table. The application solves the encoding and decoding problems of the Morse code in a dynamic environment, and can run on an edge device.

Description

Technical Field

[0001] This invention relates to the field of digital information transmission technology, and more specifically, to a Morse code encoding and decoding method and system based on artificial intelligence. Background Technology

[0002] With the rapid development of the Internet of Things (IoT), smart devices, and edge computing technologies, digital information transmission technologies are playing an increasingly important role in various application scenarios. Morse code, as a classic communication encoding method, remains valuable in emergency communications, military communications, amateur radio communications, and low-power IoT transmission due to its strong anti-interference capabilities, high transmission reliability, and low hardware requirements. Especially in edge computing scenarios, where terminal devices need to achieve efficient information encoding and decoding under resource constraints, the simplicity and robustness of Morse code make it one of the ideal communication methods.

[0003] Currently, existing Morse code encoding and decoding technologies mainly rely on manual operation or algorithmic processing based on fixed rules. In the encoding stage, traditional methods convert text information into a dot-dash sequence according to standard Morse code rules before modulation and transmission. In the decoding stage, typical methods identify characters by detecting parameters such as the dot-dash duration ratio and interval characteristics of the signal, combined with a pre-set Morse code mapping table. This traditional processing method faces many challenges in complex channel environments and scenarios with high real-time requirements. Some improved schemes have introduced signal processing techniques such as adaptive threshold adjustment and noise filtering to improve recognition accuracy in noisy environments. However, these methods are still essentially based on fixed rules and empirical parameter adjustments, making it difficult to adapt to dynamically changing signal characteristics.

[0004] However, existing technologies still have significant limitations. First, Morse code is essentially a continuous temporal signal (a combination of dot-dash-interval durations), unlike discrete binary codes. Its dot-dash recognition relies on the relative duration ratios of the continuous temporal signals, rather than absolute values. Traditional encoding and decoding algorithms based on fixed thresholds cannot handle dynamic factors such as changes in transmission rate, differences in operator habits, and signal distortion. Under complex channel conditions such as strong noise, signal attenuation, and multipath interference, the recognition accuracy drops significantly. Second, existing AI-based encoding and decoding solutions are mostly deployed in the cloud or on high-performance devices. Edge devices, limited by computing power and storage, cannot run intelligent models with large parameter scales and high computational complexity for real-time inference. Cloud-edge collaborative architectures also struggle to guarantee real-time performance and reliability in scenarios with network outages or high latency. Currently, there is a lack of lightweight AI models optimized for the temporal characteristics of Morse code, making it impossible to achieve highly robust real-time encoding and decoding at resource-constrained edge devices.

[0005] No effective solutions have yet been proposed to address the problems in the relevant technologies. Summary of the Invention

[0006] In view of the problems in related technologies, this invention proposes a Morse code encoding and decoding method and system based on artificial intelligence to overcome the above-mentioned technical problems existing in the existing related technologies.

[0007] Therefore, the specific technical solution adopted by the present invention is as follows:

[0008] According to one aspect of the present invention, an artificial intelligence-based Morse code encoding and decoding method is provided. This method includes: converting text information into an initial dot-dash symbol sequence according to a standard Morse code mapping table, and collecting channel characteristic parameters and inputting them into a lightweight decision tree model to determine a timing feature weight vector; optimizing the initial dot-dash symbol sequence to generate an optimized dot-dash symbol sequence; calculating the absolute transmission duration based on the optimized dot-dash symbol sequence and the timing feature weight vector, dynamically adjusting the transmission power and waveform envelope parameters in conjunction with the channel's multipath fading coefficient, and generating an encoded signal based on the adjusted dot-dash symbol sequence; demodulating the received signal to obtain the timing feature weight vector used by the transmitter, performing envelope detection on the baseband signal and segmenting it according to the timing feature weight vector to obtain a dot-dash symbol sequence, and converting the dot-dash symbol sequence according to a standard Morse code mapping table to obtain the decoded text information.

[0009] Furthermore, the text information is converted into an initial dot-dash symbol sequence according to a standard Morse code mapping table, and channel feature parameters are collected and input into a lightweight decision tree model to determine the temporal feature weight vector. The initial dot-dash symbol sequence is then optimized to generate an optimized dot-dash symbol sequence. This involves: converting the text information to be transmitted into a character sequence, and mapping each character in the character sequence to a corresponding dot-dash symbol combination according to a standard Morse code mapping table to generate the initial dot-dash symbol sequence; real-time acquisition of the current channel's channel feature parameters, which are then input into a pre-trained lightweight decision tree model; and determination of the current channel's temporal feature weight vector based on the output of the lightweight decision tree model. The temporal feature weight vector includes dot duration weight, dash duration weight, dot-dash interval weight, and character interval weight. The duration of each symbol in the initial dot-dash symbol sequence is then weighted and adjusted based on the temporal feature weight vector to generate the optimized dot-dash symbol sequence.

[0010] Furthermore, the absolute transmission duration is calculated based on the optimized dot-dash symbol sequence and timing feature weight vector. The transmission power and waveform envelope parameters are dynamically adjusted in conjunction with the channel's multipath fading coefficient. The generated coded signal based on the adjusted dot-dash symbol sequence includes: calculating the absolute transmission duration of each dot-dash symbol based on the optimized dot-dash symbol sequence and timing feature weight vector; dynamically adjusting the transmission power and waveform envelope parameters of each dot-dash symbol to minimize transmission power consumption while ensuring timing discriminability at the receiver, in conjunction with the channel's multipath fading coefficient; converting the adjusted dot-dash symbol sequence into a baseband signal and superimposing a preset timing synchronization preamble to generate a complete coded signal; the timing synchronization preamble carries the coded information of the timing feature weight vector.

[0011] Further, demodulating the received signal to obtain the timing feature weight vector used by the transmitter, performing envelope detection on the baseband signal and segmenting it according to the timing feature weight vector to obtain a dot-dash symbol sequence, and converting the dot-dash symbol sequence according to the standard Morse code mapping table to obtain the decoded text information includes: acquiring the received radio frequency signal and demodulating it to obtain the baseband signal, extracting the timing synchronization preamble from the baseband signal, parsing to obtain the timing feature weight vector used by the transmitter as the initial decoding reference; performing envelope detection on the baseband signal to obtain the timing waveform sequence, and statistically analyzing the duration distribution of continuous high-level segments and continuous low-level segments in the timing waveform sequence; identifying peak features in the duration distribution through probability density estimation, and using the peak duration corresponding to the peak features as candidate values ​​for dot duration and dot duration; determining the dot-dash segmentation threshold and character segmentation threshold according to the timing feature weight vector in the initial decoding reference, and segmenting the timing waveform sequence to obtain the dot-dash symbol sequence; converting the dot-dash symbol sequence into a character sequence according to the standard Morse code mapping table to obtain the decoded text information.

[0012] According to another aspect of the present invention, an artificial intelligence-based Morse code encoding and decoding system is also provided. This system includes: a symbol generation module, configured to convert text information into an initial dot-dash symbol sequence according to a standard Morse code mapping table, and collect channel characteristic parameters to input into a lightweight decision tree model to determine a timing feature weight vector; optimize the initial dot-dash symbol sequence to generate an optimized dot-dash symbol sequence; a signal encoding module, configured to calculate the absolute transmission duration based on the optimized dot-dash symbol sequence and the timing feature weight vector, dynamically adjust the transmission power and waveform envelope parameters in conjunction with the channel's multipath fading coefficient, and generate an encoded signal based on the adjusted dot-dash symbol sequence; and a signal decoding module, configured to demodulate the received signal to obtain the timing feature weight vector used by the transmitter, perform envelope detection on the baseband signal and segment it according to the timing feature weight vector to obtain a dot-dash symbol sequence, and convert the dot-dash symbol sequence according to a standard Morse code mapping table to obtain the decoded text information.

[0013] The beneficial effects of this invention are as follows:

[0014] (1) This invention solves the problem that traditional Morse code encoding and decoding cannot cope with actual changes when using fixed thresholds. In actual use, the transmission rate may be fast or slow, different operators have different habits, and the signal will be distorted during transmission. All of these will cause the duration ratio of dot and dash symbols to change. Traditional methods are prone to identification errors when using fixed thresholds. This invention uses a lightweight decision tree model to determine a suitable timing feature weight vector based on the current channel feature parameters. The encoding end uses these weights to dynamically adjust the dot duration, dash duration and various interval durations to generate symbol sequences. The decoding end extracts the weight information used by the transmitting end from the preamble, and then combines it with probability density estimation to find the duration features of dot and dash symbols from the actual waveform. Finally, it determines a suitable segmentation threshold based on this information. In this way, the encoding and decoding parameters can change with the actual situation, effectively cope with the fluctuation of the duration ratio, and greatly improve the recognition accuracy and stability of Morse code in various dynamic environments.

[0015] (2) This invention solves the problem that existing intelligent solutions have too much computation and cannot run on edge devices with limited resources. Existing intelligent solutions have too many parameters and too complex calculations, so they can only run on the cloud or high-performance devices. The computing power and storage of edge devices are insufficient. This invention adopts a lightweight decision tree model, which establishes the correspondence between channel features and time-series weights through historical data. During inference, only simple condition judgments and table lookup operations are needed, without running complex matrix operations. At the same time, the time-series weight information is directly encoded in the preamble and sent to the receiving end. The decoding end only needs to perform simple signal processing such as envelope detection and probability density estimation. The entire encoding and decoding process has very low requirements for equipment and can run in real time on edge devices with limited computing power and storage, which ensures the accuracy of recognition without consuming too many resources. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a flowchart illustrating the encoding process in an artificial intelligence-based Morse code encoding and decoding method according to an embodiment of the present invention.

[0018] Figure 2 This is a flowchart illustrating the decoding process in an artificial intelligence-based Morse code encoding and decoding method according to an embodiment of the present invention.

[0019] Figure 3 This is a specific implementation diagram of generating an optimized dot-dash symbol sequence in a Morse code encoding and decoding method based on artificial intelligence according to an embodiment of the present invention;

[0020] Figure 4 This is a schematic diagram of a Morse code encoding and decoding system based on artificial intelligence according to an embodiment of the present invention. Detailed Implementation

[0021] To further illustrate the various embodiments, the present invention provides accompanying drawings, which are part of the disclosure of the present invention. These drawings are mainly used to illustrate the embodiments and can be used in conjunction with the relevant descriptions in the specification to explain the operating principles of the embodiments. With reference to these drawings, those skilled in the art should be able to understand other possible implementation methods and the advantages of the present invention. The components in the drawings are not drawn to scale, and similar component symbols are generally used to represent similar components.

[0022] According to an embodiment of the present invention, an artificial intelligence-based Morse code encoding and decoding method and system are provided.

[0023] The present invention will now be further described in conjunction with the accompanying drawings and specific embodiments, such as... Figure 1 and Figure 2 As shown, according to an embodiment of the present invention, an artificial intelligence-based Morse code encoding and decoding method is provided, the artificial intelligence-based Morse code encoding and decoding method comprising:

[0024] S1. Convert the text information into an initial dot-dash symbol sequence according to the standard Morse code mapping table, and collect channel feature parameters to input into the lightweight decision tree model to determine the temporal feature weight vector; optimize the initial dot-dash symbol sequence to generate an optimized dot-dash symbol sequence;

[0025] S2. Calculate the absolute transmission duration based on the optimized dot-dash symbol sequence and timing feature weight vector, dynamically adjust the transmission power and waveform envelope parameters in combination with the multipath fading coefficient of the channel, and generate the coded signal based on the adjusted dot-dash symbol sequence.

[0026] S3. Demodulate the received signal to obtain the timing feature weight vector used by the transmitter, perform envelope detection on the baseband signal and segment it according to the timing feature weight vector to obtain the dot-dash symbol sequence, and convert the dot-dash symbol sequence according to the standard Morse code mapping table to obtain the decoded text information.

[0027] In one embodiment, such as Figure 1 and Figure 3As shown, the text information is converted into an initial dot-dash symbol sequence according to the standard Morse code mapping table, and channel feature parameters are collected and input into a lightweight decision tree model to determine the temporal feature weight vector; the initial dot-dash symbol sequence is optimized to generate an optimized dot-dash symbol sequence, including:

[0028] The text information to be sent is converted into a character sequence, and each character in the character sequence is mapped to a corresponding dot-dash symbol combination according to the standard Morse code mapping table to generate an initial dot-dash symbol sequence.

[0029] The channel feature parameters of the current channel are collected in real time and input into a pre-trained lightweight decision tree model;

[0030] Based on the output of the lightweight decision tree model, the temporal feature weight vector of the current channel is determined; the temporal feature weight vector includes dot duration weight, swipe duration weight, dot-swipe interval weight, and character interval weight.

[0031] The duration of each symbol in the initial dot-dash symbol sequence is weighted and adjusted based on the temporal feature weight vector to generate an optimized dot-dash symbol sequence.

[0032] In one embodiment, the channel characteristic parameters include signal-to-noise ratio, multipath fading coefficient, and interference spectrum distribution;

[0033] The lightweight decision tree model establishes a mapping relationship between channel features and coding parameters based on historical transmission success rate data, and outputs the initial weight values ​​of the time-series feature weight vector.

[0034] The initial weight values ​​of the temporal feature weight vector include the initial weight of point duration, the initial weight of stroke duration, the initial weight of point-stroke interval, and the initial weight of character interval.

[0035] In one embodiment, the output process of a lightweight decision tree model includes:

[0036] Obtain historical transmission record data; historical transmission record data includes the encoding parameter configurations for different channel characteristic parameters and the corresponding transmission success rates;

[0037] Data with a transmission success rate higher than a preset success rate threshold is selected from historical transmission record data as sample data. The range of values ​​for signal-to-noise ratio, multipath fading coefficient, and interference spectrum distribution in the sample data is extracted, and the range of values ​​is divided into several sub-intervals according to the preset segmentation interval.

[0038] Based on several sub-intervals, construct the decision node judgment conditions of the decision tree; each branch node in the decision tree corresponds to a sub-interval judgment of a channel feature parameter;

[0039] The time-series feature weight vectors corresponding to the sample data that reach the same leaf node along the decision tree path are statistically analyzed, and the average value of the time-series feature weight vectors is stored in that leaf node.

[0040] When the input channel feature parameters reach a leaf node along the decision tree path, the average value stored in that leaf node is output as the initial weight value of the time-series feature weight vector.

[0041] Specifically, the decision tree branch node determination conditions include: comparing the input signal-to-noise ratio (SNR) with the boundary values ​​of each SNR sub-interval, and determining the SNR sub-interval to which the SNR belongs based on the comparison results; comparing the input multipath fading coefficient with the boundary values ​​of each fading sub-interval, and determining the fading sub-interval to which the multipath fading coefficient belongs based on the comparison results; and comparing the input interference spectrum distribution with the boundary values ​​of each interference sub-interval, and determining the interference sub-interval to which the interference spectrum distribution belongs based on the comparison results.

[0042] In one embodiment, determining the temporal feature weight vector of the current channel based on the output of the lightweight decision tree model includes:

[0043] Based on the initial weight values ​​of the temporal feature weight vector output by the lightweight decision tree model, the difference between the signal-to-noise ratio and the preset signal-to-noise ratio threshold is calculated.

[0044] If the difference is positive, the initial weights of the point duration and the initial weights of the phantom duration are reduced by the first scaling factor; if the difference is negative, the initial weights of the point duration and the initial weights of the phantom duration are increased by the second scaling factor to obtain the point duration weights and phantom duration weights.

[0045] Calculate the ratio of the multipath fading coefficient to the preset fading threshold. If the ratio is greater than the first preset ratio, the initial weight of the dot-dash interval is amplified according to the third scaling factor to obtain the dot-dash interval weight, so as to reduce symbol adhesion caused by multipath interference.

[0046] If the ratio is less than the second preset ratio, the initial weight of the dot-dash interval remains unchanged, and the dot-dash interval weight is obtained.

[0047] Extract the energy proportion of the interference signal in the interference spectrum distribution. If the energy proportion exceeds the preset energy threshold, the initial weight of the character interval is amplified according to the fourth scaling factor to obtain the character interval weight, so as to enhance the distinguishability of the character boundary.

[0048] If the energy percentage does not exceed the preset energy threshold, the initial weight of the character interval remains unchanged, and the character interval weight is obtained.

[0049] The dot duration weight, stroke duration weight, dot-stroke interval weight, and character interval weight are combined to form the temporal feature weight vector of the current channel.

[0050] It should be noted that the present invention achieves intelligent encoding and decoding of Morse code through three steps. In the first step (S1), the text is converted into a sequence of dotted and dashed symbols and the optimal encoding parameters are determined according to the channel state.

[0051] Specifically, when preparing to send the text message "SOS" at the encoding end, the text is first converted into a character sequence ['S', 'O', 'S']. Then, the standard Morse code mapping table is consulted, where the character 'S' corresponds to the dot-dash symbol combination "···" (three dots), and the character 'O' corresponds to the dot-dash symbol combination "---" (three dashes). The dot-dash symbol combinations of each character are concatenated in sequence, and character spacers are inserted between the characters. Finally, the initial dot-dash symbol sequence "···---···" is generated. This sequence contains 9 symbols (6 dots and 3 dashes), 2 character spaces, and several dot-dash spaces within the symbols.

[0052] Specifically, while generating the initial dot-dash symbol sequence, the encoding end collects the characteristic parameters of the current communication channel in real time through the radio frequency receiving module. This includes measuring the power of the received probe signal to obtain the signal-to-noise ratio, calculating the multipath fading coefficient by analyzing the multipath component delay and power distribution of the received signal, and obtaining interference spectrum distribution data by scanning the channel frequency band with a spectrum analyzer. These channel characteristic parameters reflect the current quality status of the channel and will directly affect the selection of subsequent coding parameters.

[0053] Specifically, the encoding end inputs the collected channel characteristic parameters (signal-to-noise ratio 15dB, multipath fading coefficient 0.3, interference energy percentage 8%) into a pre-trained lightweight decision tree model for inference. During the training phase, this decision tree model has established a mapping relationship between channel characteristic parameters and the optimal temporal feature weight vector based on a large amount of historical transmission record data. The model's inference process starts from the root node, first determining which sub-interval the input signal-to-noise ratio of 15dB belongs to (in this embodiment, the first-level branch of the decision tree divides the signal-to-noise ratio into three intervals: below 10dB, 10-20dB, and above 20dB; the current 15dB belongs to the middle interval). Then, it proceeds along the corresponding branch to the next level node, continuing to determine the sub-interval to which the multipath fading coefficient of 0.3 belongs (in this embodiment, the second-level branch divides the fading coefficient into less than 0.2 and 0.2-0.5). The system first divides the interference energy percentage into three intervals: less than 5%, 5-15%, and greater than 15%, with the current 0.3 belonging to the middle interval. Then, it enters the third layer to determine the interval to which the interference energy percentage of 8% belongs (in this embodiment, the third layer branch divides the energy percentage into three intervals: less than 5%, 5-15%, and greater than 15%, with the current 8% belonging to the middle interval). Finally, it reaches a certain leaf node, which stores the average value of the temporal feature weight vector corresponding to the sample with the highest historical transmission success rate under this channel condition. In this embodiment, the initial weight values ​​stored in the leaf node are: initial weight of dot duration 1.0, initial weight of stroke duration 3.0, initial weight of dot-stroke interval 1.0, and initial weight of character interval 3.0.

[0054] Specifically, after the encoding end obtains the initial value of the temporal feature weight vector output by the decision tree model, it needs to be fine-tuned according to the current channel feature parameters to adapt to the real-time changing channel conditions. First, the current signal-to-noise ratio (SNR) of 15dB is compared with the preset SNR threshold (12dB in this embodiment), and the difference is calculated to be 3dB. Since the difference is positive, it indicates that the channel quality is good. Therefore, the initial weight of the dot duration (1.0) and the initial weight of the span duration (3.0) are multiplied by the first scaling factor (0.9) to reduce the weight, resulting in a dot duration weight of 0.9 and a span duration weight of 2.7. This can speed up the transmission speed and reduce energy consumption while ensuring the accuracy of identification. Next, the ratio of the multipath fading coefficient (0.3) to the preset fading threshold (0.25) is calculated to be 1.2. This ratio is compared with the first preset ratio (1.5). It is found that 1.2 is less than 1.5, indicating that the multipath effect is not serious. Therefore, the initial weight of the dot-span interval (1.0) is kept unchanged, resulting in a dot-span interval weight of 1.0. Then, the energy percentage of the interference signal in the interference spectrum distribution is extracted to be 8%, and compared with the preset energy threshold of 10%. It is found that 8% does not exceed the threshold, indicating that the interference is relatively light. Therefore, the initial character interval weight of 3.0 is kept unchanged, resulting in a character interval weight of 3.0. Finally, the four adjusted weight values ​​are combined into the current channel's timing feature weight vector [0.9, 2.7, 1.0, 3.0].

[0055] Specifically, the encoding end adjusts the duration of each symbol in the initial dot-dash symbol sequence "···---···" based on a determined temporal feature weight vector [0.9, 2.7, 1.0, 3.0]. Assuming the baseline duration of a standard dot symbol is 50ms, the optimized dot symbol duration is calculated to be 45ms based on a dot duration weight of 0.9, the optimized dash symbol duration is calculated to be 135ms based on a dash duration weight of 2.7, the optimized symbol interval duration is calculated to be 50ms based on a dot-dash interval weight of 1.0, and the optimized character interval duration is calculated to be 150ms based on a character interval weight of 3.0. Iterate through each symbol in the initial dot-dash symbol sequence, assigning each dot symbol a value of 45ms, each dash symbol a value of 135ms, each symbol interval a value of 50ms, and each character interval a value of 150ms. Finally, an optimized dot-dash symbol sequence is generated. This sequence not only contains the dot-dash symbols themselves, but also specifies the absolute transmission duration of each symbol and interval, providing a precise timing control basis for the subsequent generation of encoded signals.

[0056] In one embodiment, such as Figure 1 As shown, the absolute transmission duration is calculated based on the optimized dot-dash symbol sequence and timing feature weight vector. The transmission power and waveform envelope parameters are dynamically adjusted in conjunction with the channel's multipath fading coefficient. The coded signal is then generated based on the adjusted dot-dash symbol sequence, including:

[0057] The absolute transmission duration of each dot-dash symbol is calculated based on the optimized dot-dash symbol sequence and the timing feature weight vector.

[0058] By combining the multipath fading coefficient of the channel, the transmission power consumption is minimized while ensuring the timing discernibility of the receiver, and the transmission power and waveform envelope parameters of each dot and dash symbol are dynamically adjusted.

[0059] The adjusted dot-dash symbol sequence is converted into a baseband signal and superimposed with a preset timing synchronization preamble to generate a complete encoded signal; the timing synchronization preamble carries the encoded information of the timing feature weight vector.

[0060] In one embodiment, calculating the absolute transmission duration of each dot-dash symbol based on the optimized dot-dash symbol sequence and the timing feature weight vector includes:

[0061] The standard dot symbol reference duration is determined based on the optimized dot-dash symbol sequence. The standard dot symbol reference duration is multiplied by the dot duration weight to obtain the absolute transmission duration of the dot symbol.

[0062] Specifically, the standard dot symbol reference duration is determined based on the target transmission rate during Morse code encoding, as shown in Table 1; the target transmission rate is calculated by dividing the number of standard words transmitted per minute by the total number of time units of the standard words.

[0063] Table 1. Transmission Rate vs. Reference Duration

[0064] Transmission rate (wpm) Standard dot symbol reference duration (ms) 10 120 25 48 50 24 100 12

[0065] Multiply the standard dot symbol base duration by the weighted dot duration to obtain the absolute transmission duration of the dot symbol;

[0066] Multiply the standard dot symbol reference duration by the dot-dash interval weight to obtain the absolute transmission duration of the interval between symbols;

[0067] Multiply the standard dot symbol base duration by the character interval weight to obtain the absolute transmission duration of the character interval;

[0068] Iterate through each symbol in the optimized dot-dash symbol sequence and assign the corresponding absolute transmission duration based on the symbol type.

[0069] Specifically, if the current symbol is a dot symbol, the absolute transmission duration of the dot symbol is assigned; if the current symbol is a dash symbol, the absolute transmission duration of the dash symbol is assigned; if the current position is a symbol interval, the absolute transmission duration of the symbol interval is assigned; and if the current position is a character interval, the absolute transmission duration of the character interval is assigned.

[0070] In one embodiment, by combining the multipath fading coefficient of the channel and minimizing transmission power consumption while ensuring the timing discriminability of the receiver, the transmission power and waveform envelope parameters of each dot and dash symbol are dynamically adjusted, including:

[0071] Calculate the minimum identifiable power threshold corresponding to the multipath fading coefficient; the minimum identifiable power threshold is the lowest transmit power value that ensures the receiver can distinguish adjacent symbols;

[0072] Traverse and optimize the dot-dash symbol sequence, and calculate the duration redundancy based on the absolute transmission duration of each symbol; the duration redundancy is the difference between the absolute transmission duration of the symbol and the standard minimum identifiable duration.

[0073] If the duration redundancy is greater than the preset redundancy threshold, the transmission power of the symbol is set to the minimum identifiable power threshold to reduce transmission energy consumption.

[0074] If the duration redundancy is less than or equal to the preset redundancy threshold, the transmission power of the symbol is set to the first preset power value, and the first preset power value is higher than the minimum identifiable power threshold.

[0075] Based on the frequency response characteristics of the multipath fading coefficient, determine the rising and falling slopes of the waveform envelope parameters; specifically including:

[0076] If the delay spread corresponding to the multipath fading coefficient exceeds the preset delay threshold, the rising edge slope and falling edge slope are reduced to make the waveform envelope of the symbol exhibit a gradual change characteristic in order to reduce inter-symbol interference caused by multipath effect; if the delay spread corresponding to the multipath fading coefficient does not exceed the preset delay threshold, the rising edge slope and falling edge slope are kept at the preset standard slope value.

[0077] The adjusted transmission power and waveform envelope parameters of each symbol are correlated with the absolute transmission duration of that symbol to obtain the adjusted dot-dash symbol sequence.

[0078] It should also be noted that after optimizing the coding parameters, the present invention needs to convert the dot-dash symbol sequence into a transmittable physical signal. In the second step (S2), the precise transmission duration is calculated based on the optimized symbol sequence, and the power and waveform parameters are adjusted in combination with the channel fading characteristics to balance transmission reliability and energy consumption, and a complete coded signal carrying timing weight information is generated.

[0079] Specifically, the encoding end calculates the absolute transmission duration of each symbol based on the optimized dot-dash symbol sequence and the timing feature weight vector [0.9, 2.7, 1.0, 3.0]. Assuming the current target transmission rate is 25wpm, the standard dot symbol baseline duration is found to be 48ms from Table 1. This 48ms is then multiplied by the four weight values: the absolute transmission duration of the dot symbol is 43.2ms, the absolute transmission duration of the dash symbol is 129.6ms, the absolute transmission duration of the symbol interval is 48ms, and the absolute transmission duration of the character interval is 144ms. Then, the optimized dot-dash symbol sequence "···---···" is traversed, assigning 43.2ms to each dot symbol, 129.6ms to each dash symbol, 48ms to the symbol interval, and 144ms to the character interval, thus completing the duration assignment for all symbols.

[0080] Specifically, the encoder dynamically adjusts the transmission power and waveform envelope parameters of each symbol based on the multipath fading coefficient of 0.3. First, the minimum identifiable power threshold of 100mW is calculated based on the multipath fading coefficient of 0.3. Then, the duration redundancy (the difference between the absolute transmission duration and the standard minimum identifiable duration of 30ms) of each symbol is calculated by traversing the dotted symbol sequence. Taking the first dotted symbol as an example, its duration redundancy is 13.2ms, which is greater than the preset redundancy threshold of 10ms. Therefore, the transmission power of this symbol is set to 100mW to reduce energy consumption. If the duration redundancy of a symbol is less than 10ms, the transmission power is increased to 150mW to ensure reliability. Next, the delay spread is calculated to be 15μs based on the multipath fading coefficient of 0.3, which is less than the preset delay threshold of 20μs. Therefore, the rising and falling edge slopes are maintained at the preset standard values. If the delay spread exceeds 20μs, the slope needs to be reduced to make the waveform envelope exhibit a gradually changing characteristic to reduce inter-symbol interference. The transmission power and waveform envelope parameters of each symbol are correlated with its absolute transmission duration to form an adjusted dot-dash symbol sequence.

[0081] Specifically, the encoding end generates an encoded signal based on the adjusted dot-dash symbol sequence. First, the dot-dash symbol sequence is converted into a baseband signal, and a corresponding waveform segment is generated according to the parameters of each symbol: taking the first dot symbol as an example, it linearly rises from 0V to 1.0V (corresponding to 100mW power) within 5ms, maintains 1.0V for 33.2ms (43.2ms minus rise and fall time), and then linearly drops to 0V within 5ms, forming a trapezoidal pulse; waveform segments of all symbols are generated in sequence and spliced ​​together to obtain the complete baseband signal. Then, the four floating-point quantizations of the timing feature weight vector [0.9, 2.7, 1.0, 3.0] are quantized into 32-bit binary numbers (quantization factor 255), and encoded into a 100ms preamble waveform using DBPSK modulation. The preamble begins with a 16-bit synchronization sequence for easy identification by the receiver. Finally, the preamble is superimposed on the baseband signal to generate a complete coded signal (preamble approximately 100ms + Morse code baseband signal approximately 975ms, total duration approximately 1075ms). This coded signal is then sent to the RF modulation module for carrier modulation and transmitted through the antenna.

[0082] In one embodiment, such as Figure 2 As shown, the received signal is demodulated to obtain the timing feature weight vector used by the transmitter. Envelope detection is performed on the baseband signal, and the dot-dash symbol sequence is obtained by segmentation based on the timing feature weight vector. The dot-dash symbol sequence is then converted according to the standard Morse code mapping table to obtain the decoded text information, including:

[0083] The received radio frequency signal is acquired and demodulated to obtain the baseband signal. The timing synchronization preamble in the baseband signal is extracted and parsed to obtain the timing feature weight vector used by the transmitter as the initial decoding reference.

[0084] Envelope detection is performed on the baseband signal to obtain a timing waveform sequence, and the duration distribution of continuous high-level segments and continuous low-level segments in the timing waveform sequence is statistically analyzed.

[0085] Peak features in the duration distribution are identified by probability density estimation, and the peak durations corresponding to the peak features are used as candidate values ​​for point duration and wave duration.

[0086] The dot-dash segmentation threshold and character segmentation threshold are determined based on the temporal feature weight vector in the initial decoding reference, and the temporal waveform sequence is segmented to obtain the dot-dash symbol sequence;

[0087] The dotted-dash symbol sequence is converted into a character sequence according to the standard Morse code mapping table to obtain the decoded text information.

[0088] In one embodiment, the dot-dash segmentation threshold and character segmentation threshold are determined based on the temporal feature weight vector in the initial decoding reference, and the temporal waveform sequence is segmented to obtain a dot-dash symbol sequence, including:

[0089] Calculate the duration ratio between candidate point duration values ​​and candidate stroke duration values, and calculate the matching degree between the duration ratio and the point duration weight and stroke duration weight in the initial decoding reference;

[0090] If the matching degree is lower than the preset matching threshold, the temporal feature weight vector is dynamically adjusted. Specifically, this includes: analyzing the temporal waveform sequence through a sliding time window, calculating the deviation between the average actual point duration within the current window and the midpoint duration weight of the initial decoding reference; if the absolute value of the deviation exceeds the first preset deviation threshold, the point duration weight is corrected according to the fifth scaling factor; calculating the deviation between the average actual span duration within the current window and the span duration weight of the initial decoding reference; if the absolute value of the deviation exceeds the second preset deviation threshold, the span duration weight is corrected according to the sixth scaling factor.

[0091] Multiply the corrected dot duration weight by the preset standard dot symbol baseline duration to obtain the dot-dash segmentation threshold; multiply the corrected character interval weight by the preset standard dot symbol baseline duration to obtain the character segmentation threshold.

[0092] It should be noted that the present invention requires the accurate restoration of the encoding parameters of the sending end and the recovery of the original text information from the received signal at the receiving end. In the third step (S3), the timing weight information used by the sending end is extracted from the received signal as the decoding benchmark. The actual duration characteristics of the dot-dash symbols are identified through envelope detection and probability density analysis. The segmentation threshold is dynamically determined to accurately restore the dot-dash symbol sequence and finally decode it into text information.

[0093] Specifically, after the receiving antenna receives the radio frequency signal, the carrier signal is demodulated into a baseband signal through the radio frequency demodulation module. The receiving end searches for a special synchronization sequence (a 16-bit alternating 1010 sequence) at the beginning position of the baseband signal. After locating the start position of the preamble, the receiving end extracts the timing synchronization preamble waveform of the following 100ms, demodulates the preamble into 32-bit binary data, and then divides these 32 bits into four groups of 8 bits each. These groups are then divided by a quantization factor of 255 to obtain four floating-point numbers: 0.9, 2.7, 1.0, and 3.0. The timing feature weight vector [0.9, 2.7, 1.0, 3.0] used by the transmitting end is recovered as the initial decoding reference, providing a benchmark for subsequent symbol recognition.

[0094] Specifically, the receiving end performs envelope detection on the baseband signal after removing the preamble to extract timing information. The envelope detector uses rectification and low-pass filtering to binarize the envelope curve into a timing waveform sequence by using a preset detection threshold (in this embodiment, an envelope amplitude greater than 0.5V is determined as a high level, and less than 0.3V is determined as a low level). The entire timing waveform sequence is traversed, and the duration of all consecutive high-level segments (representing dots or dashes) and the duration of all consecutive low-level segments (representing various intervals) are counted to obtain duration distribution data.

[0095] Specifically, the receiver identifies the typical durations of dot and dash symbols from the duration distribution using a probability density estimation method. A histogram is constructed from the durations of all high-level segments obtained statistically, with a duration interval step size of 2ms. The frequency of occurrence within each interval is counted, and then the histogram data is processed using a peak detection algorithm: a local maximum search method is used to traverse all duration intervals. If the frequency of a certain interval is simultaneously greater than the frequencies of its left and right adjacent intervals and exceeds a preset minimum peak frequency threshold, then that interval is marked as a peak interval. The average value of all duration samples within that interval is calculated as a candidate value for the dash duration.

[0096] Specifically, the receiver determines the segmentation threshold based on the timing feature weight vector [0.9, 2.7, 1.0, 3.0] in the initial decoding reference and the candidate values ​​for dot duration (43ms) and stroke duration (130ms) extracted from the actual signal. First, the matching degree between the actual duration ratio and the theoretical ratio is calculated. In the above embodiment, the matching degree is higher than the preset matching threshold of 95%, indicating that the timing characteristics of the actual received signal are basically consistent with the encoding parameters of the transmitter, and the initial decoding reference can be used directly without adjustment. If the matching degree is lower than 95%, a dynamic adjustment mechanism is activated, using a sliding time window to statistically calculate the average actual duration of the dotted and dashed symbols. If the absolute value of the deviation from the weight value in the initial decoding reference exceeds the preset deviation threshold (dot duration deviation exceeds 5ms or stroke duration deviation exceeds 15ms), the weight value is corrected according to the corresponding scaling factor. The timing waveform sequence is traversed, and the high-level and low-level segments are classified according to the two segmentation thresholds to obtain the dotted and dashed symbol sequence "···---···".

[0097] Specifically, the receiving end converts the dot-dash symbol sequence into text information according to the standard Morse code mapping table. The dot-dash symbol sequence "···---···" is divided into three subsequences according to the character separators: "···", "---", and "···". Then, the subsequence "···" is looked up one by one in the standard Morse code mapping table: the subsequence "···" corresponds to the character 'S', the subsequence "---" corresponds to the character 'O', and the subsequence "···" corresponds to the character 'S'. These three characters are combined in order to obtain the character sequence ['S', 'O', 'S'], which is then concatenated to form the final decoded text information "SOS", completing the entire decoding process.

[0098] like Figure 4 As shown, according to another embodiment of the present invention, an artificial intelligence-based Morse code encoding and decoding system is also provided, the artificial intelligence-based Morse code encoding and decoding system comprising:

[0099] Symbol generation module 1 is used to convert text information into an initial dot-dash symbol sequence according to the standard Morse code mapping table, and to collect channel feature parameters and input them into a lightweight decision tree model to determine the temporal feature weight vector; and to optimize the initial dot-dash symbol sequence to generate an optimized dot-dash symbol sequence.

[0100] Signal encoding module 2 is used to calculate the absolute transmission duration based on the optimized dot-dash symbol sequence and timing feature weight vector, dynamically adjust the transmission power and waveform envelope parameters in combination with the multipath fading coefficient of the channel, and generate an encoded signal based on the adjusted dot-dash symbol sequence.

[0101] Signal decoding module 3 is used to demodulate the received signal to obtain the timing feature weight vector used by the transmitter, perform envelope detection on the baseband signal and segment it according to the timing feature weight vector to obtain the dot-dash symbol sequence, and convert the dot-dash symbol sequence according to the standard Morse code mapping table to obtain the decoded text information.

[0102] 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 principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A Morse code encoding and decoding method based on artificial intelligence, characterized in that, include: The text information is converted into an initial dot-dash symbol sequence according to the standard Morse code mapping table, and the channel feature parameters are collected and input into the lightweight decision tree model to determine the temporal feature weight vector; The initial dot-dash symbol sequence is optimized to generate an optimized dot-dash symbol sequence; the channel feature parameters include signal-to-noise ratio, multipath fading coefficient, and interference spectrum distribution; the temporal feature weight vector includes dot duration weight, dash duration weight, dot-dash interval weight, and character interval weight. The absolute transmission duration is calculated based on the optimized dot-dash symbol sequence and timing feature weight vector. The transmission power and waveform envelope parameters are dynamically adjusted in conjunction with the channel's multipath fading coefficient. An coded signal is then generated based on the adjusted dot-dash symbol sequence. Specifically, this includes: calculating the absolute transmission duration of each dot-dash symbol based on the optimized dot-dash symbol sequence and timing feature weight vector; dynamically adjusting the transmission power and waveform envelope parameters of each dot-dash symbol to minimize transmission power consumption while ensuring timing discriminability at the receiver, in conjunction with the channel's multipath fading coefficient; converting the adjusted dot-dash symbol sequence into a baseband signal and superimposing a preset timing synchronization preamble to generate a complete coded signal; the timing synchronization preamble carries the encoding information of the timing feature weight vector. The received signal is demodulated to obtain the timing feature weight vector used by the transmitter. The baseband signal is envelope detected and segmented according to the timing feature weight vector to obtain the dot-dash symbol sequence. The dot-dash symbol sequence is converted according to the standard Morse code mapping table to obtain the decoded text information. The step of calculating the absolute transmission duration of each dot-dash symbol based on the optimized dot-dash symbol sequence and the timing feature weight vector includes: determining the standard dot symbol baseline duration based on the optimized dot-dash symbol sequence; multiplying the standard dot symbol baseline duration by the dot duration weight to obtain the absolute transmission duration of the dot symbol; multiplying the standard dot symbol baseline duration by the dash duration weight to obtain the absolute transmission duration of the dash symbol; multiplying the standard dot symbol baseline duration by the dot-dash interval weight to obtain the absolute transmission duration of the interval between symbols; multiplying the standard dot symbol baseline duration by the character interval weight to obtain the absolute transmission duration of the interval between characters; and traversing each symbol in the optimized dot-dash symbol sequence and assigning the corresponding absolute transmission duration according to the symbol type.

2. The Morse code encoding and decoding method based on artificial intelligence according to claim 1, characterized in that, The text information is converted into an initial dot-dash symbol sequence according to a standard Morse code mapping table, and channel feature parameters are collected and input into a lightweight decision tree model to determine the temporal feature weight vector; The initial dot-dash symbol sequence is optimized to generate an optimized dot-dash symbol sequence, which includes: The text information to be sent is converted into a character sequence, and each character in the character sequence is mapped to a corresponding dot-dash symbol combination according to the standard Morse code mapping table to generate an initial dot-dash symbol sequence. The channel feature parameters of the current channel are collected in real time and input into a pre-trained lightweight decision tree model; Based on the output of the lightweight decision tree model, determine the temporal feature weight vector of the current channel; The duration of each symbol in the initial dot-dash symbol sequence is weighted and adjusted based on the temporal feature weight vector to generate an optimized dot-dash symbol sequence.

3. The Morse code encoding and decoding method based on artificial intelligence according to claim 2, characterized in that, The lightweight decision tree model establishes a mapping relationship between channel features and coding parameters based on historical transmission success rate data, and outputs the initial weight values ​​of the time-series feature weight vector. The initial weight values ​​of the temporal feature weight vector include the initial weight of point duration, the initial weight of stroke duration, the initial weight of point-stroke interval, and the initial weight of character interval.

4. The Morse code encoding and decoding method based on artificial intelligence according to claim 2, characterized in that, The output process of the lightweight decision tree model includes: Acquire historical transmission record data; the historical transmission record data includes the encoding parameter configurations for different channel characteristic parameters and the corresponding transmission success rates; Data with a transmission success rate higher than a preset success rate threshold is selected from historical transmission record data as sample data. The range of values ​​for signal-to-noise ratio, multipath fading coefficient, and interference spectrum distribution in the sample data is extracted, and the range of values ​​is divided into several sub-intervals according to the preset segmentation interval. Based on several sub-intervals, the decision tree's branch node determination conditions are constructed; each branch node in the decision tree corresponds to a sub-interval determination of a channel feature parameter. The time-series feature weight vectors corresponding to the sample data that reach the same leaf node along the decision tree path are statistically analyzed, and the average value of the time-series feature weight vectors is stored in that leaf node. When the input channel feature parameters reach a leaf node along the decision tree path, the average value stored in that leaf node is output as the initial weight value of the time-series feature weight vector.

5. The Morse code encoding and decoding method based on artificial intelligence according to claim 2, characterized in that, The step of determining the temporal feature weight vector of the current channel based on the output of the lightweight decision tree model includes: Based on the initial weight values ​​of the temporal feature weight vector output by the lightweight decision tree model, the difference between the signal-to-noise ratio and the preset signal-to-noise ratio threshold is calculated. If the difference is positive, the initial weights of the point duration and the initial weights of the phantom duration are reduced by the first scaling factor; if the difference is negative, the initial weights of the point duration and the initial weights of the phantom duration are increased by the second scaling factor to obtain the point duration weights and phantom duration weights. Calculate the ratio of the multipath fading coefficient to the preset fading threshold. If the ratio is greater than the first preset ratio, the initial weight of the dot-dash interval is amplified according to the third scaling factor to obtain the dot-dash interval weight, so as to reduce symbol adhesion caused by multipath interference. If the ratio is less than the second preset ratio, the initial weight of the dot-dash interval remains unchanged, and the dot-dash interval weight is obtained. Extract the energy proportion of the interference signal in the interference spectrum distribution. If the energy proportion exceeds the preset energy threshold, the initial weight of the character interval is amplified according to the fourth scaling factor to obtain the character interval weight, so as to enhance the distinguishability of the character boundary. If the energy percentage does not exceed the preset energy threshold, the initial weight of the character interval remains unchanged, and the character interval weight is obtained. The dot duration weight, stroke duration weight, dot-stroke interval weight, and character interval weight are combined to form the temporal feature weight vector of the current channel.

6. The Morse code encoding and decoding method based on artificial intelligence according to claim 1, characterized in that, The multipath fading coefficient of the combined channel, while ensuring the timing discriminability of the receiver, minimizes transmission power consumption, and dynamically adjusts the transmission power and waveform envelope parameters of each dot and dash symbol, including: Calculate the minimum identifiable power threshold corresponding to the multipath fading coefficient; the minimum identifiable power threshold is the lowest transmit power value that ensures the receiver can distinguish adjacent symbols; Traverse and optimize the dot-dash symbol sequence, and calculate the duration redundancy based on the absolute transmission duration of each symbol; the duration redundancy is the difference between the absolute transmission duration of the symbol and the standard minimum identifiable duration. If the duration redundancy is greater than the preset redundancy threshold, the transmission power of the symbol is set to the minimum identifiable power threshold to reduce transmission energy consumption. If the duration redundancy is less than or equal to the preset redundancy threshold, the transmission power of the symbol is set to the first preset power value, and the first preset power value is higher than the minimum identifiable power threshold. Based on the frequency response characteristics of the multipath fading coefficient, determine the rising slope and falling slope of the waveform envelope parameters; The adjusted transmission power and waveform envelope parameters of each symbol are correlated with the absolute transmission duration of that symbol to obtain the adjusted dot-dash symbol sequence.

7. The Morse code encoding and decoding method based on artificial intelligence according to claim 1, characterized in that, The demodulated received signal is used to obtain the timing feature weight vector used by the transmitter. Envelope detection is performed on the baseband signal, and the signal is segmented according to the timing feature weight vector to obtain a dotted-dash symbol sequence. The dotted-dash symbol sequence is then converted according to a standard Morse code mapping table to obtain the decoded text information, including: The received radio frequency signal is acquired and demodulated to obtain the baseband signal. The timing synchronization preamble in the baseband signal is extracted and parsed to obtain the timing feature weight vector used by the transmitter as the initial decoding reference. Envelope detection is performed on the baseband signal to obtain a timing waveform sequence, and the duration distribution of continuous high-level segments and continuous low-level segments in the timing waveform sequence is statistically analyzed. Peak features in the duration distribution are identified by probability density estimation, and the peak durations corresponding to the peak features are used as candidate values ​​for point duration and wave duration. The dot-dash segmentation threshold and character segmentation threshold are determined based on the temporal feature weight vector in the initial decoding reference, and the temporal waveform sequence is segmented to obtain the dot-dash symbol sequence; The dotted-dash symbol sequence is converted into a character sequence according to the standard Morse code mapping table to obtain the decoded text information.

8. An artificial intelligence-based Morse code encoding and decoding system, used to implement the artificial intelligence-based Morse code encoding and decoding method according to any one of claims 1-7, characterized in that, The system includes: The symbol generation module is used to convert text information into an initial dot-dash symbol sequence according to a standard Morse code mapping table, and to collect channel feature parameters into a lightweight decision tree model to determine the temporal feature weight vector; the initial dot-dash symbol sequence is optimized to generate an optimized dot-dash symbol sequence; the channel feature parameters include signal-to-noise ratio, multipath fading coefficient, and interference spectrum distribution; the temporal feature weight vector includes dot duration weight, dash duration weight, dot-dash interval weight, and character interval weight; The signal encoding module is used to calculate the absolute transmission duration based on the optimized dot-dash symbol sequence and the timing feature weight vector, dynamically adjust the transmission power and waveform envelope parameters in conjunction with the multipath fading coefficient of the channel, and generate an encoded signal based on the adjusted dot-dash symbol sequence. Specifically, it includes: calculating the absolute transmission duration of each dot-dash symbol based on the optimized dot-dash symbol sequence and the timing feature weight vector; dynamically adjusting the transmission power and waveform envelope parameters of each dot-dash symbol to minimize transmission power consumption while ensuring the timing discriminability of the receiver, in conjunction with the multipath fading coefficient of the channel; converting the adjusted dot-dash symbol sequence into a baseband signal and superimposing a preset timing synchronization preamble to generate a complete encoded signal; the timing synchronization preamble carries the encoding information of the timing feature weight vector. The signal decoding module is used to demodulate the received signal to obtain the timing feature weight vector used by the transmitter, perform envelope detection on the baseband signal and segment it according to the timing feature weight vector to obtain the dot-dash symbol sequence, and convert the dot-dash symbol sequence according to the standard Morse code mapping table to obtain the decoded text information. The step of calculating the absolute transmission duration of each dot-dash symbol based on the optimized dot-dash symbol sequence and the timing feature weight vector includes: determining the standard dot symbol baseline duration based on the optimized dot-dash symbol sequence; multiplying the standard dot symbol baseline duration by the dot duration weight to obtain the absolute transmission duration of the dot symbol; multiplying the standard dot symbol baseline duration by the dash duration weight to obtain the absolute transmission duration of the dash symbol; multiplying the standard dot symbol baseline duration by the dot-dash interval weight to obtain the absolute transmission duration of the interval between symbols; multiplying the standard dot symbol baseline duration by the character interval weight to obtain the absolute transmission duration of the interval between characters; and traversing each symbol in the optimized dot-dash symbol sequence and assigning the corresponding absolute transmission duration according to the symbol type.

Images