Ship mail adaptive coding and dynamic bandwidth allocation method

Abstract

The invention relates to a ship mail adaptive coding and dynamic bandwidth allocation method, belongs to mobile voice service, mobile data communication service and other telecommunication services, and particularly relates to the field of digital information transmission. The method comprises the following steps: intelligently predicting the total number of transmitted mails and the total number of transmitted mail data of a target ship in a current time segment by using an AI network data prediction model according to various basic data which are fully and comprehensively selected; and determining an adaptive coding strategy and a dynamic bandwidth allocation strategy for the current time segment of the target ship based on the intelligent prediction result. According to the invention, in order to solve the technical problems that effective and smooth transmission of all mail data of each future time segment of each ship is difficult to consider and the utilization rate of network transmission resources is difficult to improve, A I network data prediction models of different structures customized and designed for different ships are used; and intelligent prediction of the mail transmission state of each ship in the future time segment is completed, so that the technical problem is solved.

Description

TECHNICAL FIELD

[0001] The mobile voice service, mobile data communication service and other telecommunication services of the present application, more specifically, relate to the field of digital information transmission, and particularly relate to a ship mail adaptive coding and dynamic bandwidth allocation method. BACKGROUND

[0002] Digital information transmission is an important branch of technology in mobile voice service, mobile data communication service and other telecommunication services, and the transmission of ship mail is a specific application of digital information transmission mode, which has specific technical characteristics and requirements in its special scenario. It is particularly worth noting that due to the limited and inefficient nature of maritime communication resources, higher requirements need to be placed on the transmission coding method and dynamic bandwidth allocation method of ship mail in order to ensure the reliable and smooth transmission of each mail of each ship in different time segments in different sea areas.

[0003] For example, Chinese invention patent publication CN112653613A proposes a system and method for implementing ocean-going ship mail based on Beidou short message. Specifically, the ship terminal subsystem obtains the mail content and mail address input by the client and encapsulates it as a Beidou short message to be sent to the Beidou satellite; the Beidou command machine receives the Beidou short message from the Beidou satellite and restores it to a mail data packet, and the ground mail service center receives the mail sent by the Beidou command machine and the Internet, and according to the recipient mapping list, finds the mail address belonging to the Internet address or a certain ocean-going ship; the mail belonging to the Internet address is forwarded to the corresponding Internet address through the Internet; the mail belonging to the ocean-going ship is packaged and sent to the Beidou command machine, and forwarded to the ship terminal subsystem corresponding to the Beidou card user address through the Beidou satellite. On the ocean-going ship, without other communication satellites and communication links, through the short message communication service provided by the Beidou satellite, the user on the ship can realize the mail receiving function and improve the office efficiency.

[0004] For example, the Chinese invention patent publication CN102055515A proposes a satellite short data communication method, device and system between ship and shore, which comprises a ship satellite terminal, a satellite space segment, a ground station, a land terminal and a satellite short data communication device. The satellite space segment comprises maritime satellites covering the Indian Ocean region and the Pacific Ocean region, and maritime satellites covering the Atlantic Ocean region. The ground station comprises a first maritime satellite ground station communicating with the maritime satellites covering the Indian Ocean region and the Pacific Ocean region, and a second maritime satellite ground station communicating with the maritime satellites covering the Atlantic Ocean. The satellite short data communication device is connected with the first and second ground stations and the land user terminal, receives the email sent by the land terminal, selects to send the email to the first or second ground station according to the state of the ship satellite terminal, or sends the email received from the first or second ground station to the land terminal. The system realizes the email communication between the global navigation ship and the land user terminal.

[0005] Therefore, the communication scheme of the ship mail designed in the prior art basically adopts special processing of the maritime communication in terms of link optimization and service optimization in order to avoid the maritime communication resources with limitedness and low efficiency. Since the total number of transmitted mails and the total amount of transmitted mail data of each ship in different maritime areas in future time segments cannot be intelligently predicted, it is difficult to ensure the effective smooth transmission of all mail data of each ship in each future time segment, the utilization rate of network transmission resources cannot be improved, and the limited maritime network transmission resources are easily wasted. SUMMARY

[0006] In order to solve the technical problems in the related field, the present application provides a ship mail adaptive coding and dynamic bandwidth allocation method, which intelligently predicts the total number of transmitted mails and the total amount of transmitted mail data of a target ship in a current time segment by adopting an AI prediction mode, and determines an adaptive coding strategy and a dynamic bandwidth allocation strategy for the target ship in the current time segment based on the intelligent prediction result. The total number of network slices allocated to the target ship in the current time segment is equal to the total number of transmitted mails of the target ship in the current time segment intelligently predicted, and the mail transmission bandwidth value allocated to the target ship in the current time segment is a set multiple of the total amount of transmitted mail data of the target ship in the current time segment intelligently predicted, so as to realize the effective smooth transmission of all mail data of each ship in each future time segment and the improvement of the utilization rate of network transmission resources.

[0007] According to the present application, a ship mail adaptive coding and dynamic bandwidth allocation method is provided, which comprises:

[0008] Obtaining the number of passengers, the male and female ratio of the passengers, the average age of the passengers, the set sailing speed and the distance from the coastline of the current sea area of the target ship as the ship-related information of the target ship;

[0009] Obtaining the maximum transmission bandwidth, the average transmission bandwidth, the overall computing performance data, the average storage capacity data and the average daily online duration of each registered network terminal of the target ship as the network terminal data of the target ship;

[0010] Performing a plurality of learning operations on the deep neural network to obtain a deep neural network after the plurality of learning operations, and outputting the deep neural network after the plurality of learning operations as an AI network data prediction model;

[0011] Collecting a plurality of pieces of ship mail data corresponding to a plurality of historical time segments before the current time of the target ship, and the ship mail data corresponding to each time segment is the total number of transmitted mails and the total amount of transmitted mail data of the target ship in the time segment;

[0012] Intelligently predicting the total number of transmitted mails and the total amount of transmitted mail data of the target ship in the current time segment by using the AI network data prediction model according to the total number of each registered network terminal of the target ship, the occupied duration of each time segment, the ship-related information and the network terminal data of the target ship, and the plurality of pieces of ship mail data corresponding to a plurality of historical time segments before the current time of the target ship;

[0013] Based on the intelligent prediction result, determining the adaptive coding strategy and the dynamic bandwidth allocation strategy for the target ship in the current time segment.

[0014] Compared with the prior art, the present application has at least the following five outstanding substantive features:

[0015] Substantive feature A: an AI prediction mode is used to intelligently predict the total number of transmitted mails and the total amount of transmitted mail data of the target ship in the current time segment, and based on the intelligent prediction result, the adaptive coding strategy and the dynamic bandwidth allocation strategy for the target ship in the current time segment are determined, wherein the total number of network slices allocated to the target ship in the current time segment is equal to the total number of transmitted mails of the target ship in the current time segment intelligently predicted, and the mail transmission bandwidth value allocated to the target ship in the current time segment is a set multiple of the total amount of transmitted mail data of the target ship in the current time segment intelligently predicted, so as to ensure the effective and smooth transmission of all mail data of the target ship in each future time segment, improve the utilization rate of network transmission resources, and avoid the waste of limited network transmission resources at sea;

[0016] Essential feature B: In order to realize intelligent prediction of the total number of transmission mails and the total amount of transmission mail data of the target ship in the current time segment, an AI network data prediction model designed for the target ship is used, the AI network data prediction model is a deep neural network after multiple learning operations, the number of learning operations is positively correlated with the total number of registered network terminals of the target ship, the deep neural network includes multiple hidden layers, a single input layer and a single output layer, the hidden layers are between the input layer and the output layer, and the total number of hidden layers is positively correlated with the average daily online duration of each registered network terminal. The different custom structures of the AI network data prediction model designed for different ships ensure the reliability and stability of the intelligent prediction results.

[0017] Essential feature C: In order to realize intelligent prediction of the total number of transmission mails and the total amount of transmission mail data of the target ship in the current time segment, various basic data are used, including the total number of registered network terminals of the target ship, the occupied time length of each time segment, the various ship-related information of the target ship and the multiple network terminal data, and the multiple ship mail data corresponding to multiple historical time segments before the current time of the target ship. The sufficient and comprehensive selection of the above-mentioned various basic data further ensures the reliability and stability of the intelligent prediction results.

[0018] Essential feature D: Specifically, the various ship-related information of the target ship is the number of passengers, the male-female ratio of passengers, the average passenger age, the set sailing speed and the distance from the coastline of the current sea area of the target ship. The multiple network terminal data of the target ship is the maximum transmission bandwidth, the average transmission bandwidth, the overall operation performance data, the storage capacity data mean and the average daily online duration of each registered network terminal of the target ship. The ship mail data corresponding to each time segment is the total number of transmission mails and the total amount of transmission mail data of the target ship in the time segment. The number of historical time segments is proportional to the average transmission bandwidth of each registered network terminal of the target ship, thereby providing a specific data structure design for the sufficient and comprehensive selection of various basic data.

[0019] Essential feature E: In each learning operation performed on the deep neural network, the known total number of transmission mails and the total amount of transmission mail data of the target ship in a certain historical time segment are taken as two output contents of the deep neural network, and the total number of registered network terminals of the target ship, the occupied time length of each time segment, the various ship-related information of the target ship and the multiple network terminal data, and the multiple ship mail data corresponding to multiple historical time segments before the certain historical time segment of the target ship are taken as multiple input contents of the deep neural network. Complete this learning operation, thereby ensuring the learning effect of each learning operation of the deep neural network. BRIEF DESCRIPTION OF DRAWINGS

[0020] Embodiments of the present application will be described below with reference to the accompanying drawings, in which:

[0021] Figure 1 A working scenario diagram of a ship mail adaptive coding and dynamic bandwidth allocation method according to the present application.

[0022] Figure 2 A step flow chart of a ship mail adaptive coding and dynamic bandwidth allocation method according to Embodiment 1 of the present application.

[0023] Figure 3 A step flow chart of a ship mail adaptive coding and dynamic bandwidth allocation method according to Embodiment 2 of the present application.

[0024] Figure 4 A step flow chart of a ship mail adaptive coding and dynamic bandwidth allocation method according to Embodiment 3 of the present application.

[0025] Figure 5 A step flow chart of a ship mail adaptive coding and dynamic bandwidth allocation method according to Embodiment 4 of the present application.

[0026] Figure 6 A step flow chart of a ship mail adaptive coding and dynamic bandwidth allocation method according to Embodiment 5 of the present application. DETAILED DESCRIPTION

[0027] As shown in Figure 1 , a working scenario diagram of a ship mail adaptive coding and dynamic bandwidth allocation method according to the present application is given. The mobile voice service, mobile data communication service and other telecommunication services of the present application are more specifically related to the field of digital information transmission.

[0028] In Figure 1 , the target ship and other ships are all in the offshore field, i.e. when the distance of the sea area where the target ship and other ships are currently located from the coastline is less than or equal to the set mileage threshold value, the communication network relied on by the target ship and other ships can be a mobile communication network. Specifically, the design of the network transmission channel of the mobile communication network is achieved by using shore base station B and shore base station A respectively. Of course, a relatively high-cost satellite communication network can also be used, for example, in Figure 1 , the other ships can simultaneously use the mobile communication network and the satellite communication network.

[0029] and in Figure 1In the specific embodiment, the satellite A, the satellite B and the satellite C used by the satellite communication network represent different satellite positioning modes, such as a Beidou positioning mode, a Galileo positioning mode and a GPS positioning mode.

[0030] The specific technical process of the present application is as follows:

[0031] Technical process one: in order to realize intelligent prediction of the total number of transmission mails and the total amount of transmission mail data of the target ship in the current time segment, an AI network data prediction model with a customized structure designed for the target ship is used. Since the current time segment takes the current time as the starting time, the current time segment belongs to a future time segment.

[0032] Specifically, the customized structure design of the AI network data prediction model mainly reflects in the following aspects:

[0033] First aspect: the AI network data prediction model used is a deep neural network after multiple learning operations, which includes multiple hidden layers, a single input layer and a single output layer, and the hidden layers are between the input layer and the output layer.

[0034] Second aspect: in the deep neural network used, the total number of hidden layers is positively correlated with the average daily online duration of each registered network terminal;

[0035] For example, the average daily online duration of each registered network terminal is 5 hours, the total number of hidden layers selected is 3, the average daily online duration of each registered network terminal is 7 hours, the total number of hidden layers selected is 5, the average daily online duration of each registered network terminal is 9 hours, the total number of hidden layers selected is 7, and so on.

[0036] Third aspect: the number of learning operations of the deep neural network selected is positively correlated with the total number of registered network terminals of the target ship;

[0037] For example, the total number of registered network terminals of the target ship is 200, the number of learning operations performed on the deep neural network is 1000, the total number of registered network terminals of the target ship is 300, the number of learning operations performed on the deep neural network is 1500, the total number of registered network terminals of the target ship is 400, the number of learning operations performed on the deep neural network is 2000, the total number of registered network terminals of the target ship is 500, the number of learning operations performed on the deep neural network is 2500, and so on.

[0038] Therefore, through the second aspect and the third aspect, AI network data prediction models with different customized structures are designed for different ships.

[0039] The fourth aspect: in each learning operation performed on the deep neural network, the total number of transmission mails and the total amount of transmission mail data of the known target ship in a certain historical time segment are taken as two output contents of the deep neural network, the total number of each registered network terminal of the target ship, the occupancy time length of each time segment, the target ship-related information and the network terminal data of the target ship, and the ship mail data corresponding to multiple historical time segments before the certain historical time segment are taken as multiple input contents of the deep neural network, and the learning operation is completed, so as to ensure the learning effect of each learning operation of the deep neural network.

[0040] In this way, through the four aspects of the customization structure design of the AI network data prediction model, the reliability and stability of the intelligent prediction result of the transmission mail state of the target ship in the current time segment are ensured.

[0041] Technical process two: in order to realize intelligent prediction of the total number of transmission mails and the total amount of transmission mail data of the target ship in the current time segment, various basic data are used.

[0042] Specifically, the various basic data include the total number of each registered network terminal of the target ship, the occupancy time length of each time segment, the target ship-related information and the network terminal data of the target ship, and the ship mail data corresponding to multiple historical time segments before the current time.

[0043] Further specifically, the target ship-related information of the target ship is the number of passengers, the male-female ratio of passengers, the average passenger age, the set sailing speed, and the distance from the coastline in the current sea area, the network terminal data of the target ship is the maximum transmission bandwidth, the average transmission bandwidth, the overall operation performance data, the storage capacity data mean value, and the average daily online time length of each registered network terminal of the target ship, the ship mail data corresponding to each time segment is the total number of transmission mails and the total amount of transmission mail data of the target ship in the time segment, the number of selected historical time segments is proportional to the average transmission bandwidth of each registered network terminal of the target ship, thereby providing specific data structure design for fully comprehensive selection of various basic data.

[0044] For example, the number of selected historical time segments is proportional to the average transmission bandwidth of each registered network terminal of the target vessel, including: the average transmission bandwidth of each registered network terminal of the target vessel is 100 Mbps, and the number of selected historical time segments is 9; the average transmission bandwidth of each registered network terminal of the target vessel is 200 Mbps, and the number of selected historical time segments is 18; the average transmission bandwidth of each registered network terminal of the target vessel is 300 Mbps, and the number of selected historical time segments is 27, etc.

[0045] In this way, by fully and comprehensively selecting the above-mentioned basic data, the reliability and stability of the intelligent prediction results of the target ship's mail transmission status in the current time segment are further guaranteed.

[0046] Technical Process 3: Using the AI ​​network data prediction model designed for the target ship based on the structure customized in Technical Process 1, and based on the comprehensive selection of various basic data in Technical Process 2, the model intelligently predicts the email transmission status of the target ship in the current time segment.

[0047] Specifically, the target vessel's email transmission status in the current time segment is the total number of emails transmitted and the total amount of email data transmitted in the current time segment.

[0048] Technical Process 4: Based on the intelligent prediction of the target ship's email transmission status in the current time segment in Technical Process 3, determine the adaptive coding strategy and dynamic bandwidth allocation strategy for the target ship's current time segment.

[0049] Specifically, the total number of network slices allocated to the target vessel in the current time segment is equal to the total number of emails transmitted by the target vessel in the current time segment as predicted by the intelligent system, and the email transmission bandwidth allocated to the target vessel in the current time segment is a set multiple of the total amount of email data transmitted by the target vessel in the current time segment as predicted by the intelligent system.

[0050] It is evident that through the coordinated operation of the above four technical processes, the key step of intelligently predicting the transmission status of emails for each ship in different sea areas in future time segments was completed. This led to the determination of corresponding adaptive coding strategies and dynamic bandwidth allocation strategies, thereby achieving both effective and smooth transmission of all email data for each ship in each future time segment and improving the utilization rate of network transmission resources.

[0051] The key points of this invention are: customized structural design of different AI network data prediction models for different ships, full and comprehensive selection of various basic data for intelligent prediction, targeted design of adaptive encoding strategy and dynamic bandwidth allocation strategy, and customized design of each learning operation of deep neural network.

[0052] The adaptive coding and dynamic bandwidth allocation method for ship mail of the present invention will now be described in detail by way of an embodiment.

[0053] Example 1

[0054] Figure 2 The present invention is illustrated in Embodiment 1 of the present invention as a flowchart of a method for adaptive coding and dynamic bandwidth allocation of ship mail.

[0055] like Figure 2 As shown, the ship mail adaptive coding and dynamic bandwidth allocation method includes the following specific steps:

[0056] Step S201: Obtain the target vessel's passenger capacity, passenger gender ratio, average passenger age, set sailing speed, and distance from the current sea area to the coastline as various vessel-related information.

[0057] For example, when the target vessel is in near-shore waters, that is, when the distance between the target vessel's current location and the coastline is less than or equal to a set distance threshold, the communication network that the target vessel relies on can be a mobile communication network.

[0058] For example, when the target vessel is in the open sea, that is, when the distance between the target vessel's current sea area and the coastline is greater than a set distance threshold, the communication network that the target vessel relies on can be a satellite communication network + a mobile communication network.

[0059] Step S202: Obtain the maximum transmission bandwidth, average transmission bandwidth, overall computing performance data, average storage capacity data, and average daily online time of each registered network terminal of the target vessel as multiple network terminal data of the target vessel;

[0060] Specifically, obtaining the maximum transmission bandwidth, average transmission bandwidth, overall computing performance data, average storage capacity data, and average daily online time of each registered network terminal of the target vessel as multiple network terminal data of the target vessel includes: using different capture units to separately capture the maximum transmission bandwidth, average transmission bandwidth, overall computing performance data, average storage capacity data, and average daily online time of each registered network terminal of the target vessel;

[0061] Step S203: Perform multiple learning operations on the deep neural network to obtain a deep neural network after multiple learning operations, and output the deep neural network after multiple learning operations as the AI ​​network data prediction model.

[0062] For example, numerical simulation mode can be used to perform multiple learning operations on a deep neural network to obtain a deep neural network after multiple learning operations, and the deep neural network after multiple learning operations can be used as the model construction process for the output of the AI ​​network data prediction model for testing and simulation.

[0063] Step S204: Collect multiple pieces of ship mail data corresponding to multiple historical time segments before the current time of the target ship. The ship mail data corresponding to each time segment is the total number of emails transmitted by the target ship and the total amount of email data transmitted within the time segment.

[0064] For example, multiple pieces of ship mail data corresponding to several historical time segments prior to the current time are collected for the target ship. The ship mail data corresponding to each time segment includes the total number of emails transmitted by the target ship within the time segment and the total amount of email data transmitted. The current time is 11:00 AM, and the current time interval is a future time interval, which is from 11:00 AM to 11:20 AM. The multiple historical time segments prior to the current time can be 10:20 AM to 10:40 AM, 10:00 AM to 10:20 AM, 9:40 AM to 10:00 AM, 9:20 AM to 9:40 AM, 9:00 AM to 9:20 AM, 8:40 AM to 9:00 AM, 8:20 AM to 8:40 AM, 8:00 AM to 8:20 AM, and 7:40 AM to 8:00 AM, for a total of 9 historical time segments.

[0065] Step S205: Using an AI network data prediction model, based on the total number of registered network terminals of the target vessel, the duration of each time segment, various vessel-related information of the target vessel, multiple network terminal data, and multiple ship mail data corresponding to multiple historical time segments before the current moment, the target vessel will intelligently predict the total number of emails and the total amount of email data transmitted in the current time segment.

[0066] For example, using an AI network data prediction model, based on the total number of registered network terminals of the target vessel, the duration of each time segment, various vessel-related information of the target vessel, multiple network terminal data, and multiple ship email data corresponding to multiple historical time segments before the current moment, the total number of emails transmitted and the total amount of email data transmitted by the target vessel in the current time segment are intelligently predicted. The duration of each time segment can be 20 minutes.

[0067] Step S206: Determine the adaptive coding strategy and dynamic bandwidth allocation strategy for the current time segment of the target ship based on the intelligent prediction results;

[0068] Among them, the adaptive coding strategy and dynamic bandwidth allocation strategy for the target ship's current time segment determined based on the intelligent prediction results include: in the determined adaptive coding strategy for the target ship's current time segment, the total number of network slices allocated to the target ship in the current time segment is equal to the total number of emails transmitted by the target ship in the current time segment as predicted by the intelligent prediction.

[0069] For example, if the total number of emails transmitted by the target ship in the current time segment is 100 according to the intelligent prediction, the total number of network slices allocated to the target ship in the current time segment is equal to 100.

[0070] Among them, determining the adaptive coding strategy and dynamic bandwidth allocation strategy for the current time segment of the target ship based on the intelligent prediction results includes: in the determined dynamic bandwidth allocation strategy for the current time segment of the target ship, the email transmission bandwidth allocated to the target ship in the current time segment is a set multiple of the total amount of email data transmitted by the target ship in the current time segment as predicted by the intelligent prediction, and the value of the set multiple is between 1.10 and 1.15.

[0071] For example, in the determined dynamic bandwidth allocation strategy for the current time segment of the target vessel, the email transmission bandwidth allocated to the target vessel in the current time segment is a set multiple of the total amount of email data transmitted by the target vessel in the current time segment as intelligently predicted. The set multiple is in the range of 1.10-1.15, including: the specific value of the set multiple can be 1.12.

[0072] Among them, the number of learning operations performed on the deep neural network is positively correlated with the total number of registered network terminals on the target ship;

[0073] For example, a positive correlation between the number of learning operations performed on the deep neural network and the total number of registered network terminals on the target ship includes: 1000 learning operations performed on the deep neural network when the total number of registered network terminals on the target ship is 200; 1500 learning operations performed on the deep neural network when the total number of registered network terminals on the target ship is 300; 2000 learning operations performed on the deep neural network when the total number of registered network terminals on the target ship is 400; 2500 learning operations performed on the deep neural network when the total number of registered network terminals on the target ship is 500; and so on.

[0074] In each learning operation performed on the deep neural network, the total number of emails transmitted and the total amount of email data transmitted by the target ship in a certain historical time segment are taken as two output contents of the deep neural network. The total number of registered network terminals of the target ship, the occupation duration of each time segment, various ship association information of the target ship, multiple network terminal data, and multiple ship email data corresponding to multiple historical time segments before the certain historical time segment are taken as multiple input contents of the deep neural network to complete this learning operation.

[0075] Among them, multiple ship email data corresponding to multiple historical time segments before the current time are collected from the target ship. The ship email data corresponding to each time segment is the total number of emails transmitted by the target ship within the time segment and the total amount of email data transmitted. The number of selected multiple historical time segments is proportional to the average transmission bandwidth of each registered network terminal of the target ship.

[0076] For example, the number of selected historical time segments is proportional to the average transmission bandwidth of each registered network terminal of the target vessel, including: the average transmission bandwidth of each registered network terminal of the target vessel is 100 Mbps, the number of selected historical time segments is 9; the average transmission bandwidth of each registered network terminal of the target vessel is 200 Mbps, the number of selected historical time segments is 18; the average transmission bandwidth of each registered network terminal of the target vessel is 300 Mbps, the number of selected historical time segments is 27, and so on.

[0077] Example 2

[0078] Figure 3 The following is a flowchart illustrating the steps of a ship mail adaptive coding and dynamic bandwidth allocation method according to Embodiment 2 of the present invention.

[0079] like Figure 3 As shown, with Figure 2 Unlike the embodiments described above, in the ship mail adaptive coding and dynamic bandwidth allocation method, after determining the adaptive coding strategy and dynamic bandwidth allocation strategy for the current time segment of the target ship based on the intelligent prediction results, i.e. after step S206, the method further includes:

[0080] Step S207: Use a dynamic storage chip to receive the adaptive coding strategy and dynamic bandwidth allocation strategy for the current time segment of the target ship;

[0081] The method of using a dynamic storage chip to receive the adaptive coding strategy and dynamic bandwidth allocation strategy for the current time segment of the target ship includes: storing the adaptive coding strategy and dynamic bandwidth allocation strategy for the current time segment of the target ship in different physical addresses in the dynamic storage chip.

[0082] For example, the dynamic storage chip can be implemented using FLASH flash memory, SD storage chip or MMC storage chip.

[0083] Example 3

[0084] Figure 4 The following is a flowchart illustrating the steps of a ship mail adaptive coding and dynamic bandwidth allocation method according to Embodiment 3 of the present invention.

[0085] like Figure 4 As shown, with Figure 2 Unlike the previous embodiment, in the ship mail adaptive coding and dynamic bandwidth allocation method, before obtaining the target ship's passenger count, passenger gender ratio, average passenger age, set sailing speed, and distance from the coastline to the current sea area as various ship-related information, i.e. before step S201, the method further includes:

[0086] Step S208: Use satellite positioning mode to obtain satellite positioning data of the current sea area where the target vessel is located. The satellite positioning data of the current sea area where the target vessel is located is used to determine the distance between the current sea area where the target vessel is located and the coastline.

[0087] Among them, satellite positioning data of the current sea area of ​​the target vessel is obtained by using satellite positioning mode. The satellite positioning data of the current sea area of ​​the target vessel is used to determine the distance of the current sea area of ​​the target vessel from the coastline. The satellite positioning mode is one of Beidou positioning mode, Galileo positioning mode and GPS positioning mode.

[0088] For example, the satellite positioning mode is one of the BeiDou positioning mode, Galileo positioning mode, and GPS positioning mode, including providing different mode code values ​​for different positioning modes.

[0089] Example 4

[0090] Figure 5 The following is a flowchart illustrating the steps of a ship mail adaptive coding and dynamic bandwidth allocation method according to Embodiment 4 of the present invention.

[0091] like Figure 5 As shown, with Figure 4Unlike the embodiments described above, in the ship mail adaptive coding and dynamic bandwidth allocation method, after determining the adaptive coding strategy and dynamic bandwidth allocation strategy for the current time segment of the target ship based on the intelligent prediction results, i.e. after step S206, the method further includes:

[0092] Step S209: Receive the adaptive coding strategy and dynamic bandwidth allocation strategy for the current time segment of the target vessel, and wirelessly transmit the adaptive coding strategy and dynamic bandwidth allocation strategy for the current time segment of the target vessel to the remote ship management server through the mobile communication network.

[0093] The adaptive coding strategy and dynamic bandwidth allocation strategy for the current time segment of the target ship are wirelessly transmitted to the remote ship management server via a mobile communication network, including: the mobile communication network is based on time division duplex transmission mode or frequency division duplex transmission mode.

[0094] For example, the adaptive coding strategy and dynamic bandwidth allocation strategy for the current time segment of the target vessel are wirelessly transmitted to a remote ship management server via a mobile communication network. The ship management server is a blockchain management node, a cloud computing management node, or a big data management node.

[0095] Example 5

[0096] Figure 6 The present invention is illustrated in Embodiment 5 of the present invention as a flowchart of a method for adaptive coding and dynamic bandwidth allocation of ship mail.

[0097] like Figure 6 As shown, with Figure 2 Unlike the previous embodiment, in the ship mail adaptive coding and dynamic bandwidth allocation method, before obtaining the target ship's passenger count, passenger gender ratio, average passenger age, set sailing speed, and distance from the coastline to the current sea area as various ship-related information, i.e. before step S201, the method further includes:

[0098] Step S210: Use a quartz oscillator to provide timing service to the target ship, wherein the quartz oscillator has a built-in pulse generation unit for generating a reference pulse waveform for providing timing service;

[0099] Among them, a quartz oscillator is used to provide timing services for the target ship. The quartz oscillator has a built-in pulse generation unit to generate a reference pulse waveform for providing timing services, including: the reference pulse waveform is a rectangular wave of a set frequency;

[0100] For example, the reference pulse waveform is a rectangular wave of a set frequency, including a set frequency ranging from 5000Hz to 10kHz.

[0101] Next, the various method embodiments of the present invention will be described in detail.

[0102] In a method for adaptive coding and dynamic bandwidth allocation of ship mail according to various embodiments of the present invention:

[0103] The acquisition of the maximum transmission bandwidth, average transmission bandwidth, overall computing performance data, average storage capacity data, and average daily online time of each registered network terminal of the target vessel as multiple network terminal data of the target vessel includes: after removing the maximum and minimum values ​​of each upper limit of computing power per second corresponding to each registered network terminal of the target vessel, the arithmetic mean of the remaining upper limits of computing power per second is used as the overall computing performance data of each registered network terminal of the target vessel.

[0104] The process of collecting multiple ship mail data corresponding to multiple historical time segments before the current time of the target ship includes the total number of emails transmitted by the target ship within the time segment and the total amount of email data transmitted. It also includes the following: the occupation duration of each time segment is equal, the current time segment starts from the current time, and the current time segment and multiple historical time segments before the current time form a complete time interval on the time axis.

[0105] For example, each time segment occupies an equal duration, and the current time segment starts at the current moment. The current time segment and multiple historical time segments preceding the current moment form a complete time interval on the time axis, including: each time segment occupies 20 minutes, the current moment is 11:00 AM, the current time interval is a future time interval from 11:00 AM to 11:20 AM, and the multiple historical time segments preceding the current moment can be 10:20 AM to 10:40 AM, 10:00 AM to 10:20 AM, 9:40 AM to 10:00 AM, 9:20 AM to 9:40 AM, 9:00 AM to 9:20 AM, 8:40 AM to 9:00 AM, 8:20 AM to 8:40 AM, 8:00 AM to 8:20 AM, and 7:40 AM to 8:00 AM, for a total of 9 historical time segments;

[0106] Among them, the information obtained about the target vessel includes the number of passengers, the male-to-female ratio of passengers, the average age of passengers, the set sailing speed, and the distance from the current sea area to the coastline. The passenger information of the target vessel includes the crew and passengers of the target vessel.

[0107] In a method for adaptive coding and dynamic bandwidth allocation of ship mail according to various embodiments of the present invention:

[0108] The positive correlation between the number of learning operations performed on the deep neural network and the total number of registered network terminals on the target ship includes: expressing the parameter transformation relationship between the number of learning operations performed on the deep neural network and the total number of registered network terminals on the target ship using a parameter transformation formula;

[0109] Specifically, the MATLAB toolbox can be used to simulate and model the implementation process of the parameter transformation formula;

[0110] The parameter transformation relationship, which uses a parameter transformation formula to express the positive correlation between the number of learning operations performed on the deep neural network and the total number of registered network terminals of the target ship, includes: in the parameter transformation formula, the total number of registered network terminals of the target ship is the input parameter of the parameter transformation formula;

[0111] The parameter transformation formula used to express the positive correlation between the number of learning operations performed on the deep neural network and the total number of registered network terminals of the target ship includes: in the parameter transformation formula, the number of learning operations performed on the deep neural network that is positively correlated with the total number of registered network terminals of the target ship is the output parameter of the parameter transformation formula.

[0112] And in a method for adaptive coding and dynamic bandwidth allocation of ship mail according to various embodiments of the present invention:

[0113] The AI ​​network data prediction model is used to intelligently predict the total number of emails and the total amount of email data transmitted by the target vessel in the current time segment based on the total number of registered network terminals of the target vessel, the duration of each time segment, various ship-related information and multiple network terminal data of the target vessel, as well as multiple ship email data corresponding to multiple historical time segments before the current time. This includes converting the total number of registered network terminals of the target vessel, the duration of each time segment, various ship-related information and multiple network terminal data of the target vessel, as well as multiple ship email data corresponding to multiple historical time segments before the current time, into octal values ​​and then synchronously inputting them into the AI ​​network data prediction model.

[0114] Specifically, for the input content such as the total number of registered network terminals of the target vessel, the duration of each time segment, various ship-related information of the target vessel, multiple network terminal data, and multiple ship email data corresponding to multiple historical time segments of the target vessel before the current moment, the octal value conversion is skipped for the input content originally represented by octal values.

[0115] The AI ​​network data prediction model intelligently predicts the total number of emails and the total amount of email data transmitted by the target ship in the current time segment based on the total number of registered network terminals of the target ship, the duration of each time segment, various ship-related information of the target ship, multiple network terminal data, and multiple ship email data corresponding to multiple historical time segments before the current moment. It also includes the fact that the total number of emails and the total amount of email data transmitted by the target ship in the current time segment output by the AI ​​network data prediction model are in octal numerical form.

[0116] The process of converting the total number of registered network terminals of the target vessel, the duration of each time segment, various ship-related information of the target vessel, multiple network terminal data, and multiple ship email data corresponding to multiple historical time segments of the target vessel before the current moment into octal values ​​and then synchronously inputting them into the AI ​​network data prediction model includes: using a first ASIC chip to implement octal value conversion and using a second ASIC chip to implement synchronous input.

[0117] The process of converting the total number of registered network terminals of the target vessel, the duration of each time segment, various ship-related information of the target vessel, multiple network terminal data, and multiple ship email data corresponding to multiple historical time segments of the target vessel before the current moment into octal values ​​and then synchronously inputting them into the AI ​​network data prediction model also includes: the first AS IC chip and the second AS IC chip are connected and share the same parameter configuration interface.

[0118] For example, the first ASIC chip and the second ASIC chip being connected and sharing the same parameter configuration interface include: the same parameter configuration interface being an IIC serial configuration interface.

[0119] In addition, the present invention may also refer to the following technical contents to highlight the significant technical progress of the present invention:

[0120] Performing multiple learning operations on a deep neural network to obtain a deep neural network after multiple learning operations, and outputting the deep neural network after multiple learning operations as an AI network data prediction model includes: the deep neural network used includes multiple hidden layers, and the total number of hidden layers is positively correlated with the average daily online time of each registered network terminal;

[0121] For example, the deep neural network used includes multiple hidden layers, and the total number of hidden layers is positively correlated with the average daily online time of each registered network terminal, including: if the average daily online time of each registered network terminal is 5 hours, the total number of hidden layers selected is 3; if the average daily online time of each registered network terminal is 7 hours, the total number of hidden layers selected is 5; if the average daily online time of each registered network terminal is 9 hours, the total number of hidden layers selected is 7, and so on.

[0122] The deep neural network used includes multiple hidden layers, and the total number of hidden layers is positively correlated with the average daily online time of each registered network terminal. This includes selecting a numerical mapping function to represent the numerical mapping relationship between the total number of hidden layers and the average daily online time of each registered network terminal.

[0123] For example, selecting a numerical mapping function to represent the positive correlation between the total number of hidden layers and the average daily online time of each registered network terminal includes: in the numerical mapping function, the average daily online time of each registered network terminal is used as the input parameter of the numerical mapping function, and the total number of hidden layers positively correlated with the average daily online time of each registered network terminal is used as the output parameter of the numerical mapping function;

[0124] The deep neural network used includes multiple hidden layers, and the total number of hidden layers is positively correlated with the average daily online time of each registered network terminal. Furthermore, the deep neural network used includes multiple hidden layers, a single input layer, and a single output layer, wherein the single input layer is connected to the multiple hidden layers, and the single output layer is connected to the multiple hidden layers.

[0125] Although the invention has been described with reference to specific exemplary embodiments thereof, those skilled in the art will understand that various adjustments and variations can be made to the invention without departing from the spirit or scope of the invention as defined in the appended claims and their equivalents.

Claims

1. A method for adaptive coding and dynamic bandwidth allocation of ship mail, characterized in that, The method includes: The system obtains information about the target vessel, including its passenger capacity, male-to-female passenger ratio, average passenger age, set sailing speed, and distance from the coastline to the current sea area. The maximum transmission bandwidth, average transmission bandwidth, overall computing performance data, average storage capacity data, and average daily online time of each registered network terminal of the target vessel are obtained as multiple network terminal data of the target vessel. Perform multiple learning operations on a deep neural network to obtain a deep neural network after multiple learning operations, and output the deep neural network after multiple learning operations as an AI network data prediction model. Collect multiple pieces of ship mail data corresponding to several historical time segments before the current moment for the target ship. The ship mail data corresponding to each time segment is the total number of emails transmitted by the target ship and the total amount of email data transmitted within the time segment. The AI ​​network data prediction model uses the total number of registered network terminals of the target vessel, the duration of each time segment, various ship-related information of the target vessel, multiple network terminal data, and multiple ship email data corresponding to multiple historical time segments before the current moment to intelligently predict the total number of emails and the total amount of email data transmitted by the target vessel in the current time segment. Based on the intelligent prediction results, an adaptive coding strategy and a dynamic bandwidth allocation strategy are determined for the current time segment of the target ship.

2. The ship mail adaptive coding and dynamic bandwidth allocation method as described in claim 1, characterized in that: The adaptive coding strategy and dynamic bandwidth allocation strategy determined based on the intelligent prediction results for the target ship's current time segment include: in the determined adaptive coding strategy for the target ship's current time segment, the total number of network slices allocated to the target ship in the current time segment is equal to the total number of emails transmitted by the target ship in the current time segment as predicted by the intelligent prediction. Among them, determining the adaptive coding strategy and dynamic bandwidth allocation strategy for the current time segment of the target ship based on the intelligent prediction results includes: in the determined dynamic bandwidth allocation strategy for the current time segment of the target ship, the email transmission bandwidth allocated to the target ship in the current time segment is a set multiple of the total amount of email data transmitted by the target ship in the current time segment as predicted by the intelligent prediction, and the value of the set multiple is between 1.10 and 1.

15.

3. The adaptive coding and dynamic bandwidth allocation method for ship mail as described in claim 2, characterized in that: The number of learning operations performed on the deep neural network is positively correlated with the total number of registered network terminals on the target ship. In each learning operation performed on the deep neural network, the total number of emails transmitted and the total amount of email data transmitted by the target ship in a certain historical time segment are taken as two output contents of the deep neural network. The total number of registered network terminals of the target ship, the occupation duration of each time segment, various ship association information of the target ship, multiple network terminal data, and multiple ship email data corresponding to multiple historical time segments before the certain historical time segment are taken as multiple input contents of the deep neural network to complete this learning operation. The process involves collecting multiple pieces of ship mail data corresponding to several historical time segments prior to the current moment. The ship mail data corresponding to each time segment includes the total number of emails transmitted by the target ship within that time segment and the total amount of transmitted email data. The number of selected historical time segments is proportional to the average transmission bandwidth of each registered network terminal of the target ship.

4. The ship mail adaptive coding and dynamic bandwidth allocation method as described in claim 3, characterized in that, After determining the adaptive coding strategy and dynamic bandwidth allocation strategy for the current time segment of the target ship based on the intelligent prediction results, the method further includes: The system employs a dynamic storage chip to receive adaptive coding and dynamic bandwidth allocation strategies tailored to the current time segment of the target ship. The method of using a dynamic storage chip to receive the adaptive coding strategy and dynamic bandwidth allocation strategy for the current time segment of the target ship includes: storing the adaptive coding strategy and dynamic bandwidth allocation strategy for the current time segment of the target ship in different physical addresses in the dynamic storage chip.

5. The ship mail adaptive coding and dynamic bandwidth allocation method as described in claim 3, characterized in that, Before acquiring the target vessel's passenger capacity, male-to-female passenger ratio, average passenger age, set sailing speed, and distance from the coastline to the current sea area as various vessel-related information, the method further includes: The satellite positioning data of the target vessel's current sea area is obtained using satellite positioning mode. The satellite positioning data of the target vessel's current sea area is used to determine the distance between the target vessel's current sea area and the coastline. Specifically, the satellite positioning data of the target vessel's current sea area is obtained using a satellite positioning mode. The satellite positioning data of the target vessel's current sea area is used to determine the distance between the target vessel's current sea area and the coastline. The satellite positioning mode is one of the following: Beidou positioning mode, Galileo positioning mode, and GPS positioning mode.

6. The adaptive coding and dynamic bandwidth allocation method for ship mail as described in claim 3, characterized in that, After determining the adaptive coding strategy and dynamic bandwidth allocation strategy for the current time segment of the target ship based on the intelligent prediction results, the method further includes: It receives the adaptive coding strategy and dynamic bandwidth allocation strategy for the current time segment of the target vessel, and wirelessly transmits the adaptive coding strategy and dynamic bandwidth allocation strategy for the current time segment of the target vessel to the remote ship management server through the mobile communication network. The process of wirelessly transmitting the adaptive coding strategy and dynamic bandwidth allocation strategy for the current time segment of the target vessel to a remote ship management server via a mobile communication network includes: the mobile communication network being based on time-division duplex transmission mode or frequency-division duplex transmission mode.

7. The adaptive coding and dynamic bandwidth allocation method for ship mail as described in claim 3, characterized in that, Before acquiring the target vessel's passenger capacity, male-to-female passenger ratio, average passenger age, set sailing speed, and distance from the coastline to the current sea area as various vessel-related information, the method further includes: A quartz oscillator is used to provide timing services for the target ship. The quartz oscillator has a built-in pulse generation unit for generating a reference pulse waveform for providing timing services. Among them, a quartz oscillator is used to provide timing services for the target ship. The quartz oscillator has a built-in pulse generation unit for generating a reference pulse waveform for providing timing services, including: the reference pulse waveform is a rectangular wave of a set frequency.

8. The ship mail adaptive coding and dynamic bandwidth allocation method as described in any one of claims 3-7, characterized in that: The acquisition of the maximum transmission bandwidth, average transmission bandwidth, overall computing performance data, average storage capacity data, and average daily online time of each registered network terminal of the target vessel as multiple network terminal data of the target vessel includes: after removing the maximum and minimum values ​​of each upper limit of computing power per second corresponding to each registered network terminal of the target vessel, the arithmetic mean of the remaining upper limits of computing power per second is used as the overall computing performance data of each registered network terminal of the target vessel. The process of collecting multiple ship mail data corresponding to multiple historical time segments before the current time of the target ship includes the total number of emails transmitted by the target ship within the time segment and the total amount of email data transmitted. It also includes the following: the occupation duration of each time segment is equal, the current time segment starts from the current time, and the current time segment and multiple historical time segments before the current time form a complete time interval on the time axis. Among them, the information obtained about the target vessel includes the number of passengers, the male-to-female ratio of passengers, the average age of passengers, the set sailing speed, and the distance from the current sea area to the coastline. The passenger information of the target vessel includes the crew and passengers of the target vessel.

9. The ship mail adaptive coding and dynamic bandwidth allocation method as described in any one of claims 3-7, characterized in that: The positive correlation between the number of learning operations performed on the deep neural network and the total number of registered network terminals on the target ship includes: expressing the parameter transformation relationship between the number of learning operations performed on the deep neural network and the total number of registered network terminals on the target ship using a parameter transformation formula; The parameter conversion formula used to express the positive correlation between the number of learning operations performed on the deep neural network and the total number of registered network terminals of the target ship includes: in the parameter conversion formula, the total number of registered network terminals of the target ship is the input parameter of the parameter conversion formula; The parameter transformation formula used to express the positive correlation between the number of learning operations performed on the deep neural network and the total number of registered network terminals of the target ship includes: in the parameter transformation formula, the number of learning operations performed on the deep neural network that is positively correlated with the total number of registered network terminals of the target ship is the output parameter of the parameter transformation formula.

10. The ship mail adaptive coding and dynamic bandwidth allocation method as described in any one of claims 3-7, characterized in that: The AI ​​network data prediction model is used to intelligently predict the total number of emails and the total amount of email data transmitted by the target vessel in the current time segment based on the total number of registered network terminals of the target vessel, the duration of each time segment, various ship-related information and multiple network terminal data of the target vessel, as well as multiple ship email data corresponding to multiple historical time segments before the current time. This includes converting the total number of registered network terminals of the target vessel, the duration of each time segment, various ship-related information and multiple network terminal data of the target vessel, as well as multiple ship email data corresponding to multiple historical time segments before the current time, into octal values ​​and then synchronously inputting them into the AI ​​network data prediction model. The AI ​​network data prediction model intelligently predicts the total number of emails and the total amount of email data transmitted by the target ship in the current time segment based on the total number of registered network terminals of the target ship, the duration of each time segment, various ship-related information of the target ship, multiple network terminal data, and multiple ship email data corresponding to multiple historical time segments before the current moment. It also includes the fact that the total number of emails and the total amount of email data transmitted by the target ship in the current time segment output by the AI ​​network data prediction model are in octal numerical form. The process of converting the total number of registered network terminals of the target vessel, the duration of each time segment, various ship-related information of the target vessel, multiple network terminal data, and multiple ship email data corresponding to multiple historical time segments of the target vessel before the current moment into octal values ​​and then synchronously inputting them into the AI ​​network data prediction model includes: using a first ASIC chip to implement octal value conversion and using a second ASIC chip to implement synchronous input. The process of converting the total number of registered network terminals of the target vessel, the duration of each time segment, various ship-related information of the target vessel, multiple network terminal data, and multiple ship email data corresponding to multiple historical time segments before the current moment into octal values ​​and then synchronously inputting them into the AI ​​network data prediction model also includes: the first AS IC chip and the second AS IC chip are connected and share the same parameter configuration interface.

Images