Large file subpackage transmission method based on real-time bit error rate

Abstract

The invention discloses a large file subpackage transmission method based on a real-time bit error rate, and relates to the technical field of satellite communication. In the transmission process, if the bit error rate changes, a fixed subpackage strategy determined based on the initial bit error rate is still adopted, so that the transmission delay is increased. The change of the bit error rate can influence the loss rate and retransmission times of data packets, and when the bit error rate is increased, a fixed subpackage strategy may not adapt to a new network condition, so that the packet loss is frequent, the retransmission is increased, and the transmission efficiency is obviously reduced. Transmission is carried out according to the error rate calculation subpackage strategy in real time, so that the transmission delay is effectively reduced, and the transmission efficiency and reliability are improved.

Description

Technical Field

[0001] The present invention relates to the field of satellite communication technology, and particularly designs a method for reducing transmission delay by sub-packetizing large files based on real-time bit error rate in a network transmission environment. Background Art

[0002] With the rapid development of satellite communication technology, satellite data transmission tasks have become increasingly important, and a large amount of data needs to be transmitted between different devices and networks. Especially when dealing with large file transmission, improving transmission efficiency and reliability becomes a key issue. Traditional file transmission methods usually sub-packetize based on fixed parameters, such as determining the sub-packetization strategy according to the initial bit error rate and keeping the sub-packetization strategy unchanged throughout the transmission process. However, the actual network environment is complex and variable, and the bit error rate will change in real time with factors such as time, network congestion degree, and signal interference. During the transmission process, if the bit error rate changes and the fixed sub-packetization strategy determined based on the initial bit error rate is still used, it will lead to an increase in transmission delay. This is because the change in the bit error rate will affect the packet loss rate and the number of retransmissions. When the bit error rate increases, the fixed sub-packetization strategy may not be able to adapt to the new network conditions, resulting in frequent packet loss and increased retransmissions, thus significantly reducing the transmission efficiency. Summary of the Invention

[0003] In view of the above existing problems, the present invention proposes a method for sub-packetizing large files based on real-time bit error rate, aiming to calculate the sub-packetization strategy in real time according to the bit error rate for transmission, thereby effectively reducing the transmission delay and improving the transmission efficiency and reliability.

[0004] Step 1: The user end receives a data transmission task, which requires transmitting a file (File G) of a specific size to a certain target address in the Internet.

[0005] Step 2: The user end detects the status of each currently accessed network in real time, including information such as bandwidth and bit error rate.

[0006] Step 3: Calculate the number of file packets into which File G needs to be divided to minimize the transmission delay of the transmitted data under the current network status.

[0007] Assume that the bit error rate of the communication network at the current moment t is p i ∈[p min , p max , where p min is the minimum value of the bit error rate, p max is the maximum value of the bit error rate, the bit error rate at the initial moment t0 is p0, the large file G is divided into N file packets, and the size g of each file packet is n is the optimal number of information packets into which the file packet g is divided, then the size m of each information packet is With a bandwidth of r, LT codes are used to encode data, and each information packet requires an additional parameter value. Define the encoding parameter value of a single information packet as a, then the size of the additional parameter information required in each encoded packet is an. Due to the instability of the satellite transmission link, check information also needs to be added to the encoded packet to determine whether a single data packet is correctly transmitted. Add a cyclic redundancy check to each encoded packet, and the size of the check code is b bytes. Therefore, it can be concluded that the size of a single encoded data packet is (m + an + b), and the size of each piece of data must satisfy

[0008] m + an + b ≤ h#(1)

[0009] where h is the maximum value of the data part in a single data frame in the MAC layer protocol.

[0010] The known sub-packetization strategy determined based on the initial fixed bit error rate p0 is where

[0011] The transmission delay T1(p i ) formula based on the initial fixed bit error rate is

[0012]

[0013] The transmission delay T2(p i ) formula for real-time adjustment of the sub-packetization strategy is

[0014]

[0015] where

[0016] Then, during the entire transmission process, the total transmission delay based on the initial fixed bit error rate is The transmission delay for real-time adjustment of the sub-packetization strategy is Let the difference in transmission delay between the two sub-packetization strategies during the transmission process be

[0017] ΔT(p i ) = T1(p i ) - T2(p i ) = T(n0, p i ) - T(n(p i ), p i )#(4)

[0018] Then, during the entire transmission process, the total difference in transmission delay between the two sub-packetization strategies is

[0019]

[0020] Since So T(n0, pi ) ≥ T(n(p i ), p i ), that is, ΔT(p i ) ≥ 0, obtaining ΔT total ≥ 0.

[0021] At the same time, since Therefore Then there is

[0022]

[0023] Denote

[0024]

[0025] Taking the derivative of F(n, p) with respect to p gives:

[0026]

[0027] It can be obtained that:

[0028]

[0029] Since Therefore Also, since p ∈ [0, 1), then ln(1 - p) ≤ 0,

[0030] Taking the derivative of the logarithm of T(n, p) with respect to p gives:

[0031]

[0032] Since and p → 0, therefore The transmission delay T will increase as p increases, so it is obtained that:

[0033] ΔT(p i ) max = max((T(n min , p max ) - T(n max , p max ), (T(n max , p min ) - T(n min , p min ))) # (11)

[0034] Since the actual delay of dynamically adjusting the packet splitting strategy is less than the delay of the fixed initial bit error rate strategy, when transmitting a large file, the large file G is split into N identical small files Then, the small file g is transmitted according to the packet splitting algorithm, and the packet splitting strategy is calculated in real time based on the bit error rate, which can reduce the actual transmission delay.

[0035] After splitting the large file G to be transmitted into N small files g of the same size, according to the real-time bit error rate p i , i = 1, 2, …, N, calculate the optimal number of split packets Assume that during the transmission process, the bit error rate of each small file g follows a normal distribution Its probability density function is f(p), and its average transmission delay is:

[0036]

[0037] Then the total transmission delay T total (N) is:

[0038]

[0039] For solving the value of N that minimizes T total , the differential approximation optimal N algorithm for splitting large files into packets is proposed

[0040] Initialize N = N0, and calculate ΔT(N)

[0041] ΔT(N) = T total (N + 1) - T total (N) #(14)

[0042] If |ΔT(N)| > ε, update N to

[0043] N = N - ΔT(N)α #(15)

[0044] Otherwise, it is considered that the optimal N has been found.

[0045] Step 4: Split the file G according to the number of file packets N that need to be split calculated, and the size of each split file packet is g.

[0046] Step 5: Monitor the network status in real time, and calculate the optimal packet splitting strategy for the split file packets g according to the packet splitting algorithm.

[0047] Step 6: According to the calculated packet splitting strategy, split the small file g into several information packets of size m i and transmit them in the order of packet splitting.

[0048] Step 7: Repeat Steps 5 and 6 to adjust the packet splitting strategy in real time until the task sending ends.

[0049] Step 8: After the encoded packets arrive at the receiving end, the decoder sorts the received packets in the encoding order and performs verification to determine whether there are problems such as packet loss and errors during transmission. If there are problems during transmission, the problems are fed back to the sending end for retransmission.

[0050] Step 9: According to the encoding method of the fountain code, perform decoding operations on the received encoded packets, and then splice the decoded data in the splitting order of the original file to restore the complete file content.

[0051] Step 10: The receiving end sends a confirmation message to the sending end to inform that the file has been successfully received and stored, and the file transmission task is completed. Description of the Drawings

[0052] Figure 1 Data transmission scenario diagram.

[0053] Figure 2 Schematic diagram of data packet transmission strategy.

[0054] Figure 3 Flowchart for optimizing the data encoding strategy of the packet splitting algorithm. Detailed Implementation Manner

[0055] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the invention.

[0056] The present invention discloses a large file packet splitting transmission method based on real-time bit error rate. The method is based on data transmission of the Tianyuan constellation, and the data transmitted is an image, as shown in the following: Figure 1 as follows:

[0057] Step 1: The remote sensing satellite needs to transmit an image file (file size G is 1.1 GB) to the ground station. The Tianyuan satellite calculates the optimal packet splitting strategy in real time through the differential approximation large file packet splitting optimal N algorithm. This algorithm calculates the optimal number of packets N = 108 based on the current bit error rate (1e-8 to 5e-6) and available bandwidth (100 Mbps) of the satellite network, and splits the image file into file packets g = 1.1 GB / 108 ≈ 10.43 MB.

[0058] Step 2: The Tianyuan satellite adopts the encoding method of the fountain code, encodes the 108 file packets in sequence to generate a file packet g' containing redundant information, with a size of 12.5 MB, and queues up for transmission.

[0059] Step 3: Calculate the packet splitting strategy for the file packet g' containing redundant information according to the real-time link environment, and evenly split the file packet g' into 709 data packets m = 12.5 MB / 709 = 18.02 KB.

[0060] Step 4: The Tianyuan satellite continues to use the fountain code encoding method to encode 709 data packets and transmit them to the ground station.

[0061] Step 5: After the data packets arrive at the ground station, the ground station sorts the received packets according to the encoding order and performs verification to determine whether there are problems such as packet loss and errors during the transmission process. If there are problems during the transmission process, the ground station will feedback the problems to the satellite for retransmission.

[0062] Step 6: The ground station decodes the received encoded packets, then splices the decoded data according to the order carried by the redundant information to restore the original blocks (108 file packets), and then restores the complete initial file content based on all the file packets.

[0063] Step 7: The ground station sends a confirmation message to the satellite, informing that the file has been successfully received and stored, and the file transmission task is completed.

[0064] This solution can reduce the transmission delay to 78% of the traditional solution in the space-ground link environment, and is especially suitable for reliable data transmission in the space-ground cooperation scenario.

Claims

1. A method for transmitting large files in packets based on real-time bit error rate, characterized in that It includes the following steps: Step 1: The client receives a data transmission task, which requires transmitting a file G of a specific size to a certain target address in the Internet; Step 2: The client detects the status of each currently connected network in real time. The status includes bandwidth and bit error rate; Step 3: Calculate the number of file packets into which the file G needs to be divided to minimize the transmission delay of the data after dividing the file G under the current network status; The bit error rate of the communication network at the current moment t is p i ∈ [p min , p max , where p min is the minimum value of the bit error rate, p max is the maximum value of the bit error rate. The bit error rate at the initial moment t0 is p0. The large file G is divided into N file packets, and the size g of each file packet is a is the optimal number of information packets into which the file packet g is divided, then the size m of each information packet is The bandwidth is r, and the LT code is used to encode the data. Each information packet needs to add an additional parameter value; Define the encoding parameter value of a single information packet as a, then the size of the additional parameter information required in each encoding group is an; Due to the instability of the satellite transmission link, it is also necessary to add check information to the encoding group to determine whether a single data packet is correctly transmitted; Add a cyclic redundancy check to each encoding group, and the size of the check code is b bytes. Therefore, the size of a single encoded data packet is (m + an + b), and the size of each data must satisfy m + an + b ≤ h#(1) where h is the maximum value of the data part in a single data frame in the MAC layer protocol; The known sub-packetization strategy determined based on the initial fixed bit error rate p0 is where Transmission delay T1(p based on the initial fixed bit error rate i ) The formula is The transmission delay T2(p i ) of the real-time adjusted sub-packet strategy is given by the formula Among them Then, during the entire transmission process, the total transmission delay based on the initial fixed bit error rate is The transmission delay of the sub-packet strategy adjusted in real time is Let the difference in transmission delay between the two sub-packet strategies during the transmission process be ΔT(p i ) = T1(p i ) - T2(p i ) = T(n0, p i ) - T(n(p i ), p i ) #(4) Then the difference in the total transmission delay between the two sub-packet strategies in the entire transmission process is Since so T(n0,p i )≥T(n(p i ),p i ), that is, ΔT(p i )≥0, and we get ΔT total ≥0; Meanwhile, since Therefore Then there is Denote Take the derivative of F(n, p) with respect to p to get: Get: Since Therefore Also, since p ∈ [0, 1), then ln(1 - p) ≤ 0, Take the derivative of the logarithm of T(n, p) with respect to p to get: Since and p→0, thus the transmission delay T will increase as p increases, so we get: ΔT(p i ) max =max((T(n min ,p max )-T(n max ,p max )),(T(n max ,p min )-T(n min ,p min ))) #(11) Since the actual delay of dynamically adjusting the sub-packet strategy is less than the delay of the fixed initial bit error rate strategy, when transmitting a large file, the large file G is split into N identical small files Then, the small file g is transmitted according to the sub-packet algorithm, and the sub-packet strategy is calculated in real time according to the bit error rate; After splitting the large file G to be transmitted into N small files g of the same size, according to the real-time bit error rate p i , i = 1, 2, …, N, calculate the optimal number of sub-packets Assume that during the transmission process, the bit error rate of each small file g during transmission follows a normal distribution Its probability density function is f(p), and its average transmission delay is: Then the total transmission delay T total (N) for N small files is as follows: For solving the N value that minimizes T total A differential approximation optimal N algorithm for large file subcontracting is proposed Initialize N = N0 and calculate ΔT(N) ΔT(N) = T tptal (N + 1) - T total (N) # (14) If |ΔT(N)| > ε, update N to N = N - ΔT(N)α#(15) Otherwise, it is considered that the optimal N has been found; Step 4: Divide the file G according to the calculated number of file packets N to be divided. The size of each divided file packet is g; Step 5: Monitor the network status in real time, and calculate the optimal sub-packet strategy for the divided file packets g according to the sub-packet algorithm; Step 6: According to the calculated subcontracting strategy, split the small file g into several information packets of size m i and transmit them in the subcontracting order; Step 7: Repeat Steps 5 and 6 to adjust the sub-packet strategy in real time until the task sending ends; Step 8: After the encoding packets arrive at the receiving end, the decoder sorts the received packets in the encoding order and performs verification to determine whether there are problems such as packet loss and errors during the transmission process; If there are problems during the transmission process, feedback the problems to the sending end for retransmission; Step 9: According to the encoding method of the fountain code, perform decoding operations on the received encoding packets, and then splice the decoded data in the splitting order of the original file to restore the complete file content; Step 10: The receiving end sends a confirmation message to the sending end, informing that the file has been successfully received and stored, and the file transmission task is completed.

Images