Cross-layer multi-path network transmission evaluation method and system

Abstract

The invention discloses a cross-layer multi-path network transmission evaluation method and system, and the method comprises the steps: calculating a weighted signal-to-interference-noise ratio according to the internal noise intensity, signal intensity and external noise intensity corresponding to each transmission path in a state that each transmission path is marked as active and non-blocking; calculating a substream weighted aging SINR corresponding to each transmission path according to the interference-to-noise ratio corresponding to each transmission path; calculating a comprehensive score corresponding to each transmission path according to the sub-stream weighted aging SINR and the data transmission RTT corresponding to each transmission path; and selecting the transmission path with the highest comprehensive score from the comprehensive scores corresponding to the plurality of transmission paths in the MPQUIC as the main transmission path of the next round of data transmission. According to the method, the SINR and the RTT are combined to comprehensively evaluate the path quality, and the optimal transmission path is selected as a main transmission path for data transmission in the jth round of path scheduling, so that the data transmission efficiency and stability under an unstable network condition are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of network transmission evaluation, and in particular to a cross-layer multi-path network transmission evaluation method and system. Background Art

[0002] With the rapid development of digitalization and informatization, Web video streaming services have become an indispensable part of daily life. This service is of great significance for spreading knowledge, promoting cultural exchanges, and supporting global business operations. In recent years, the Multipath Quick UDP Internet Connections (MPQUIC) protocol has played an increasingly important role in Web video streaming services. By supporting multipath transmission and more flexible congestion control, it significantly improves the stability and transmission efficiency of video streaming, and further enhances the user experience and service quality of video services.

[0003] MPQUIC is a multipath transmission protocol developed by Google and based on Quick UDP Internet Connections (QUIC). MPQUIC allows data to be transmitted in parallel on multiple network paths and uses TLS1.3 as its encryption mechanism. These measures significantly improve the reliability and efficiency of MPQUIC data transmission, thereby enhancing the speed and security of network transmission. At the same time, MPQUIC also inherits the low latency characteristics of UDP. Compared with the TCP-based Multipath TCP (MPTCP) protocol, MPQUIC's innovative 0-RTT handshake mechanism significantly reduces the time for initial connection establishment and maintenance, making data transmission faster. Therefore, MPQUIC is particularly suitable for applications with high real-time requirements, such as Web video streaming services.

[0004] Figure 1 The use scenarios based on MPQUIC communication are shown. Figure 1 As shown, STA nodes located within the coverage area of ​​a wireless network can communicate through multiple paths. This multipath approach helps devices aggregate the bandwidth of multiple paths and improve effective throughput. When STA nodes are within the coverage of two or more wireless networks, they can simultaneously use the connection migration technology implemented by MPQUIC. When the original path performs poorly or is unavailable due to network congestion, link failure, or other reasons, the connection migration technology allows data flows to be transferred to other network paths to maintain or enhance the continuity and performance of communications, thereby enhancing the robustness of the communication system.

[0005] Although the MPQUIC protocol plays an important role in today's Web video streaming services, due to the instability of wireless network environments, such as user mobility, signal blocking and frequent interference, as well as bandwidth limitations and signal attenuation under weak network conditions, video streaming services are often plagued by problems such as low data transmission rate, high latency and unstable connection. These problems not only reduce the user experience, but also limit the function and application scope of video streaming services. Especially in the context of the increasing number of high-definition video and real-time interactive applications, traditional multi-path transmission protocols are even more difficult to meet the growing network performance requirements in weak network environments. Summary of the invention

[0006] The purpose of the present invention is to provide a cross-layer multi-path network transmission evaluation method and system to improve the data transmission efficiency and stability under unstable network conditions.

[0007] To achieve the above object, the present invention provides a cross-layer multi-path network transmission evaluation method, the method comprising:

[0008] Step S1: Determine whether each transmission path in the MPQUIC in the j-th round of path scheduling is in an active state; if in an active state, store each transmission path corresponding to the active state in the MPQUIC into the set of paths to be determined, and execute "step S2"; otherwise, mark each transmission path as an inactive state; wherein j is a positive integer greater than or equal to 1;

[0009] Step S2: Selecting a transmission path pi within the MPQUIC from the set of paths to be determined;

[0010] Step S3: Calculate the weighted signal-to-interference-noise ratio corresponding to the transmission path pi in the MPQUIC in the j-th round of path scheduling according to the internal noise intensity, signal intensity and external noise intensity corresponding to the transmission path pi in the MPQUIC in the j-th round of path scheduling;

[0011] Step 4: Determine whether the transmission path pi within the MPQUIC is in a blocked state in the j-th round of path scheduling; if the transmission path pi within the MPQUIC is in a blocked state in the j-th round of path scheduling, it means that the transmission path pi within the MPQUIC will not be used as the main transmission path in the next path scheduling, and execute "step S7"; if the transmission path pi within the MPQUIC is in an unblocked state in the j-th round of path scheduling, execute "step S5";

[0012] Step S5: Calculate the substream weighted effective SINR corresponding to the transmission path pi in the MPQUIC in the j-th round of path scheduling according to the interference to noise ratio of the transmission path pi in the MPQUIC in the j-th round of path scheduling;

[0013] Step S6: Calculate the comprehensive score corresponding to the transmission path pi within the MPQUIC in the j-th round of path scheduling based on the sub-flow weighted time-effective SINR and data transmission RTT corresponding to the transmission path pi within the MPQUIC in the j-th round of path scheduling;

[0014] Step S7: Determine whether i is greater than or equal to n; if i is greater than or equal to n, execute "step S8"; if i is less than n, set i=i+1 and execute "step S2"; wherein n is the total number of active transmission paths in the MPQUIC in the path set to be determined;

[0015] Step S8: Select the transmission path with the highest comprehensive score from the comprehensive scores corresponding to multiple transmission paths in the MPQUIC in the j-th round of path scheduling as the main transmission path for data transmission in the j-th round of path scheduling.

[0016] Optionally, the weighted signal to interference noise ratio corresponding to the transmission path pi within the MPQUIC in the j-th round of path scheduling is calculated according to the internal noise intensity, signal strength and external noise intensity corresponding to the transmission path pi within the MPQUIC in the j-th round of path scheduling, and the specific formula is:

[0017]

[0018] in, represents the weighted signal-to-interference-noise ratio corresponding to the transmission path pi in the MPQUIC in the j-th round of path scheduling, Represents the j-th round of path scheduling middle MPQUIC internal transmission path P i The corresponding signal strength, Represents the j-th round of path scheduling middle MPQUIC internal transmission path P i The corresponding internal noise intensity, Represents the j-th round of path scheduling middle MPQUIC internal transmission path P i The corresponding external noise intensity, β i With γ i They represent the transmission path P in MPQUIC respectively. i The signal attenuation coefficient and environmental noise coefficient.

[0019] Optionally, the substream weighted effective SINR corresponding to the transmission path pi in the MPQUIC in the j-th round of path scheduling is calculated according to the interference to noise ratio of the transmission path pi in the MPQUIC in the j-th round of path scheduling, and the specific formula is:

[0020]

[0021] in, represents the freshness-weighted SINR of the substream corresponding to the transmission path pi in the MPQUIC in the jth round of path scheduling, t cur Indicates the current calculation The time point, t prev It means that the receiving end successfully receives the data packet and extracts the interference noise ratio of the transmission path Pi in the MPQUIC multipath transmission protocol in the jth round of path scheduling. time point.

[0022] Optionally, the comprehensive score corresponding to the transmission path pi within the MPQUIC in the j-th round of path scheduling is calculated based on the sub-flow weighted time-effective SINR and data transmission RTT corresponding to the transmission path pi within the MPQUIC in the j-th round of path scheduling, and the specific formula is:

[0023]

[0024] in, represents the comprehensive score corresponding to the transmission path pi in the MPQUIC in the jth round of path scheduling based on the sub-flow weighted time-sensitive SINR and RTT, τ represents the threshold used to balance the freshness weight of SINR and the influence of RTT in path selection, represents the weighted time-sensitive SINR of the substream corresponding to the transmission path pi in MPQUIC, represents the maximum value of the sub-stream weighted time-sensitive SINR corresponding to the transmission path pi in MPQUIC, represents the round-trip delay (RTT) of data transmission corresponding to the transmission path pi in MPQUIC in the jth round of path scheduling, It represents the maximum round-trip delay (RTT) of data transmission corresponding to the transmission path pi in MPQUIC in the j-th round of path scheduling.

[0025] Optionally, the transmission path with the highest comprehensive score is selected from multiple transmission paths in the MPQUIC in the j-th round of path scheduling as the main transmission path for data transmission in the j+1-th round of path scheduling. The specific formula is:

[0026]

[0027] in, Represents the comprehensive score of all transmission paths in the j-th round of path scheduling The optimal path selected by the value ensures that data is sent through the best path under current network conditions.

[0028] The present invention also provides a cross-layer multi-path network transmission evaluation system, the system comprising:

[0029] The first judgment module is used to judge whether each transmission path in the MPQUIC is in an active state in the j-th round of path scheduling; if it is in an active state, each transmission path corresponding to the active state in the MPQUIC is stored in the path set to be judged, and the "transmission path selection module" is executed; otherwise, each transmission path is marked as an inactive state; wherein j is a positive integer greater than or equal to 1;

[0030] A transmission path selection module, used to select a transmission path pi within the MPQUIC from a set of paths to be determined;

[0031] A weighted signal interference and noise ratio calculation module, used to calculate the weighted signal interference and noise ratio corresponding to the transmission path pi in the MPQUIC in the j-th round of path scheduling according to the internal noise intensity, signal intensity and external noise intensity corresponding to the transmission path pi in the MPQUIC in the j-th round of path scheduling;

[0032] The second judgment module is used to judge whether the transmission path pi within the MPQUIC is in a blocked state in the j-th round of path scheduling; if the transmission path pi within the MPQUIC is in a blocked state in the j-th round of path scheduling, it means that the transmission path pi within the MPQUIC will not be used as the main transmission path in the next path scheduling, and the "third judgment module" is executed; if the transmission path pi within the MPQUIC is in an unblocked state in the j-th round of path scheduling, the "sub-flow weighted time-effective SINR calculation module" is executed;

[0033] A sub-stream weighted time-sensitive SINR calculation module, configured to calculate a sub-stream weighted time-sensitive SINR corresponding to a transmission path pi within the MPQUIC in the j-th round of path scheduling according to an interference noise ratio of a transmission path pi within the MPQUIC in the j-th round of path scheduling;

[0034] A comprehensive score calculation module, used to calculate the comprehensive score corresponding to the transmission path pi in the MPQUIC in the j-th round of path scheduling based on the sub-flow weighted time-effective SINR and data transmission RTT corresponding to the transmission path pi in the MPQUIC in the j-th round of path scheduling;

[0035] The third judgment module is used to judge whether i is greater than or equal to n; if i is greater than or equal to n, the "main path selection module" is executed; if i is less than n, i=i+1 is set, and the "transmission path selection module" is executed; wherein n is the total number of active transmission paths in the MPQUIC in the path set to be judged;

[0036] The main path selection module is used to select the transmission path with the highest comprehensive score from the comprehensive scores corresponding to multiple transmission paths in MPQUIC in the j-th round of path scheduling as the main path transmission path for data transmission in the j-th round of path scheduling.

[0037] Optionally, the weighted signal to interference noise ratio corresponding to the transmission path pi within the MPQUIC is calculated according to the internal noise intensity, signal strength and external noise intensity corresponding to the transmission path pi within the MPQUIC, and the specific formula is:

[0038]

[0039] in, Represents the j-th round of path scheduling middle The weighted signal-to-interference-noise ratio corresponding to the transmission path pi in MPQUIC, Represents the j-th round of path scheduling middle MPQUIC internal transmission path P i The corresponding signal strength, Represents the j-th round of path scheduling middle MPQUIC internal transmission path P i The corresponding internal noise intensity, Represents the j-th round of path scheduling middle MPQUIC internal transmission path P i The corresponding external noise intensity, β i With γ i They represent the transmission path P in MPQUIC respectively. i The signal attenuation coefficient and environmental noise coefficient.

[0040] Optionally, the substream weighted effective SINR corresponding to the transmission path pi in the MPQUIC in the j-th round of path scheduling is calculated according to the interference to noise ratio of the transmission path pi in the MPQUIC in the j-th round of path scheduling, and the specific formula is:

[0041]

[0042] in, represents the freshness-weighted SINR of the substream corresponding to the transmission path pi in the MPQUIC in the jth round of path scheduling, t cur Indicates the current calculation The time point, t prev It means that the receiving end successfully receives the data packet and extracts the interference noise ratio of the transmission path Pi in the MPQUIC multipath transmission protocol in the jth round of path scheduling. time point.

[0043] Optionally, the comprehensive score corresponding to the transmission path pi within the MPQUIC in the j-th round of path scheduling is calculated based on the sub-flow weighted time-effective SINR and data transmission RTT corresponding to the transmission path pi within the MPQUIC in the j-th round of path scheduling, and the specific formula is:

[0044]

[0045] in, represents the comprehensive score corresponding to the transmission path pi in the MPQUIC in the jth round of path scheduling based on the sub-flow weighted time-sensitive SINR and RTT, τ represents the threshold used to balance the freshness weight of SINR and the influence of RTT in path selection, represents the weighted time-sensitive SINR of the substream corresponding to the transmission path pi in MPQUIC, represents the maximum value of the sub-stream weighted time-sensitive SINR corresponding to the transmission path pi in MPQUIC, represents the round-trip delay (RTT) of data transmission corresponding to the transmission path pi in MPQUIC in the jth round of path scheduling, It represents the maximum round-trip delay (RTT) of data transmission corresponding to the transmission path pi in MPQUIC in the j-th round of path scheduling.

[0046] The present invention also provides a cross-layer multi-path network transmission evaluation system, the system comprising:

[0047] MPQUIC sender and MPQUIC receiver;

[0048] The MPQUIC sender includes: a socket send buffer, a first scheduler, a plurality of sub-stream channels, a stream send buffer, and an ACK processing module; the first scheduler is a scheduler corresponding to the MPQUIC sender;

[0049] The MPQUIC receiving end includes: a socket receiving buffer, a second scheduler, a plurality of sub-stream channels, a stream receiving buffer and a cross-layer module; the second scheduler is a scheduler corresponding to the MPQUIC receiving end;

[0050] The multiple sub-stream channels in the MPQUIC transmitter and the multiple sub-stream channels in the MPQUIC receiver are connected via a wireless network;

[0051] The first scheduler is used for steps S1-S2 and steps S4-S8 in the above-mentioned cross-layer multi-path network transmission evaluation method, and the cross-layer module is used to execute step S3 in the above-mentioned cross-layer multi-path network transmission evaluation method.

[0052] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0053] The present invention discloses a cross-layer multi-path network transmission evaluation method and system. When each transmission path is marked as active and non-blocked, a weighted signal-to-interference-noise ratio is calculated according to the internal noise intensity, signal intensity and external noise intensity corresponding to each transmission path; a substream weighted time-effective SINR corresponding to each transmission path is calculated according to the interference-to-noise ratio corresponding to each transmission path; a comprehensive score corresponding to each transmission path is calculated according to the substream weighted time-effective SINR corresponding to each transmission path and the data transmission RTT; and a transmission path with the highest comprehensive score is selected from the comprehensive scores corresponding to multiple transmission paths in MPQUIC as the main transmission path for the next round of data transmission. The present invention combines SINR and RTT to comprehensively evaluate the path quality, select the optimal transmission path as the main transmission path for data transmission in the jth round of path scheduling, and thereby improve the data transmission efficiency and stability under unstable network conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0055] Figure 1 It is a schematic diagram of the existing usage scenario based on MPQUIC communication;

[0056] Figure 2 This is a flow chart of a cross-layer multi-path network transmission evaluation method according to an embodiment of the present invention;

[0057] Figure 3 This is a structural diagram of a cross-layer multi-path network transmission evaluation based on MPSI according to an embodiment of the present invention;

[0058] Figure 4 A schematic diagram of a wireless communication scenario according to an embodiment of the present invention;

[0059] Figure 5 The figure is a schematic diagram of fairness indexes of four different path scheduling algorithms calculated using the Jain fairness index during the experiment of the embodiment of the present invention. DETAILED DESCRIPTION

[0060] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0061] The purpose of the present invention is to provide a cross-layer multi-path network transmission evaluation method and system to improve the data transmission efficiency and stability under unstable network conditions.

[0062] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0063] Example 1

[0064] like Figure 2 As shown, the present invention discloses a cross-layer multi-path network transmission evaluation method, the method comprising:

[0065] Step S1: Determine whether each transmission path in the MPQUIC in the j-th round of path scheduling is in an active state; if it is in an active state, store each transmission path corresponding to the active state in the MPQUIC into the set of paths to be determined, and execute "Step S2"; otherwise, mark each transmission path as an inactive state.

[0066] Step S2: Select the MPQUIC intra-transmission path pi from the set of paths to be determined.

[0067] Step S3: Calculate the weighted signal to interference noise ratio corresponding to the transmission path pi within the MPQUIC in the j-th round of path scheduling according to the internal noise intensity, signal intensity and external noise intensity corresponding to the transmission path pi within the MPQUIC in the j-th round of path scheduling.

[0068] Step 4: Determine whether the transmission path pi within the MPQUIC is in a blocked state in the j-th round of path scheduling; if the transmission path pi within the MPQUIC is in a blocked state in the j-th round of path scheduling, it means that the transmission path pi within the MPQUIC will not be used as the main transmission path in the next path scheduling, and execute "Step S7"; if the transmission path pi within the MPQUIC is in an unblocked state in the j-th round of path scheduling, execute "Step S5".

[0069] Step S5: Calculate the substream weighted time-sensitive SINR corresponding to the transmission path pi within the MPQUIC in the j-th round of path scheduling according to the interference to noise ratio of the transmission path pi within the MPQUIC in the j-th round of path scheduling.

[0070] Step S6: Calculate the comprehensive score corresponding to the transmission path pi within the MPQUIC in the j-th round of path scheduling based on the sub-flow weighted time-sensitive SINR and data transmission RTT corresponding to the transmission path pi within the MPQUIC in the j-th round of path scheduling.

[0071] Step S7: Determine whether i is greater than or equal to n; if i is greater than or equal to n, execute "step S8"; if i is less than n, set i=i+1 and execute "step S2"; wherein n is the total number of active transmission paths in the MPQUIC in the path set to be determined.

[0072] Step S8: Select the transmission path with the highest comprehensive score from the comprehensive scores corresponding to multiple transmission paths in the MPQUIC in the j-th round of path scheduling as the main transmission path for data transmission in the j+1-th round of path scheduling.

[0073] The following is a detailed discussion of each step:

[0074] Step S1: Determine whether each transmission path in the MPQUIC in the j-th round of path scheduling is in an active state; if it is in an active state, store each transmission path corresponding to the active state in the MPQUIC into the set of paths to be determined, and execute "step S2"; otherwise, mark each transmission path as an inactive state, indicating that each transmission path is faulty, and after a period of time, a tentative transmission path transmission judgment will be performed again to confirm whether it is still in a faulty state.

[0075] In this embodiment, if a path challenge frame or a path response frame sent by the sender is transmitted and correctly received on a transmission path, the transmission path is marked as active; the path challenge frame is used to verify the reachability of the new network path. When a QUIC endpoint wants to confirm whether its data packet can reach the other party through a new path, it sends a path challenge frame. The path response frame is a response to the path challenge frame. After receiving the path challenge frame, the receiving end will send a path response frame to confirm the received path challenge.

[0076] Step S2: Select the MPQUIC intra-transmission path pi from the set of paths to be determined.

[0077] In multipath transmission, when multiple paths with greatly different quality are used to send data at the same time, there is a high probability that packets will arrive out of order, resulting in a decrease in path transmission performance. Especially in a weak network environment, the robustness and adaptability of the transmission protocol are tested more. However, compared with a weak network environment, multipath substreams with asymmetric and random quality changes are more common. In a wireless network environment, it is unrealistic for all substreams in an MPQUIC connection to provide stable performance for data transmission. In order for MPQUIC to better utilize different network paths, a common approach is to monitor the transmission quality of each path and then select the path with the best overall quality as the main transmission path for future data packets.

[0078] Thanks to the fact that the connection identifier of MPQUIC is a 64-bit connection ID, when the user switches between WLAN and LTE, whether the IP or port changes, the connection ID in the QUIC connection can remain unchanged, so there is no need to re-create the connection. Therefore, in the case where many mobile devices such as smartphones have multiple network transmission ports, the present invention can connect multiple sub-streams of MPQUIC to the LTE and WLAN networks respectively. When the transmission conditions of the LTE or WLAN network are poor, the intelligent scheduler will switch the main transmission path for sending data packets to the sub-stream with better communication conditions, and the SINR obtained by the cross-layer model combined with other network performance indicators can judge the quality of path transmission in a certain time period in the future in real time and accurately.

[0079] SINR is critical to multiple wireless communication fields, including cellular networks, satellite communications, wireless LANs, and broadcast systems. It is directly related to the selection of modulation and coding schemes, power control strategies, antenna design, and spectrum allocation. SINR is defined as the ratio of the power of the signal strength to the total power of the external interference signal and the internal noise. It is a benchmark for evaluating the reliability of a wireless link and is usually expressed in decibels (dB). A high SINR value generally means better link quality and higher data transmission rates. Recent studies have shown that using SINR as the main performance indicator for path scheduling may be a better choice in real-time wireless Internet applications. Therefore, the present invention defines an indicator of SINR While receiving a data packet, the cross-layer model extracts the MPQUIC intra-transmission path P related to sending the packet. i (also known as substream ID), and the weighted signal strength of the transmission path Weighted internal noise intensity The intensity of external interference with the channel Since the weighted signal strength and noise intensity extracted from the ACK frame are expressed in decibels (dB), these values ​​need to be converted into power units for processing to evaluate the weighted signal-to-interference-to-noise ratio of the path.

[0080] Step S3: Calculate the weighted signal-to-interference-noise ratio corresponding to the transmission path pi in the MPQUIC in the j-th round of path scheduling according to the internal noise intensity, signal intensity and external noise intensity corresponding to the transmission path pi in the MPQUIC in the j-th round of path scheduling

[0081]

[0082] in, Represents the j-th round of path scheduling middle The weighted signal-to-interference-noise ratio corresponding to the transmission path pi in MPQUIC, Represents the j-th round of path scheduling middle MPQUIC internal transmission path P i The corresponding signal strength, Represents the j-th round of path scheduling middle MPQUIC internal transmission path P i The corresponding internal noise intensity, Represents the j-th round of path scheduling middle MPQUIC internal transmission path P i The corresponding external noise intensity, β i With γ i They represent the transmission path P in MPQUIC respectively. i The signal attenuation coefficient and environmental noise coefficient.

[0083] In this embodiment, the specific calculation formula for the signal strength corresponding to the transmission path pi in the MPQUIC in the j-th round of path scheduling is:

[0084]

[0085] Among them, P ref It is the reference power value. In network transmission, 1 milliwatt (mW) is usually used as the reference power.

[0086] In this embodiment, the calculation method of the internal noise intensity and the external noise intensity corresponding to the transmission path pi in the MPQUIC in the j-th round of path scheduling is the same as formula (2), and will not be repeated here.

[0087] By utilizing MPQUIC for multipath transmission, data can be transmitted simultaneously on multiple network paths. This approach can not only increase the bandwidth of the network and improve the reliability of data transmission, but also reduce the overall delay. However, multipath transmission introduces a new challenge: packet reordering. Since different paths may have different transmission characteristics, such as delay, bandwidth, and packet loss rate, this will result in batches of out-of-order packets being received at the receiving end. This is especially important when the MPQUIC receiver has limited buffer resources. Therefore, MPQUIC needs to effectively handle this reordering problem to ensure correct and efficient data transmission.

[0088] In the standard multipath QUIC protocol, each path of the MPQUIC connection is classified as "active" or "inactive". When the sender's scheduler finds that none of the data streams sent on a path return ACK confirmation frames, it will mark the path as inactive, indicating that the path is faulty. After a period of time, it will try to perform a tentative transmission on the path again to confirm whether it is still in a faulty state. Under this mechanism, the sender distributes data on all paths marked as active without considering the performance quality of these paths. This practice may cause problems and usually leads to a decrease in MPQUIC performance, because MPQUIC sub-flows with low SINR may interfere with the overall data transmission process. For example, when the receiver buffer has N out-of-order queues waiting for their respective related sub-packets, if these sub-packets to be received are transmitted by an active path with poor performance, the risk of data transmission errors or packet loss will increase significantly. Before the receiving end successfully receives and correctly sorts these data packets in the buffer, other out-of-order queues that rely on these to-be-sorted packets for reordering must continue to occupy the receiving end buffer until all related sub-packets arrive at the receiving end. However, the long wait for sub-packets may cause the number of out-of-order queues to accumulate, thereby occupying the limited space of the receiving buffer. In the worst case, such accumulation may cause a receiving buffer overflow error and may lead to more serious consequences, such as the premature closure of the protocol stack connection. This situation is unacceptable for maintaining an efficient and stable communication protocol, so the state management of the MPQUIC sub-flow needs to be carefully designed to ensure that the link quality is not degraded or the connection reliability is not reduced due to buffer limitations at the receiving end. As shown in previous studies. In order to identify and avoid using low-performance paths in multi-path transmission, the path manager of MPSI introduces a new path state defined by the receiving end, called "blocking state". The present invention can use SINR as the main weight to determine whether the sub-flow has entered a blocking state, as shown in formulas (3)-(5).

[0089] Step S4: Determine whether the transmission path pi within the MPQUIC is in a blocked state in the j-th round of path scheduling; if the transmission path pi within the MPQUIC is in a blocked state in the j-th round of path scheduling, it means that the transmission path pi within the MPQUIC will not be used as the main transmission path in the next path scheduling, and execute "Step S7"; if the transmission path pi within the MPQUIC is in an unblocked state in the j-th round of path scheduling, execute "Step S5";

[0090] There are two ways to judge whether the transmission path pi in MPQUIC is in a blocked state in each round of path scheduling. In the following example, the weighted signal interference noise ratio corresponding to the transmission path pi in MPQUIC in the jth round of path scheduling is used. The specific steps are as follows:

[0091] The first method is to determine whether the transmission path pi in the MPQUIC in the j-th round of path scheduling is in a blocked state, specifically including:

[0092] Step S41: Select the minimum SINR value of the substream in the j-th round of path scheduling from all substreams in the j-1th round of path scheduling and the maximum SINR of the subflow in the j-th round of path scheduling The specific calculation formula is:

[0093]

[0094] in, is the pth path in the j-1th round of path scheduling n The SINR value corresponding to the sub-flow is also called the transmission path p in the MPQUIC in the j-1th round of path scheduling. m The corresponding weighted signal-to-interference-noise ratio.

[0095] Step S42: Calculate the arithmetic mean of the substream SINRs corresponding to all active transmission paths in the j-th round of path scheduling according to the minimum substream SINR value in the j-th round of path scheduling and the maximum substream SINR value in the j-th round of path scheduling. The specific calculation formula is:

[0096]

[0097] Step S43: Determination Is it less than if Less than This means that the transmission path pi in MPQUIC in the jth round of path scheduling is in a blocked state within the future set time; if Greater than or equal to This means that the transmission path pi in MPQUIC in the jth round of path scheduling is in an unblocked state; is the weighted signal-to-interference-noise ratio corresponding to the transmission path pi in MPQUIC in the j-th round of path scheduling.

[0098] The second method is to determine whether the transmission path pi in the MPQUIC in the j-th round of path scheduling is in a blocked state, specifically including:

[0099] Step S44: Determination Is it greater than if Greater than Update if Less than or equal to Then judge Is it less than if Less than Update And execute steps S42-S43 in the first method; if Greater than or equal to Then directly execute steps S42-S43 in the first method. and It can be either a set value or Make a selection based on actual needs.

[0100] The second solution is the same as the first solution, which will not be discussed in detail here. Please refer to the first solution for details.

[0101] Due to the inherent uncertainty in the process of wireless network data propagation, the time sensitivity of SINR becomes an important consideration. For each data packet received at the receiving end, the SINR index measured during transmission may quickly reduce the relevance and accuracy of the future SINR. This phenomenon is mainly attributed to factors such as multipath propagation, channel fading, and environmental interference, which can significantly change the propagation conditions of wireless signals in a short period of time. Therefore, as time goes by, the SINR measured during this transmission may become less and less reliable in predicting future channel quality, thus posing challenges to real-time or predictive network performance optimization strategies. Therefore, in order to enable the sender to obtain the latest SINR as quickly as possible, after the receiver calculates the SINR of the substream, it will immediately update the SINR of the relevant substream and send it to the sender as the basis for the next path scheduling in the future. However, even so, the timeliness of the SINR of each substream at the sender cannot be guaranteed. Therefore, the present invention uses the timeliness of SINR as a weight, reduces the transmission priority of the substream with low timeliness SINR, and selects the substream with high SINR timeliness and high SINR as the main path for data transmission in a certain time period in the future as much as possible during scheduling, as shown in formula (6). The specific steps are as follows:

[0102] Step S5: According to the interference noise ratio of the transmission path pi in the MPQUIC in the j-th round of path scheduling Calculate the sub-flow weighted time-effective SINR corresponding to the transmission path pi in the MPQUIC in the j-th round of path scheduling. The specific calculation formula is:

[0103]

[0104] in, represents the freshness-weighted SINR of the substream corresponding to the transmission path pi in the MPQUIC in the jth round of path scheduling, t cur Indicates the current calculation The time point, t prev It means that the receiving end successfully receives the data packet and extracts the interference noise ratio of the transmission path Pi in the MPQUIC multipath transmission protocol in the jth round of path scheduling. time point.

[0105] It can be seen from formula (6) that, under the premise of scheduling and selecting high SINR sub-streams, it is also necessary to ensure that the SINR of the selected sub-streams is sufficiently timely to maintain the accuracy and effectiveness of the network performance optimization strategy.

[0106] The present invention will comprehensively judge the SINR and RTT of each sub-flow (also known as the transmission path), and schedule the sub-flow with the highest comprehensive score, which is in an active state and not judged to be in a blocked state, as the main transmission path for the next data packet to be sent. The specific steps are as follows:

[0107] Step S6: Calculate the comprehensive score corresponding to the transmission path pi within the MPQUIC in the j-th round of path scheduling based on the sub-flow weighted time-effective SINR and data transmission RTT corresponding to the transmission path pi within the MPQUIC in the j-th round of path scheduling. The specific calculation formula is:

[0108]

[0109] in, represents the comprehensive score corresponding to the transmission path pi in the MPQUIC in the jth round of path scheduling based on the sub-flow weighted time-sensitive SINR and RTT, τ represents the threshold used to balance the freshness weight of SINR and the influence of RTT in path selection, represents the weighted time-sensitive SINR of the substream corresponding to the transmission path pi in MPQUIC, represents the maximum value of the sub-stream weighted time-sensitive SINR corresponding to the transmission path pi in MPQUIC, represents the round-trip delay (RTT) of data transmission corresponding to the transmission path pi in MPQUIC in the jth round of path scheduling, It represents the maximum round-trip delay (RTT) of data transmission corresponding to the transmission path pi in MPQUIC in the j-th round of path scheduling.

[0110] In this embodiment, It represents the comprehensive score corresponding to the transmission path pi in MPQUIC in the jth round of path scheduling calculated based on the timeliness-weighted SINR and RTT. This score is used to evaluate and compare the transmission quality of different paths and decide which path is most suitable for sending the next data packet.

[0111] In formula (7), the threshold τ can be adjusted dynamically. For example, when the SINR difference of each sub-flow is small, the threshold τ can be lowered to give priority to the path with lower RTT to achieve fast data transmission. On the contrary, when the SINR difference of most non-blocking sub-flows is significant, the threshold τ can be increased to ignore the RTT weight to ensure the channel quality and guarantee the correct reception of data packets at the receiving end.

[0112] Step S7: Determine whether i is greater than or equal to n; if i is greater than or equal to n, execute "step S8"; if i is less than n, set i=i+1 and execute "step S2"; wherein n is the total number of active transmission paths in the MPQUIC in the path set to be determined.

[0113] Step S8: Select the transmission path with the highest comprehensive score from multiple transmission paths in the MPQUIC in the j-th round of path scheduling as the main transmission path for data transmission in the j+1-th round of path scheduling. The specific calculation formula is:

[0114]

[0115] in, Represents the comprehensive score of all transmission paths in the j-th round of path scheduling The optimal path selected by the value ensures that data is sent through the best path under current network conditions.

[0116] In order to meet the challenges of Web video streaming services in wireless network environments, especially the special needs under weak network conditions, the present invention proposes a cross-layer multipath network transmission evaluation method and system based on Multipath QuickUDP Internet Connections with SignalInterference Noise Ratio (MPSI for short). Compared with traditional methods, MPSI focuses on the optimization of weak network environments, and uses intelligent path selection and network quality evaluation to effectively improve the data transmission efficiency and stability under unstable network conditions, thereby significantly improving user experience and service quality. The present invention has the following advantages:

[0117] (1) The present invention achieves accurate reflection of real-time changes in wireless link conditions and identification of substreams with low weighted SINR values ​​by extending the weighted time-sensitive SINR mechanism and the substream state model (i.e., inactive state, active state, and non-blocking).

[0118] (2) The present invention introduces a comprehensive evaluation mechanism that combines SINR and RTT to comprehensively evaluate path quality. This approach provides detailed control for the optimization of substream selection, while considering signal quality, stability, and path delay to adapt to the rapid changes in the wireless network environment.

[0119] (3) The MPSI-based cross-layer multi-path network transmission evaluation method of the present invention not only optimizes the data transmission process, but also establishes an effective information exchange mechanism, so that the cross-layer path evaluation can dynamically adjust the data flow allocation according to the real-time network status. At the same time, a large number of simulation test results show that the MPSI strategy performs well in terms of effective throughput performance and fairness index.

[0120] Example 2

[0121] The present invention also discloses a cross-layer multi-path network transmission evaluation system, the system comprising:

[0122] The first judgment module is used to judge whether each transmission path in the MPQUIC is in an active state in the j-th round of path scheduling; if it is in an active state, each transmission path corresponding to the active state in the MPQUIC is stored in the path set to be judged, and the "transmission path selection module" is executed; otherwise, each transmission path is marked as an inactive state; wherein j is a positive integer greater than or equal to 1;

[0123] A transmission path selection module, used to select a transmission path pi within the MPQUIC from a set of paths to be determined;

[0124] A weighted signal interference and noise ratio calculation module, used to calculate the weighted signal interference and noise ratio corresponding to the transmission path pi in the MPQUIC in the j-th round of path scheduling according to the internal noise intensity, signal intensity and external noise intensity corresponding to the transmission path pi in the MPQUIC in the j-th round of path scheduling;

[0125] The second judgment module is used to judge whether the transmission path pi within the MPQUIC is in a blocked state in the j-th round of path scheduling; if the transmission path pi within the MPQUIC is in a blocked state in the j-th round of path scheduling, it means that the transmission path pi within the MPQUIC will not be used as the main transmission path in the next path scheduling, and the "third judgment module" is executed; if the transmission path pi within the MPQUIC is in an unblocked state in the j-th round of path scheduling, the "sub-flow weighted time-effective SINR calculation module" is executed;

[0126] A sub-stream weighted time-sensitive SINR calculation module, configured to calculate a sub-stream weighted time-sensitive SINR corresponding to a transmission path pi within the MPQUIC in the j-th round of path scheduling according to an interference noise ratio of a transmission path pi within the MPQUIC in the j-th round of path scheduling;

[0127] A comprehensive score calculation module, used to calculate the comprehensive score corresponding to the transmission path pi in the MPQUIC in the j-th round of path scheduling based on the sub-flow weighted time-effective SINR and data transmission RTT corresponding to the transmission path pi in the MPQUIC in the j-th round of path scheduling;

[0128] The third judgment module is used to judge whether i is greater than or equal to n; if i is greater than or equal to n, the "main path selection module" is executed; if i is less than n, i=i+1 is set, and the "transmission path selection module" is executed; wherein n is the total number of active transmission paths in the MPQUIC in the path set to be judged;

[0129] The main path selection module is used to select the transmission path with the highest comprehensive score from the comprehensive scores corresponding to multiple transmission paths in MPQUIC in the j-th round of path scheduling as the main path transmission path for data transmission in the j-th round of path scheduling.

[0130] The details of the parts that are the same as those in Example 1 are given in Example 1 and will not be repeated here.

[0131] Example 3

[0132] like Figure 3 As shown, the present invention also discloses a cross-layer multi-path network transmission evaluation system, the system comprising:

[0133] MPQUIC sender and MPQUIC receiver.

[0134] The MPQUIC sender includes: a socket send buffer, a first scheduler, multiple sub-stream channels, a stream send buffer and an ACK processing module; the first scheduler is a scheduler corresponding to the MPQUIC sender.

[0135] The MPQUIC receiving end includes: a socket receiving buffer, a second scheduler, multiple sub-stream channels, a stream receiving buffer and a cross-layer module; the second scheduler is a scheduler corresponding to the MPQUIC receiving end; the multiple sub-stream channels in the MPQUIC sending end and the multiple sub-stream channels in the MPQUIC receiving end are connected through a wireless network.

[0136] The working principle of each device is discussed below:

[0137] The socket send buffer in the MPQUIC sender is used to store data generated by the application layer pending further processing.

[0138] The first scheduler in the MPQUIC sender is used to schedule data from the socket send buffer and allocate data to different sub-stream channels (for example: sub-stream #1, sub-stream #2, sub-stream #3) according to network conditions, available bandwidth, etc.

[0139] The stream sending buffer in the MPQUIC sending end is used to store the data allocated to each sub-stream channel by the first scheduler.

[0140] The ACK processing model in the MPQUIC sender is used to process the confirmation (ACK) information, adjust the data transmission strategy according to the information frame attached to the received ACK, and optimize the data transmission.

[0141] The socket receive buffer on the MPQUIC receiver side is used to receive data coming from the network and pass the data to the scheduler.

[0142] The stream receive buffer of the MPQUIC receiver is responsible for storing data received from the scheduler.

[0143] The second scheduler at the MPQUIC receiving end is used to obtain data from the stream receiving buffer, reorder the obtained out-of-order data packets, and submit them to the upper layer.

[0144] The cross-layer model of the MPQUIC receiver is used to analyze and process the wireless channel indicators transmitted from the physical layer, thereby evaluating the signal quality of each sub-stream, and then optimizing the data reception process and improving the efficiency and stability of data transmission.

[0145] In the MPSI system, data is first generated at the application layer of the sender and stored in the sender's socket send buffer. The sender scheduler intelligently distributes the data in the socket send buffer to the send buffer of each stream based on the wireless channel indicators of each substream. These data are then transmitted to the receiver through the wireless network (IP). At the receiver, the scheduler collects the out-of-order data from the receive buffer of each stream, reorders it, and assembles it into a complete data packet and stores it in the socket receive buffer. When the application layer prepares enough buffer, the data packet is delivered to the upper layer for processing. The cross-layer model at the receiver is responsible for evaluating and processing the wireless link signal strength and interference noise from the physical layer, converting this information into link quality evaluation indicators, and attaching these indicators to the ACK frame as the basis for the sender's next data transmission decision.

[0146] The sender performs the same operations as native MPQUIC, and customizes a first scheduler that combines the latest SINR of each substream and the minimum RTT obtained from the receiver as the scheduling strategy. For the receiver, in addition to performing the same operations as native MPQUIC, a cross-layer model is added. The basic functions of these two modules can be summarized as follows:

[0147] Cross-layer model: The cross-layer model obtains the signal strength and internal and external noise strength of each MPQUIC sub-stream at the physical layer, calculates the SINR value of each sub-stream based on the first-hand knowledge obtained from the receiving end, and determines whether the sub-stream is in a blocked state.

[0148] The first scheduler is used to adaptively and intelligently schedule the use of multiple sub-streams in the MPQUIC connection, and through the received ACK frame with link status information calculated by the cross-layer device, combined with the RTT indicator of the current sub-stream, effectively identify and use the MPQUIC sub-stream with the highest overall quality as the main transmission path for sending the next data packet.

[0149] When the first scheduler sends a path challenge frame or a path response frame and the receiving end correctly receives this frame, the first scheduler will set the transmission path of the transmission path challenge frame or the path response frame to an active state. Specifically, step S1 and step S2 are executed; after the cross-layer model obtains the wireless channel indicators of each channel (i.e., internal noise intensity, signal strength and external noise intensity), step 3 is executed, and each link quality (weighted signal interference noise ratio) indicator is attached to the ACK frame, and the second scheduler schedules the book path to send the frame to the sending end; the sending end hands over the received ACK frame with the link quality indicator attached to the ACK processing module for link quality indicator extraction, and then hands over the extracted indicator to the first scheduler, and then the first scheduler executes steps S4-S8.

[0150] Example 4

[0151] The present invention evaluates the performance of MPSI on the Network Simulator 3 (NS-3) platform and compares its performance with other mainstream path scheduling algorithms. The experimental topology designs a typical wireless communication scenario, such as Figure 4 As shown, the sender and receiver use MPQUIC as the transport protocol.

[0152] The sender and receiver are interconnected through four independent MPQUIC paths (#1, #2, #3, and #4). In order to simulate the instability and propagation loss of the wireless link, in addition to the logarithmic distance path loss model commonly used in wireless transmission network simulation, a random propagation loss model is introduced to simulate the propagation loss of wireless signals of four MPQUIC sub-streams at different distances, thereby more accurately reflecting the quality of the wireless link in the network simulation. Unlike the distance path loss model based on a deterministic formula, the random propagation loss model increases the uncertainty of the simulation by introducing random variables, making the simulated wireless environment closer to the uncertainty and variability of the real world.

[0153] In order to more accurately simulate the random fluctuations in signal propagation path loss and highlight the robustness and adaptability of the path scheduling algorithm, this study selected the normal distribution as the probability distribution model of random path loss. The parameter setting of the normal distribution is designed to simulate a weak network environment with poor network stability, high latency (long time required for data packets to be transmitted from the sender to the receiver), and high packet loss rate (high probability of packet loss during transmission), while allowing random fluctuations within a specific range.

[0154] Since the simulation experiment mainly simulates a weak network environment, after a large number of tests, the present invention temporarily sets the τ value in Formula 7 in the subsequent experiments to 0.85. This adjustment is intended to maximize the efficiency of the MPSI scheduling algorithm under weak network conditions. The relevant parameter settings of the network topology used in the experiment are shown in Table 1.

[0155] Table 1. Related parameters of network topology

[0156]

[0157] The experiment will compare and analyze with three other mainstream path scheduling algorithms ECF, BLEST and MinRTT: By simulating the target network conditions and scenarios on the NS3 network simulator, the present invention evaluates the performance of MPSI and the other three path scheduling algorithms under the above experimental conditions.

[0158] Ideally, each subflow should have an equal opportunity to utilize network resources such as bandwidth and cache to ensure that no subflow is biased during the entire transmission process. The level of fairness directly affects the user experience and network efficiency, especially in environments with limited bandwidth or significantly changing network conditions. The fairness index of the subflow reflects the degree of equality of different data flows in the allocation of network resources during multi-path transmission. When considering network availability, fairness is more important than transmission efficiency. Therefore, fairness is also an important criterion for evaluating the performance of path scheduling algorithms. Jain's fairness index can measure the fairness of the path scheduling algorithm, as shown in formula (9).

[0159]

[0160] Throughput i represents the throughput of the sub-flow with sequence number . Through this formula, the present invention can evaluate the fairness index of each algorithm at a specific moment.

[0161] Figure 5The fairness indexes of four different path scheduling algorithms calculated using the Jain fairness index during the experiment are shown. It can be seen that the fairness indexes of the ECF and BLEST scheduling algorithms after stabilization are approximately 0.506 and 0.509, respectively. On the one hand, the ECF algorithm tends to select the path that is expected to complete the transmission first, which may cause a specific path that was previously judged to be fast to be frequently selected. In contrast, slower paths are used less frequently, thereby reducing fairness. On the other hand, the main goal of the BLEST scheduling algorithm is to alleviate buffer congestion and head-of-line blocking (HOL) phenomena. However, this approach may ignore certain subflows when transmitting larger data packets, leaving other subflows with available bandwidth idle and unable to fully realize their transmission potential. Ultimately, this affects the overall fairness between subflows.

[0162] The fairness index of the MPSI-based cross-layer multi-path network transmission evaluation algorithm disclosed in the present invention reaches 0.98 at 60 seconds and remains relatively stable at the subsequent mark points of 90 seconds and 120 seconds. This phenomenon is partly attributed to the fact that the original signal-to-noise ratio (SINR) of the four sub-streams is randomly generated by processing the same normal distribution and uniform distribution of losses, which can reflect most of the characteristics of the high-density environment and ensure that each sub-stream has the same loss probability within a specific period. In addition, the high sensitivity of the MPSI algorithm to the change of sub-stream SINR is an important factor. It can process SINR values ​​affected by random losses in real time and adjust the path selection accordingly.

[0163] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the system disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.

[0164] The principles and implementation methods of the present invention are described in this article using specific examples. The description of the above embodiments is only used to help understand the method and core idea of ​​the present invention. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting the present invention.

Claims

1. A cross-layer multi-path network transmission evaluation method, characterized in that: The method comprises: Step S1: determine whether each transmission path in the MPQUIC in the j-th round of path scheduling is in an active state; if in an active state, store each transmission path corresponding to the active state in the MPQUIC into a set of paths to be determined, and execute "step S2"; otherwise, mark each transmission path as an inactive state; wherein j is a positive integer greater than or equal to 1; Step S2: Selecting a transmission path pi within the MPQUIC from the set of paths to be determined; Step S3: Calculate the weighted signal-to-interference-noise ratio corresponding to the transmission path pi in the MPQUIC in the j-th round of path scheduling according to the internal noise intensity, signal intensity and external noise intensity corresponding to the transmission path pi in the MPQUIC in the j-th round of path scheduling; Step 4: Determine whether the transmission path pi within the MPQUIC is in a blocked state in the j-th round of path scheduling; if the transmission path pi within the MPQUIC is in a blocked state in the j-th round of path scheduling, it means that the transmission path pi within the MPQUIC will not be used as the main transmission path in the next path scheduling, and execute "step S7"; if the transmission path pi within the MPQUIC is in an unblocked state in the j-th round of path scheduling, execute "step S5"; Step S5: Calculate the substream weighted effective SINR corresponding to the transmission path pi in the MPQUIC in the j-th round of path scheduling according to the interference to noise ratio of the transmission path pi in the MPQUIC in the j-th round of path scheduling; Step S6: Calculate the comprehensive score corresponding to the transmission path pi within the MPQUIC in the j-th round of path scheduling based on the sub-flow weighted time-effective SINR and data transmission RTT corresponding to the transmission path pi within the MPQUIC in the j-th round of path scheduling; Step S7: Determine whether i is greater than or equal to n; if i is greater than or equal to n, execute "step S8"; if i is less than n, set i=i+1 and execute "step S2"; wherein n is the total number of active transmission paths in the MPQUIC in the path set to be determined; Step S8: Select the transmission path with the highest comprehensive score from the comprehensive scores corresponding to multiple transmission paths in the MPQUIC in the j-th round of path scheduling as the main transmission path for data transmission in the j-th round of path scheduling.

2. The cross-layer multipath network transmission evaluation method according to claim 1, characterized in that: The weighted signal to noise ratio corresponding to the transmission path pi within the MPQUIC in the j-th round of path scheduling is calculated according to the internal noise intensity, signal strength and external noise intensity corresponding to the transmission path pi within the MPQUIC in the j-th round of path scheduling. The specific formula is: in, represents the weighted signal-to-interference-noise ratio corresponding to the transmission path pi in the MPQUIC in the j-th round of path scheduling, represents the transmission path P in MPQUIC in the jth round of path scheduling i The corresponding signal strength, represents the transmission path P in MPQUIC in the jth round of path scheduling i The corresponding internal noise intensity, represents the transmission path P in MPQUIC in the jth round of path scheduling i The corresponding external noise intensity, β i With γ i They represent the transmission path P in MPQUIC respectively. i The signal attenuation coefficient and environmental noise coefficient.

3. The cross-layer multipath network transmission evaluation method according to claim 1, characterized in that: The substream weighted effective SINR corresponding to the transmission path pi in the MPQUIC in the j-th round of path scheduling is calculated according to the interference to noise ratio of the transmission path pi in the MPQUIC in the j-th round of path scheduling, and the specific formula is: in, represents the freshness-weighted SINR of the substream corresponding to the transmission path pi in the MPQUIC in the jth round of path scheduling, t cur Indicates the current calculation The time point, t prev It means that the receiving end successfully receives the data packet and extracts the interference noise ratio of the transmission path Pi in the MPQUIC multipath transmission protocol in the jth round of path scheduling. time point.

4. The cross-layer multi-path network transmission evaluation method according to claim 1, characterized in that: The comprehensive score corresponding to the transmission path pi within the MPQUIC in the j-th round of path scheduling is calculated based on the sub-flow weighted time-effective SINR and data transmission RTT corresponding to the transmission path pi within the MPQUIC in the j-th round of path scheduling: in, represents the comprehensive score corresponding to the transmission path pi in the MPQUIC in the jth round of path scheduling based on the sub-flow weighted time-sensitive SINR and RTT, τ represents the threshold used to balance the freshness weight of SINR and the influence of RTT in path selection, represents the weighted time-sensitive SINR of the substream corresponding to the transmission path pi in MPQUIC, represents the maximum value of the sub-stream weighted time-sensitive SINR corresponding to the transmission path pi in MPQUIC, represents the round-trip delay (RTT) of data transmission corresponding to the transmission path pi in MPQUIC in the jth round of path scheduling, It represents the maximum round-trip delay (RTT) of data transmission corresponding to the transmission path pi in MPQUIC in the j-th round of path scheduling.

5. The cross-layer multi-path network transmission evaluation method according to claim 4 is characterized in that: The specific formula for selecting the transmission path with the highest comprehensive score from multiple transmission paths in the MPQUIC in the j-th round of path scheduling as the main transmission path for data transmission in the j+1-th round of path scheduling is: in, Represents the comprehensive score of all transmission paths in the j-th round of path scheduling The optimal path selected by the value ensures that data is sent through the best path under current network conditions.

6. A cross-layer multi-path network transmission evaluation system, characterized in that: The system comprises: The first judgment module is used to judge whether each transmission path in the MPQUIC is in an active state in the j-th round of path scheduling; if it is in an active state, each transmission path corresponding to the active state in the MPQUIC is stored in the set of paths to be judged, and the "transmission path selection module" is executed; otherwise, each transmission path is marked as an inactive state; wherein j is a positive integer greater than or equal to 1; A transmission path selection module, used to select a transmission path pi within the MPQUIC from a set of paths to be determined; A weighted signal interference and noise ratio calculation module, used to calculate the weighted signal interference and noise ratio corresponding to the transmission path pi in the MPQUIC in the j-th round of path scheduling according to the internal noise intensity, signal intensity and external noise intensity corresponding to the transmission path pi in the MPQUIC in the j-th round of path scheduling; The second judgment module is used to judge whether the transmission path pi within the MPQUIC is in a blocked state in the j-th round of path scheduling; if the transmission path pi within the MPQUIC is in a blocked state in the j-th round of path scheduling, it means that the transmission path pi within the MPQUIC will not be used as the main transmission path in the next path scheduling, and the "third judgment module" is executed; if the transmission path pi within the MPQUIC is in an unblocked state in the j-th round of path scheduling, the "sub-flow weighted time-effective SINR calculation module" is executed; A sub-stream weighted time-sensitive SINR calculation module, configured to calculate a sub-stream weighted time-sensitive SINR corresponding to a transmission path pi within the MPQUIC in the j-th round of path scheduling according to an interference noise ratio of a transmission path pi within the MPQUIC in the j-th round of path scheduling; A comprehensive score calculation module, used to calculate the comprehensive score corresponding to the transmission path pi in the MPQUIC in the j-th round of path scheduling based on the sub-flow weighted time-effective SINR and data transmission RTT corresponding to the transmission path pi in the MPQUIC in the j-th round of path scheduling; The third judgment module is used to judge whether i is greater than or equal to n; if i is greater than or equal to n, the "main path selection module" is executed; if i is less than n, i=i+1 is set, and the "transmission path selection module" is executed; wherein n is the total number of active transmission paths in the MPQUIC in the path set to be judged; The main path selection module is used to select the transmission path with the highest comprehensive score from the comprehensive scores corresponding to multiple transmission paths in MPQUIC in the j-th round of path scheduling as the main path transmission path for data transmission in the j-th round of path scheduling.

7. The cross-layer multi-path network transmission evaluation system according to claim 6, characterized in that: The weighted signal to noise ratio corresponding to the transmission path pi within the MPQUIC is calculated according to the internal noise intensity, signal strength and external noise intensity corresponding to the transmission path pi within the MPQUIC, and the specific formula is: in, represents the weighted signal-to-interference-noise ratio corresponding to the transmission path pi in the MPQUIC in the j-th round of path scheduling, represents the transmission path P in MPQUIC in the jth round of path scheduling i The corresponding signal strength, represents the transmission path P in MPQUIC in the jth round of path scheduling i The corresponding internal noise intensity, represents the transmission path P in MPQUIC in the jth round of path scheduling i The corresponding external noise intensity, β i With γ i They represent the transmission path P in MPQUIC respectively. i The signal attenuation coefficient and environmental noise coefficient.

8. The cross-layer multi-path network transmission evaluation system according to claim 6, characterized in that: The substream weighted effective SINR corresponding to the transmission path pi in the MPQUIC in the j-th round of path scheduling is calculated according to the interference to noise ratio of the transmission path pi in the MPQUIC in the j-th round of path scheduling, and the specific formula is: in, represents the freshness-weighted SINR of the substream corresponding to the transmission path pi in the MPQUIC in the jth round of path scheduling, t cur Indicates the current calculation The time point, t prev It means that the receiving end successfully receives the data packet and extracts the interference noise ratio of the transmission path Pi in the MPQUIC multipath transmission protocol in the jth round of path scheduling. time point.

9. The cross-layer multi-path network transmission evaluation system according to claim 6, characterized in that: The comprehensive score corresponding to the transmission path pi within the MPQUIC in the j-th round of path scheduling is calculated based on the sub-flow weighted time-effective SINR and data transmission RTT corresponding to the transmission path pi within the MPQUIC in the j-th round of path scheduling: in, represents the comprehensive score corresponding to the transmission path pi in the MPQUIC in the jth round of path scheduling based on the sub-flow weighted time-sensitive SINR and RTT, τ represents the threshold used to balance the freshness weight of SINR and the influence of RTT in path selection, represents the weighted time-sensitive SINR of the substream corresponding to the transmission path pi in MPQUIC, represents the maximum value of the sub-stream weighted time-sensitive SINR corresponding to the transmission path pi in MPQUIC, represents the round-trip delay (RTT) of data transmission corresponding to the transmission path pi in MPQUIC in the jth round of path scheduling, It represents the maximum round-trip delay (RTT) of data transmission corresponding to the transmission path pi in MPQUIC in the j-th round of path scheduling.

10. A cross-layer multi-path network transmission evaluation system, characterized in that: The system comprises: MPQUIC sender and MPQUIC receiver; The MPQUIC sender includes: a socket send buffer, a first scheduler, a plurality of sub-stream channels, a stream send buffer, and an ACK processing module; the first scheduler is a scheduler corresponding to the MPQUIC sender; The MPQUIC receiving end includes: a socket receiving buffer, a second scheduler, a plurality of sub-stream channels, a stream receiving buffer and a cross-layer module; the second scheduler is a scheduler corresponding to the MPQUIC receiving end; The multiple sub-stream channels in the MPQUIC transmitter and the multiple sub-stream channels in the MPQUIC receiver are connected via a wireless network; The first scheduler is used to execute steps S1-S2 and steps S4-S8 in the cross-layer multi-path network transmission evaluation method described in any one of claims 1-7, and the cross-layer module is used to execute step S3 in the cross-layer multi-path network transmission evaluation method described in any one of claims 1-7.

Images