TWI902227B - Adaptive cost and reward scheduling algorithms for multiple types of data flow of 6g/b5g/5g and low-orbit satellite wireless access network system and method thereof - Google Patents

Abstract

An adaptive cost and reward scheduling algorithm for multiple types of data flow of 6G/B5G/5G and low-orbit satellite wireless access network system and a method thereof are provided. Cost of each types of flow are calculated through total number of vRBs within latency range for each type of flow and vRB status for each type of flow. Reward for arriving packets for each type of flow are initialized. Predicted number of packets arriving for each type of flow at current time or actual number of packets arriving for each type of flow at current time and predicted number of vRBs that will be allocated and occupied at current time. Rate of arrival of packets for each type of flow between current time and previous time and dynamic slope of dynamic linear reward function are calculated to calculate the reward for each type of flow arriving at the packet. Reward for packet arriving for that type of flow minus cost of that type of flow as net profit value. Packet flow of this type with net profit value greater than 0 as candidate scheduled flow. Flows of this type are selected and added to schedule according to net profit value from large to small. Therefore, the efficiency of providing flow scheduling based on flow cost and flow reward may be achieved.

Description

Adaptive cost and revenue scheduling algorithm and method for 6G/B5G/5G and LEO satellite wireless access network systems with multiple data streams

A wireless access network system and method thereof, particularly an adaptive cost and revenue scheduling algorithm for 6G/B5G/5G and LEO satellite wireless access network systems and methods for multiple types of data streams.

Several mechanisms are specified in 6G/B5G/5G communications to distinguish different slices and traffic to ensure the Quality of Service (QoS) for different types of traffic, such as eMERGENCY, eV2X, uRLLC, eMBB, mMTC, etc. Important mechanisms and specifications include: 5G New Radio (5G NR) frequency multimode sub-carrier spacing modes (numerology SCS) using different time slots and sub-carrier spacings; Type 0, Type 1, and Dynamic Switch allocation of three types of Resource Blocks (RBs); Network Slicing (NS)/Service Function Chain (SFC); and bandwidth allocation for downlink (DL) and uplink (UL) transmissions. Pre-configured allocation of Parts (BWP), as well as Network Function Virtualization (VNF) and Software-Defined Networking (SDN), etc.

Regarding packet scheduling techniques in 6G/B5G/5G, the operating costs of the communication system and the revenue (reward) from users are generally ignored. Even if the QoS requested by the user equipment (UE) can be improved, the net profit of the communication system is not optimal. In the scheduling algorithms for different traffic queues in 6G/B5G/5G, the status of available traffic blocks (RBs) is ignored, making it difficult to maximize RB utilization and scheduling performance. Different types of traffic using different SCS numerologies are ignored; considering only a single SCS cannot maximize system performance. 5G uses pre-configured BWPs for fully partitioned RB allocation, such as Type 0, Type 1, and Dynamic Switch, which is a completely static allocation mechanism. However, RB allocation is significantly affected by dynamic traffic, thus reducing 5G performance. The impact of AI-based deep learning models on process scheduling and RB allocation is not considered. Only a single Numerology SCS is considered, thus inevitably ignoring the mutual exclusion between all SCSs resulting from the allocation or cancellation of RBs by the SCS.

In summary, it can be seen that previous technologies have long suffered from the problem that the existing 6G/B5G/5G communication process does not take into account the cost and revenue of traffic packets, resulting in suboptimal net profit in traffic packet scheduling. Therefore, it is necessary to propose improved technical means to solve this problem.

In view of the problem that prior art does not consider the cost and revenue of data packets in existing 6G/B5G/5G communication processes, resulting in suboptimal net profit in data packet scheduling, this invention discloses an adaptive cost and revenue scheduling algorithm and method for 6G/B5G/5G and LEO satellite wireless access network systems with multiple data streams, wherein:

The adaptive cost and revenue scheduling algorithm disclosed in this invention is used in 6G/B5G/5G and LEO satellite wireless access network systems with multiple types of data streams for terminal devices of 6G/B5G/5G communication. It includes: a traffic packet cost calculation module, a traffic packet revenue calculation module, a net profit calculation module, and a scheduling module.

The traffic packet cost calculation module calculates the cost of each type of traffic packet by using the total number of virtual resource blocks (vRBs) within the latency range and the vRB status of each type of traffic packet. The vRB status represents the number of allocated and occupied resource blocks (RBs). The traffic packet revenue calculation module initializes the reward for each type of traffic packet arrival, predicts the number of packets of each type arriving at the current time or the actual number of packets of each type arriving at the current time, and predicts the number of allocated and occupied vRBs at the current time. It then calculates the arrival rate of each type of traffic packet between the current time and previous times and calculates dynamic linearity. The dynamic slope of the revenue function is used to calculate the revenue of each type of traffic arrival packet, where the previous time is the time point before the current time; the net profit calculation module calculates the net profit value by subtracting the cost of the traffic packet from the revenue of the traffic arrival packet of that type; and the scheduling module selects traffic packets of that type with a net profit value greater than 0 as candidate scheduling traffic packets, and then selects the corresponding traffic packets of that type in descending order of net profit value and adds them to the schedule.

The adaptive cost and revenue scheduling algorithm disclosed in this invention is applicable to 6G/B5G/5G and LEO satellite wireless access networks with various data streams, and is suitable for terminal devices of 6G/B5G/5G communication. It includes the following steps:

First, the terminal device calculates the cost of each type of traffic packet by using the total number of vRBs within the latency range and the vRB status of each type of traffic packet, where the vRB status represents the number of allocated and occupied vRBs. Next, the terminal device initializes the revenue for each type of traffic packet arrival: then, the terminal device predicts the number of packets of each type of traffic packet arriving at the current time or the actual number of packets of each type of traffic packet arriving at the current time, and predicts the number of allocated and occupied vRBs at the current time. Then, the terminal device... The system calculates the arrival rate of each type of traffic packet between the current time and the previous time, and calculates the dynamic slope of the dynamic linear revenue function to calculate the revenue of each type of traffic packet, where the previous time is the time point before the current time. Next, the terminal device calculates the net profit by subtracting the cost of the traffic packet from the revenue of that type of traffic packet. Finally, the terminal device selects traffic packets of that type with a net profit greater than 0 as candidate scheduling traffic packets, and then selects the corresponding traffic packets of that type in descending order of net profit and adds them to the schedule.

The system and method disclosed in this invention, as described above, calculate the cost of each type of traffic packet by using the total number of VPNs within the latency range and the VPN status of each type of traffic packet. It initializes the revenue of each type of traffic packet arrival, predicts the number of packets of each type arriving at the current time or the actual number of packets of each type arriving at the current time, and predicts the number of VPNs allocated and occupied at the current time. The arrival rate of each type of traffic packet and the dynamic slope of the dynamic linear revenue function are calculated between the current time and the previous time to calculate the revenue of each type of traffic packet. The net profit is calculated by subtracting the cost of the traffic packet from the revenue of the traffic packet. Traffic packets of the type with a net profit greater than 0 are selected as candidate scheduling traffic packets. Then, the corresponding traffic packets of the type are selected and added to the schedule in descending order of net profit.

Through the above-mentioned technical means, the present invention can achieve the technical effect of providing traffic packet scheduling based on traffic packet cost and traffic packet revenue.

The following will explain in detail the implementation of the present invention with the aid of diagrams and examples, so as to fully understand how the present invention uses technical means to solve technical problems and achieve technical effects and to implement it accordingly.

The following section will first explain the adaptive cost and revenue scheduling algorithm disclosed in this invention in a 6G/B5G/5G and LEO satellite wireless access network system with multiple data streams. Please refer to Figure 1, which is a system block diagram of the adaptive cost and revenue scheduling algorithm of this invention in a 6G/B5G/5G and LEO satellite wireless access network system with multiple data streams.

The adaptive cost and revenue scheduling algorithm disclosed in this invention is used in 6G/B5G/5G and LEO satellite wireless access network systems for various types of data streams. The terminal device 10 for 6G/B5G/5G communication includes: a traffic packet cost calculation module 11, a traffic packet revenue calculation module 12, a net profit calculation module 13, and a scheduling module 14. It is worth noting that the terminal devices 10 can communicate with each other through physical networks, physical links, virtual networks, and combinations of virtual links in 6G/B5G/5G communication. The terminal devices 10 are, for example, vehicle-mounted devices, mobile devices, etc., which are only examples and are not intended to limit the application scope of this invention.

Traffic packet cost calculation module 11 calculates the cost of each type of traffic packet by using the total number of virtual resource blocks (vRBs) within the latency range and the vRB status of each type of traffic packet, where the vRB status represents the number of allocated and occupied RBs. Traffic packet revenue calculation module 12 initializes the revenue (reward) for each type of traffic packet arrival, predicts the number of packets of each type of traffic packet arriving at the current time or the actual number of packets of each type of traffic packet arriving at the current time, and calculates the cost of each type of traffic packet based on the predicted number of allocated and occupied vRBs at the current time. The system calculates the revenue of each type of traffic arrival packet by measuring the arrival rate between the current time and the previous time and calculating the dynamic slope of the dynamic linear revenue function. The previous time is the time point before the current time. The net profit calculation module 13 calculates the net profit value by subtracting the cost of the traffic packet from the revenue of the traffic packet of that type. The scheduling module 14 selects traffic packets of that type with a net profit value greater than 0 as candidate scheduling traffic packets and then selects the corresponding traffic packets of that type in descending order of net profit value and adds them to the schedule.

In 6G/B5G/5G communications, seven different SCS modes are defined to support subframe (1ms) time slots with different transmission latency requirements. For each type of traffic packet's SCS, the algorithm for determining the slot time in the time domain is defined as follows: The algorithm for bandwidth in the frequency domain is defined as follows: That is, we can obtain the SCS It boasts the longest time slot duration but the shortest bandwidth, making it suitable for non-real-time (NRT) applications with the lowest priority traffic, such as mMTC, best-effort, etc. Relatively speaking, for SCS... It has the shortest time slot but the longest bandwidth, making it suitable for real-time (RT) applications with the highest priority traffic, such as uRLLC, eMERGENCY, etc.

The characteristics of RBs within a specific bandwidth include: the maximum number of RBs for each SCS within a given spectral bandwidth is specified and limited. Specifically, for a 30kHz SCS at 100MHz, the maximum number of RBs is 273. This is merely an example and does not limit the application scope of this invention. Furthermore, for a given spectral bandwidth, the maximum number of RBs varies for different SCSs. Specifically, at a bandwidth of 20MHz, for an SCS of... , as well as At that time, the maximum number of RBs that can be used are 106 RBs, 51 RBs and 24 RBs respectively. These are only examples for illustration and are not intended to limit the application scope of this invention.

Therefore, 6G/B5G/5G has the following characteristics in terms of traffic packet scheduling and RB allocation:

Using a higher SCS results in a smaller maximum number of RBs, therefore the carrying cost of RBs with a higher SCS is greater than that with RBs with a lower SCS. When there is a higher number of available RBs, the carrying cost is lower than that with a lower number of available RBs. In other words, the carrying cost of a higher number of available RBs is equal to the required lower carrying cost. Higher priority terminal device (UE) traffic brings higher access profitability to 6G/B5G/5G network providers, and vice versa.

The traffic packet cost calculation module 11 calculates the cost of each type of traffic packet by using the total number of VPNs within the latency range and the VPN status of each type of traffic packet. The total number of VPNs within the latency range and the VPN status of each type of traffic packet can be presented using an adaptive exponential cost function, as shown below:

in, This represents the total number of VRRBs for each type of traffic packet within the latency range. For each type of traffic packet, the vRB status is defined by the number of allocated and occupied RBs. Let be the integer factor of the exponential function, and It is the inverse function of the sum of the remaining usable RBs in vRB, as shown below:

Specifically, in the calculation of the adaptive exponential cost function This shows that as the total number of allocated and occupied RBs increases, the carrying cost of RBs increases exponentially. This is reflected in the calculation of the adaptive exponential cost function. As the number of available RBs decreases, the carrying cost of RBs also increases exponentially. Therefore, under the above circumstances, the carrying cost of RBs... Significant increase.

Please refer to Figure 2. Figure 2 is a schematic diagram of the carrying cost of different SCSs under different numbers of RBs used within the specified delay range of this invention. It shows the adaptive cost function of different types of traffic packets under different frequency numberology SCSs, using different numbers of vRBs (i.e., the sum of allocated and occupied vRBs). The traffic cost 21 is shown in Figure 2.

like The delay limit is 5 milliseconds. The delay threshold is 10 milliseconds, and The delay threshold is 50 milliseconds. With a bandwidth of 10 MHz, based on the aforementioned delay thresholds, the maximum number of available RBs for different SCSs is obtained, i.e. The maximum number of available RBs is 2600 (52) 1 50), The maximum number of available RBs is 480 (24 2 10), The maximum number of available RBs is 220 (11) 4 5) This is merely an example and is not intended to limit the scope of application of this invention.

Further illustrating the cost of traffic packets with practical examples, The delay threshold is 5 milliseconds and Taking a maximum available number of RBs of 220 as an example, if all allocated and occupied vRBs (i.e. The remaining number of RBs that can be allocated is 20 (i.e. Integer factor ,get as follows:

Further obtain The carrying cost is as follows:

As shown in Figure 2, the adaptation cost increases exponentially with the increase of the number of vRBs used in different SCS modes. For each vRB state in an SCS mode, when the number of RBs used increases to the maximum number of available RBs, the carrying cost of the RBs will increase to the maximum cost of 1.0. Conversely, in the initial state, the number of RBs used is 0, and the carrying cost of the RBs will be the minimum cost of 0.0.

In 6G/B5G/5G communications, different types of traffic packets will have different revenues. Higher-priority traffic packets have higher revenues in communications. The traffic packet revenue calculation module 12 formulates dynamic linear reward functions for different types of traffic packets based on key factors, and initializes the revenue of each type of traffic arrival packet. Predict the current time for each type of traffic packet ( The number of packets that arrived ( ) or the actual traffic packets of each type at the current time ( The number of packets that arrived ( ) and predictions at present ( The number of vRBs allocated and occupied ( ), calculate the current time for each type of traffic packet ( ) and previous time ( The arrival rate between () ) and calculate the dynamic slope of the dynamic linear revenue function ( When the number of RBs used increases or the number of RBs available decreases, the revenue from communication traffic packets will increase dynamically and linearly, but the carrying cost will grow exponentially for the adaptive exponential cost function.

Please refer to Figure 3, which is a schematic diagram of the dynamic slope of the dynamic linear revenue function of this invention. First, set the initial point to 31 (0, ), and set the slope of the dynamic linear revenue function to 0, that is During initialization, the dynamic linear revenue function is set to a horizontal revenue function, and the dynamic linear revenue function and the adaptive exponential cost function have an intersection point 321. , The dynamic linear yield line function is derived and determined according to the following formula:

Next, the state of vRB and all influencing factors will be dynamically updated, causing the revenue function and costs to change dynamically. The dynamic linear revenue function should be updated according to the latest state, and the slope of the dynamic linear revenue function... The arrival rate increases with the number of VRRBs used or the arrival rate of traffic packets. The arrival rate of traffic packets is calculated using the following formula:

in, To predict in the current time ( The number of packets arriving in the traffic packet or the actual number of packets arriving at the current time. The number of packets that the traffic packets arrive at.

Predictions at present ( The number of vRBs that have been allocated or occupied is calculated using the following formula:

If new When estimated, a new crossover point 322 can be obtained by combining it with the adaptive exponential cost function. , The slope of the dynamic linear revenue function is updated based on the new crossover point 322, in the current time ( The adaptive exponential cost function of the latest state in the algorithm is as follows:

Further illustrating the revenue generated from traffic delivery packets using practical examples, assuming the traffic type... In the previous time ( ) and current time ( The number of packets arriving were 450 ( ) and 475 ( ) , in the previous time ( The number of vRBs used is 200. ), arrival rate of traffic packets as follows:

In the current time ( Predicted number of vRBs as follows:

When obtained At that time, the new cost value is as follows:

The new intersection point 322 is obtained by further calculation. And based on the new crossover point 322, the slope of the dynamic linear revenue function is calculated. as follows:

In the current time ( The flow type is The dynamic linear revenue function is as follows:

Please refer to Figure 4 for the adaptive exponential cost function and dynamic linear revenue function of different SCS modes with different vRB numbers. Figure 4 is a schematic diagram of the adaptive exponential cost function and dynamic linear revenue function of different SCS modes of this invention with different vRB numbers.

Based on the adaptive exponential cost function and the dynamic linear revenue function, the scheduling of traffic packets to obtain the highest profit value is the highest scheduling priority for optimal decision-making. The formula is as follows:

Based on traffic packetization and vRB minimizing the queuing latency and packet loss rate of the highest and higher priority traffic, while maximizing vRB utilization and completely sharing vRB resources among all SCSs, and only when the revenue from the arrival of a certain type of traffic packet is greater than the cost of that type of traffic packet (i.e., This approach prioritizes candidate traffic packets for scheduling, thus ensuring positive revenue and net profit while simultaneously guaranteeing the Quality of Service (QoS) of the traffic packets. It prioritizes the traffic packets with the highest net profit for scheduling, thereby maximizing communication net profit. This optimal traffic packet scheduling contribution, based on the state of the VRRB, is completely different from existing optimal traffic packet scheduling methods and scheduling methods that only focus on one or more parameters or fairness indicators.

Next, the operation method of this invention will be explained below, and please refer to "Figure 5". "Figure 5" is a flowchart of the method of the adaptive cost and revenue scheduling algorithm of this invention for 6G/B5G/5G and low-orbit satellite wireless access networks with multiple types of data streams.

The adaptive cost and revenue scheduling algorithm disclosed in this invention is applicable to 6G/B5G/5G and LEO satellite wireless access networks with various data streams, and is suitable for terminal devices of 6G/B5G/5G communication. It includes the following steps:

First, the terminal device calculates the cost of each type of traffic packet by using the total number of vRBs within the latency range and the vRB status of each type of traffic packet, where the vRB status represents the number of allocated and occupied vRBs (step 401); next, the terminal device initializes the revenue of each type of traffic packet arrival (step 402); then, the terminal device predicts the number of packets of each type of traffic packet arriving at the current time or the actual number of packets of each type of traffic packet arriving at the current time and predicts the number of allocated and occupied vRBs at the current time (step 403); then, the terminal device... Calculate the arrival rate of each type of traffic packet between the current time and the previous time, and calculate the dynamic slope of the dynamic linear revenue function to calculate the revenue of each type of traffic packet, where the previous time is the time point before the current time (step 404); next, the terminal device calculates the net profit by subtracting the cost of the traffic packet from the revenue of the traffic packet of that type (step 405); finally, the terminal device selects the traffic packets of that type with a net profit greater than 0 as candidate scheduling traffic packets, and then selects the corresponding traffic packets of that type in descending order of net profit and adds them to the schedule (step 406).

In summary, by analyzing the total number of VPNs for each type of traffic packet within the latency range and the VPN status of each type of traffic packet, the cost of each type of traffic packet is calculated. The revenue of each type of traffic packet is initialized, and the number of packets of each type arriving at the current time is predicted, or the actual number of packets of each type arriving at the current time is determined, along with the predicted number of VPNs allocated and occupied at the current time. This allows for the calculation of the cost of each type of traffic packet. The arrival rate of each type of traffic packet between the current time and the previous time, as well as the dynamic slope of the dynamic linear revenue function, are used to calculate the revenue of each type of traffic packet. The net profit is calculated by subtracting the cost of the traffic packet from the revenue of that type of traffic packet. Traffic packets of that type with a net profit greater than 0 are selected as candidate scheduling traffic packets. Then, traffic packets of that type are selected and added to the schedule in descending order of net profit.

This technology can solve the problem that existing technologies do not take into account the cost and revenue of data packets in the current 6G/B5G/5G communication process, resulting in suboptimal net profit in data packet scheduling. It can then provide adaptive cost and revenue scheduling algorithms for 6G/B5G/5G and LEO satellite wireless access networks with various data streams.

Although the embodiments disclosed in this invention are as described above, the content described is not intended to directly limit the scope of patent protection of this invention. Anyone skilled in the art to which this invention pertains may make some modifications in form and details of the implementation without departing from the spirit and scope disclosed in this invention. The scope of patent protection of this invention shall still be determined by the scope of the attached patent application.

10: Terminal Device 11: Traffic Packet Cost Calculation Module 12: Traffic Packet Revenue Calculation Module 13: Net Profit Calculation Module 14: Scheduling Module 21: Traffic Cost 22: Traffic Revenue 31: Initial Point 321: Crossover Point 322: Crossover Point Step 401: The terminal device calculates the cost of each type of traffic packet by using the total number of vRBs within the latency range and the vRB status of each type of traffic packet, where the vRB status represents the number of allocated and occupied RBs. 402: The terminal device initializes the revenue of each type of traffic arrival packet. 403: The terminal device predicts the number of packets of each type arriving at the current time, or the actual number of packets of each type arriving at the current time, and predicts the number of VRBs allocated and occupied at the current time. 404: The terminal device calculates the arrival rate of each type of traffic packet between the current time and the previous time, and calculates the dynamic slope of the dynamic linear revenue function to calculate the revenue for each type of traffic packet, where the previous time is the time point preceding the current time. 405: The terminal device calculates the net profit by subtracting the cost of that type of traffic packet from the revenue for that type of traffic packet. 406: The terminal device selects traffic packets of this type with a net profit value greater than 0 as candidate scheduling traffic packets, and then selects the corresponding traffic packets of this type in descending order of net profit value and adds them to the schedule.

Figure 1 is a system block diagram of the adaptive cost and revenue scheduling algorithm of the present invention applied to 6G/B5G/5G and LEO satellite wireless access network systems with multiple data streams. Figure 2 is a schematic diagram of the carrying cost of different SCSs under different numbers of RBs used within the specified latency range of the present invention. Figure 3 is a schematic diagram of the dynamic slope of the dynamic linear revenue function of the present invention. Figure 4 is a schematic diagram of the adaptive exponential cost function and dynamic linear revenue function of different SCS modes of the present invention with different numbers of vRBs. Figure 5 is a flowchart of the adaptive cost and revenue scheduling algorithm of the present invention applied to 6G/B5G/5G and LEO satellite wireless access network systems with multiple data streams.

10: Terminal Devices

11: Traffic Packet Cost Calculation Module

12: Traffic Packet Revenue Calculation Module

13: Net Profit Calculation Module

14: Scheduling Module

Claims (10)

An adaptive cost and revenue scheduling algorithm for 6G/B5G/5G and LEO satellite wireless access network systems with multiple data streams, applicable to terminal devices for 6G/B5G/5G communication, includes: a traffic packet cost calculation module, which calculates the cost of each type of traffic packet within the latency range using virtual wireless resource blocks (VRRPs). The total number of blocks (vRBs) and the vRB status of each type of traffic packet are used to calculate the cost of each type of traffic packet. The vRB status represents the number of allocated and occupied vRBs. A traffic packet revenue calculation module initializes the revenue of each type of traffic packet arrival, predicts the number of packets of each type of traffic packet arriving at the current time or the actual number of packets of each type of traffic packet arriving at the current time, and predicts the number of allocated and occupied vRBs at the current time. It then calculates the cost of each type of traffic packet at the current time and... The system includes: a packet arrival rate over a previous time period and a dynamic slope of a dynamic linear revenue function to calculate the revenue of each type of traffic packet, where the previous time period is the time point before the current time period; a net profit calculation module to calculate the net profit value by subtracting the cost of the traffic packet from the revenue of the traffic packet of that type; and a scheduling module to select traffic packets of that type with a net profit value greater than 0 as candidate scheduling traffic packets, and then sequentially select the corresponding traffic packets of that type in descending order of net profit value and add them to the schedule. As described in Request 1, the adaptive cost and revenue scheduling algorithm is applied to 6G/B5G/5G and LORDS wireless access network systems with multiple data streams. The traffic packet cost calculation module presents the total number of virtual redundancies (VRBs) for each type of traffic packet within the latency range and the VRB status for each type of traffic packet using an adaptive exponential cost function. The following formula is used: in, This represents the total number of VRRBs for each type of traffic packet within the latency range. For each type of traffic packet, the vRB status is defined by the number of allocated and occupied RBs. is the integer factor of the exponential function. This is the sharpness factor parameter of the exponential function. The adaptive cost and revenue scheduling algorithm described in claim 1 is applied to 6G/B5G/5G and LEO satellite wireless access network systems with multiple data streams, wherein the arrival packet rate for each type of traffic packet is calculated by the traffic packet revenue calculation module using the following formula: in, To predict in the current time ( The number of packets arriving in the traffic packet or the actual number of packets arriving at the current time. The number of packets that the traffic packets arrive at. The adaptive cost and revenue scheduling algorithm described in claim 1 for 6G/B5G/5G and LEO satellite wireless access network systems with multiple data streams, wherein the measured packet revenue calculation module predicts the number of vRBs allocated and occupied at the current time using the following formula: . The adaptive cost and revenue scheduling algorithm described in claim 1 is applied to 6G/B5G/5G and LORDS wireless access network systems with multiple data streams, wherein the scheduling module determines the highest scheduling priority based on optimality. The following formula is used to calculate: in, Revenue generated from this type of traffic arrival packets. The cost of this type of traffic packet. An adaptive cost and revenue scheduling algorithm for 6G/B5G/5G and LEO satellite wireless access networks with multiple data streams, applicable to terminal devices of 6G/B5G/5G communication, includes the following steps: the terminal device uses virtual wireless resource blocks (VRRPs) within the latency range for each type of traffic packet. The terminal device calculates the cost of each type of traffic packet by including the total number of blocks (vRBs) and the vRB status of each type of traffic packet, where the vRB status represents the number of allocated and occupied vRBs; the terminal device initializes the revenue of each type of traffic packet arrival; the terminal device predicts the number of packets of each type of traffic packet arriving at the current time or the actual number of packets of each type of traffic packet arriving at the current time, and predicts the number of allocated and occupied vRBs at the current time; the terminal device calculates the cost of each type of traffic packet in the current... The arrival rate of packets between previous and previous times is used to calculate the dynamic slope of the dynamic linear revenue function, thereby calculating the revenue of each type of traffic arrival packet, where the previous time is the time point before the current time; the terminal device calculates the net profit value by subtracting the cost of the traffic packet from the revenue of the traffic arrival packet of that type; and the terminal device selects the traffic packets of that type with a net profit value greater than 0 as candidate scheduling traffic packets, and then selects the corresponding traffic packets of that type in descending order of net profit value and adds them to the schedule. The adaptive cost and revenue scheduling algorithm described in claim 6 is applied to 6G/B5G/5G and LEO satellite wireless access networks with multiple data streams. The terminal device presents the total number of virtual redundancies (VRBs) for each type of traffic packet within the latency range and the VRB status for each type of traffic packet using an adaptive exponential cost function. The following formula is used: in, This represents the total number of VRRBs for each type of traffic packet within the latency range. For each type of traffic packet, the vRB status is defined by the number of allocated and occupied RBs. is the integer factor of the exponential function. This is the sharpness factor parameter of the exponential function. The adaptive cost and revenue scheduling algorithm described in claim 6 is applied to 6G/B5G/5G and LEO satellite wireless access networks with multiple data streams, wherein the arrival packet rate of the terminal device for each type of traffic packet is calculated by the following formula: in, To predict in the current time ( The number of packets arriving in the traffic packet or the actual number of packets arriving at the current time. The number of packets that the traffic packets arrive at. The adaptive cost and revenue scheduling algorithm described in claim 6 is applied to 6G/B5G/5G and LEO satellite wireless access networks with multiple data streams, wherein the terminal device predicts the number of vRBs allocated and occupied at the current time using the following formula: . The adaptive cost and revenue scheduling algorithm described in claim 6 is applied to 6G/B5G/5G and LEO satellite wireless access networks with multiple data streams, wherein the terminal device is assigned the highest scheduling priority based on the optimal decision. The following formula is used to calculate: in, Revenue generated from this type of traffic arrival packets. The cost of this type of traffic packet.