Discrete time-based 5G private network scheduling method and system, and 5G base station

Abstract

The invention discloses a 5G private network scheduling method and system based on discrete time and a 5G base station, and relates to the technical field of data communication. A DCS (Distributed Control System) scheduler based on discrete time information is deployed between an MAC (Media Access Control) layer and an RLC (Radio Link Control) layer at a base station side, the DCS scheduler is used for dynamically managing a credit value of a UE (User Equipment) queue, and the queue with qualified credit is submitted to the MAC layer based on a credit threshold by dynamically updating the credit value of the UE queue. According to the invention, the DCS scheduler is introduced between the MAC layer and the RLC layer, the UE queue forwarding is quantitatively controlled by combining the 5G wireless side air interface characteristics and based on a credit authorization mechanism of discrete time slots, the end-to-end deterministic transmission performance is improved, and the air interface resource utilization rate is improved; besides, through partial authorization means, the credit of the UE transmission queue can be recovered more quickly, the queuing delay of the queue is reduced, and credit border crossing and conflicts caused by traffic burst are reduced.

Description

Technical Field

[0001] The present invention relates to the field of data communication technology, and in particular to a discrete-time-based 5G private network scheduling method, system, and 5G base station. Background Art

[0002] One of the core goals of fifth-generation mobile communication technology (5G) is to meet the stringent requirements of ultra-reliable low-latency communication (URLLC) scenarios, such as industrial automation control, remote surgery, and autonomous driving. This requires achieving end-to-end latency below 1ms and reliability up to 99.99%. However, the uplink scheduling mechanisms of traditional 4G / LTE and current 5G systems suffer from key bottlenecks such as high scheduling latency, rigid resource allocation, and insufficient predictive capabilities.

[0003] Existing TSN converters mainly map Ethernet priorities to 5G Quality of Service (QoS) identifiers in 5G network scheduling, but keep the wireless scheduler unchanged. As a result, frame-level accuracy is lost during data transmission on the wireless side, and end-to-end deterministic transmission cannot be met. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a discrete-time based 5G private network scheduling method, system and 5G base station to improve the end-to-end deterministic transmission performance.

[0005] In order to solve the above technical problems, the present invention adopts the following technical solutions: The first aspect of the present invention provides a discrete-time-based 5G private network scheduling method, in which a DCS scheduler based on discrete-time information is deployed between the MAC layer and the RLC layer on the base station side. The DCS scheduler dynamically updates the UE queue credit value and submits the credit-qualified queue to the MAC layer based on the credit threshold. A second aspect of the present invention provides a 5G base station, including an RLC layer and a MAC layer, wherein a DCS scheduler based on discrete time information is deployed between the MAC layer and the RLC layer, the DCS scheduler being used to dynamically update a UE queue credit value and submit a credit-qualified queue to the MAC layer based on a credit threshold; A third aspect of the present invention provides a discrete-time-based 5G private network scheduling system, comprising the above-mentioned 5G base station; The beneficial technical effect of the present invention is that the above-mentioned discrete time-based 5G private network scheduling method, system and 5G base station, by introducing a DCS scheduler between the MAC layer and the RLC layer, combined with the 5G wireless side air interface characteristics, based on the credit authorization mechanism of discrete time slots, quantizes the UE queue forwarding, improves the end-to-end deterministic transmission performance and improves the air interface resource utilization. BRIEF DESCRIPTION OF THE DRAWINGS

[0006] Figure 1 This is a schematic structural diagram of a 5G base station according to the present invention; Figure 2 Schematic diagram of the process of the discrete time-based 5G private network scheduling method of the present invention; Figure 3 This is the credit value change curve in the discrete time-based 5G private network scheduling method of the present invention. DETAILED DESCRIPTION

[0007] In order to enable those skilled in the art to more clearly understand the objectives, technical solutions and advantages of the present invention, the present invention is further described below with reference to the accompanying drawings and embodiments.

[0008] The present invention provides a 5G private network scheduling method, system and 5G base station based on discrete time, which normally connect the user equipment (UE) and the core network at the service layer. A DCS (Discrete Credit Schedule Unit) scheduler based on discrete time information is deployed between the MAC (Media Access Control) layer and the RLC (Radio Link Control) layer on the base station side. The DCS scheduler is used to dynamically manage the credit value of the UE queue, dynamically update the UE queue credit value, and submit the credit-qualified queue to the MAC layer based on the credit threshold. The present invention introduces a DCS scheduler between the MAC layer and the RLC layer, combines the air interface characteristics of the 5G wireless side, and performs quantitative control on the UE queue forwarding based on a credit authorization mechanism based on discrete time slots, thereby improving end-to-end deterministic transmission performance and improving air interface resource utilization.

[0009] Figure 1 The structural diagram of the 5G base station of the present invention is shown. Figure 1 As shown in Figure 1, a 5G base station (gNB) consists of the RLC and MAC layers, with a discrete-time information-based DCS scheduler deployed between them. During data transmission, the existing MAC layer scheduling strategy remains unchanged. Within a time slot cycle with a fixed subcarrier spacing (e.g., subcarrier spacing = 15kHz, time slot = 1ms), data flows through the DCS scheduler, where queue credit shaping is enhanced. This controls the qualified queues output by the DCS scheduler for each TTI (Transmission Time Interval) cycle. This is equivalent to adding a queue credit-based logic gate to the standard MAC layer scheduling logic for enhanced control, enabling quantitative control of UE queue forwarding.

[0010] The 5G private network scheduling method based on discrete time of the present invention is based on Figure 1 The gNB optimizes the scheduling of the data flow of the UE queue, such as Figure 2As shown, the discrete time-based 5G private network scheduling method includes the following steps S1 to S4: S1, 5G new air interface sends downlink data to the 5G base station.

[0011] Specifically, the 5G DN (New Radio) sends downlink data, which is forwarded to the gNB via the UPF (User Plane Function) on the user plane.

[0012] S2. The RLC layer of the 5G base station forwards the downlink data to the DCS scheduler.

[0013] Specifically, downlink data is forwarded to the RLC layer via the PDCP (Packet Data Convergence Protocol) layer at the gNB, mapped to the UM entity in the Unacknowledged (UM) mode, and then forwarded to the DCS scheduler before the MAC layer via the PDU (Protocol Data Unit) unit of the RLC layer.

[0014] During data flow transmission at the RLC layer, the RLC layer operates in UM mode to avoid uncontrolled delays caused by ARQ (Automatic Repeat reQuest) retransmissions. After data is transferred from the RLC layer's PDU unit to the MAC layer, subsequent retransmissions are performed solely via the MAC layer's HARQ (Hybrid Automatic Repeat reQuest) protocol, without revisiting the RLC layer. The DCS scheduler, acting as an intermediary between the MAC and RLC layers, manages transmission queue credits and optimizes MAC layer scheduling.

[0015] S3. The DCS scheduler performs UE queue credit management on downlink data. Within each time slot period under a fixed subcarrier interval, the DCS scheduler dynamically updates the UE queue credit value and submits the queue with qualified credit to the MAC layer based on the credit threshold.

[0016] After the data stream reaches the DCS scheduler, the DCS scheduler performs credit management on the queue buffer based on the NR discrete time slot information, establishes credit gating for admission control, and quantifies the control of the next forwarding time point. The eligible queue submitted to the MAC layer within each TTI (Transmission Time Interval) transmission cycle is determined and optimized by the DCS scheduler.

[0017] The step S3 further comprises: S31. Establish an idle state accumulation and working state deduction mechanism for UE queue credit.

[0018] Define the credit dynamics of UE queues. When a UE queue has data to transmit and its credit value is non-negative, the queue enters the working state, starts the hardware network card queue, and begins transmission based on priority. The transmission process consumes the corresponding queue's credit value. When the UE queue is idle, it automatically switches to the idle state, and the queue's credit value accumulates. This dynamic credit consumption and accumulation prevents queues from starving due to long wait times and improves the fairness of UE queue transmission. Specifically, the queue state is divided into Idle State and Work State. The idle state credit accumulation method is: idleValue = (t2-t1) * B, where idleValue is the credit value accumulated in the idle state, B is the bandwidth share reserved for traffic over time, and (t2-t1) is the time interval between any two moments in the idle state. The working state credit deduction method is: workValue = (t´2-t´1) * S, where workValue is the credit value deducted in the working state, (t´2-t´1) is the time interval between any two moments in the working state, and S is the credit consumption generated by the actual transmission rate over time within (t´2-t´1). S = B - L, and L is the loss during transmission based on the physical link transmission rate.

[0019] S32. Configure the maximum credit value and the minimum credit value to limit the upper and lower thresholds of multi-queue transmission to avoid abnormal credit crossing and excessive consumption.

[0020] Specifically, the maximum credit value is defined as max_credit, and the minimum credit value is defined as min_credit. Figure 3 As shown in the figure, when the UE queue credit value reaches the maximum credit value max_credit, the queue credit value stops accumulating, limiting the maximum burst value transmitted by a queue after the idle state to avoid abnormal credit crossing the limit; when the UE queue credit value drops to min_credit, the queue exits the working state and the queue credit value stops consuming, ensuring that there is still credit value after the working state to avoid excessive credit consumption.

[0021] S33: Update the UE queue credit value and perform queue transmission authorization.

[0022] During data stream transmission at the PHY (Physical) layer, the standard defines the subcarrier spacing and time slot as the time slot period with a fixed subcarrier spacing. The bandwidth share within this period (used to calculate the credit accumulation value in the idle state) is hashed to the number of resource blocks (RBs). Within the slot period, the MCS (Modulation and Coding Scheme) index is determined based on the standard CQI to calculate the TBS (Transport Block Size).

[0023] When quantizing the forwarding time of a UE queue, the cumulative available transmission capacity at the start of time slot n is defined as C(n), which determines the UE's eligibility to receive new transmission grants. Within time slot n, C(n) -> C(n+1) follows the idle state credit accumulation and active state credit expenditure mechanism (i.e., the idle state accumulation and active state deduction mechanism mentioned above). Deductions are made based on the reserved bandwidth and credit consumption during transmission. ΔC(n) = idleValue * T(slot), where T(slot) is the time slot period with a fixed subcarrier spacing. After the end of time slot n, the UE queue credit value is updated using the formula C(n+1) = f(C(n), ΔC(n), Q(n)) - D(n).

[0024] in, ; Where D(n) is the credit deduction amount within time slot n, which is equivalent to the resources granted within time slot n in the MAC layer. It is the TBS block size, which is related to channel quality and random access and is a random value, that is, D(n) = TBS(n); Q(n) is the buffer of the UE queue.

[0025] During actual data transmission, the randomness of D(n) can lead to unstable transmission. Furthermore, if the TBS block value exceeds the Q(u) buffer size, a credit deficit can occur, resulting in excessively penalizing the UE and delaying the next event trigger. Therefore, embodiments of the present invention quantify D(n) using credit values. Based on the proportional relationship between the actual link data flow and the D(n) value, partial authorization is performed by monitoring the port traffic volume of the physical network interface card (NIC). This dynamically adjusts D(n) based on the physical NIC port traffic flow, satisfying D(n) = k * min[TBS(n), Q(n)], where k is a proportional coefficient related to the link traffic flow. Through partial authorization, after determining the TBS block size, the minimum of TBS(n) and Q(n) is taken based on the current buffer size. This ensures that the reduction in D(n) credit within the granted resources is proportional to the actual data transmitted, rather than issuing all credits at once. This avoids excessive credit deficit expenditure, helps accelerate credit recovery in the UE's transmission queue, reduces queueing delays, and mitigates credit violations and conflicts caused by traffic bursts.

[0026] S4. After receiving the data output by the DCS scheduler, the MAC layer generates downlink control information (DCI) and sends it to the user equipment (UE).

[0027] The discrete-time 5G private network scheduling method provided by the present invention introduces a DCS scheduler between the MAC and RLC layers. In combination with the characteristics of the 5G wireless air interface, it uses a discrete-time slot-based credit authorization mechanism to quantitatively control UE queue forwarding, thereby improving end-to-end deterministic transmission performance and air interface resource utilization. Furthermore, partial authorization facilitates faster credit recovery for UE transmission queues, reduces queue delays, and mitigates credit violations and conflicts caused by traffic bursts.

[0028] The present invention also provides a 5G private network scheduling system based on discrete time, including Figure 1 The 5G base station (gNB) shown in the figure, the discrete time-based 5G private network scheduling system can achieve Figure 2 The various steps of the discrete time-based 5G private network scheduling method are shown.

[0029] The above description is merely a preferred embodiment of the present invention and does not constitute any form of limitation to the present invention. Those skilled in the art may make various equivalent changes and improvements based on the above embodiment. Any equivalent changes or modifications made within the scope of the claims shall fall within the scope of protection of the present invention.

Claims

1. A 5G private network scheduling method based on discrete time, characterized in that: A DCS scheduler based on discrete time information is deployed between the MAC layer and the RLC layer on the base station side. The DCS scheduler dynamically updates the UE queue credit value and submits the queue with qualified credit to the MAC layer based on the credit threshold.

2. The discrete-time-based 5G private network scheduling method according to claim 1, wherein: The discrete-time-based 5G private network scheduling method comprises the following steps: The 5G new air interface sends downlink data to the 5G base station; The RLC layer of the 5G base station forwards the downlink data to the DCS scheduler; The DCS scheduler manages UE queue credit for downlink data. Within each time slot period with a fixed subcarrier spacing, the DCS scheduler dynamically updates the UE queue credit value and submits the queues with qualified credit to the MAC layer based on the credit threshold. After receiving the data output by the DCS scheduler, the MAC layer generates downlink control information and sends it to the user equipment.

3. The discrete-time based 5G private network scheduling method according to claim 2, wherein: The DCS scheduler updates the UE queue credit value based on the idle state accumulation and working state deduction mechanism, in, The idle state credit accumulation method is: idleValue = (t2-t1) * B, where idleValue is the idle state accumulated credit value, B is the bandwidth share reserved for traffic over time, and (t2-t1) is the time interval between any two moments in the idle state. The credit deduction method for the working state is: workValue = (t´2 - t´1) * S, where workValue is the credit value deducted in the working state, (t´2 - t´1) is the time interval between any two moments in the working state, S is the credit consumption generated by the actual transmission rate within (t´2 - t´1) over time, and S = B - L, where L is the loss during transmission based on the physical link transmission rate.

4. The discrete-time-based 5G private network scheduling method according to claim 3, wherein: When the UE queue credit value is ≥ 0 and there is data to be transmitted, the queue enters the working state; When the UE queue is idle, it automatically switches to the idle state.

5. The discrete time-based 5G private network scheduling method according to claim 4, wherein: When the UE queue credit value reaches the maximum credit value max_credit, the queue credit value stops accumulating; When the UE queue credit value falls below the minimum credit value min_credit, the queue exits the working state.

6. The discrete-time based 5G private network scheduling method according to claim 3, wherein: The credit value update formula of the UE queue in time slot n is: C(n+1)=f(C(n),ΔC(n),Q(n))-D(n); ; Where C(n) is the available cumulative transmission amount at the beginning of time slot n, ΔC(n) = idleValue*T(slot), T(slot) is the time slot cycle time of the fixed subcarrier spacing, D(n) is the credit deduction amount within the time slot n interval, and Q(n) is the buffer of the UE queue.

7. The discrete-time based 5G private network scheduling method according to claim 6, wherein: The D(n) is dynamically adjusted according to the port traffic, satisfying D(n)=k*min[TBS(n),Q(n)], where TBS(n) is the TBS block size within the time slot n, and k is a proportional coefficient related to the link traffic.

8. The discrete time-based 5G private network scheduling method according to claim 2, wherein: The RLC layer operates in UM mode, and retransmission is only processed by the HARQ of the MAC layer.

9. A 5G base station, comprising an RLC layer and a MAC layer, characterized in that: A discrete time information based DCS scheduler is deployed between the MAC layer and the RLC layer. The DCS scheduler is used to dynamically update the UE queue credit value and submit the credit-qualified queues to the MAC layer based on the credit threshold.

10. A 5G private network scheduling system based on discrete time, characterized in that: Including the 5G base station as described in claim 9.

Images