JP6792544B2 - Communication device and communication method - Google Patents

Description

The present invention relates to a communication device and a communication method.

In the access network, a packet transfer device (hereinafter, referred to as “transfer device”) that performs data traffic transfer processing collects data traffic of a plurality of services. The transfer device multiplexes the data traffic of multiple services and transfers it to the core network through the edge router. The core network is a large-capacity core communication network that connects telecommunications carriers. The transfer device identifies the priority described in the layer 2 frame and identifies the source user using the user identifier.

Mobile data traffic is a new service for data traffic that concentrates on lines. In addition, as one of the configurations of a wireless access network, a large number of optical overhanging stations (DU: Distributed Unit) are arranged at high density, and wireless signals are aggregated and signal processed by a wireless control device (CU: Central Unit). , There is a centralized radio access network (C-RAN). The CU and DU are communicated and connected by an optical access network called a mobile fronthaul (MFH). A proposal to configure this MFH at low cost by a transfer device (layer 2 switch) as shown in FIG. 9 is being discussed in the standardization organization IEEE (Institute of Electrical and Electronics Engineers) 802.1CM.

In MFH data traffic, the requirements for the delay required for data transfer between CU and DU (hereinafter referred to as "e2e (end-to-end) delay") are strict. For example, in Option 6 of 3GPP (Third Generation Partnership Project) TR 38.801, the worst value of e2e delay (hereinafter referred to as "worst e2e delay") is defined as 250 microseconds (μs). The worst e2e delay is a value different for each option.

The reason why the worst e2e delay is defined as 250 μs is that the hybrid ARQ (Automatic repeat request) performed between the wireless terminal (UE: User Equipment) and the CU is performed at high speed. Specifically, the main signal within the 250 μs delay request time provided for low delay transfer of the acknowledgment (ACK) signal and the negative response (NAK: Negative-acknowledgement) signal, which are the control signals of the hybrid ARQ. This is due to the need for a transfer of.

The delays that are the components of the e2e delay include the processing delay caused by the transfer processing in the layer 2 switch, the propagation delay that is the time required for physical transmission between the layer 2 switches, and the conflict delay that occurs between packets. There is. The contention delay referred to here is a queuing delay that occurs between flows having the same priority. For example, if two packets of the same priority arrive at the layer 2 switch at the same time and these two packets are output from the same port, the other packet is forwarded until the forwarding of one packet is completed. Is not performed and it goes into a standby state. The delay in this case is called the conflict delay.

In addition, the 5th generation mobile communication system (hereinafter referred to as "5G") is expected to adopt a Time Division Duplex (TDD) system. In this case, the DUs arranged at high density synchronize the time, and the data traffic flows into the optical access network at the same timing. As a result, the probability of occurrence of competition delay increases.

Regarding the configuration of MFH by layer 2 switch, in addition to the ring topology using the ring protocol, IS-IS (Intermediate System to Intermediate System) such as SPB (Shortest Path Bridging) and IEEE 802.1Qca are used. A method of realizing a flexible network configuration can be considered by using it and exchanging route information between nodes to determine a transfer route. In particular, when IEEE 802.1Qca is used, it is possible to set and use a transfer route other than the simple shortest route according to the route design algorithm to be used.

Conventionally, Non-Patent Document 1 has proposed a route design algorithm that determines a priority based on the e2e delay of each flow and designs a route so as to minimize the value of the e2e delay that becomes the maximum among all the flows.

Yu Nakayama, Daisuke Hisano, Takahiro Kubo, Tatsuya Shimizu, Hirotaka Nakamura, Jun Terada, and Akihiro Otaka, "Novel Rank-Based Low-Latency Scheduling for Maximum Fronthaul Accommodation in Bridged Network", Optical Fiber Communication Conference 2017, 19-23 March 2017

By using the route design algorithm of Non-Patent Document 1, the number of DUs that can be collected increases. However, in the route design algorithm of Non-Patent Document 1, the number of DUs that can be accommodated is rate-determined by the worst delay time, and the statistical multiplexing effect is not expected. Therefore, in order to concentrate more DUs and reduce the cost of the optical access network, it is required to establish a method for further increasing the number of DUs that can be accommodated by anticipating the statistical multiplexing effect.

In view of the above circumstances, an object of the present invention is to provide a communication device and a communication method capable of increasing the number of optical overhang stations capable of concentrating lines.

In one aspect of the present invention, an optical overhang station of a radio base station that employs a time-division duplex system and an aggregation station that aggregates the optical overhang stations are connected via a plurality of communication devices, and the plurality of the opticals are connected. A communication device in a communication network that transfers a flow of an overhanging station to the centralized station, and is a plurality of flows transmitted from the optical overhanging station or the centralized station, or a route to be transferred in a transfer path of the communication network. A receiver that receives a plurality of flows transmitted from another communication device connected to the opposite side of the above, and the other receiver that is connected to a route that transfers the flows received by the receiver in the transfer path. An output unit that outputs to the communication device according to the priority of each flow, a separation unit that separates a data signal and a positive response signal or a negative response signal, and the transfer from the receiving unit to the output unit. The route is designed, and the priority for each flow is set to at least the minimum value, maximum value, and average value of the delay time allowed for the time-divided duplex communication system signal transmitted / received by the radio base station. A control unit that determines based on statistical information including one and sets the end-to-end delay of all the flows so as not to exceed the delay time allowed for the time-divided duplex signal. It is a communication device provided with.

Further, one aspect of the present invention is the above-mentioned communication device, and the control unit is allowed for each radio frame of the new flow when there is a request for adding a new flow to the plurality of the flows. Based on the statistical information including at least one of the minimum value, the maximum value, and the average value of the delay time, the vicinity of the corresponding priority is searched and inserted, and the order of the flow corresponding to the lower priority is obtained. To move down.

Further, one aspect of the present invention is the above-mentioned communication device, and when the control unit receives a request to delete a specific flow for a plurality of the flows, the information of the flow is given from the corresponding priority. Is deleted, and the priority of the other flow registered in the lower priority is raised.

Further, one aspect of the present invention is the above-mentioned communication device, which includes a queue for each priority, and the queue is the total number of queues queued by another queue having a higher priority than the own queue. If the amount of data exceeds the predetermined amount of data, the data in the own queue is discarded.

Further, one aspect of the present invention is to connect a plurality of optical overhanging stations of a radio base station adopting a time-division duplex system and an centralized station that aggregates the optical overhanging stations via a plurality of communication devices. A communication method performed by a computer of the communication device in a communication network that transfers the flow of the optical overhang station to the centralized station, and is a plurality of flows transmitted from the optical overhang station or the centralized station, or the communication network. A reception step that receives a plurality of flows transmitted from other communication devices connected to the opposite side of the transfer route in the transfer path of the above, and a method that transfers the flow received by the reception step in the transfer path. An output step for outputting to the other communication device connected to the road according to the priority of each flow, a separation step for separating a data signal and a positive response signal or a negative response signal, and the reception step to the above. The transfer path up to the output step is designed, and the priority for each flow is set to the minimum value of the delay time allowed for the time-divided duplex signal transmitted / received by the radio base station. Determined based on statistics including at least one of the maximum and average values, the end-to-end delay of all the flows does not exceed the delay time allowed for the time-divided duplex signal. It is a communication method having a control step set so as to be.

According to the present invention, it is possible to increase the number of light overhanging stations capable of concentrating lines.

A configuration example of a conventional layer 2 switch is shown. It is a figure which shows the transmission procedure in the downlink. It is a figure which shows the vertical link ratio of 7 kinds of TDD schemes defined by 3GPP. It is a figure which shows the list of the reply timing of the ACK signal and the NAK signal. It is a time chart which shows the data transmission procedure of an uplink. It is a figure which shows the transmission timing of a data signal and the transmission timing of an ACK signal and a NAK signal. A configuration example of a layer 2 switch when a separation unit for separating a data signal, an ACK signal, and a NAK signal is provided is shown. A flowchart showing the route design process is shown. It is a figure which shows the configuration example of the conventional optical access network which used the layer 2 switch.

<First Embodiment>
Hereinafter, the first embodiment of the present invention will be described.

FIG. 1 shows a configuration example of a conventional layer 2 switch. As shown in FIG. 1, the layer 2 switch 1 has m receiving units 11 (11-1, 11-2, ..., 11-m) for receiving packets and m units that can be stored according to priority. Queue 12 (12-1, 12-2, ..., 12-m) and m output units 13 (12-1, 12-2, ..., 12-m) that transfer packets to the next layer 2 switch in order from the queue 12 with the higher priority. 13-1, 13-2, ..., 13-m).

Further, as shown in FIG. 1, the layer 2 switch 1 has a control unit 14 that calculates and manages the priority order. The control unit 14 can be configured to be provided in the layer 2 switch 1 as shown in FIG. 1, but is not limited to this configuration. For example, a node that collectively manages all layer 2 switches may be provided, and the control unit 14 may be provided in the node.

In the conventional method, the control unit 14 determines the priority and estimates the delay time of each flow. Then, the route design of the transfer route is performed so that the maximum delay time among the estimated delay times of each flow is equal to or less than the required condition.
Since the TDD system is adopted for the radio base station in 5G, in the present invention, the priority is determined by using the communication configuration by the TDD system. The prerequisites are described below.

The delay requirements in the MFH are rate-determined by the hybrid ARQ. FIG. 2 shows a transmission procedure for the downlink.
When the uplink data signal is transmitted from the UE on the uplink, the CU determines whether the reception of the uplink data signal is successful or unsuccessful during the processing time of 3 subframes. When the CU determines that the reception is successful, the CU sends an ACK signal back to the UE. If the CU determines that the reception has failed, the CU sends back the NAK signal to the UE.

The UE must receive the ACK or NAK signal within 4 subframes of transmitting the uplink data signal. Here, the subframe represents an element constituting a wireless frame. One radio frame is composed of 10 subframes. For example, if the radio frame length is 10 μs, one subframe is 1 μs.

The downlink data transmission procedure is the same as the uplink data transmission procedure. Since the time from the reception of the ACK signal or the NAK signal to the retransmission is not standardized, it depends on the mounting method.

The above data transmission procedure is a data transmission procedure in the case of the frequency division duplex (FDD) method. The data transmission procedure in the case of the TDD method will be described below. FIG. 3 is a diagram showing the vertical link ratios of seven types of TDD schemes defined by 3GPP.

In FIG. 3, "D" represents a downlink signal and "U" represents an uplink signal. Further, in FIG. 3, “S” indicates that both the uplink and downlink signals and the guard interval are included.
For example, No. It is shown in FIG. 3 that the subframe of item 3 of item 1 is an uplink data signal.

Regarding the uplink data transmission procedure, the basic operation is the same as that of the FDD method, but the reply timings of the ACK signal and the NAK signal are not all timings after 4 subframes. This is because, in the TDD method, the upper and lower links are switched in time, so that the subframe after 4 subframes is not always the subframe that can be transmitted.

A list of reply timings of the ACK signal and the NAK signal is shown in FIG. For example, in this list, No. When the CU receives the uplink data signal in the subframe of item 3 of item 1, it is indicated that the CU should send back the ACK signal or the NAK signal within 6 subframes.
As shown in FIG. 4, the reply timing of the ACK signal and the NAK signal is a value of 4 subframes or more.

FIG. 5 shows a time chart of the uplink data transmission procedure. The “processing time in the CU” shown in FIG. 5 is the time required to issue the ACK signal and the NAK signal. After that, there is a waiting time until the transmission possible timing is reached.

In view of the above-mentioned preconditions, it can be seen that the data signal can be separated from the ACK signal and the NAK signal. FIG. 6 shows the transmission timing of the data signal and the transmission timing of the ACK signal and the NAK signal in the radio base station according to the present invention.

As shown in FIG. 6, in a radio base station, the CU first transmits an ACK signal and a downlink signal toward the EU. The ACK signal or NAK signal for this downlink signal needs to be returned within the processing time of 4 subframes. The UE transmits an ACK signal or a NAK signal (ACK signal in FIG. 6) and an uplink signal toward the CU in the subframe of item No. 4. At this time, the ACK signal or the NAK signal is transferred without delay exceeding the required condition. Here, the uplink signal does not need to be transmitted within the processing time of 4 subframes. This is because the ACK signal or NAK signal for this uplink signal need only be transmitted within 6 subframes.

Therefore, even if the uplink signal is buffered by the layer 2 switch and a delay is given, a delay of 2 μs is further given to the transfer. FIG. 7 shows a configuration example of a layer 2 switch 2 (communication device) provided with a separation unit 15 for separating the data signal, the ACK signal, and the NAK signal.

The layer 2 switch 2 (communication device) shown in FIG. 7 is composed of a DU (optical overhang station) of a radio base station that employs a TDD system (time division duplex system) and a CU (aggregate station) that aggregates DUs. It is a transfer device in MFH (communication network) that connects the spaces via a plurality of layer 2 switches 2 and transfers the flows of a plurality of DUs to the CU.

The receiving unit 11 (11-1, 11-2, ... 11-m) is connected to a plurality of flows transmitted from the DU or CU, or to the opposite side of the forwarding route in the MFH forwarding path. Receives multiple flows transmitted from other layer 2 switches.

The output unit 13 (11-1, 11-2, ... 11-m) transfers the flow received by the receiving unit 11 to another layer 2 switch connected to the route to be transferred in the transfer path. Output according to the priority of each.

The control unit 14 designs the transfer path from the reception unit 11 to the output unit 13, and sets the priority for each flow to the minimum allowable delay time for the TDD system signal transmitted / received by the radio base station. Determined based on statistics including at least one of a value, a maximum value and an average value, and set so that the e2e delay of all flows does not exceed the delay time allowed for the TDD signal.

The separation unit 15 separates the data signal from the ACK signal (affirmative response signal) or NAK signal (negative response signal).

The separation unit 15 can be configured to be provided in the layer 2 switch 2 as shown in FIG. 7, but is not limited to this configuration. For example, the separation unit 15 may be configured to be mounted only on the DU or a part of the layer 2 switches 2.

Separation of the data signal, the ACK signal, and the NAK signal can be realized as a simple method, for example, by dividing a VLAN (Virtual Local Area Network).

In determining the priority in Non-Patent Document 1, the value of the e2e delay, which is the maximum among the e2e delays (end-to-end delays) of each flow when the route is set, is minimized, and the total flow is determined. The e2e delay was determined not to exceed the worst e2e delay. On the other hand, in the present invention, it is possible to set the allowable delay more slowly with respect to the transfer of the uplink signal. As shown in FIG. 4, when an ACK signal or a NAK signal is returned after 4 + k subframes, an additional allowable time is given by the processing time for the k subframes. Therefore, the present invention has this additional allowable time. Consider.

FIG. 8 shows a flowchart showing the route design process. The route design process shown in FIG. 8 is implemented in a route design controller (not shown) that designs a route in a bridge network. The route design controller can be implemented regardless of whether it is inside or outside the layer 2 switch 2. Further, the route design controller may be configured to be provided in the control unit 14 of all layer 2 switches 2. In that case, the same result will be obtained for all layer 2 switches 2.

First, the route design controller refers to the accommodation flow (FIG. 8, step S01) and acquires the TDD method vertical link ratio of each flow or information indicating which of the seven types the vertical link ratio corresponds to. (FIG. 8, step S02). Based on this information, the route design controller calculates the e2e delay for each flow (FIG. 8, step S03). The route design controller saves the calculation result in the memory together with the flow ID (Identifier) (FIG. 8, step S04).

In the conventional method, the worst e2e delay is defined by the functional division point in 3GPP TR 38.801 or the like. For example, in Option 6 of 3GPP TR 38.801, the e2e delay is specified as 250 μs. On the other hand, in the present invention, the worst e2e delay value is determined based on statistical information such as the maximum value, the minimum value, and the average value in the subframe.

When the reference of all the flows is completed (FIG. 8, step S05, YES), the route design controller shifts to the next operation. If there is a flow that has not been referenced (FIG. 8, step S05, NO), the route design controller refers to the containment flow that has not yet been referenced (FIG. 8, step S01).

When the reference of all the flows is completed (FIG. 8, step S05, YES), the route design controller determines the priority based on the worst e2e delay of each flow (FIG. 8, step S06). After determining the priority order, the route design controller performs the route design (FIG. 8, step S07). As for the method of setting the route design, various algorithms (for example, linear programming method, Monte Carlo method, etc.) have been conventionally devised, but any method may be used.

The route design controller estimates the e2e delay of each flow for the designed route. The estimation is performed by further adding the total value obtained by adding the processing delay, the propagation delay, and the like, and the contention delay (= priority x packet burst length / transmission speed). When the route is designed for all the flows so that the e2e delay of each flow estimated by this does not exceed the worst e2e delay of each flow (FIG. 8, step S08, YES), the processing shown in this flowchart is performed. Is finished. If the estimated e2e delay of each flow exceeds the worst e2e delay of each flow (FIG. 8, step S08, NO), the route design controller redoes the route design.

<Second embodiment>
The configuration of the layer 2 switch according to the second embodiment is a configuration that reflects, when a flow is added during operation, in addition to the configuration of the layer 2 switch 2 according to the first embodiment described above. Specifically, the control unit manages the priority, and when there is a request to add a new flow for a plurality of flows, the minimum value, the maximum value, and the maximum value of the delay time allowed for each radio frame of the new flow. Based on the statistical information including at least one of the average values, the neighborhood of the corresponding priority is searched and inserted, and the order of the flow corresponding to the lower priority is moved down.

At this time, if there is a vacant route in the operating route and it is possible to concentrate the line without affecting other flows even if the flow is inserted, the flow is added as it is. If there is no vacant route in the operating route, the route design process shown in the flowchart in FIG. 8 is executed again, and the path into which an additional flow can be inserted is reset to all flows. To do.

<Third embodiment>
The configuration of the layer 2 switch according to the third embodiment is a configuration that reflects the configuration of the layer 2 switch 2 according to the first embodiment described above when the flow is deleted during operation. Specifically, the control unit manages the priority, and when there is a request to delete a specific flow for a plurality of flows, the flow information is deleted from the corresponding priority and registered in the lower priority. Raise the priority of other flows that have been made.

At this time, if the deletion of the flow only does not forward the packet and does not affect other flows, but the control unit does not recognize the deletion of the flow, it becomes impossible to insert an additional flow. there is a possibility. Therefore, when a certain number of flows are deleted, it is preferable to recalculate the priority order and redo the route design.

<Fourth Embodiment>
When the packet of the target flow is stored in the queue, if the queue with a higher priority than the own queue is in a congested state, there is a case where the transfer cannot be performed within the worst e2e delay. .. In the configuration of the layer 2 switch according to the fourth embodiment, in addition to the configuration of the layer 2 switch 2 according to the first embodiment described above, the number of accumulated packets of the queue having a higher priority than the own queue is counted. When the total amount of data queued by other higher-level queues exceeds a predetermined amount of data and is congested, the transfer of packets stored in the own queue is discarded. This prevents extra packets from staying on the network.

As described above, the layer 2 switch (communication device) according to the first to fourth embodiments of the present invention distinguishes between a main signal (data signal) and a control signal (ACK signal and NAK signal). The route is designed by separating and transferring the data and calculating the e2e delay in consideration of the reply timing of the ACK signal and the NAK signal in the subframe.

Conventionally, even if the route design is devised, the number of DUs that can be accommodated is limited by the worst e2e delay. According to the present invention, the worst e2e delay of the data signal is extended by separating and transferring the data signal, the ACK signal and the NAK signal. In the worst case, the e2e delay is extended, which gives a grace period to the time that can be buffered in the layer 2 switch. According to the present invention, the extended delay time is added in the route design process, so that the number of accommodated DUs can be increased.

Although the embodiments of the present invention have been described in detail with reference to the drawings, the specific configuration is not limited to this embodiment, and includes designs and the like within a range that does not deviate from the gist of the present invention.

1 ... Layer 2 switch, 2 ... Layer 2 switch, 11 ... Receiver, 12 ... Queue, 13 ... Output, 14 ... Control, 15 ... Separation

Claims (7)

The optical overhang station of the radio base station adopting the time division duplex system and the aggregation station that aggregates the optical overhang stations are connected via a plurality of communication devices, and the flows of the plurality of optical overhang stations are aggregated. The communication device in the communication network to transfer to the station.
Receives a plurality of flows transmitted from the optical overhang station or the aggregation station, or a plurality of flows transmitted from another communication device connected to the opposite side of the transfer route in the transfer path of the communication network. Receiver and
An output unit that outputs the flow received by the receiving unit to the other communication device connected to the transfer route in the transfer path according to the priority of each flow.
A separator that separates the data signal from the acknowledgment signal or the negative response signal,
The transfer path from the receiving unit to the output unit is designed, and the priority order for each flow is set to the delay allowed for the time division duplex system signal transmitted and received by the radio base station. At least one of the minimum, maximum and average values of the delay time , which is the time and includes the stretch time indicating the time stretched according to the separation of the data signal from the positive response signal or the negative response signal. A control unit that determines based on statistical information including one and sets that the end-to-end delay of all the flows does not exceed the delay time allowed for the signal of the time division duplex system.
A communication device equipped with. When a new flow is requested to be added to the plurality of flows, the control unit sets at least one of the minimum, maximum, and average delay times allowed for each radio frame of the new flow. The communication device according to claim 1, wherein the vicinity of the corresponding priority is searched for and inserted based on the included statistical information, and the order of the flow corresponding to the lower priority is lowered. When the control unit receives a request to delete a specific flow for a plurality of the flows, the control unit deletes the information of the flow from the corresponding priority, and the other said other registered in the lower priority. The communication device according to claim 1, wherein the priority of the flow is raised. With a queue for each of the priorities
The queue discards the data of the own queue when the total amount of data queued by another queue having a higher priority than the own queue exceeds a predetermined amount of data. The communication device described in. The extension time corresponds to the length of the waiting time from the issuance of the positive response signal or the negative response signal corresponding to the data signal to the transmission possible timing of the positive response signal or the negative response signal. It's time to do
The communication device according to claim 1. Queue for each priority
With
The output unit outputs the flow stored in the queue according to the priority.
The communication device according to claim 1. The optical overhang station of the radio base station adopting the time division duplex system and the aggregation station that aggregates the optical overhang stations are connected via a plurality of communication devices, and the flows of the plurality of optical overhang stations are aggregated. A communication method performed by a computer of the communication device in a communication network to transfer to a station.
Receives a plurality of flows transmitted from the optical overhang station or the aggregation station, or a plurality of flows transmitted from another communication device connected to the opposite side of the transfer route in the transfer path of the communication network. Receive step and
An output step in which the flow received by the reception step is output to another communication device connected to the transfer route in the transfer path according to the priority of each flow.
A separation step that separates the data signal from the acknowledgment signal or the negative response signal,
The transfer path from the reception step to the output step is designed, and the priority for each flow is set to the delay allowed for the time division duplex system signal transmitted and received by the radio base station. At least one of the minimum, maximum and average values of the delay time , including the stretch time, which indicates the time stretched by the separation of the data signal from the positive response signal or the negative response signal. A control step that determines based on the including statistical information and sets that the end-to-end delay of all the flows does not exceed the delay time allowed for the time division duplex signal.
Communication method with.

Images