JP7599446B2 - COMMUNICATION METHOD AND SYSTEM FOR ESTABLISHING MULTIPLE SESSIONS - Patent application - Google Patents

Description

The present invention relates to a technique for establishing a session in a mobile network core system of a mobile communication system.

Conventionally, in a mobile communication system, a terminal is accommodated in a mobile network core system of a communication carrier via a Radio Access Network (RAN) (see, for example, Non-Patent Documents 1 and 2). This system is composed of a control plane and a user plane.
In recent years, the practical application of "Cell-free Massive MIMO (Multi Input Multi Output)" is expected as a 6th generation mobile communication system (see, for example, Non-Patent Documents 3 and 4). Massive MIMO, which is also provided in 5G, creates a coverage area like a spotlight for a moving terminal. Massive MIMO is further realized by the concept of Cell-free, which is created by one base station CPU (Central Processing Unit) and multiple access points.

3GPP TS 23.501 V17.2.0, “3rd Generation Partnership Project; Technical Specification Group Services and System Aspects; System architecture for the 5G System (5GS); Stage 2 (Release 17)” 3GPP TS 23.502 V17.2.1, “3rd Generation Partnership Project; Technical Specification Group Services and System Aspects; Procedures for the 5G System (5GS); Stage 2 (Release 17)” IEEE Transactions on Wireless Communications (Volume: 16, Issue: 3, March 2017), "Cell-Free Massive MIMO Versus Small Cells" "What are the technologies that will realize lifestyles in the Beyond 5G/6G era?", KDDI Research, Inc., [online], [searched on February 23, 2022], Internet <URL: https://www.itmedia.co.jp/mobile/articles/2106/16/news052.html>

According to the above-mentioned Cell-free Massive MIMO, a terminal can simultaneously connect wireless links to a plurality of access points managed by a base station CPU.
However, the question of which gateway a session with a terminal should be established with via a plurality of access points is merely a matter of implementation.
The node through which a session between a terminal and an application server passes has a large effect on throughput. In particular, when multiple sessions can be established simultaneously, the effect is not only on the number of sessions but also on the combination of access points and gateways through which sessions pass.
For example, if a terminal is to achieve a throughput of about 1 Tbps, it is generally assumed that multiple data flows are required. However, the throughput is greatly affected by which nodes in the user plane of the mobile network core system these data flows pass through.

Therefore, according to the present invention, it is an object of the present invention to provide a communication method and system that establishes multiple sessions via multiple access points and gateways, which is suitable for realizing high-volume communication between a terminal and a single application server.

According to the present invention, there is provided a communication method in which a terminal communicates with an application server using a control device via a plurality of access points, a plurality of gateways, and a proxy server, comprising the steps of:
A terminal transmits a session establishment request including identifiers of a plurality of connectable access points to a control device in order to communicate with one application server;
The control device selects an arbitrary combination of an access point and a gateway, and transmits an access point connection request including a list of the combination to the terminal;
a first step in which the terminal establishes a session with the proxy server via the access point and the gateway for each combination ;
a second step in which the terminal branches one stream of packets into a plurality of sessions and transmits each session to the proxy server via the access point and the gateway;
and a third step in which the proxy server aggregates packets received from the multiple gateways via the multiple sessions into one stream, and transfers the aggregated single stream to the application server.

According to another embodiment of the communication method of the present invention,
The control device further includes:
Regarding the first step,
A terminal transmits a session establishment request including identifiers of a plurality of connectable access points to a control device;
The control device selects an arbitrary combination of an access point and a gateway, and transmits an access point connection request including a list of the combination to the terminal;
It is also preferable that the terminal establishes a session with the proxy server via the access point and the gateway for each combination.

According to another embodiment of the communication method of the present invention,
Regarding the first step,
The session establishment request sent by the terminal further includes an application server address;
The control device notifies the proxy server of the application server address,
The proxy server establishes a connection with the application server based on the application server address;
For the third step, the proxy server also preferably forwards the integrated single stream to the application server via the established connection.

According to another embodiment of the communication method of the present invention,
Regarding the first step,
It is also preferable that the control device selects all or some of the combinations (full mesh) of multiple access points to which the terminal can connect and multiple gateways that can relay between the access points and the proxy server.

According to another embodiment of the communication method of the present invention,
It is also preferable that the number of any combination of access points and gateways is selected so as not to exceed the upper limit of sessions per terminal.

According to another embodiment of the communication method of the present invention,
It is also preferable to select, from among the combinations of access points and gateways, a combination that does not include an access point or gateway whose processing load rate or communication error rate exceeds a predetermined threshold.

According to another embodiment of the communication method of the present invention,
It is also preferable to select, from among the combinations of access points and gateways, a combination that does not include an access point or gateway whose processing load rate or communication error rate deviates from the average value by a predetermined rate or more.

According to another embodiment of the communication method of the present invention,
One stream is composed of multiple substreams with redundancy n,
The multiple substreams are copies of a packet sequence of one stream,
Each of the multiple sub-streams is transmitted to a session having a different combination of the access point and the gateway;
It is also preferable that the proxy server, upon first receipt of a packet with a given sequence number from any of the different sub-streams, considers that reception of packets with that sequence number has been completed.

According to another embodiment of the communication method of the present invention,
The transmission speed is measured for each session consisting of a combination of an access point and a gateway.
When the substream redundancy n is smaller than the number of sessions k (k>n), it is also preferable to select, from among _{the k} C _n combinations, the combination with the highest sum of the transmission rates in multiple sessions per stream.

According to the present invention, in a system in which a terminal is configured to communicate with an application server using a control device via a plurality of access points and a plurality of gateways and a proxy server,
A terminal transmits a session establishment request including identifiers of a plurality of connectable access points to a control device in order to communicate with one application server;
The control device selects an arbitrary combination of an access point and a gateway, and transmits an access point connection request including a list of the combination to the terminal;
the terminal establishes a session with the proxy server via the access point and the gateway for each combination ;
A terminal establishes a session with a proxy server for each combination of an access point and a gateway in order to communicate with one application server;
A terminal branches one stream of packets into a plurality of sessions and transmits each session to a proxy server via an access point and a gateway;
The proxy server combines packets received from a plurality of gateways via a plurality of sessions into one stream, and transfers the combined single stream to the application server.

According to the communication method and system of the present invention, a terminal can establish multiple sessions via multiple access points and gateways, which is suitable for achieving high-volume communication between the terminal and a single application server.

1 is a configuration diagram of a mobile communication system according to the present invention. FIG. 2 is an explanatory diagram illustrating a session in the mobile communication system of FIG. FIG. 2 is a sequence diagram according to the present invention. FIG. 11 is an explanatory diagram showing selection of a session according to the load factor of an access point and a gateway. FIG. 2 is an explanatory diagram showing the relationship between streams and sub-streams. FIG. 11 is an explanatory diagram showing a combination of multiple substreams.

The following describes in detail the embodiment of the present invention with reference to the drawings.

Figure 1 is a configuration diagram of a mobile communication system according to the present invention.

As shown in FIG. 1, a terminal (User Equipment) 2 is accommodated in a mobile network core system via multiple access points (Access Points) 3 in a RAN (Radio Access Network). The multiple access points 3 are connected to one base station CPU. The terminal 2 communicates with one application server 6 via the mobile network core system.

Moreover, according to the mobile network core system of FIG. 1, a control device 1 serving as a control plane, and a plurality of gateways 4 and a proxy server 5 serving as a user plane are arranged.
A terminal 2 simultaneously communicates with one application server 6 via multiple access points 3 and multiple gateways 4 and a proxy server 5 .

The terminal 2 can simultaneously connect wireless links to multiple access points 3 and simultaneously establish multiple sessions. Such a function pre-installed in the terminal 2 becomes the terminal-side end node for multiple sessions, and is also called, for example, a "bulk TCP client." This function is activated when the user initiates a high-volume communication operation with one application server 6.

The control device 1 is a group of network devices that are configured as a control plane by a plurality of functions and transmit and receive control signals for establishing communication and the like.
The control device 1 is responsible for session management between the terminal 2 and the application server 6. Of course, it is not limited to this, but also centrally manages registration, connection and mobility of the terminals, and assigns IP (Internet Protocol) addresses to the terminals.
In particular, according to the present invention, the control device 1 controls the establishment of multiple sessions via multiple access points and gateways when a terminal communicates with one application server 6 .

The multiple gateways 4 relay between any multiple access points 3 and the proxy server 5 via sessions. A session is specified by a combination of an access point 3 and a gateway 4. According to FIG. 1, the number of access points is 3 and the number of gateways is 4.

The proxy server 5 terminates multiple sessions established with the terminal 2, and establishes a TCP (Transmission Control Protocol) connection with the application server 6. It then transfers user data received from the terminal 2 via the multiple sessions to the application server 6, and transfers user data received from the application server 6 to the terminal 2 via the multiple sessions. The proxy server 5 serves as a mobile core side termination node for the multiple sessions established with the terminal 2, and is also referred to as, for example, a "bulk TCP proxy."

Also, according to FIG. 1, there are 12 combinations (full mesh) of multiple access points 3 to which the terminal 2 can connect and multiple gateways 4 that can relay between the access points 3 and the proxy server 5. Multiple sessions are selected from these 12 combinations.

Figure 2 is an explanatory diagram showing a session in the mobile communication system of Figure 1.

The terminal 2 transmits and receives user data to and from one application server 6 via a plurality of PDU (Protocol Data Unit) sessions. The PDU session is identified by an IP address assigned to the terminal 2.
According to FIG. 2, three PDU sessions #1 to #3 are established as follows:
Terminal - AP#1 - GW#1 - Proxy server - Application server
-AP#2-GW#2-
-AP#2-GW#4-
From the perspective of the access point 3, at least only the gateways 4 that are physically connected are candidate nodes.

Figure 3 is a sequence diagram of the present invention.

<First step>
(S11) The terminal 2 is assumed to have acquired the IP address of the application server 6 in advance. For example, the IP address may be preset in an application pre-installed in the terminal 2. Also, for example, the terminal 2 may acquire the IP address of the application server 6 by accessing a Domain Name Server (DNS). The terminal 2 is assumed to have at least one PDU session established at all times when the terminal 2 is powered on or when the airplane mode is released, and to be in a state where it can connect to the Internet.

The terminal 2 detects access point IDs (identifiers) and radio wave strengths from multiple access points that can be connected as a RAN. Then, the terminal 2 transmits a "session establishment request" including those access point IDs to the control device 1 via one access point 3. This session establishment request is also called a "bulk PDU session establishment request." The session establishment request may include, for example, the following information:
- Application server address - Desired communication speed - Direction (upstream/downstream/both directions)
- List [Connectable access point ID, connection status, signal strength]

Note that multiple sessions established simultaneously from terminal 2 are also referred to as "bulk PDU sessions." Communications from terminal 2 via bulk PDU sessions are also referred to as "bulk PDU communications."

(S12) The control device 1 determines the number of PDU sessions to be established and selects an arbitrary combination of access points 3 and gateways 4.
The control device 1 selects all or some of the combinations from among all combinations (full mesh) of a plurality of access points 3 to which the terminal 2 can be connected and a plurality of gateways 4 that can relay between the access points 3 and the proxy server 5. For example, according to the above-mentioned Fig. 1, between three access points 3 and four gateways 4, there are 12 full mesh session candidates.

(S13) The control device 1 sends an "access point connection request" including a list of combinations of multiple access points 3 and multiple gateways 4 to the terminal 2.

(S14) Terminal 2 simultaneously connects wireless links to multiple access points 3 listed in the list included in the access point connection request.

(S15) Then, terminal 2 returns an "access point connection response" to the control device 1.

(S16) In order to communicate with one application server 6, the terminal 2 establishes a session with the proxy server 5 via the access point 3 and the gateway 4 for each combination in the list. According to Fig. 3, similar to Fig. 2, three PDU sessions #1 to #3 are established as follows:
Terminal - AP#1 - GW#1 - Proxy server - Application server
-AP#2-GW#2-
-AP#2-GW#4-

(S17) The control device 1 notifies the proxy server 5 of the application server address.

(S18) The proxy server 5 establishes a connection with the application server 6 based on the application server address requested by the terminal 2. This connection may be, for example, TCP.

<Second step>
The terminal 2 branches one stream of packets into a plurality of sessions, and transmits each session to the proxy server 5 via the access point 3 and the gateway 4 .

<Third step>
The proxy server 5 integrates packets received from a plurality of gateways 4 via a plurality of sessions into one stream, and transfers the integrated single stream to the application server 6 via the established connection.

Figure 4 is an explanatory diagram showing session selection according to the load factor of the access point and gateway.

The control device 1 selects the number of any combination of access points 3 and gateways 4 so as not to exceed the upper limit of sessions per terminal (e.g., 15). The upper limit of sessions may be set as a static value by the operator, or may be updated as a dynamic value depending on the capacity of the access points 3 and/or gateways 4.

The control device 1 is also assumed to constantly detect the load rate of the access points 3 and gateways 4. The load rate can be recognized, for example, from the frequency of congestion or the percentage of available bandwidth.

(1) For any access point 3 and/or gateway 4, the access point 3 and/or gateway 4 whose processing load rate or communication error rate exceeds a predetermined threshold is excluded from the candidate nodes of the PDU session. The predetermined threshold may be set as a static value by an operator, or may be updated as a dynamic value according to the accommodation status of the access point 3 and/or gateway 4.
In addition, as another embodiment, a combination of access points 3 and gateways 4 that does not include an access point 3 or gateway 4 whose processing load rate or communication error rate deviates from the average value by a predetermined rate or more may be selected.

(2) Next, calculate the number of paths when the access point 3 of the candidate node and the gateway 4 are connected in a full mesh. If the number of full mesh paths exceeds the upper limit of the number of sessions per terminal, select the PDU session in ascending order of the communication path with the lowest load factor.

(3) If the number of full meshes is less than the maximum number of sessions per terminal, a PDU session is established in the full mesh.

According to FIG. 4, it is assumed that the load factors for the access points AP#1 to AP#3 are as follows.
AP#1: Load factor 30%
AP#2: Load factor 50%
AP#3: Load factor 70%
Average load factor = (30% + 50% + 70%) / 3 = 50%
Also, assume that the deviation threshold from the average load factor is set to "+10%."
In this case, AP#3, whose load factor exceeds 60% (= 50% + 10%), is excluded from the PDU session.

Also, according to FIG. 4, the load factors of the gateways GW#1 to GW#4 are as follows:
GW#1: Load factor 30%
GW#2: Load factor 50%
GW#3: Load factor 70%
GW#4: Load factor 40%
Average load factor = (30% + 50% + 70% + 40%) / 4 = 47.5%
Also, assume that the deviation threshold from the average load factor is set to "+10%."
In this case, GW#3, whose load factor exceeds 57.5% (= 47.5% + 10%), is excluded from the PDU session.

As shown in Figure 4, even in a wireless communication environment where random packet loss is likely to occur, TCP-type (AIMD-type delivery data loss prevention and order maintenance) multipath communication can be realized while suppressing the amount of information managed by the gateway in the mobile network core system and reducing the relay processing load as much as possible.

<Another embodiment in which a stream is branched into multiple substreams>
When high-volume communication is performed between the terminal 2 and the application server 6, the data flow increases. This places a heavy load on the gateway 4 in the user plane of the mobile network core system. The gateway 4 manages a flow table for identifying data flows, but the size of this flow table increases. If the number of data flows that the gateway 4 must manage and identify increases, a heavy load is placed on the flow table search process during relay.
Therefore, as another embodiment, a sequence is applied in which a stream is branched into multiple substreams and a PDU session with as low a load as possible is passed through.

Figure 5 is an explanatory diagram showing the relationship between streams and substreams.

5, one stream to be transmitted by terminal 2 is made up of packets 1 to 5. The stream is identified by the source port number of terminal 2. Terminal 2 generates multiple sub-streams from one stream.
Here, one stream is composed of n substreams with redundancy of n. The substreams are copies of the packet sequence of one stream. The substreams are identified by a combination of the IP address of the terminal and the source port number.

According to FIG. 5, one stream is composed of two substreams with a redundancy of two.
The terminal 2 copies one stream into two substreams and transmits each of the substreams via a PDU session to the proxy server 5. The two substreams copied from the same stream are each branched into a different PDU session. This separates the communication paths, making it possible to reduce the correlation between the timing of loss occurrence between the substreams.

In addition, terminal 2 allocates large volumes of traffic for one application to multiple streams. It then manages the window size on a stream-by-stream basis. For example, it may allocate the traffic of an application in order of the stream with the largest available window size. Then, multiple sub-streams are generated from each stream.

When the proxy server 5, which is the receiver of the substream, receives a packet with a given sequence number from one of the different substreams for the first time, it considers that reception of the packet with that sequence number is complete. At this time, it operates to increase the window size for one stream linked to that substream. This reduces the apparent packet loss rate, realizes high-volume communication with a smaller number of data flows, and reduces the processing load on communication devices such as the gateway 4.

In the next embodiment, the problem is how to arrange multiple substreams generated from one stream for multiple established PDU sessions.

Figure 6 is an explanatory diagram showing a combination of multiple substreams.

According to FIG. 6, three patterns are shown as follows for arranging two substreams with redundancy of 2 for three PDU sessions.
(First Arrangement Pattern)
Sub-stream #1: AP#1 - GW#1
Sub-stream #2: AP#2 - GW#2
(Second Arrangement Pattern)
Sub-stream #1: AP#1 - GW#1
Sub-stream #2: AP#2 - GW#4
(Third Arrangement Pattern)
Sub-stream #1: AP#2 - GW#2
Sub-stream #2: AP#2 - GW#4
As shown in FIG. 4, it is assumed that AP#3 and GW#3, whose load rates are higher than the deviation rate, have already been excluded.

In this way, terminal 2 transmits multiple sub-streams #1 and #2 to PDU sessions with different combinations of access points 3 and gateways 4, respectively.
When the proxy server 5 receives a packet with a given sequence number from either of the different sub-streams #1 and #2 for the first time, it considers that reception of the packet with that sequence number has been completed.

It is also assumed that the control device 1 measures the transmission speed for each PDU session consisting of a combination of the access point 3 and the gateway 4.
(First arrangement pattern): Sum of transmission speeds Substream #1: AP#1-GW#1: 100Mbps: 150Mbps
Sub-stream #2: AP#2 - GW#2: 50Mbps
(Second Arrangement Pattern)
Sub-stream #1: AP#1－GW#1 ：100Mbps ：110Mbps
Sub-stream #2: AP#2 - GW#4: 10Mbps
(Third Arrangement Pattern)
Sub-stream #1: AP#2 - GW#2: 50Mbps: 60Mbps
Sub-stream #2: AP#2 - GW#4: 10Mbps
It should be noted that the transmission speed is an "expected speed value" and is constantly measured and updated by the control device 1.
It is also assumed that the sum of the transmission rates of a plurality of substreams is proportional to the expected rate per stream.

In addition, if the substream redundancy n (=2) is smaller than the number of sessions k (=3) (k>n), the combination with the highest sum of the transmission rates in the multiple sessions per stream is selected from the _k C _n combinations. According to Fig. 6, the first arrangement pattern with the highest sum of the transmission rates is selected.
Of course, it is not limited to the transmission speed, but may be the delay time or the packet loss rate.

The substream arrangement determined as shown in FIG. 6 is notified from the control device 1 to the terminal 2 and the proxy server 5 .
Stream Port number Redundancy (2)
- PDU session list
Sub-stream #1: AP#1 - GW#1
Sub-stream #2: AP#2 - GW#2

As described in detail above, the communication method and system of the present invention make it possible to establish multiple sessions via multiple access points and gateways, which is suitable for realizing high-volume communication between a terminal and a single application server.

As a result, for example, "even when a terminal communicates with an application server, it is possible to reduce the relay processing load in the mobile network core system," which will contribute to Goal 9 of the United Nations-led Sustainable Development Goals (SDGs) "Build resilient infrastructure, promote sustainable industrialization and foster innovation."

With respect to the various embodiments of the present invention described above, various changes, modifications, and omissions within the scope of the technical ideas and viewpoints of the present invention can be easily made by a person skilled in the art. The above description is merely an example and is not intended to be restrictive in any way. The present invention is limited only by the scope of the claims and their equivalents.

Reference Signs List 1 Control device 2 Terminal 3 Access point 4 Gateway 5 Proxy server 6 Application server

Claims (9)

A communication method in which a terminal communicates with an application server via a plurality of access points, a plurality of gateways, and a proxy server by using a control device, comprising :
A terminal transmits a session establishment request including identifiers of a plurality of connectable access points to a control device in order to communicate with one application server;
The control device selects an arbitrary combination of an access point and a gateway, and transmits an access point connection request including a list of the combination to the terminal;
a first step in which the terminal establishes a session with the proxy server via the access point and the gateway for each combination ;
a second step in which the terminal branches one stream of packets into a plurality of sessions and transmits each session to the proxy server via the access point and the gateway;
and a third step of the proxy server aggregating packets received from the multiple gateways via the multiple sessions into one stream and transferring the aggregated single stream to the application server. Regarding the first step,
The session establishment request sent by the terminal further includes an application server address;
The control device notifies the proxy server of the application server address,
The proxy server establishes a connection with the application server based on the application server address;
2. The communication method according to claim 1 , wherein in the third step, the proxy server transfers the integrated single stream to the application server via the established connection. Regarding the first step,
The communication method according to claim 1 or 2, characterized in that the control device selects all or some of the combinations (full mesh) of a plurality of access points to which the terminal can be connected and a plurality of gateways that can relay between the access points and the proxy server. 4. The method of claim 3 , wherein the number of any combination of access points and gateways is selected so as not to exceed an upper limit of sessions per terminal. 5. The communication method according to claim 3 , further comprising selecting a combination of access points and gateways that does not include an access point or gateway whose processing load rate or communication error rate exceeds a predetermined threshold value from among the combinations of access points and gateways. The communication method according to claim 5, characterized in that, from among combinations of access points and gateways, a combination is selected that does not include an access point or gateway whose processing load rate or communication error rate deviates from the average value by a predetermined rate or more. One stream is composed of multiple substreams with redundancy n,
The multiple substreams are copies of a packet sequence of one stream,
Each of the multiple sub-streams is transmitted to a session having a different combination of the access point and the gateway;
The communication method according to any one of claims 1 to 6 , characterized in that, when the proxy server first receives a packet with a given sequence number from any of the different substreams, it considers that reception of the packet with that sequence number has been completed. The transmission speed is measured for each session consisting of a combination of an access point and a gateway.
The communication method according to claim ₇ , characterized in that, when the substream redundancy n is smaller than the number of sessions _k (k>n), a combination having the highest sum of transmission rates in multiple sessions per stream is selected from among the k C n combinations. A system in which a terminal is configured to communicate with an application server via a plurality of access points and a plurality of gateways and a proxy server using a control device ,
A terminal transmits a session establishment request including identifiers of a plurality of connectable access points to a control device in order to communicate with one application server;
The control device selects an arbitrary combination of an access point and a gateway, and transmits an access point connection request including a list of the combination to the terminal;
the terminal establishes a session with the proxy server via the access point and the gateway for each combination ;
A terminal establishes a session with a proxy server for each combination of an access point and a gateway in order to communicate with one application server;
A terminal branches one stream of packets into a plurality of sessions and transmits each session to a proxy server via an access point and a gateway;
A system in which a proxy server combines packets received from a plurality of gateways via a plurality of sessions into one stream, and transfers the combined single stream to an application server.

Images