RU2373654C1 - Method for making peer-to-peer connection and system designed for it - Google Patents

Abstract

FIELD: information technologies. ^ SUBSTANCE: invention is related to technology of communication between units of computer network. Method for making a peer-to-peer connection (P2P connection) includes transfer by server, after reception of synchronisation query from the first host, to each of the first and second hosts of synchronisation data packets, comprising at least information about address of another host from specified first and second hosts; and transfer of UDP protocol data packet, after reception of specified synchronisation data packet, by the first and second hosts to each other. Besides system is suggested for making of P2P connection. According to the present invention, P2P connection between two hosts may be successfully made in most cases, because data packet has already been transferred by each host from specified first and second hosts, when each of them receives data packet sent by another host from specified first and second hosts. ^ EFFECT: improved extent of P2P connection making successfulness. ^ 12 cl, 6 dwg

Description

Technical field

The present invention relates to communication technology between nodes of a computer network. More specifically, the invention relates to a method for establishing a peer-to-peer connection (P2P connection) and to an intended system.

State of the art

Connections between two network nodes are usually made in two modes. The first mode is “client-server”, according to which the client requests a service, and the server provides this service, as a rule, according to the “one to many” scheme. As shown in figure 1, the server can receive requests from many clients and provide them with services, while the connection between them is not peer-to-peer. The second mode is peer-to-peer (P2P mode), in which any two nodes of the computer network are connected directly to each other (see figure 2), i.e. every two nodes form a peer-to-peer pair.

Since in the "client-server" mode the servers are responsible for providing services to many clients, high-performance and, as a result, expensive computers are usually used as servers. In P2P mode, the two communication terminals are peer-to-peer and provide services to each other, which provides the P2P mode with noticeable advantages, since multiple PCs of average performance can be combined for sharing resources and communicating with each other, while the costs are significantly reduced. This and other reasons influenced the fact that when developing networks began to use the P2P mode. Currently, applications using this mode include, for example, file sharing applications, distributed computing applications, electronic commerce, etc.

Despite these advantages, the further development of the P2P mode is to some extent limited by the fact that more and more computers operate using a network address translator (NAT translator) that protects computers from network virus attacks. A NAT translator is a mechanism commonly used in routers that is responsible for checking and forwarding packets between a private network and a public network. Since NAT translators can change the IP address and host port information of a private network, they limit the direct connection between any two computers and allow you to establish communication only in accordance with certain rules. Therefore, when implementing the P2P mode, the main task is to overcome the limitations of NAT translators.

NAT translators can be broadly divided into three categories: Basic NAT, Cone NAT, and Symmetric NAT.

The Basic NAT translator usually associates a pair {local IP address, local port} with a pair {global IP address, global port} and translates the addresses between these associated ports.

The Cone NAT translator typically associates a pair {local IP address, local port} with a pair {global IP address, global port} and reuses this associated port to service connections from the same combination of local IP address and port number. The Cone NAT translator keeps the associated port active as long as at least one such connection remains active.

Cone NAT translators, in turn, are divided into several subcategories: Full Cone NAT, Restricted Cone NAT, and Port-Restricted Cone NAT. The Full Cone NAT translator associates a global port with a local port when establishing a new communication session with an external terminal and then receives communication data from any external terminal on the public network through the global port. The Restricted Cone NAT translator does not accept any data packets from an external host with an external IP address until the local host sends one or more data packets to the IP address of the external host. The Port-Restricted Cone NAT translator does not accept any data packets from an external host with a given combination of IP address and port until the local host sends one or more data packets to the IP address and port number of the external host. It is easy to see that the Restricted Cone NAT and Port-Restricted Cone NAT translators can effectively protect local hosts from receiving unwanted data packets.

The Symmetric NAT translator does not support the association between the {local IP address, local port number} pair and the {global IP address, global port number} pair, which is unchanged for all communication sessions. Instead, for each new session, it uses a new global port or global IP address.

Since in practice all types of NAT translators are used and each of them has its own specific communication mode, when establishing a P2P connection between two nodes with different types of NAT translators, it is first necessary to determine the types of NAT translators used and then establish a P2P connection using different modes communications according to these types of NAT translators.

The following describes a method for establishing a P2P connection according to the prior art.

When host A of a network tries to establish a P2P connection with host B, an external server with two IP addresses is usually used to determine the type of NAT translator of its network and then transfer data about this type to the server. Accordingly, this server saves the type of NAT translator transmitted by host B. Host A receives from the server information about the type of NAT translator used in the network where host B is located. If host B is located in a public network or in a network where the type of translator is Full Cone NAT, then host A must use active mode (see FIG. 3). As shown in FIG. 3, host A sends a data packet directly to host B, and host B returns a response packet to host A. Thus, a P2P connection is established. Otherwise, host A uses passive mode (see FIG. 4). As shown in FIG. 4, host A first sends a command through the server to host B to transmit a data packet to host A, then host A returns a response packet to host B, thus establishing a P2P connection.

If in the described passive mode host A is located in a public network or in a network where the type of translator Full Cone NAT is used, then the P2P connection is established successfully. However, if the network where host A is located uses one of the types Restricted Cone NAT, Port-Restricted Cone NAT, or Symmetric NAT, then NAT translators of these types will reject unwanted data packets or even drop packets sent from a certain period of time combinations of IP address and port number, due to the fact that they will be identified as having signs of attack. All this makes it impossible to establish P2P connections. It is easy to see that the technical solution of the prior art provides a low degree of success in establishing P2P connections.

SUMMARY OF THE INVENTION

The objective of the invention is to provide a method and system for establishing connections P2P, which can increase the percentage of success in establishing such connections.

The invention is represented by the following embodiments.

The claimed method of establishing a connection P2P includes the following steps:

transmitting by the server after receiving a synchronization request from the first host to each of the first and second hosts a synchronization data packet containing at least address information of another host from said first and second hosts;

and transmitting, upon receipt of the specified packet of synchronization data, by the first and second hosts to each other, a UDP protocol data packet (User Datagram Protocol) in accordance with the received address information of another host from said first and second hosts.

In addition, the method comprises:

registering the first and second hosts on the server before sending the synchronization request to the server by the first host,

and storing the server network identifiers and address information of the first and second hosts, respectively.

The synchronization request from the first host contains at least a network identifier and address information of the second host. In addition, the method comprises:

determining by the second host whether to establish a P2P connection with the first host after receiving the synchronization data packet, and sending the UDP protocol data packet to the first host, if a decision was made to establish a P2P connection to the first host, or completion of the process if it was decided not to establish a P2P connection to the first host .

In addition, the method comprises:

the first and second hosts transmitting a response data packet to each other after receiving a UDP protocol data packet transmitted by another host from said first and second hosts.

If the first host is located in a domain that uses the Restricted Cone NAT translator, Port-Restricted Cone NAT or Symmetric NAT, and the second host is located on a public network or in If the domain uses the Full Cone NAT translator, then the transfer of the UDP protocol data packet to each other by the first and second hosts after receiving the synchronization data packet contains:

transmitting by the first host after receiving the synchronization data packet a UDP protocol data packet to the second host in accordance with the received address information of the second host;

and transmitting by the second host after receiving the synchronization data packet a UDP protocol data packet to the first host in accordance with the obtained address information of the first host.

If the first host is located in a public network or in a domain using the Full Cone NAT translator, and the second host is located in a domain using the Restricted Cone NAT, Port-Restricted Cone NAT, or Symmetric NAT translators, the data packet is transmitted by the first and second hosts to each other UDP protocol after receiving a synchronization data packet contains:

transmitting by the second host after receiving the synchronization data packet a UDP protocol data packet to the first host in accordance with the received address information of the first host;

and transmitting by the first host after receiving the synchronization data packet a UDP protocol data packet to the second host in accordance with the obtained address information of the second host.

If the first host is located in a domain that uses the Restricted Cone NAT translator or Port-Restricted Cone NAT translator, and the second host is located in a domain that uses the Restricted Cone NAT translator or Port-Restricted Cone NAT translator, the data packet is transmitted by the first and second hosts to each other UDP protocol after receiving a synchronization data packet contains:

transmitting by the first host after receiving the synchronization data packet a UDP protocol data packet to the second host in accordance with the received address information of the second host;

and transmitting by the second host after receiving the synchronization data packet a UDP protocol data packet to the first host in accordance with the obtained address information of the first host.

If the first host is located in a domain that uses the Restricted Cone NAT translator or Port-Restricted Cone NAT translator, and the second host is located in a domain that uses the Restricted Cone NAT translator or Port-Restricted Cone NAT translator, the data packet is transmitted by the first and second hosts to each other UDP protocol after receiving a synchronization data packet contains:

transmitting by the second host after receiving the synchronization data packet a UDP protocol data packet to the first host in accordance with the received address information of the first host;

and transmitting by the first host after receiving the synchronization data packet a UDP protocol data packet to the second host in accordance with the obtained address information of the second host.

The address information contains at least an IP address in the public network and a port number.

In addition, the address information includes the local IP address and port number.

The claimed P2P connection establishment system comprises a server, a first host and a second host;

moreover, the server after receiving a synchronization request transmitted by the first host, transmits to each of said first and second hosts a synchronization data packet containing at least address information of another host from said first and second hosts;

and the first and second hosts, after receiving the synchronization data packet, transmit to each other a UDP protocol data packet (User Datagram Protocol) in accordance with the received address information of another host from the indicated first and second hosts.

The claimed method of establishing a connection P2P contains:

receiving the first host from the server address information of the second host and information about the NAT translator (network address translator) used in the network where the second host is located;

and if the second host is located on a shared network or in a domain using the Full Cone NAT translator, then the first host transmits a UDP protocol data packet to the second host in accordance with the received address information of the second host;

and if the second host is neither located in the public network nor in the domain using the Full Cone NAT translator, then the first host determines the information about the NAT translator used in the network where the first host is located;

if the first host is located on a public network or in a domain using the Full Cone NAT translator, then the first host sends a command to the second host to transmit the UDP protocol data packet to the first host through the server;

and if the first host is neither located in the public network, nor in the domain using the Full Cone NAT translator, then the following steps:

the first host transmits a synchronization request to the server after receiving this synchronization request;

a server transmitting to each of said first and second hosts a synchronization data packet containing at least address information of another host from said first and second hosts;

and transmitting after receiving the synchronization data packet by the first and second hosts to each other, a UDP protocol data packet in accordance with the received address information of another host from the indicated first and second hosts.

It can be seen from the above diagram that according to the claimed method, both hosts send each other a UDP protocol data packet, therefore, the NAT translator considers the received data packet to be valid, because the data packet was transferred to another host from the indicated first and second hosts while receiving the data packet from this other the host. Therefore, a P2P connection can be successfully established between the two hosts. This solves the problem of the prior art related to the impossibility of successfully establishing a P2P connection if a host is located in a domain using Restricted Cone NAT or Port-Restricted Cone NAT translators that reject unwanted data packets. In addition, when one host is located in a domain using a Symmetric NAT translator with port replacement, and another host is located in a domain using a Restricted Cone NAT translator, a P2P connection can also be successfully established. As a result, the degree of success of establishing P2P compounds increases significantly.

Brief Description of the Drawings

Figure 1 illustrates the connection diagram of the network hosts in the client / server mode.

Figure 2 illustrates a connection diagram of network hosts in P2P mode.

Figure 3 illustrates a connection diagram of a P2P connection in the active mode in accordance with the prior art.

Figure 4 illustrates a P2P connection establishment scheme in a passive mode in accordance with the prior art.

Figure 5 illustrates a P2P connection establishment scheme in synchronous mode in accordance with the present invention.

6 depicts an algorithm for establishing a P2P connection in synchronous mode in accordance with the present invention.

Description of embodiments of the invention

Further, the invention is described in more detail by the example of some variants of its implementation, which are disclosed with reference to the attached drawings and allow a more complete understanding of the objectives, technical essence and advantages of this invention. It must be borne in mind that the proposed embodiments are used only to explain the essence of the invention and should not be construed as limiting the scope of its legal protection.

The invention provides a method for establishing a P2P connection in synchronous mode (see FIG. 5). If you want to establish a P2P connection with host B, then host A first sends a synchronization request data packet to the server. Upon receipt of the synchronization request data packet, the server transmits to each of the hosts A and B a synchronization data packet containing the network information of the other host. And after receiving these synchronization data packets, hosts A and B transmit to each other a UDP protocol data packet (User Datagram Protocol). After receiving the UDP protocol data packet transmitted by the other side, hosts A and B transmit to each other a UDP protocol response data packet.

6 depicts an algorithm for establishing a P2P connection in synchronous mode in accordance with the present invention. As follows from Fig.6, the method includes the following steps.

At step 601, host A and host B are registered on a server that stores global IP addresses and global port numbers, local IP addresses and local port numbers, and network identifiers of host A and host B.

At 602, when attempting to establish a P2P connection with host B, host A transmits to the server a synchronization request data packet containing at least the local IP address and port number of host A, the global IP address and port number of host A, and the network identifier of host B.

At 603, after receiving the synchronization request data packet from host A, the server sends each of the hosts A and B a synchronization data packet. The synchronization data packet transmitted to host A contains at least a global IP address and port number of host B and the local IP address and port number of host B, while the synchronization data packet transmitted to host B contains at least a global IP address and port number of host A and the local IP address and port number of host A. Next, steps 604 and 606 are performed.

At 604, after receiving the synchronization data packet transmitted by the server, host A sends UDP protocol data packet to host B using the IP addresses and port numbers of host B on the public network and on the private network.

In step 605, after receiving the UDP protocol data packet transmitted by host A, host B returns to host A a response data packet received by the NAT translator of the network where host A is located, since host A has already transmitted the UDP protocol data packet to host B. Then, host A completes process.

At step 606, upon receipt of the synchronization data packet transmitted by the server, host B decides whether to establish a P2P connection with host A, and performs step 607 if a positive decision is made, otherwise the process ends and the P2P connection is not established.

At step 607, host B sends the UDP protocol data packet to host A using the IP addresses and port numbers of the public network and private network of host A.

At 608, after receiving the UDP protocol data packet transmitted by host B, host A returns to host B a response data packet, also received by the NAT translator of the network where host B is located, because host B has already transmitted the UDP protocol data packet to host A.

Thus, the process of establishing a P2P connection between hosts A and B is completed.

Next, the results provided by the above scheme are analyzed.

If hosts A and B are located on a public network or in domains using a Full Cone NAT translator, then in accordance with the present invention, a P2P connection can be successfully established in a synchronous P2P connection establishment mode.

If host A is located in a domain using Restricted Cone NAT, Port-Restricted Cone NAT, or Symmetric NAT translators, and host B is located on a public network or in a domain using Full Cone NAT translator, then a P2P connection can be successfully established if the step 604 is performed before step 608, i.e. before host A receives the UDP protocol data packet transmitted by host B, host A should already transmit the UDP protocol data packet to host B. In practice, to ensure that host A transmits the UDP protocol data packet to host B before host A receives the UDP protocol data packet host B needs to prevent host B from transmitting the UDP protocol data packet to host A until host A sends the UDP protocol data packet to host B.

If host A is located on a public network or in a domain using the Full Cone NAT translator, and host B is located in a domain using the Restricted Cone NAT, Port-Restricted Cone NAT, or Symmetric NAT translators, then a P2P connection can be successfully established if the step 607 is performed before step 605, i.e. host B should already transmit the UDP protocol data packet to host A before host B receives the UDP protocol data packet transmitted by host A. In practice, to ensure that host B transmits the UDP protocol data packet to host A before receiving the UDP protocol data packet from host A prevent host A from transmitting the UDP protocol data packet to host B until host B sends the UDP protocol data packet to host A.

If host A is located in a domain using Restricted Cone NAT or Port-Restricted Cone NAT translators, and host B is located in a domain using Restricted Cone NAT or Port-Restricted Cone NAT translators, then a P2P connection can be successfully established if step 607 is performed before step 605 or step 604 is performed before step 608. In practice, you can prevent host B from transmitting a UDP protocol data packet to host A before host A sends a UDP protocol data packet to host B or block host A from transmitting a UDP protocol data packet to host B before packet B sends protocol data UDP to host A.

If any of their hosts A or B is located in a domain using the Symmetric NAT translator, then a P2P connection can also be successfully established in the following two cases.

Case 1. If one of the hosts A or B is located in the domain using the Symmetric NAT translator with port replacement, and the other host is in the domain using the Restricted NAT translator, then the P2P connection between these hosts can be established because the Symmetric NAT translator is not changes the global IP address of the source from which data packets are transmitted, and therefore, the Restricted NAT translator will accept data packets as the opposite side.

Case 2: A P2P connection between hosts can be established on systems that display the IP address and port number in the Port-Restricted Cone NAT translator mode, preferably in the case of a Symmetric NAT translator, and in which Symmetric modes, such as iptables, are used only with the likelihood of a conflict.

The analysis shows that the establishment of a P2P connection in synchronous mode according to the present invention allows to solve at least the problem of the impossibility of establishing a P2P connection when two hosts are located in domains using Restricted Cone NAT or Port-Restricted Cone NAT translators, if each of both hosts is already sent a data packet to another host while receiving a data packet from this other host. As follows from figure 5, the condition is that the sum of the time required by the server to transmit the synchronization data packet to host A, and the time required by host A to transmit the UDP protocol data packet to host B, should exceed the time required by the server to transmit the data packet synchronization to host B. In practice, this condition is easy to implement, therefore, the present invention provides more frequently successful establishment of P2P connections, i.e., increases the degree of success of establishing P2P connections.

In addition, the invention provides a combined circuit combining the above-described simultaneous packet transmission scheme with a P2P connection setup scheme according to the prior art. This scheme is described in detail below.

Host A receives from the server the address information of host B and the type of NAT translator used in the network where host B is located. If host B is located in a public network or in a network using the Full Cone NAT translator type, then host A uses active mode (see Fig. 3). As shown in FIG. 3, host A sends a data packet directly to host B using the address information of host B. Host B returns a response packet to host A. Thus, a P2P connection is established.

If host B is located neither in the public network nor in the domain using the Full Cone NAT translator, then host A determines whether host A is located in the public network or in the domain using the Full Cone NAT translator.

If host A is located in a public network, but not in a domain using the Full Cone NAT translator, then host A uses passive mode (see FIG. 4). As shown in FIG. 4, host A through the server instructs host B to first transmit the data packet to host A. Host A returns a response packet to host B, thus establishing a P2P connection.

If host A is neither in the public network nor in the domain using the Full Cone NAT translator, then the above-described scheme of simultaneous transmission of data packets is applied. In accordance with this scheme, host A transmits a synchronization request to the server, then the server simultaneously transmits to each of the hosts A and B a synchronization data packet containing at least the address information of the other host. Upon receipt of the synchronization data packet, hosts A and B transmit each other a data packet.

Only preferred embodiments of the invention are described in this application, however, the scope of legal protection of the invention is not limited to these options only. All changes, equivalent replacements or improvements, not going beyond the principles of the present invention, should be construed as falling within the scope of its legal protection.

List of symbols used

basic NAT - Basic NAT

Cone NAT - Cone NAT

Full Cone NAT - Full Cone NAT

NAT - Network Address Translator

P2P - Peer to Peer

Port-Restricted Cone NAT - Port-Restricted Cone NAT

Restricted Cone NAT - Limited Cone NAT

Symmetric NAT - Symmetric NAT

UDP - User Data Protocol

Claims (12)

1. A method for establishing a peer-to-peer connection (P2P connection), comprising the following steps:
transmitting by the server, after receiving a synchronization request from the first host, to each of the first and second hosts a synchronization data packet containing at least the address information of another host from said first and second hosts;
and transmitting, after receiving said synchronization data packet, the first and second hosts to each other a UDP protocol data packet (User Datagram Protocol) in accordance with the received address information of another host from said first and second hosts. 2. The method according to claim 1, additionally containing:
registering the first and second hosts on the server before transferring the synchronization request from the first host to the server,
and storing the server network identifiers and address information of the first and second hosts, respectively. 3. The method according to claim 1, wherein the synchronization request from the first host comprises at least a network identifier and address information of the second host. 4. The method according to claim 1, additionally containing:
determining by the second host whether to establish a P2P connection with the first host after receiving the synchronization data packet, and sending the UDP protocol data packet to the first host, if it was decided to establish a P2P connection to the first host, or to end the process if it was decided not to establish a P2P connection to the first host . 5. The method according to claim 1, additionally containing:
the first and second hosts transmitting a response data packet to each other after receiving a UDP protocol data packet transmitted by another host from said first and second hosts. 6. The method according to claim 1, in which if the first host is located in a domain using the Restricted Cone NAT translator, Port-Restricted Cone NAT (Port Cone NAT limited) or Symmetric NAT (Symmetric NAT) , and the second host is located in a public network or in a domain using the Full Cone NAT translator, then the transmission of the UDP protocol data packet to each other by the first and second hosts, after receiving the synchronization data packet, contains:
transmitting, by the first host, after receiving the synchronization data packet, a UDP protocol data packet to the second host in accordance with the received address information of the second host;
and transmitting by the second host, after receiving the synchronization data packet, the UDP protocol data packet to the first host in accordance with the received address information of the first host. 7. The method according to claim 1, wherein if the first host is located in a public network or in a domain using a Full Cone NAT translator, and the second host is located in a domain using a Restricted Cone NAT or Port-Restricted Cone NAT or Symmetric NAT translator , the transmission of the UDP protocol data packet by the first and second hosts to each other, after receiving the synchronization data packet, contains:
transmitting by the second host, after receiving the synchronization data packet, a UDP protocol data packet to the first host in accordance with the received address information of the first host;
and transmitting by the first host, upon receipt of the synchronization data packet, a UDP protocol data packet to the second host in accordance with the obtained address information of the second host. 8. The method according to claim 1, wherein if the first host is located in a domain using a Restricted Cone NAT translator or Port-Restricted Cone NAT translator, and the second host is located in a domain using a Restricted Cone NAT translator or Port-Restricted Cone NAT translator , the transmission of the UDP protocol data packet by the first and second hosts to each other, after receiving the synchronization data packet, contains:
transmitting, by the first host, after receiving the synchronization data packet, a UDP protocol data packet to the second host in accordance with the received address information of the second host;
and transmitting by the second host, after receiving the synchronization data packet, the UDP protocol data packet to the first host in accordance with the received address information of the first host. 9. The method according to any one of claims 1 to 8, wherein said address information comprises at least an IP address in a public network and a port number. 10. The method according to claim 9, in which the address information further comprises a local IP address and port number. 11. A system for establishing a peer-to-peer connection (P2P connection), comprising a server, a first host and a second host;
moreover, the server after receiving a synchronization request transmitted by the first host, transmits to each of said first and second hosts a synchronization data packet containing at least the address information of another host from said first and second hosts;
and the first and second hosts, after receiving the synchronization data packet, transmit to each other a UDP protocol data packet (User Datagram Protocol) in accordance with the received address information of the other host from the indicated second and first hosts. 12. A method for establishing a peer-to-peer connection (P2P connection), comprising:
receiving the first host from the server address information of the second host and information about the NAT translator (network address translator) used in the network where the second host is located;
and if the second host is located on a shared network or in a domain using the Full Cone NAT translator, then the first host transmits a UDP protocol data packet to the second host in accordance with the received address information of the second host;
but if the second host is neither located on the public network nor in the domain using the Full Cone NAT translator, then the first host determines the information about the NAT translator used in the network where the first host is located;
if the first host is located on a public network or in a domain using the Full Cone NAT translator, then the first host sends a command to the second host to transmit the UDP protocol data packet to the first host through the server;
but if the first host is located neither in the public network, nor in the domain using the Full Cone NAT translator, then perform the following actions:
the first host transmits a synchronization request to the server after receiving this synchronization request;
a server transmitting to each of said first and second hosts a synchronization data packet containing at least address information of another host from said first and second hosts;
and transmitting, after receiving the synchronization data packet, the first and second hosts to each other the UDP protocol data packet in accordance with the received address information of the other host from said first and second hosts.

Images