CN1099794C - Address Allocation Method in Telecommunications Network Nodes - Google Patents

Abstract

The present invention relates to a method of allocating addresses in a telecommunications network node. The telecommunication network comprises a number of nodes (a.. G) which are connected together by transport connections, whereby the network forms a tree structure, wherein one of the nodes is a root node (a). The nodes send to each other signals containing address data for use by the nodes. In order to provide a flexible method which is easy to use, the root node (A) of the network assigns a unique address identifier to all transport connections it uses and sends each address identifier over a transport connection to its neighbouring node, after receiving the address identifier, the other nodes of the network a) assign unique identifiers to all transport connections they use (except the connection from which they received the address identifier), b) send to each of said transport connections an address identifier which comprises the address identifier itself received by the node and the identifier corresponding to said connection, whereby each node assigns its own address according to the address identifier it received.

Description

Address Allocation Method in Telecommunications Network Nodes

technical field

The invention relates to an address allocation method in a node device of a tree topology telecommunication network.

Background technique

In this application, a point of intersection of transport connections of a telecommunication system is referred to as a node device or node. A node can be any device or facility that can affect clock synchronization, such as a branch or cross-connect device.

Nodes of a telecommunications network must know their own address (ie location) in the network, since the role (configuration, etc.) of a node in the network depends on the address. Address information is also required in order for nodes to be able to route messages in the network.

Many modern telecommunications protocols use logical addresses that do not explicitly indicate the physical address of a device. In an Internet address, for example, the host part indicates which device in a network is being discussed, but the physical address of the device in said network cannot be derived from this. A logical address needs to be assigned to the node device during the installation phase of the network and when the node device moves to another location in the network. Generally, addresses are allocated by manually configuring a logical address for each device. This can be done, for example, from the network management workstation of the network operator through the network management system, or firstly when commissioning, each node device is assigned an address in a centralized manner corresponding to the location of the specified node in the network, and then each node is moved to its own s position.

A disadvantage of this approach is that it requires a lot of work, especially if a larger network should be installed (which means a large number of nodes).

Another disadvantage is the inflexibility associated with this method when changing networks, since a new address has to be manually configured for all such nodes whose position in the network changes.

Contents of the invention

The object of the present invention is to achieve improvements to overcome the above-mentioned drawbacks by providing a novel method of address assignment in nodes of a telecommunication network having a tree topology and comprising point-to-point connections. This object is achieved by a method according to the invention, characterized in that the root node of the network generates unique address identifiers for all transport connections it uses and sends each address identifier over a transport connection to its neighboring nodes, and After receiving this address identifier, the other nodes of the network a) generate unique identifiers for all transport connections they use except the one from which they received the address identifier, b) send An address identifier is sent which includes the address identifier received by the node itself and the identifier corresponding to the connection, so that each node assigns its own address based on the address identifier it received.

The idea of the invention is to automatically generate node addresses in the nodes of a tree-shaped network structure by assigning unique identifiers to all (point-to-point) transport connections (except the connection to the parent node), and sending Each node (child node) located lower in the tree structure sends an address identifier which includes both the address identifier received from the parent node and the identifier assigned to the connection. Since the root node of the network has no parent node, the address identifier it sends includes either only the identifier assigned to the connection, or its own fixed root node address identifier and the identifier assigned to the connection.

If the method according to the invention is used, only one address needs to be assigned to the root node of the network at the installation (or some other stage) of said node device. Thereafter, other nodes (node devices) of the network will automatically receive their location information relative to the root node. Accordingly, the address of the node is automatically assigned in a predetermined manner based on the location of the node. In addition, addresses do not have to be reconfigured when changing equipment or installing new equipment.

The solution according to the invention is particularly advantageous in networks where the physical location of the node devices makes access difficult.

Description of drawings

Describe the present invention and its preferred embodiment in detail below in conjunction with accompanying drawing example, in accompanying drawing:

Figure 1 illustrates the general principle of the method according to the invention;

Figures 2a...2f illustrate the operation of the method according to the invention when a network change occurs;

The flowchart of Figure 3 describes the operation of a single node device;

Figure 4 is a high-level block diagram of those components in a node that are important to the invention, and

Figure 5 illustrates the components of a node device that are essential to the invention.

Detailed ways

The starting point for the method according to the invention is that only the address of the root node of the network is assigned (manually) during installation of the node device in question. On the other hand, other individual node devices do not need to know their own address (ie location) in the network at the beginning. Other single node devices only know that they have multiple point-to-point connections to other nodes. One of these connections is the connection to the parent node of said node device in the tree structure, but in the initial situation (when a node is just started) the node does not need to know which of its connections is the relevant connection.

The principle of the method according to the invention is described below with reference to FIG. 1 . The ring nodes form a tree hierarchy, the root node of which is indicated by a reference sign R. Connections between nodes (ie transport connections) are indicated by arrows (transport connections are naturally bidirectional, the direction of the arrows indicates the transmission of address identifiers). Important to the invention is that the network forms a tree-like hierarchy. In other words, the nodes of the network and the connections between nodes form an acyclic graph (no nodes are created in the network).

The root node R adopts a predetermined method, such as numbering connections from 0 forward, and determines an identifier for each connection. Thereafter, the root node sends with each connection the identifier of said connection. In the example of FIG. 1 , integers are used as identifiers, denoted by reference numerals 0 to N. Referring to FIG.

If (except the root node) any node (i.e. node X) is started and a connection is established, node X waits for its location information on a certain connection. Optionally, a node may send an inquiry to a neighboring node. If location information is received, node X registers the connection as a connection to its parent node in the tree network structure. The received location information includes a set of identifiers, the first identifier specifying which connection can be used to access the next node on the path from the root node of the network to node X. The next identifier states which connection from said next node is closer to node X, and so on. In the example in the figure, identifiers are separated by dots.

After receiving the location information, node X determines (according to some predetermined method) the identifiers of all connections other than the one from which the location information was received. Thereafter, node X sends over each connection location information consisting of its own location (which is the same as node X received earlier) plus the connection identifier.

Correspondingly, the length of the received location information depends on the level of the receiving node in the tree. For example, in the example of FIG. 1 , identifiers are integers separated from each other by dots, node A1 receives position information 0 (zero) from the root node, and node B1 of the next layer receives position information 0.0 from node A1. The location information sent is indicated on the corresponding link in the figure.

When establishing new connections to node X, the node provides these connection identifiers (numbers them) and sends location information on these connections. If new location information is received from the parent node of node X, the new location information with the connection identifier is further sent through other connections.

Figures 2a...2f illustrate how the method according to the invention works in the case of nodes E, F and G being connected to a network in which nodes A...D are already operating (Figure 2a). The figure shows nodes that are not working as dotted lines. Active nodes are shown in solid lines, if an address identifier is marked in the node, the node also knows its address (ie its address in the network). Correspondingly, empty nodes do not know their addresses. From this example it is clear that this method works even if the nodes are started in an arbitrary order.

In the case of Figure 2a, only node F has started up among the many nodes that need to join, but it is disconnected from the network because its parent node (node E) is not started. In the case of Figure 2b, node E has already started up, thus receiving address 0.1 from node B. Node E thus knows the connection to its parent node in the network. Node E then numbers all other existing links starting from zero and sends its own address over these links, appending the number of said link after its own address. In the example case, there is only one existing link, which thus gets number 0, on which address 0.1.0 is sent. This is illustrated in Figure 2c. Thereafter, node F knows its address (Fig. 2d). In the case of Fig. 2e, node G starts up, so that its parent node (E) observes the "startup" of the link and provides it with a number (1) according to some predetermined numbering method, updating the list of active links. Node E then sends node G an identifier 0.1.1 (its own address plus the link identifier). In the last case (Fig. 2f), all nodes know their address (location in the network).

Link identifiers can be assigned in many different ways. It is not necessary to give the numbers in the order in which the links were started, because the numbers don't give any useful additional information. For example points on a compass can be used for numbering, a link facing north can have number 1 and a link facing south can have number 0. Another solution, which is more practically possible, is for example to determine the identifier in such a way that if the device is viewed from a certain angle, for example from the front, the identifier changes in a selected direction, for example from left to right or from top to bottom.

The flowchart of Figure 3 illustrates the operation of a single node. After startup, the node is in phase 31, where it waits for address messages and link events. Nodes of the network can constantly send their address information, so that the node receives the address immediately after startup. A link event indicates the break or start of a link of a node in the connection. Phase 31 is the start phase, in which no normal data transfer takes place. If the node finds an address message (stage 32), it stores the received address and the identifier of the link, assigns the identifier to other links in operation, and sends the address to its children (stage 34). Thereafter, the node changes to its normal state (phase 35), in which it performs normal data transmission.

If the node observes a link event in stage 32, it updates the link list it maintains (stage 33) and returns to stage 31 where normal state cannot be accessed until an address is received.

In the normal state (stage 35), a node is able to observe:

a) The link event of the child node (the link to the child node is interrupted or the link to the new child node is established);

b) new address messages if there is a change in the network above the node (closer to the parent node), or

c) The link to the parent node is broken

In case a) if a link to a new child node is established, the node updates the link list and sends the new address (phase 37). In case b), the node enters phase 34, which stores the received address and the identifier of the link, determines the identifiers of other links in operation and sends the addresses to its child nodes. In case c), the node returns to the start phase (phase 31), waiting for a new address from the parent node.

A network node using the aforementioned method could be, for example, a digital cross-connect device, which can actually be implemented in many different ways. The implementation may vary, for example, depending on the location of the node control part (the part that controls the operation of the node device). Figure 4 shows a high-level block diagram of those components of a node device that are important to the invention. This node (indicated by reference number N) has in this example 4 bidirectional connections (L1...L4) to neighboring nodes. In this exemplary case, the control part CTRL of a node is an independent unit which is common to all interface units IF of the node. The node is connected to the network via an interface unit IF, which may for example be a 2Mbit/sPCM interface. The mode of transfer between nodes is not critical to the invention; for example any frame-based method can be used as the mode of transfer.

It is important for the node that the address processing part (control unit CTRL) receives from each interface unit whether the link is working (ie whether the signal received is serviceable). This message is indicated in the figure with the reference CR (carrier). Reference sign DATA indicates bidirectional data transmission between each interface unit IF and the control unit.

Figure 5 details the components of the control section CTRL which are important to the invention. The input to the control part is generated by a decoding unit 51 which receives eg a frame from the interface unit for which the checksum has been calculated. The decoding unit decodes the protocol data units (PDUs) contained therein from these frames and sends them further to the connection unit 53 . The data unit contains information on the type of frame received, the address that may be included in the frame, and the data included in the frame. If the type is eg normal (normal frame), the connection unit further connects the data unit either directly to another connection unit or to other components of the node device. In the sending direction, the encoding unit 52 generates a frame from the data unit PDU.

If the received frame contains address information, the connection unit connects the data unit to the address processing unit 54, which stores the received address in the memory area M1 in which the node also stores its own address. In addition, the processing unit stores the identifier of the link leading to the parent node in the memory area M2, and stores the identifiers of other links in the memory area M3, which may also contain information about which link is active/deactivated at each specific moment.

The invention has been described above in the form of an embodiment in which the root node only sends the identifier of said transfer link to its neighbor nodes. However, the root node may send its own address in addition to the connection identifier. Thus, if the root node address is eg 0 in Fig. 1, it sends the identifier 0.0 to node A1, which further sends the identifier 0.0.0 to node B1. Accordingly, the address identifier to be sent in this case is one level longer than described above.

Although the invention has been described above with reference to the example of the accompanying drawings, it is clear that the invention is not restricted thereto but modifications can be made within the scope of the innovative idea presented above and described in the appended claims. For example, instead of directly using the address identifier received from the parent node as the address of the node itself, the node assigns its own address according to the address identifier received from the parent node (by processing the received address identifier in a predetermined way, so that The end result is a unique address identifier).

Claims (7)

1. A method of assigning addresses in nodes of a telecommunication network comprising several nodes (A...G) connected together by transport connections so that the network forms a tree structure, one of which is the root node (A) , according to which nodes send each other signals containing address data used by the nodes, The method is characterized in that, - the root node (A) of the network generates unique address identifiers for all transit connections it uses and sends each address identifier over transit connections to its neighbors, and - After receiving this address identifier, the other nodes of the network a) generate unique identifiers for all transport connections they use, except the one from which they received the address identifier, and b) send a message to each of said transport The connection sends an address identifier comprising the address identifier received by the node itself and the identifier corresponding to the connection, so that each node assigns its own address according to the address identifier it received. 2. A method according to claim 1, characterized in that each node directly uses the address identifier it has received as its own address. 3. A method according to claim 1, characterized in that the root node of the network sends its own address to its neighboring nodes in addition to the address identifier of the transport connection. 4. A method according to claim 1, characterized in that integers are used as identifiers. 5. The method according to claim 1, characterized in that the identifiers of the transfer connections are generated according to predetermined principles, independent of the order in which the transfer connections are started. 6. A method according to claim 5, characterized in that the identifier is generated from the physical location of the transport connection in the node device. 7. A method according to claim 1, characterized in that the address identifier received by the node is separated from the transmitting connection identifier using a predetermined delimiter, whereby the address identifier sent by the node is generated in serial form.

Images