RU2281617C2 - Addressing and routing in wireless cell networks - Google Patents

Abstract

FIELD: device and method for addressing and/or routing in wireless cell communication system.

SUBSTANCE: method includes adding to packet being transmitted of address information of protocol level being lower than network level, which information points out geographical position of destination node, to which packet can be routed, and this address information may change during routing of aforementioned packet through wireless communication network; difference between information about geographical position of wireless router or node being the source of packet transfer, and information about geographical position of aforementioned destination node, is determined, and aforementioned packet is transferred inside wireless com network to aforementioned destination node completely on aforementioned lower level through neighboring node, having closest position and direction relatively to aforementioned destination node, while aforementioned neighboring node is selected on basis of certain difference.

EFFECT: engineering of system for addressing and/or routing in wireless cell network, providing changing of address and nodes of routing in operative mode.

3 cl, 6 dwg

Description

FIELD OF THE INVENTION

The present invention relates to a method and apparatus for addressing and / or routing in a wireless mesh network (a network with multiple routes between any pair of nodes).

State of the art

Wireless broadband networks enable high-speed Internet access, where wired broadband infrastructure is not practical. However, such a wireless broadband network will be successful mainly only for residence-related and small business markets, if the infrastructure is provided at low cost, resilient to a changing environment, easy to deploy, and scalable on demand.

New wireless networks with wireless routers as network nodes based on a mesh network emulate the topology and protocols of the Internet, but are optimized for wireless high-speed data transfer. Such a mesh network consists of wireless routers connected to each other via radio links. These radio links can be created and cut off, and the network can adapt to the current state of the radio links. Also, individual devices can be turned on and off without undue disruption to network functions. The routing protocol monitors the location of nodes in the network, thereby creating the ability to route packets between remote nodes, using other nodes in the network as intermediate nodes. Currently, there are several routing protocols that can be used in a cellular mobile communication environment.

Routing protocols can control mobility on a small scale. Problems nevertheless arise when most nodes are mobile, because each node must know the location of any other node within certain limits. This becomes a more serious problem when mesh networks become larger, because changes in the network structure must be communicated to a larger number of nodes and because the number of changes in the network, of course, the higher, the larger the network. Also, the routing tables become larger and larger if the cell becomes larger, because the routes in this network cannot be grouped in the way used in traditional wired networks.

Figure 1 shows a wireless mesh network (SNN, WMN) comprising several wireless routers (BM, WR) connected via radio links. Not all BM wireless routers hear each other. Consequently, packets between two wireless BM routers in the same SNF can be routed through several wireless BM routers and wireless communication lines before reaching their destination. The BM wireless router can also have interfaces different from those in the mesh network, for example, wireless local area network (WLAN), Ethernet, and / or Bluetooth interfaces to which other network devices are connected. Moreover, there may also be routers having other subnets with laptop computers (laptops) (LT, LT), wired routers (M, R) and servers (CB, SV). One or more wireless BM routers can act as a trunk access node (AirHead, AH) configured to connect the SNF to other large networks, such as the Internet. Terminals, such as LT laptops, can also include the functions of a wireless router, i.e. they can act as part of a nuclear warhead.

SNF networks are characterized by the negative impact of constantly changing conditions that cause failure and restoration of communication lines according to current conditions. These networks cannot have any specific administrator to monitor them, which means that they must self-organize and self-repair. In SNF networks, BM wireless routers can also move, which means that the network structure is changing all the time.

Currently, the existing mesh networks are absolutely static with predominantly fixed BM wireless routers, but in the near future cars, trains, buses and other means of public transport may also enter the NFC. According to the most daring assumptions, all people walking in the streets have their own small BM wireless router in their pocket, and the entire network will consist of their own devices. In such a situation, the network structure changes really quickly, and there is no way to track the exact location of each device. Also, the size of the SNC both in the number of wireless BS routers and in the surface area can be huge. In extreme cases, a single nuclear warhead can cover the whole world. This suggests that the methods proposed for use in future SNF should be as scalable as possible so that they can be used effectively no matter how large the network is. The large address space of the new Internet Protocol Version 6 (IPv6) makes it possible to use the Internet Protocol (IP) in this new environment. However, since the IP addresses of routers cannot easily be changed online without disconnecting, this is not an optimal solution for mobile wireless networks. Changing addresses in the online mode will be necessary in order to maintain the totality of routes in IP networks.

New routing protocols are designed specifically for wireless networks without a fixed topology. These protocols try to reduce the routing traffic needed to prevent other nodes from reporting less important changes in the network structure. However, the location of the other nodes must be known to some extent in order to be able to direct traffic in the right direction. These protocols work quite well when the network structure remains the same, and only the radio communication quality of the communication lines changes slightly. But when the routers begin to move, the used network bandwidth is violated because the routing protocol is updated and the packets do not find their destinations.

Mobile IP protocol (Mobile IP) solves the problems of terminal mobility, which are caused by the fact that mobile terminals change their location on the network. This converts the mobility problem into a routing problem, without, therefore, solving the mobility problem in wireless mesh networks in which routers can also move and in which the main problems actually lie in the field of routing.

Moreover, traditional switching in accordance with the Media Access Control Protocol (MAC) in fixed wired networks is based on recognizing the MAC addresses based on the MAC addresses of the packet source received from the interface. It also requires that broadcast and multicast packets are repeated for each link. This method works well in traditional fixed wired networks, but it cannot be used in SNF networks due to the excellent network structure and network bandwidth wasted due to radio broadcasts.

SUMMARY OF THE INVENTION

For this reason, it is an object of the present invention to provide a method and apparatus for providing addressing and / or routing functions in dynamic wireless mobile mesh networks.

This problem is solved by the method according to claim 1 of the claims and the device according to clause 12.

Accordingly, lower level addressing (e.g., link layer, layer 2 or MAC layer) is provided based on the location, which is very convenient in networks with a mesh topology in which routing nodes are also able to move. Location-based addressing makes packet transmission easier on the network because each node can decide in which direction to forward incoming packets only based on the information in the packet header and their own location. This means that nodes do not need to maintain large routing tables and perform time-consuming searches in these tables, which must be really huge in large mesh networks. Thereby, it is possible to change addresses on-line based on the global location without disconnecting and / or without having to update a gigantic amount of routing information. This is possible because there is no need to change the addresses of the upper level (for example, the network level, level 3 or IP level) when the equipment moves inside the cell, because packet forwarding in the network is based on the lower level addresses. Instead, the lower-level address of the device changes as it moves, while the upper-level address remains unchanged. If the location has changed to another network, then higher-level methods such as Mobile IP can be used.

Location-based lower-level addressing also provides the ability to forward packets on mesh networks at the data link layer, which is simpler and faster than when each packet needs to be processed at an upstream level.

In this way, the amount of routing information required for exchange on the network is minimized. This means that when the power of the equipment is turned on, this equipment can almost immediately begin to communicate through the network, which is impossible with the help of previous solutions, because the routing tables of other devices must be updated before that. Also, device movements cause less signaling traffic. Signaling traffic is minimized because each node does not need to be informed about the route to each other node.

Additional advantageous solutions are defined in the dependent claims.

The present invention will now be described in more detail based on preferred embodiments with reference to the accompanying drawings, in which:

Figure 1 is a schematic representation of a wireless mesh network;

2A is a block diagram of the logical structure of a wireless mesh network with data forwarding at level 3 from the point of view of level 3;

Figv is a block diagram of the logical structure of a wireless mesh network with data forwarding at level 2 from the point of view of level 3;

Figure 3 is an example of routing according to preferred embodiments;

4 is a direction-based packet routing according to a first preferred embodiment;

5 is a distance based packet routing according to a second preferred embodiment.

Detailed Description of Preferred Embodiments

Preferred embodiments of the present invention will be described based on the BNC shown in FIG.

2A shows a block diagram of the logical structure of a wireless mesh network with data forwarding at the network layer from a network layer perspective. Here, the network layer routing protocols should provide the information required for routing packets within the SNF, i.e. every wireless router must know the functional route to every other wireless router. BM wireless routers are interconnected by separate communication lines, and packets are routed through various wireless routers inside the base.

FIG. 2B shows a block diagram of a logical structure of a wireless mesh network with data forwarding at a link layer in terms of a network layer according to a preferred embodiment. Logically, from the point of view of the network layer, the BM wireless routers form a network in which they are interconnected by a single communication line. Data transfer at the data link layer is easier and faster than data transfer at the data link network if switching can be implemented in a rather simple way. Therefore, data transfer at the data link layer consumes much less processing power of the central processor (CPU) than data transfer at the network layer, and the cost of the router is significantly reduced.

According to a preferred embodiment, location-based or location-based link layer addressing and switching (e.g., MAC addressing) can be used to forward data at the link layer to provide a simple addressing and routing method. Such a location-based method can provide faster, simpler, and more optimal data transfer, thereby reducing the cost of a wireless router. Essentially, location-based addressing means that link layer addressing is based on device location. In other words, the link layer address of the device indicates what its current location on the globe is.

Thanks to location-based addressing, at the data link layer you can find out, for example, from the MAC address where the devices are located in relation to other devices. When the network device compares the MAC address of the incoming packet with its own address, it will immediately know in which direction the packet destination node is located, and it can transmit the packet accordingly. Therefore, there is no need for large routing tables, because only the addresses of the nearest neighbors should be known.

FIG. 3 shows an example routing according to preferred embodiments based on the SNF with seven nodes Y from b to g. The node Y on the left is the source node, which wants to send a packet to the node g, located in the lower right corner. The destination MAC address is placed, for example, in the packet address field (for example, the MAC Protocol Data Unit (PDU)) and thus informs about the location of the destination. Therefore, the intermediate nodes c and e do not need to know the network topology in order to route the packet to its destination, because they know the geographical location of the destination (from the MAC address of the destination) and their own geographical location. In this way, they can route the packet in the right direction, and at the end the packet reaches node g, which is its destination.

The addresses used may be unique on a global scale or they may be unique only in a specific area or only in a specific type of network. If the addresses are unique only in a certain area, then they are all compared with the same fixed point. There are also various ways of getting location based addresses. If more than one method is used in the same network at the same time, there must be a way to determine which method is used in which device.

Information about the location of the device can be obtained using the existing global navigation and positioning system (GPS) or any other known or future methods of positioning or determining the location, for example, positioning systems of the global GSM mobile communication system, which must work one way or another way, even indoors. Other future positioning systems with sufficient accuracy may also be used. Devices can also use information about the location of neighboring devices and delays in the propagation of radio signals to neighbors when calculating their own location. If the device knows the location of its three neighbors and the distance to them, then it can calculate its own location in two-dimensional space based on this information.

Forwarding a packet to the SNF can be done using only location-based addressing or together with traditional routing methods, combining the advantages of both approaches.

According to preferred embodiments, the forwarding or routing of packets is carried out entirely at level 2. When a packet reaches a node, this node determines the location of the packet's destination based on the MAC address of the destination.

In a first preferred embodiment, the direction of the destination is determined in the node based on its own exact location and location of the destination. After that, the node forwards the packet to the neighboring node, which is the closest to this direction. This principle of direction-based routing is shown in FIG. 4, where the source node or wireless router 10 determines or calculates the direction to the destination node 30 based on its own location and the location of the destination node determined based on the corresponding MAC address of the MAC PDU . Then, it calculates the differences a and b between the direction of the destination and the known directions of neighboring nodes or wireless routers 20 and 40, respectively. After that, he selects a neighboring node with a smaller difference and routes the packet to this node, as shown by the routing paths M in FIG. 4.

Figure 5 shows an alternative distance-based routing, where the source node 10 determines or calculates the distances L, S between the location node 30 and all its neighboring nodes 20, 40 based on the specific location of the destination node 30. Then it simply forwards the packet to a neighboring node that is closest to the destination node 30. In the present example of FIG. 5, the distance S between the lower neighboring node 40 and the destination node is smaller, and the routing path M passes through the lower neighboring node 40.

In some cases, the two alternative methods described above according to the first and second preferred embodiments can lead to different behaviors, but in practice the differences should be minimal. The method according to the first embodiment leads to shorter jumps, thereby reducing power consumption due to the lower transmission power required. The method according to the second embodiment leads to longer jumps, minimizing the total number of jumps to reach the destination. But it also means that more transmit power is required, which reduces the battery life, and has a greater effect on other neighboring devices due to the higher transmit power used.

Note that other parameters may also be considered when deciding on forwarding or routing. They may include the throughput of communication lines, bit error rates, different loads on different communication lines, etc.

Since addressing is based on location, devices must always know their current location. The location information must be accurate enough to be able to send data based on it, but it is not required that it be unnecessarily accurate. Since all nodes in the network know who their neighbors are, accurate location information becomes unnecessary when packets approach their destinations. This means that the required accuracy corresponds to the distance between neighbors. Additionally, accuracy is not even necessary to such an extent, if the main task is to ensure that it is more accurate when compared with neighbors. In other words, a systematic error in location information is not critical. A systematic error can occur, for example, when the location of a device is determined based on the locations of neighboring devices and based on distances to neighbors.

Since the MAC address of the device is based on the location of the device, it must change when the equipment moves. Changes in the address must be reported to at least neighboring devices so that they can correctly forward packets destined for this node. It also affects the routing of other packets. Despite the change in address, the host may still receive packets with the old address for some time. In addition, the nodes with which the node has been communicating recently can be notified of a change in address in order to avoid packet loss. There must also be special nodes on the network to which MAC addresses are reported and from which MAC addresses corresponding to some IP addresses are requested. In other words, nodes must exist that act as a neighbor discovery cache. Therefore, one (nearest) node from these specific nodes should inform about the change of address, after which the rest can receive information from this particular node.

Changing the address is really only necessary when it significantly affects routing. This means that if a moving node does not pass close to any other nodes, moving does not have a very large effect on routing. In fact, the system can be designed so that neighboring nodes know the location of the node more accurately than other nodes in the network, and the decision to forward could be made according to this information. The MAC address will be used only when making decisions about routing at a greater distance from the node and, therefore, the MAC address should only change when neighbors change (the node passes near other nodes, new neighbors are found). Nodes that move at a constant speed can also predict their future movement and begin the process of changing the address in advance.

A packet is switched through the network until it reaches its destination or the switching algorithm cannot make a conclusion about where to send the next packet. In these situations, the packet moves to the upstream protocol layer (network layer), which then decides what to do with this packet. Situations when the switching algorithm must transfer the packet to the higher level, and the operations performed by the node, if the packet does not reach its destination by normal forwarding methods, are defined as follows.

A packet that has reached its destination at the data link layer is processed according to the normal operation of the router at the IP layer. The link layer destination is not, of course, necessarily the destination of the IP packet, but may be a router in the middle of a route on the network. If the source and destination of the received IP packet are on the same network, the packets are forwarded normally, and a redirection message (Internet Control Message Protocol (ICMP) / Neighbor Discovery Protocol) can be used to instruct the source to send the packet directly to the destination MAC address.

There are at least three situations where the switching algorithm fails to deliver the packet to its destination and when the packet must be transferred to the higher level:

1) destination does not exist,

2) it is not heard due to radio interference or

3) there is no direct route to the destination.

The switching algorithm can determine these situations based on two signs:

1) the packet comes from the interface to which it should be sent according to the switching algorithm, or

2) the wireless router can hear another wireless router located behind the wireless router of the destination, but it cannot hear the destination itself.

MAC addresses can be resolved in IPv4 and IPv6 networks by using the address resolution protocol and neighbor discovery protocol, respectively. In SNF networks, these protocols consume too much frequency band due to the broadcast and multicast nature of the use of MAC addresses. Therefore, these packages can only be delivered to certain selected nodes, which then perform the intermediary function of providing information to those who requested it. A node that broadcasts multicast and broadcast packets to a selected node can also intercept packets and act as an intermediary for a while. The selected nodes then maintain the latest information among themselves. If these selected nodes are connected to some core network, then it can be used to update information instead of SNF networks. Other protocols that use multicast and / or broadcast messages can also be implemented using the method described above.

If a node changes its location within the Nuclear Forces and its GPA-based MAC address changes, it reports the new MAC address to the selected node and / or also to all other nodes with which it has been communicating recently inside the NNS, and / or any other network node. Since the MAC address of a node can often change in the SNF, the MAC addresses of nodes cannot be cached for a long time by other nodes. The node must receive packets arriving with the old MAC address (s) for as long as other nodes can cache this address. Of course, if the node changes its location quickly enough, the packet may not find its destination at the old MAC address. In this case, the packets are re-routed to a new location or discarded by neighboring routers in the previous location or by any other network node. In order to ensure re-routing, the node can also communicate its new MAC address to all previous neighbors that it can no longer hear.

Note that the present invention is not limited to location-based MAC addressing. Implementations of the invention are also possible in other systems where location or location based variable addresses can be provided at a lower protocol level, while static or fixed addresses can be provided at an upper protocol level.

Claims (21)

1. A method of forwarding a packet in a wireless network, comprising the steps of: a) add to said packet address information of a protocol layer located lower than the network layer, which indicates the geographical location of the destination node to which the packet should be routed, and this address information may change while the packet is being forwarded through the wireless network, b) determine the difference between the geographic location information of the wireless router or node (20, 40) that is the source of the packet transmission and the geographic location information of the destination node (30) and c) forwarding said packet inside said wireless network to said destination node completely at said lower level through an adjacent node having a closest distance and direction with respect to said destination node, wherein said neighboring node is selected based on said certain difference. 2. The method according to claim 1, wherein said packet is forwarded using the smallest difference between the direction of said destination node (30) and the directions of neighboring nodes (20, 40). 3. The method according to claim 1, wherein said packet is forwarded using the calculated nearest neighboring node determined by calculating the distances between said destination node (30) and neighboring nodes (20, 40). 4. The method according to claim 1, wherein said address information is a link layer address. 5. The method of claim 4, wherein said link layer address is an address of a medium access control (MAC) protocol. 6. The method according to claim 1, wherein said packet is transmitted to an upstream protocol layer if said destination node does not exist in said wireless network, if said destination node cannot be heard, or if there is no direct route to said wireless network to said destination node. 7. The method of claim 6, wherein said network layer is an Internet Protocol (IP) layer. 8. The method of claim 6, wherein said network layer is configured to discard said packet if said packet is transmitted to said network layer. 9. The method according to claim 1, in which the new address of the downstream protocol level is reported to neighboring nodes, nodes with which communication was recently made, and / or any other nodes in the network if the geographical location of the network node changes. 10. The method of claim 1, wherein said downstream protocol layer address is a location-based address. 11. The method according to any one of claims 1 to 10, in which said packet is rerouted or discarded by a neighboring or any other node if the old destination node address corresponding to the protocol level below can no longer be detected. 12. A device for forwarding packets in a wireless network, comprising: a) adding means for adding to the packet the address information of the protocol layer located lower than the network layer, which indicates the geographical location of the destination node to which the packet should be routed, and this address information may change while the packet is forwarded through the packet wireless network b) determination means for determining a difference between the geographical location information of the wireless router or node (20, 40) that is the source of the packet transmission and the geographical location information of said destination node (30), and c) forwarding means for forwarding said packet inside said wireless network to said destination node completely at said lower level through an adjacent node having a closest distance and direction with respect to said destination node, wherein the selection of said neighboring node is based on said certain difference . 13. The device according to item 12, further comprising means for calculating the smallest difference between the direction of said node (30) of the destination and the directions of neighboring nodes (20, 40). 14. The device according to item 12, further comprising means for determining the nearest neighboring node by calculating the distances between said destination node (30) and neighboring nodes (20, 40). 15. The device according to item 12, in which the said address information is the address of the data link level. 16. The device according to clause 15, in which said link layer address is the address of the media access control protocol (MAC). 17. The device according to item 12, in which the aforementioned device is configured to transmit the said packet to the upstream protocol layer, if said destination node does not exist in said wireless network, if said destination node cannot be heard, or if there is no direct route to said wireless network to said destination node. 18. The device according to 17, in which said network layer is the level of the Internet Protocol (IP). 19. The device according to item 12, in which the aforementioned device is configured to report a new address of the lower level of the protocol to neighboring nodes, nodes that have recently been connected, and / or any other nodes in the network if the geographical location of the network node changes. 20. The device according to any one of paragraphs.12-19, wherein said address of the downstream protocol layer is a location-based address. 21. A data packet intended to be transmitted over a wireless network, embodied as at least a portion of a signal transmitted by the wireless network, the header of said data packet containing information about the address of the node that is the destination of this data packet, wherein said address corresponds to the protocol layer below than the network layer, and is based on the geographical location of the said node, and the mentioned information may change during the forwarding of the said data packet to omyanutomu node.

Images