JP2002529017A - Broadband wireless mesh topology network - Google Patents

Abstract

(57) [Summary] A broadband wireless network based on a mesh-type topology is provided to the premises of consumers, small businesses and large enterprises, and to an IP standard interface and a consumer-side multimedia digital bus interface (BAP-1, BAP). -2) is a broadband wireless network that is optimized to bring data, Internet, voice and multimedia using a data interface such as -2). With the ability to route data between multiple wireless nodes using route diversity as well as adding and dropping data via standard interfaces, the consumer premises radio transceivers (SR10-SR21) are based on a mesh network. Increase availability, reliability, and enable load balancing.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

【Technical field】

The present invention relates to the field of wireless communications in general, and more particularly to the field of high capacity broadband wireless networks designed for packet data transmission in the high frequency band of the spectrum.

[0002]

[Background technology]

The definition of a high capacity broadband system is T1 (1.5 MB / s) or E1 (2 MB / s).
/ S), typically operating at transmission rates up to 100 MB / s or more per link, and in the frequency range above 6 GHz, typically 10.15 G
Hz to 10.65 GHz, 24 GHz to 31 GHz and 37 GHz to 4
A system operating in a licensed band of the 1 GHz spectral portion. New bands are opened from 40 GHz to 60 GHz, and these can be used without a license. At those frequencies, strong transmission is achieved via an open line of sight (LOS) between the transmitting and receiving radio stations.

[0003] Wireless communication networks allow competing access to fibers, hybrid fibers, coaxial cables, and twisted pair networks. T1 (
Low bit rate wireless systems up to 1.5 MB / s, US) and E1 (2 MB / s, ETSI) have been developed and deployed. These wireless systems are used in the lower part of the spectrum, usually below 6 GHz, where the bandwidth allowed for use is limited. Examples of such systems include 900 MHz, 1.8
To 2 GHz cellular systems, 2.3 GHz WLL (wireless local loop) and unlicensed wireless systems in the 900 MHz, 2.4 GHz and 5.7 GHz ISM (Industrial, Scientific and Medical) bands. Due to the limited bandwidth of these bands, such systems operate in a narrow band and have low capacity per link. Not only because of the expansion of Internet services, but also due to the deregulation of the telecom industry, which has allowed CELECs (competitive local switching telecommunications carriers) to create their own access infrastructure (1MB / s The demand for high capacity (above 2 MB / s) has increased dramatically during the past few years. As a result, regulators around the world have opened up new bands above 10 GHz for use. These bands are typically licensed for use by a single operator in a given area. Against the previous situation, where national laws required that radio links, such as the United States FCC, be pre-approved for radio links, operators would try to implement their own radio system. Topology. Maximize operator spectrum usage due to the main difference in operation between per-link PP (point-to-point) frequency allocation and allocation of broadband spectral range per distinct region to be used by a single operator And their newly opened millimeter-wave spectrum opens up a market for a completely new topology that can create higher capacity and more advanced wireless networks. Some of these spectra are described in the literature as LMD
S (Local Multipoint Distribution System) or LMCS
(Local multipoint communication system) or PMP (point to multipoint) system. The present invention relates to new topologies that go further from these spectrum ranges licensed for use based on territory, where new customer radio stations can be added to the system anywhere without prior approval.

A topology for a typical prior art broadband wireless communication network is shown in FIG. Each base station covers 360 ° using four 90 ° sectors.
The two base stations are connected via wireless connections to other base station locations where information is transferred from all base stations to the main switch by fiber connections. The main switch provides connections between subscribers in the area covered by wireless access as well as other subscribers either through the fiber trunking system (SONET or SDH), either in the same area or in remote external areas. Play a role. The network is a point-to-point (P) network such as wireless links 9, 10 or fiber (link 8).
Based on a point-to-multipoint (PMP) access network connected via P). In FIG. 1, base stations BS11, BS12 and BS13 are located at terrain elevation so as to achieve a linear line of sight (LOS) with subscriber radio stations SR14-SR18. Each base station includes multiple sector antennas transmitting to distinctly fixed sectors such as 19, 20, and 21. To avoid interference between adjacent sectors, different frequencies are assigned to sectors that can interfere with each other. Thus, sector 20 uses frequency 122, adjacent sector 19 uses a different frequency 121, and frequencies 121 and 122 are any frequencies. Sectors 21 belonging to different base stations are assigned a frequency 111 that should be different from frequency 122 in some situations to avoid signals from sector 21 that interfere with communication in sector 20. However, under proper circumstances, frequency 111 may be similar to frequency 121 if interference is not likely to occur. Many locations do not have a LOS like SR18 behind SR17.

[0005] The subscriber SR17 of the sector 21 is divided into the subscriber SR16 of the nearby sector 24 and the SR21.
17 to BS 11, through link 10 to BS 13, through link 8 to switch 22 and back through link 8, BS 13, link 10, BS 11, and S
Communication can be performed via a line that follows a route to R16. Similarly, sector 21
The subscriber SR17 of the different base station includes the subscriber SR23 of the sector 20 of the different base station and the SR17.
To BS 11, through link 10 to BS 13, through link 8 and switch 2
2, the link 8 returns to the BS 13, and the link 9 can communicate with the BS 12 and the SR 23 of the sector 20 via the line allocated by the switch 22.

[0006] In all these cases, a central switch, such as an asynchronous transfer mode (ATM) switch 22, may serve as the center of the switching fabric through which all traffic must pass. The same switch is a trunking WAN
(Wide area network) It may include an additional port connected to the fiber link 7 to the backbone. Therefore, the subscriber who receives the service of the wireless PMP access system can access the world. The base station can convert the wireless protocol used to communicate wirelessly with the subscriber to a standard transmission protocol such as ATM.

[0007] The data from the base station to the subscriber radio station (downstream direction) is sent to a number of subscriber-oriented sectors, provided in sectors that are clearly fixed, eg at angles between 15 ° and 90 °. It is transmitted in broadcast mode via an antenna. Data is addressed to a particular customer by assigning a different address to each of the data packets at a particular broadband frequency, for example using TDMA (Time Division Multiple Access). In other prior art implementations, the PMP system allocates certain narrower frequencies within the operating spectrum band for each subscriber while communication with the subscriber lasts. To maintain transmission in the upstream direction, each base station synchronizes transmission on the uplink from each of a number of subscribers in each sector. A scheduling algorithm defines the timing and length of transmitted packets at a well-defined frequency (TDMA) for each individual subscriber wireless station. This synchronizes the arrival of successive packets from different subscribers to the base station. In a different prior art scheme, the FDMA (Frequency Division Multiple Access) scheme, the base station relies on a request for transmission received on a service channel and a specific frequency channel with a defined bandwidth based on a priority scheduling algorithm. Is assigned to each subscriber. Other modes of operation such as code division multiple access (CDMA) may also be used for PMP systems.

[0008] The advantages of the prior art wireless PMP system are as follows. 1. Low cost of subscriber unit. Most control and system information is provided in the base station.

[0009] 2. In particular, for virtual circuit allocation systems such as ATM or frequency allocation,
Simple media access control (MAC) layer.

[0010] 3. A simpler base station. It is an external switch from a known subscriber (eg, A
TM) acts as a smart multiplexer for data to and from TM).

[0011] 4. Relay based on standard existing technology for switching. For example, ATM
Such switches are external to the system.

The disadvantages of the prior art wireless PMP system are as follows. 1. The base radio station is located in an elevated location to maximize LOS coverage (
Expensive).

[0013] 2. Due to the shadowing effects of trees and buildings, only partial coverage can be achieved.

[0014] 3. High cost of securing and maintaining base station equipment and location. 4. The cost of trunking between the base station and the switch.

[0015] 5. The need for one or more base station switches to provide system connectivity between subscribers and themselves or the system with the outside world.

[0016] 6. As the density of subscribers in a sector increases, the bandwidth per subscriber decreases.

[0017] 7. Sector availability is limited based on the longest link from the base station. Longer distances mean that less signals are received from the subscriber, and thus the packet signal levels vary significantly.

[0018] 8. The base station needs to be connected to the switch via an expensive and high capacity link such as, for example, wireless or OC-3 or higher fiber.

[0019] 9. The need for a switch to effect the connection. 10. Each packet of data needs to be routed all the way to the switch and vice versa, thus consuming large bandwidth resources of the system.

11. Limited capacity of base stations. Current technology capacity is 155 MB / s per base station
Lower than.

12. No route diversity. 13. Without repeater functionality, all units communicate with the central base station.

14. As the need for higher capacity wireless access increases, more base stations must be located in the same area in closer proximity, so that the same frequency may be reused many times.

[0023]

Summary of the Invention

The wireless mesh network of the present invention includes a plurality of nodes, each of which includes a radio frequency transmitting circuit and a radio frequency receiving circuit. Each node is a LAN
Alternatively, it further includes a local port for connection to a user provided system, including a WAN connection or a computing system. Each node further includes transmitting and receiving circuitry and a data processing unit coupled to the local port for exchanging data between the circuitry and the local port. There are spatial flow determination means for transmitting and receiving data stream transmissions from particular nodes of the network. In particular, each node has a line-of-sight relationship to a subset of the nodes in the network. The means for determining the spatial flow for a given node selects among its subset nodes for either transmitting and / or receiving data streams. The data processing unit includes extraction means for extracting selected data from an incoming transmission intended for a given node. The remaining data stream is then transmitted to the next node or nodes. Similarly, there is insertion means for inserting data into the stream being transmitted. Thus, the data of the incoming data stream is extracted (dropped) by the nodes
And is output (dropped) to its local port, while the data stream is inserted (added) with data from a given node as it is being transmitted.

In a preferred embodiment of the invention, at least some nodes include status information about other nodes in the network. Thus, the node has some information about the network. This information is sent to the spatial flow decision means, and based on such information, selects a node to transmit to or receive from.

At least some nodes also have means for evaluating characteristics of the received transmission. Means are provided for transmitting information about this feature to the sending node. The sending node adjusts its transmission parameters in response to such information.

The nodes include means for providing connectionless transmission of data, connection-oriented transmission of data, or a combination where one segment of the network is connectionless and another segment is connection-oriented.

[0027]

BEST MODE FOR CARRYING OUT THE INVENTION

Referring to FIG. 2, a mesh topology wireless network based on operation at the millimeter wavelength frequency is assumed. Wireless networks are based on the realization of network topologies, with wireless mesh networks enabling packet communication over mesh networks and optimized radio stations designed for operation in the MAC (media access layer). Optimized for high capacity packet communications. According to the present invention, the network is deployed in the vicinity of the customer premises and internal data provided by the subscriber, such as a LAN (local area network), WAN (wide area network) or a typical PC as well as with each other. Includes a number of subscriber radio station (SR) nodes that communicate with the system. A linear line of sight (LOS) between subscribers exists between each node in the topology and a subset of the nodes. Multiple reflections are negligible. An SR node acts as a network node with the ability to receive data packets from neighboring nodes and analyze the destination of the packets of data. The SR node can drop data addressed to the one or more local ports and add data from the local port. The SR node, based on the selection of routes available in the network towards the final destination of the packet, receives received data packets addressed to other network nodes and packets received from its local port by its neighboring nodes. Orient to. Nodes always receive status information from other nodes,
Thus they can evolve a scheduling mechanism for real-time wireless virtual routing of data packets over mesh networks.

Thus, a mesh network can be viewed as a virtual router distributed over geographically diverse areas where the SR local ports are router ports. An important aspect of the present invention is the ability of nodes to determine data routing of received data packets by sharing status information among themselves via real-time scheduling algorithms. This scheduling algorithm allows the SR node to select the route of the packet available at each moment. For example, packets between the same source and destination may be carried over many different routes or via the same route depending on the available routes. To overcome the varying delays due to different routing, SR nodes may reorder certain data packets arriving at different times to create a well-defined data stream at the local port.

Another important aspect of the present invention is that it was not originally connected to a network,
An automatic admission process for new nodes joining the system. S
R nodes have the ability to examine their local area and find out with which SR nodes they can make communication links. SR nodes periodically update their own status information about the surrounding nodes where they have LOS communication.

The SR node simultaneously communicates with other nodes in the vicinity and communicates with other nodes.
It can provide directional communication with certain nodes. This directivity of data reception and data transmission is achieved by a combination of time division access combined with transmission in selected sector areas. This choice is based on the use of fixed angle sector antennas (30 °, 45 °, 90 °) or one-dimensional for terrestrial use only, by using beam steering performance through the choice of sector antenna. Antenna or ground and 2 for very high altitude or space use
This is achieved either by using a phased array antenna, which can be a two-dimensional antenna.

Subscriber Radio Stations (SR) Each subscriber radio station (SR) acts as a network node and, in accordance with the present invention, communicates with at least two other SR network nodes on which the LOS resides (eg, above 6 GHz). It has the unique ability to communicate at high frequencies. For example, SR16 is SR17, SR18, SR19 and SR2.
0 and thus serves as a repeater for information transmitted from, for example, WAN access BAP-1 (backbone access point) to SR18 or from SR17 to SR19. The SR network node also
Packets of data can also be added and dropped via a local interface such as 100 / 10BaseT, IEEE 1394 or other standard interfaces. In another variation of the invention, the SR may change its frequency channel when the SR communicates with a different SR nearby. This increases network throughput as well as reducing mutual interference. (A detailed description of a subscriber radio station implementation is shown in FIG. 3 and described below.) Line of Sight (LOS) In current wireless mesh topologies, SRs are used to bridge data packets and bridge them. It can perform the function of partial routing. As a result, data may be routed from one SR to another SR.
Therefore, the SR must be in line of sight with at least one other SR.
This network topology enables the construction of a new type of wireless network, where subscriber access is achieved by bypassing LOS obstacles. FIG.
For example, if SR 18 does not have a LOS with SR 19, communication with BAP1 may still be established on another route to SR 19 via SR 16 or SR 15 via SR 20. This arrangement has the advantageous property that as the density of the system increases, the ability to establish a LOS or multiple LOSs for each subscriber increases. It is noted that at the beginning of the initial network deployment, the initial sowing of areas with a certain amount of SR provided in mutually visible locations may be advantageous.

Throughput The throughput of the wireless mesh network of the present invention has the unique property of increasing as subscriber density increases. The reason for the increased capacity is due to the addition of available high speed links that allow the routing of data packets between multiple paths in the network. This phenomenon is unique to the high-frequency millimeter-wave mesh network of the present invention, where multiple links operating at the same millimeter-wave frequency require the MAC layer to synchronize their beams and create spatial interference with each other. Can operate in the same area as avoiding. The MAC layer also prevents temporary interference by synchronizing the transmission timing of potentially interfering SRs located in the same area and operating with the same beam overlap at the same frequency. The present invention is advantageous in that the propagation of these high frequency beams requires a linear LOS between the nodes. As a result, all reflected beams are absorbed and therefore do not interfere with the operation of the network. This also means that multiple reflections through non-line-of-sight reception greatly interfere with network performance, reducing network capacity as the number of subscribers increases in the area, eg, 6 GHz,
This is a significant advantage for any installation of a wireless mesh topology, especially at low frequencies below 3 GHz. Another significant advantage exists for millimeter wavelength prior art PMP wireless networks. In prior art systems, the throughput per sector is fixed, and as more subscribers join the network, the network capacity per subscriber decreases linearly.

MAC Layer According to the present invention, the subscriber wireless station includes a MAC layer that identifies the location of the nearby SR and its address, taking into account the surrounding network topology. The MAC layer of each SR reads packets received from each of its neighboring SR network nodes and from the SR's local interface port.
For packets arriving from local ports, the MAC layer specifies the routing of each packet based on the destination address of the packet. It adds additional information bits used for FEC (Forward Error Correction) and, optionally, encryption of wireless communications. Data packets arriving wirelessly from nearby SRs are analyzed based on their addresses and retransmitted or dropped locally. For example, a possible arrangement for a broadcast transmission from SR16 to SR17 and SR18 at the same time could be by a single transmission where both SRs are provided in the same beam footprint or by repeating the transmission in the direction of each of those SRs. Can be achieved. This unique combination of wireless packet radio stations and internal packet data routing and / or bridging capabilities allows for the implementation of wireless mesh network topologies.

The route diversity MAC layer is also responsible for identifying available routing paths from that address to the destination address. For blocked routes, the MAC layer may route the packet via an alternative route. For example, SR15 may receive data from BAP1 via SR12 and SR13 or via SR19 and SR16 or via SR21. It can also communicate outside the network via multiple BAPs, such as BAP-1 and BAP-2. This results in a significant increase in network reliability and resistance to rain fading, which can block certain paths for short periods of time.

An SR entering the network using an automatic admittance grant process also scans the frequency channels and, if it has beam steering capabilities, spatially different directions. It tries to be accepted by the network. Once it is registered with the system as a valid network node, the network allows it to join and updates its information with other nodes.

A channel selection network may select a particular frequency channel for different areas or links. In addition, the optimized spectrum allocation can be based on network data flow information monitored and investigated by external systems.
It can be downloaded from an MS (Network Management System). This input allows for better optimization as well as manual definition of channels used in different areas. In FIG. 2, the link from SR12 may be assigned frequency channel 102, SR15 may use frequency channel 101 to communicate with SR21, and channels 101 and 102 may be in any frequency band.

The multiple WAN access network includes BAP-1 and B, shown in FIG.
It is designed to operate with multiple WAN accesses, such as AP-2. This requires the PM to continue the backbone trunking system (SONET / SDH) towards base stations typically located at terrain elevated locations
For the P architecture, convenient access to a mesh-type wireless network in which only the SR needs to be added to the BAP can be provided. An advantage over the prior art is due to the fact that a large number of high-speed trunking connections are used to increase the total data flow into and out of the network, and thus can dramatically increase its total throughput capacity. Occurs. Route diversity into and out of the network is another by-product of multiple WAN access points.

The result of the operation through the frequency reuse nearby neighbor, the possibility of frequency reuse is greatly increased. This results in a dramatic bandwidth increase for non-LOS systems at lower frequencies (<6 GHz) and for PMP systems where transmissions are made into large sector areas.

Due to the shorter distance between subscribers, which gets shorter as the higher bandwidth density per link increases, the signal strength reaching the receiver is significantly higher compared to the PMP system. As a result, the wireless link can operate at a higher bit rate in the same frequency band. This can be achieved by using a high QAM modulation technique. In addition, since the operation of one link is independent of the nearby links, each link adjusts its bit rate based on the signal level received at each instant to optimize the BER (Bit Error Rate) Can be This is in contrast to PMP systems where the bit rate in a sector is defined based on the longest link in the sector.

Availability of Mesh Networks The availability of a wireless mesh network according to the present invention may be due to other topologies (because the distance between multiple SRs is shorter).
For example, PMP). The further increase in availability is due to the network's ability to change (increase) the power transmitted in each segment of the network independently of the other segments of the network. In addition, the network can increase the sensitivity (and reduce the bit rate) of each segment of the network without significantly changing the overall network performance. Thus, even if heavy rain causes fading to propagate through the mesh network, some links may increase their power to increase signal strength and increase their sensitivity by reducing their bit rate. , Reliability can be improved. As a result, the SR gets its data packets from the same link (perhaps at a lower bit rate) or from other links that are less susceptible to rain fading.

Mesh Network Reliability The mesh network of the present invention can operate over multiple routes simultaneously by allowing each SR to receive a stream of packets from multiple directions. Through its MAC layer, the SR can rearrange the order of arriving data and drop packets arriving via longer routes in the correct order. This route balancing capability allows the network to operate reliably even when some nodes are not running.

Backup Local reliability can be increased by using two or more SRs at the same location (node) connected to the same subscriber LAN. For example, two SRs can operate simultaneously to route information to the same location. 1
If one SR fails, the second will continue to operate at the increased bit rate to make up for the loss of the first radio station.

Increasing WAN Access Capacity If it is necessary to increase the capacity from the network to the BAP, connecting its local interface to the second 100BaseT interface of the WAN router provided in the BAP allows additional SRs to be in the same location. Can be added to In this case, double SR not only provides backup redundancy if one of them fails, but also increases the bit rate from and to the wireless network.

The low orbit satellite and high flight communication transponder mesh radio topology can also be used to communicate with network radio stations that move slowly in a particular orbit over the earth. The high-flying transponders on these low-Earth satellites or aircraft act as repeaters / routers, thus allowing access to multiple points of the mesh network. For such operation, the wireless station may communicate with the satellite by adjusting its beam and scanning a known orbit in the air. Benefits come from the ability to establish LOS almost anywhere.

Polarization Diversity This system uses different signal polarizations (vertical or horizontal) for different subgroups within the same coverage area to increase total capacity and / or improve interference immunity. obtain.

Refer to FIG. 3 for further discussion of the subscriber station (SR) according to the present invention. The SR includes a plurality of components as shown in the schematic diagram of FIG. As can be seen, the wireless station uses a time division duplex (TDD) scheme. T in mesh topology
One of the main advantages of using DD radio stations is that TDD radio stations can operate asymmetrically, i.e. have different transmission and reception speeds, which is more suitable for the packet network of the present invention. Because data packets can flow in different directions, transmitting a segment for a longer time in one direction may require more capacity than transmitting in the opposite direction. TDD radio stations do not have a diplexer that separates transmitted and received signals in two different frequencies. Thus, the wireless station can scan a larger spectrum. By eliminating the diplexer, this reduces the cost of the SR as compared to the FDD scheme which requires a diplexer.

The main components of the TDD radio station shown in FIG. 3 are as follows. R10 Drop Interface This is the local interface port to the customer premises. Local ports may include 10 / 100BaseT, ATM, IEEE 1394 or custom interfaces for connection to a LAN or WAN or computer. In addition, power can be supplied on the same cable.

R20 Cable Interface This provides security and power conditioning circuitry.

R30 Digital Transmitter This typically includes standard circuitry known in the art. They include variable level modulators, FEC encoders and channel synthesis circuitry.

[0050] R40 wireless mesh media access controller and the link layer controller this layer, link layer ARQ, a packet size adaptation, priority queue, any encrypted data from the scheduling and routing as well as the local port of the virtual circuit and Provides decryption of data dropped to a local port.

R50 Digital Receiver This typically includes standard circuitry known in the art. They include digital tuners, variable level demodulators and FEC decoder circuitry.

R60 Digital-to-Analog Converter R70 Transmit / Receive TDD Switch This switch, together with switch R130, switches the radio station between transmit mode and receive mode.

R80 analog-to-digital converter R90 mixer R100 linear power amplifier R110 local oscillator R120 low noise amplifier R130 transmit / receive TDD switch This switch, together with switch R70, switches the radio station between transmit mode and receive mode.

R140 Antenna Selection Switch This switch, under the control of the MRMAC routing circuit, performs directional antenna selection and performs data stream transmissions, ie, certain specifics so that other nearby nodes do not receive such transmissions. Provides (directional) control of the spatial flow of transmissions directed to and from nodes.

Antennas ANT1 to ANT5 These antennas are directional antennas, ie sector antennas, and transmit and receive narrow beams from a selected spatial orientation or altitude in the case of satellite communications. FIG. 3 shows 5 for illustration purposes only.
One antenna is shown. The present invention is intended to support any number of antennas. For phased array antennas, different circuit configurations are required to control the relative phase shifts of elements including such antennas that provide beam steering performance. Directed transmission performance is known in the art. In addition to the above,
Any equivalent selective transmission scheme discusses a scheme that works equally well in the present invention.

MAC (Media Access Control) Layer Implementation of the present invention is based on the introduction of a new MAC layer designed and specified to accommodate the requirements of wireless mesh networks. A summary of the MAC layer operational requirements and the main modes of operation is provided below.

FIG. 4 shows that either scanning the beam or selectively effecting communication with one of the SRs in that sector, such as by using a directional or phased array antenna as described above. 2 illustrates some principles governing the operation of a wireless mesh network, assuming wireless stations capable of communicating within a sector. Other SRs in the sector that may interfere with the communication are controlled by the MAC layer and remain silent during the scheduled transmission.

The high-level MAC layer of the present invention and its main rules are described below. Directional transmission Each node can select a spatial direction to transmit in based on the location of the receiving node of its LOS. The receiving node selects a direction from which to receive based on the location of the transmitting node in the LOS.

Scheduled Transmissions Each node can transmit only when its transmissions do not interfere with other transmissions. Transmission schedules established at all nodes provide for synchronization of such transmissions.

Parallel Transmissions Multiple wireless nodes can transmit in parallel when they do not cause interference. In FIG. 4, while wireless node 4 is transmitting to wireless node 1, wireless node 2 can transmit to wireless node 0.

Mesh Wireless Media Access Control (MRMAC) MRMAC is a set of procedures and algorithms responsible for routing and scheduling packet communications in the mesh wireless network of the present invention. All wireless nodes participating in the wireless mesh network realize the MRMAC function. MRMA
C is distributed intelligence that bundles a collection of wireless nodes into a highly efficient wireless packet network.

Mesh Radio Physical Layer (MRPHY) MRPHY is a radio physical layer that is compatible with MRMAC and provides the necessary performance to enable scheduled transmission and routing of data packets in a mesh radio network. give.

Mesh type radio link layer control (MRLLC) The MRLLC is a protocol layer that plays a role of adapting a user's protocol transfer to MRMAC. It performs functions such as priority classification, automatic repeat request (ARQ) and packet size adaptation.

The protocol data unit MRMAC has the MRPH as described in FIG.
It is a protocol layer that exists between Y and MLLLC. FIG. 5 shows the relationship between the PHY, MAC and LLC protocol layers.

Support for transport protocols above layers of MRMAC . MR
MAC supports both fixed PDU transport protocols and variable PDU transport protocols such as ATM, Frame Relay and IP. M
RMAC provides QoS (Quality of Service) specifications suitable for ATM and IP.

Transfer Types Both asynchronous and isochronous transfers are supported. Transfer Size MRMAC supports both fixed size and variable size packet transfers. The minimum packet size is 48 octets and the maximum packet size is 2000 octets.

Throughput The throughput of MRMAC is at least 80 times the MRPHY capacity.
%. All wireless mesh traffic is sent to the WAN backbone (BA
In the special case directed to and from one wireless node connected to P), the throughput is at least 80% of the WAN link capacity.

The isochronous delay to and from the delay BAP for any node in the subnet is less than 6 ms. The asymmetric delay is less than 15 ms.

[0069] delay variation with respect to the variation isochronous transfer of the delay does not exceed the 2ms. Wireless Network Configuration MRMAC accepts wireless nodes joining and leaving a wireless mesh network without causing excessive fragmentation.

The optimized wireless nodes must be able to self-organize in a fully optimized manner without the need for external intervention. The wireless node has an interface that allows the centralized control function to further optimize its topology and routing choices.

Path Diversity Alternative physical paths available are exploited for redundancy and load balancing.

Plug and Play Installation of a new node does not require equipment adjustment by the installer.

An implementation example of the MRMAC is shown below. Radio Node Initialization Process Each neighbor node's node contains a map of all other nodes in the subnet where it may have linear (LOS) communication. The information in the following table is included.

[0074]

[Table 1]

Routes Each node has a route table that describes how to deliver a packet to any MAC destination. The information in the following table is included.

[0076]

[Table 2]

Process Overview After the initialization of the nodes and the wireless mesh network, each node has the information listed in Tables 1 and 2. A queue of packets is formed at each node from both traffic originating from the node and traffic being transmitted through the node at its local port. The node consults its table for each packet destination and determines where to send the packet transmission and when such transmission should be attempted based on the scheduled handshake and priority weighting. At the scheduled handshake time, the wireless nodes reach a mutual decision and the transmission is scheduled. The decision takes into account potential interference between the nodes. It would be advantageous to include the scheduled handshake of the neighbors in each node. This facilitates the interference avoidance procedure.

[0078] [Quantitative Discussion handshake handshake procedure, 62 octet to responders from the caller, from another of 64 octets and the caller to the caller from the responder of the final 62 octet to responders Includes sending packets. There is a two octet guard time between transmissions. The 62 octet handshake packet includes FEC and other overhead. Total time for handshake is 11.71875
(Μsec), which is 192 octets. The frequency of the handshake depends on the number of neighbors, load and other factors. Clearly, the optimal frequency is maintained to maximize throughput.

Payload Packet Size The packet size for asynchronous transfer varies between 72 and 1784 octets, including FEC and other overhead. The packet size for isochronous transfer is 64 and 25 including FEC and overhead.
It varies between 6 octets. The range of the timing is as shown in the table below.

[0080]

[Table 3]

Table 3 shows that if all packet transfers require a handshake, the average throughput through a node is well below 90% of the link. Obviously, some prior scheduling is required. On the other hand, assuming a queue of, for example, 500 μs, the throughput will be more reasonable.

Handshake Timing The handshake varies based on the size of the queue of the neighboring node. Packet forwarding carries the same type of information as the handshake process, and periodic handshakes need to be scheduled only if the use of links between neighboring nodes limits their use.

Priority Weight In order to reflect the state of the route with high accuracy, each node can maintain the priority weight of the route. A long-term monitoring process, external to the wireless mesh network, is used to further optimize the routing priority weight.

Handshake Scheduling It is preferable to have a simple scheme for handshake scheduling between nearby neighbors, and first assume that it is necessary to support variable length packets. Then you immediately face a problem. If a handshake with a nearby neighbor is pre-scheduled, the interval must allow for a maximum length of packet transfer. However, this can cause the node to waste bandwidth between handshakes for a significant amount of time, i.e., some handshake intervals do not require packet transfer. A preferred approach is to schedule the handshake at a higher frequency, which allows the node to miss the handshake. The handshake scheduling algorithm of the present invention allows a wireless node to miss its schedule with one neighbor when involved in a packet transfer with another neighbor. The affected node continues its scheduling, and the delayed node resumes its handshake scheduling once its packet transfer is complete.

FIG. 6 shows a network of 25 nodes with a maximum of 8 connectivity, each node being connected to a maximum of 8 neighboring nodes. Each node performs a handshake with one of its neighbors every τ seconds and the complete handshake cycle is 8τ
Assume that it takes seconds. To satisfy the isochronous requirement, 8τ should be 125 μ
Let it be equal to some integer multiple of seconds.

[0086]

[Table 4]

Table 4 is a table of possible handshake schedules for the network shown in FIG. The scheduling algorithm prevents a node from missing a neighbor too many times.

High-Level Node Scheduling Finite State Machine (FSM) FIG. 7 shows a table of finite state machines. Each state and transition is described below.

State A0IDLE State This is the MRMAC default state. The MRMAC remains in this state and waits for any external event or timer expiration that causes a transition to one of the other states. IdleAction () is a procedure executed when entering an idle state, and performs housekeeping work such as initialization of a timer and initialization of a table.

The schedule reservation timer of the A0 → A2 adjacent station causes this transition. This transition occurs for a reservation of a neighboring station that was previously scheduled to receive a packet from a neighboring node.

A0 → A3 The schedule reservation timer of the adjacent station causes this transition. This transition occurs for a reservation of a neighboring station that was previously scheduled to send a packet to a neighboring node.

The MLLLC causes this transition by using the Mac.Tx () procedure that requests the transmission of A0 → A0a packets. The QueuePkt () procedure results in a queue of packets according to the required priority. At this time, a preferred route is associated with the packet based on routing and status information gathered from the wireless network. This route is assumed to be the best route based on the load factors present at neighboring nodes. When a preferred route is selected, not only the priority of the packet,
The quality of the radio link with neighboring stations is taken into account. The nature of the requested transfer, i.e., isochronous or asynchronous transfer, determines the scheduling priority.

A0 → A1 The schedule reservation timer of the adjacent station causes this transition. This transition occurs for a reservation of a neighboring station that was previously scheduled for a situation exchange with a neighboring node.

State A1 HANDSHAKE State In this state, the node exchanges status with neighboring nodes. The exchange updates both nodes' perspectives on other things, including packet queuing, route availability, and future schedules. HandShakeAct
The ions () procedure is performed upon entering this state and performs any relevant housekeeping work.

A1 → A0a This transition is caused by the occurrence of a handshake process failure. The UpdateNeighborTables () procedure records this failure and may be required to schedule future handshake reservations.

These transitions are triggered by the success of the A1 → A0b and A1 → A0c handshake exchanges. The procedures ScheduleRx () and ScheduleTx () may be required to schedule future reservations for sending or receiving packets. UpdateNeighborTables () is required to record the updated neighbor status information.

State A2 RECEIVE State The node enters this state when the node is ready to receive a previously scheduled packet transmission. Step ReceiveActions (
) Is required to perform necessary housekeeping functions, such as buffer allocation and timer maintenance.

This transition is triggered by the successful reception of a scheduled packet containing an update of the A2 → A1 status. The Link.Ind () procedure is required to indicate to the MLLLC that a packet has arrived.

A2 → A0a This transition is caused by a failure to receive a scheduled packet. The UpdateNeighborTables () procedure is required to record this and possibly schedule another reservation with this neighbor.

A2 → A0b This transition is triggered by the successful reception of a scheduled packet that does not include a status update. The Link.Ind () procedure is required to indicate to the MLLLC that a packet has arrived.

A3 TRANSMIT State A node enters this state when a scheduled reservation arrives to send a packet to a neighbor. The TransmitActions () procedure is required to perform the necessary housekeeping functions such as buffer and timer management.

A3 → A1 This transition is triggered by a successful transmission including the exchange of updated status information.

This transition is caused by the failure of A3 → A0a packet transmission. UpdateNeig
The hborTables () procedure is required to log the failure and possibly schedule a later attempt to send the packet.

A3 → A0b This transition is triggered by a successful transmission that did not include the exchange of updated status information.

Admission Process Mesh wireless networks incorporate a distributed admission protocol. The purpose of the admission protocol is to not interfere with the ongoing packet transfer,
The goal is to enable wireless nodes to join the network and become efficient subscribers to mesh wireless packet forwarding.

Overview of Admission Process When a wireless node boots and discovers that it has not joined the wireless network, it transitions to an admission pending state. The wireless node refers to its configuration database stored in non-volatile RAM to determine its wireless network subscription. In the admission pending state, the wireless node waits for an admission invitation signal from a potential neighbor station. Once the wireless station receives the invitation signal, it starts exchanging status information with its first neighbor, through which it learns about the scheduling of the immediate neighbor, and thus exchanges status information with further neighbors. start. Once the wireless node has accumulated sufficient routing and scheduling knowledge through its exchange with its neighbors, it is ready for normal packet forwarding and thus transitions to the normal scheduling state.

FIG. 8 shows a sequence that occurs when an unjoined node starts up and attempts to join the network. A description of the states and transitions, including the admission finite state machine that defines the sequence, follows. The wireless node starts up by scanning the adjacent station for an admission transmission invitation signal. It then attempts to connect with the neighbor and continues learning about the neighbor by establishing a schedule of status exchanges via its first contact.

I0LISTEN This is the initial state of the admission FSM, and unsubscribed wireless nodes enter this state from the activation process. The ListenActions () procedure performs housekeeping functions such as initializing the table and preparing the process for frequency and spatial domain scanning.

The I0 → I1 scan timer causes this transition. The wireless node waits for a spatially oriented invitation at a given frequency for a sufficient amount of time. If no invitation signal is received, the next frequency and direction must be scanned.

Receipt of the I0 → I2 invitation signal causes this transition. The wireless node receives an invitation including a scheduling of a first exchange with the inviting wireless node.

An I1SCAN wireless node enters this state when its scan timer expires. It then sets the next listening frequency and spatial orientation.

When the I1 → I0 wireless node establishes its next listening parameter, it transitions back to the listening state.

An I2CONNECT wireless node enters this state upon receiving a solicitation signal from its first neighbor. Data exchange occurs from the first neighbor station to the new node.

The expiration of the I2 → I2a exchange timer causes this transition. This timer is based on the initial schedule received as part of the invitation signal.
Set by the ns () procedure. The wireless node attempts to exchange status and route information with its neighbors.

The failure of the initial exchange I2 → I0 causes this transition. The wireless node returns to the listening state.

I2 → I3 The success of the first exchange causes this transition. The wireless node updates its neighbors and route table based on the information gathered in the first switching and transition to the learning state.

An I3LEARN wireless node enters this state after successfully completing its first exchange with a neighbor. The LearnActions () procedure uses the route, status, and schedule information gathered in the first exchange to schedule subsequent exchanges with other wireless nodes in the neighborhood.

The I3 → I3a exchange timer triggers this transition. The wireless node makes an initial exchange with a potential new neighbor via some other neighbor with which it has already established a schedule. If the first exchange with the new neighbor is successful, the wireless node updates its table to reflect the neighbor's status, routing and scheduling information.

The termination of the I3 → A0 learning process causes this transition to the node scheduling FSM of FIG. The wireless node has completed the admission sequence and is now part of the wireless network and is ready to begin normal packet forwarding.

[Wireless Network as Virtual Router with Local Interface Port at Wireless Node (see FIG. 9)] FIG. 9 shows a wireless network as a virtual router with an interface at each wireless node. The wireless network shows the appearance of a router, and the details of packet transfer between wireless nodes are managed by an internal protocol of the mesh type wireless. Each of the wireless nodes has a local port, which can provide narrow beam transmission to neighboring stations having a line of sight (LOS). This is indicated by the lines between the nodes. Either the beam steering phased array antenna or the switched antenna selection scheme as described above achieves this. The fan indicates the radio coverage of the node, and the radial lines in the fan indicate the possibility of a directional antenna beam present between the radial lines.

[Features] The behavior outside the wireless network (also called “virtual router”) depends on the IETF RFC-
1812 (IP version 4 router requirement). The virtual router uses a mesh wireless media access control (MRMAC) protocol and a mesh wireless link layer control (MRLLC) protocol to provide IP between its interfaces.
Forward the packet.

[Internet ^R Architecture] As per RFC-1812 [Link Layer] As per RFC-1812 [Internet Layer-Protocol] As per RFC-1812 [Internet Layer-Transport] As per RFC-1812 [Transport Layer] RFC As per -1812 [Application Layer-Routing Protocol] As per RFC-1812 [Load Balancing] Network Internal Load Balancing MRMAC incorporates a powerful mechanism for scheduling packet transfer according to the load status of neighboring nodes as well as the service type and priority of IP.

[0124] 2. External Interface Load Balancing When multiple routes are available through different interfaces, the virtual wireless network router (VRMR) may
4 (IETF RFC-1771) to achieve policy-based routing, including load balancing between interfaces.

[Route Diversity and Self-Healing] Intra-network route diversity MRMAC incorporates a powerful mechanism for scheduling packet transfers according to the availability of radio links as well as the service type and priority of IP.

[0125] 2. External Interface Route Diversity Virtual Wireless Network Router (VRM
R) uses a standard routing protocol such as BGP-4 (IETF RFC-1771) to learn from its equivalents the availability of alternative routes.

Lastly, the wireless mesh topology network of the present invention has the following unique advantages. That is, by routing packets of data through multiple hops, a line of sight (LOS) can be achieved through a nearby adjacent radio station. Thus, the coverage range can be much higher than in a point-to-multipoint cell topology. The bit rate capacity and availability of wireless mesh networks increase as subscriber density increases.

Each connection can send more data due to the shorter distance between wireless stations. The signal-to-noise ratio received over the air is improved, and more advanced modulation techniques are used to allow faster bit rate transmission.

WAN access to multiple trunking mesh networks. An SR network node can act as a WAN access point via its local interface port, eg, via 100BaseT. This connectivity increases the capacity of the entire network by simultaneously transmitting data into and out of the network via WAN access points. (See WAN access point 1 of SR 14 and 2 of SR 10 in FIG. 2.) The access point is 100 Ba.
at least one port connected to a wireless network node via a seT interface and a WA via an OC-3 or OC-48 IP over ATM
It may include a standard Internet router with other nodes connected to the N area.

[0129] Route diversity is achieved through the use of a MAC layer that can route packets between wireless stations based on the routes available in the network.

[Brief description of the drawings]

FIG. 1 is a schematic top view of the area covered by three PMP (prior art) base stations.

FIG. 2 is a schematic top view of a broadband wireless mesh topology network according to the present invention.

FIG. 3 is a diagram showing a T that can be used as an SR (subscriber station) in the present invention
FIG. 2 is a schematic diagram of a DD (time division duplex) radio station.

FIG. 4 is a schematic diagram of a mesh-type wireless network that enables a spatial flow of information, for example, between nodes 1 and 4 and between nodes 2 and 0 simultaneously.

FIG. 5 is a schematic diagram showing the relationship between PHY, MAC and LLC protocol layers.

FIG. 6 is a schematic diagram of 25 mesh nodes.

FIG. 7 is a flowchart of a high-level node scheduling FSM implemented by a finite state machine.

FIG. 8 is a flowchart of a process of an admission FSM finite state machine of a mesh network node.

FIG. 9 is a schematic diagram of a wireless mesh network configured as a virtual router with local interface ports and dynamic spatial route selection capabilities at wireless nodes.

[Procedural Amendment] Submission of translation of Article 34 Amendment of the Patent Cooperation Treaty

[Submission date] July 10, 2000 (2000.7.10)

[Procedure amendment 1]

[Document name to be amended] Drawing

[Correction target item name] All figures

[Correction method] Change

[Correction contents]

FIG.

FIG. 2

FIG. 3

FIG. 4

FIG. 5

FIG. 6

FIG. 7

FIG. 8

FIG. 9

[Procedural Amendment] Submission of translation of Article 34 Amendment of the Patent Cooperation Treaty

[Submission date] October 16, 2000 (2000.10.16)

[Procedure amendment 1]

[Document name to be amended] Statement

[Correction target item name] Claims

[Correction method] Change

[Correction contents]

[Claims]

──────────────────────────────────────────────────続 き Continuation of front page (81) Designated country EP (AT, BE, CH, CY, DE, DK, ES, FI, FR, GB, GR, IE, IT, LU, MC, NL, PT, SE ), BR, CN, IL, JP, KR, RUF term (reference)

Claims (49)

[Claims] 1. A wireless mesh network comprising a plurality of communication nodes, each node in a line-of-sight with a corresponding subset of nodes, each node transmitting a data stream transmission to another node. A radio frequency transmitting circuit for receiving a data stream transmission from another node; a local port for connecting to a data system provided by a user; and the transmitting circuit and the receiving circuit; A data processing unit coupled to the local port; and a means for determining a flow of space coupled to the transmitting circuit, the means for transmitting a data stream only to selected ones of the subset of the corresponding nodes. The means for determining the flow of space are further coupled to the receiving circuit, and Receiving data stream transmissions only from selected nodes of the set, wherein the data processing unit extracts selected data from the received data streams to generate a second data stream and the selected to the local port Extracting means for providing the received data, the data processing unit inserting data received at the local port into the second data stream to generate a third data stream and the third data stream
Wireless mesh network further comprising insertion means for providing said data stream to said transmitting circuit for its transmission. 2. The method of claim 2, wherein at least some of the nodes include status information and scheduling information of the other nodes, and the means for determining the spatial flow is based on the status information and the scheduling information. The wireless communication network of claim 1, wherein a node is selected from the corresponding subset of nodes for reception of the communication, so that communication between the nodes can occur without collision. 3. The apparatus of claim 2, wherein at least some of the nodes further include means for periodically transmitting local node information with each other to constantly update the status information and scheduling information. Wireless communication network. 4. For some of said nodes, said data processing unit comprises means for evaluating characteristics of a data stream transmission received from a first node, and said first node with respect to said characteristics. Means for communicating with the information of the first node and, in response to receiving the information from the first node, adjusting operating parameters of the transmitting and receiving circuits to achieve a change in the characteristic of data stream transmission. Means for causing the wireless communication network to communicate. 5. The operating parameters of the transmitting circuit and the receiving circuit include: frequency channel allocation, spatial direction, code rate, modulation level, modulation rate, power level, beam bandwidth, forward error correction (FEC) scheme, and 5. The wireless communication network of claim 4, including encryption and decryption. 6. The data stream is organized as packets of data,
At least some of the nodes may be in a connectionless manner, in a connection-oriented manner, or in some of the data packets may be transmitted in a connectionless manner and other of the data packets may be transmitted in a connection-oriented manner. The wireless communication network according to claim 1, further comprising a transmission mode means for transmitting the data packet between the first and second nodes in one aspect. 7. The wireless network of claim 1, wherein some of the nodes include a media access control (MAC) layer that enables a packet switched mode, a circuit switched mode, or both modes of data stream transmission. 8. The wireless network according to claim 7, wherein the MAC layer further enables asynchronous packet transmission and isochronous packet transmission. 9. The wireless communication network according to claim 1, wherein at least some of the nodes communicate using one of an isochronous transmission mode and an asynchronous transmission mode or both of the isochronous transmission mode and the asynchronous transmission mode. 10. The wireless network according to claim 9, wherein the node transfers information using a MAC layer that maps to Internet Protocol quality of service (QoS) and enables isochronous transfer. 11. The data stream includes one or more packets of data, and at least some of the data packets of the node are selectively transmitted based on either an isochronous transmission scheme or an asynchronous transmission scheme. The wireless communication network of claim 1, wherein 12. The wireless communication network according to claim 1, wherein the data stream transmission of at least some of the nodes is based on a time division duplex. 13. The wireless communication network according to claim 1, wherein the data stream transmission of at least some of the nodes is based on a code division multiple access scheme. 14. For each node of a first subset node, the transmitting circuit operates at a first frequency and the receiving circuit operates at a second frequency, wherein each node of the second subset node The transmitting circuit operates at the second frequency and the receiving circuit operates at the first frequency, the data stream between the first subset node and the second subset node The wireless communication network according to claim 1, wherein the transmission is based on frequency division duplex. 15. The system according to claim 1, wherein one of said local ports is coupled to a local area network (LAN), a wide area network (WAN), a multimedia bus or a user provided computer system. A wireless network according to claim 1. 16. The wireless network according to claim 1, 2, 4, or 6, wherein one of the local ports is connected to an ATM switch, an IP router, an IP switch, or an Ethernet hub. 17. The transmission circuit and the reception circuit have a frequency of 0.5 GHz and 100 GHz.
7. A wireless network according to claim 1, operating at a frequency between 0 GHz. 18. The method of claim 1, 2, 4 or 6, wherein at least one of the nodes includes means for adjusting a bit rate of its data stream transmission based on contextual information about a subset of the nodes. The wireless network as described. 19. The method of claim 1, 2, 4 or 6, wherein the transmitting and receiving circuits include circuitry to adjust the operating frequency spectrum in real time, thereby reducing interference and increasing frequency reuse. Wireless network. 20. The data stream is organized into a plurality of data packets, and the spatial flow determining means determines a node from the nodes of the corresponding subset based on information included in the data packet. 7. The wireless network according to claim 1, 2, 4 or 6, wherein the selection is made, whereby the nodes are selected in real time. 21. The wireless network according to claim 1, 2, 4, or 6, wherein the local port includes a 10 / 100BaseT interface or an IEEE 1394 interface. 22. The one of the nodes, wherein one of the nodes can receive its power via a separate cable, a separate wire inside the cable, or a cable coupled to the local port. 7. The wireless network according to 4, 6 or 7. 23. The data stream includes routing information, and the spatial flow determining means selects a node from the corresponding subset node based on the routing information, so that the network serves as a virtual wireless router. 7. The wireless network according to claim 1,2,4 or 6, which works. 24. The spatial flow determining means selects a node from the nodes of the corresponding subset based on external information transmitted from another node or provided from the local port. Item 7. The wireless network according to item 1, 2, 4, or 6. 25. The wireless network of claim 24, wherein said local port is coupled to a network management system to define and optimize system parameters. 26. The spatial flow determining means is coupled to one of an array of beam steering phased array antennas and a number of sector antennas.
7. The wireless network node according to 2, 4, or 6. 27. The beam steering phased array antenna includes a one-dimensional antenna array for limiting beam steering to a single plane, wherein the array of multiple sector antennas limits beam steering to a single plane. 1 to do
27. The wireless network node according to claim 26, comprising an array of three-dimensional antennas. 28. The wireless network of claim 26, wherein the beam steering phasing array antenna includes means for rapidly changing the phase coefficient of an element including the array antenna to change its steering direction. node. 29. The wireless network node according to claim 26, wherein the plane of the emitted beam includes an orbit of at least one low earth orbit satellite. 30. The space flow judging means has a first angle and a first angle.
And a beam circuit for generating a first beam propagating in a plane of the first plane and a second beam having an angle smaller than the first angle and propagating in a plane different from the first plane. Item 7. The wireless network node according to item 1, 2, 4, or 6. 31. One of the planes of the first and second beams has at least one
31. The network node of claim 30, comprising an orbit of two low-Earth orbit satellites. 32. The transmission circuitry of at least one of the nodes includes circuitry for selecting polarization of data stream transmission.
Or a wireless network according to 6. 33. Some of said nodes include means for recording routing information over said network, and said means for determining spatial flow comprises means for determining a node from said corresponding subset node in response to said routing information. The wireless network according to claim 1, 2, 4 or 6, wherein routing tasks including load balancing and self-healing can be achieved. 34. The wireless network according to claim 1, 2, 4 or 6, wherein the data stream comprises packets of data in the form of Internet Protocol data. 35. The data stream includes packets of data in the form of Internet Protocol data, and at least some of the nodes include a wireless MAC layer for coordinating transmission of the IP data between the nodes. Item 1,
7. The wireless network according to 2, 4, or 6. 36. The wireless network of claim 1, 2, 4, or 6, wherein the data processing unit includes means for encrypting and means for decrypting the data stream. 37. Some of the nodes include means for decrypting the selected data sent to the local port, and means for encrypting the data received from the local port. 7. The wireless network of claim 1, 2, 4, or 6, further comprising: 38. At least one of said nodes includes means for establishing communication with a new node and means for exchanging information with said new node.
7. The wireless network according to claim 1, 2, 4 or 6, whereby the new node knows about the network based on information obtained from the at least one node. 39. The data processing unit of the subset node according to claim 1, 2, 4 or 6, wherein the data processing unit of the subset node includes means for receiving externally provided information to configure the subset node. Wireless network. 40. A mesh wireless network comprising a plurality of spaced wireless station nodes, each node having a transmitter, a receiver, a cross-connecting data switch and a router, all of which are interconnected. Each of the wireless nodes is of a type that is connected and handles data packets, wherein each wireless node further comprises: a) at least a first, a second, and a third port, wherein the first port is associated with a local user; The second and third ports communicate with a remote wireless node, and b) a user provided data system connected to the first port and having a communication link to the second and third ports; c) A first directional antenna element connected to the second port and directed to a first remote radio node in line-of-sight relationship; And d) having a second directional antenna element connected to the third port and directed to a second remote radio node in line-of-sight relation, the first and the second Transmitting a data packet communicated from the second port, and e) reading a data packet associated with each router, the address and the synchronization information, and extracting the associated data in the packet to obtain a first packet. Means for deciding whether to be sent to a port or forwarded to a remote wireless node via a second or third port, wherein the synchronization information is transmitted to another wireless node in a data packet exchanged between the nodes. F) forming a data packet associated with each router and including address and synchronization information associated with each router; To insert data from the first port, to be routed to a remote node via one, and to be routed to the first port or to a wireless node communicating via the second and third ports , Means for receiving data from wireless nodes communicating via the second and third ports, and g) processing means for communicating directivity and synchronization to the cross-connecting data switch. So that the data switch transfers the address for the incoming data packet from either the second or third port, on the one hand to the first port or on the other hand to either the second or third port. Linking and addressing for outgoing data packets from one of the three ports to the appropriate other one of the three ports; The data switch uses the synchronization information to avoid data packet collisions, so that each node communicates in line-of-sight with another node, which exchanges data packets for local users or to other nodes. A wireless mesh network that relays packets from and to them. 41. The apparatus of claim 40, wherein the antenna elements are part of an antenna array having a number of sector antennas or a plurality of antenna elements with a wide angle narrow sensitivity pattern. 42. The cross-connecting data switch assembly,
A plurality of smaller switches having m inputs connected to the transmitter and receiver portions of the wireless node and n outputs connected to n of the plurality of antenna elements. 42. The device according to claim 41. 43. The first and second directional antenna elements of a selected node are grouped in a sector, wherein said nodes of the sector have second and third ports linked to wireless communications. Item 41. The apparatus according to Item 40. 44. The apparatus of claim 43, wherein the nodes grouped in a sector are grouped in a time division duplex configuration. 45. The apparatus of claim 44, wherein the nodes grouped in a sector are grouped in a frequency division duplex configuration. 46. A method of wireless packet radio communication, comprising the steps of: providing a plurality of spaced wireless packet radio nodes in a mesh-type arrangement; and responding to differences in traffic flow and flow in different directions. Routing packets between nodes on a packet-by-packet basis, wherein instantaneous connections are made between various nodes to drop, receive, and forward the packets. 47. A directional antenna element further defined by providing three ports for each node and directed to a first port associated with a local user and other nodes in line-of-sight communication relationship. 47. The method of claim 46, comprising a second and third port associated with the first and second ports. 48. The method of claim 47, wherein the plurality of nodes have directional antenna elements that provide radio coverage over spatial sectors. 49. First and second addresses based on an address and schedule information.
48. The method of claim 47, further defined by routing data packets arriving from a third port to a different port.