WO2003105499A1

WO2003105499A1 - Virtual switch

Info

Publication number: WO2003105499A1
Application number: PCT/US2003/017916
Authority: WO
Inventors: Zeljko Bajic
Original assignee: Airflow Networks
Priority date: 2002-06-05
Filing date: 2003-06-05
Publication date: 2003-12-18
Also published as: US8355358B2; US7643460B2; AU2003237450A1; US7113498B2; US20030227893A1; EP1525759A4; EP1525759A1; US20070025349A1; CN1659899B; CN1659899A; US20100177677A1

Abstract

A method and apparatus for communicating between devices (802) is described. In one embodiment, the method comprises running two or more instances of a switch MAC sublayer on a switch (802) and managing the two or more instances of the switch MAC sublayer (801) as multiple logical access points inside the switch (802).

Description

VIRTUAL SWITCH

FIELD OF THE INVENTION

[0001] The present invention relates to the field of wireless communications; more particularly, the present invention relates to a switch that handles multiple groups of repeaters separately from each other.

BACKGROUND OF THE INVENTION

[0002] Figure 1 illustrates an exemplary network environment used today. Referring to Figure 1, a corporate Local Area Network (LAN) backbone 102 interfaces to a number of desktop computers 103ι-103_n and may interface to Internet 101. Corporate LAN backbone 102 may comprise a firewall 102A, corporate server 102B, and a standard Ethernet switch 102C. Ethernet switch 102C includes an interface by which desktops 103_-103ι_n are coupled to the corporate LAN backbone 102 and may access corporate sever 102B and Internet 101 (via firewall 102A).

[0003] More recently, Wireless LANs (WLANs) are being installed.

Many of the recently implemented WLANs operate according to the protocol set forth in the 802.11 Standard, particularly as more enterprises are adopting the 802.11 Standard. ISO | IEC DIS 8802.11

[0004] Figure 2 illustrates one embodiment of an 802.11 based WLAN

(LAN) system. Referring to Figure 2, the Internet or other LAN 201 is coupled to an 802.11 server 203 via firewall (FW) 202. Server 203 communicates with mobile stations in a number of 802.11 cells 206_-206_n using an access point in each of cells 206_-206_n, such as access point 204. Server 203 is coupled to access points such as access point 204, via an Ethernet connection. There is one access point for each of the 802.11 cells 206„-206n. Mobile stations in each of the 802.11 cells, such as laptops 205. and 205₂ in cell 206_, communicate wirelessly with the access points via the 802.11 protocol. The communications from mobile stations in the 802.11 cells to the access points are forwarded through to server 203 and potentially to Internet/LAN 201, while communications from Internet/LAN 201 are forwarded through server 203 to the mobile stations via the access points.

[0005] There are a number of problems associated with the current implementations of 802.11 networks. For example, in order to set up an 802.11 network such as shown in Figure 2, a site survey is required in order to determine where each of the access points are placed to ensure that the 802.11 cells provide complete coverage over a particular geographic area. This may be costly. Also, the cost of each of the access points is approximately $500.00. Generally, such a high cost is a deterrent to having a large number of access points. However, by reducing the number of access points, coverage diminishes and the 802.11 network is less effective. Furthermore, there are a number of mobility problems associated with the current 802.11 network deployments. For example, the 802.11 standard sets forth a number of solutions to handle the issue of mobility of mobile stations between the 802.11 cells. However, these schemes do not work effectively as there is no standard solution in place and users haven't indicated a desire for long-term proprietary solutions.

SUMMARY OF THE INVENTION

[0006] A method and apparatus for communicating between devices is described. In one embodiment, the method comprises running two or more instances of a switch MAC sublayer on a switch and managing the two or more instances of the switch MAC sublayer as multiple logical access points inside the switch.

BRIEF DESCRIPTION OF THE DRAWINGS

[0007] The present invention will be understood more fully from the detailed description given below and from the accompanying drawings of various embodiments of the invention, which, however, should not be taken to limit the invention to the specific embodiments, but are for explanation and understanding only. [0008] Figure 1 illustrates an exemplary network environment used today.

[0009] Figure 2 illustrates one embodiment of an 802.11 based wireless

LAN-based (LAN) system.

[0010] Figure 3 illustrates one embodiment of a network architecture.

[0011] Figure 4A is a flow diagram of one embodiment of a receiver diversity processing performed by a repeater.

[0012] Figure 4B is a flow diagram of one embodiment of a receiver diversity processing performed by a switch.

[0013] Figure 4C is a process for managing repeaters using a token- based mechanism.

[0014] Figure 4D is one embodiment of a token-based process for handling packets.

[0015] Figure 5A illustrates one technique for location tracking by RSSI.

[0016] Figure 5B is a flow diagram of one embodiment of a process for performing location tracking by a switch.

[0017] Figure 6 illustrates mobility supported by routing. There is no

Figure 6, only Figure 6A

[0018] Figure 7 illustrates one embodiment of a network system.

[0019] Figure 8 illustrates one embodiment of a protocol architecture.

[0020] Figure 9A illustrates one embodiment of a rotation tracking system.

[0021] Figure 9B illustrates one embodiment of a repeater.

[0022] Figure 10 illustrates one embodiment of a hardware architecture for a repeater.

[0023] Figure 11 is a block diagram of one embodiment of the base stand processor of a repeater.

[0024] Figure 12 is a block diagram of one embodiment of a switch.

[0025] Figure 13 is one embodiment of a distributed MAC architecture.

[0026] Figure 14 illustrates one embodiment of the switching plane. [0027] Figures 15 illustrates the communication network and exemplary data traffic process.

[0028] Figure 16 illustrates an exemplary process for transferring data traffic from a mobile station to a desktop.

[0029] Figure 17 illustrates an exemplary process for transferring data traffic between two mobile stations.

[0030] Figure 18 illustrates an exemplary process for transferring data traffic from a desktop to a mobile station.

[0031] Figure 19 is a data flow diagram of one embodiment of an association and token assignment process.

[0032] Figure 20 is a block diagram of two MAC sublayer instances in a switch.

[0033] Figure 21 is a data flow diagram of one embodiment of a re- association process.

[0034] Figure 22 is a flow diagram on one embodiment of a disassociation process.

DETAILED DESCRIPTION OF THE PRESENT INVENTION

[0035] A communication system is described. In one embodiment, the communication system comprises a mobile station having a transmitter to transmit packets wirelessly according to a protocol and multiple repeaters communicably coupled with the mobile station. Each of the plurality of repeaters receives one or more packets of the wirelessly transmitted packets from the mobile station. Each of the repeaters receives an indication of which of the wirelessly transmitted packets were received without errors by other repeaters and a received signal strength for those packets. The communication system also includes a switch coupled to the repeaters. Each of the repeaters forwards to the switch each packet of the wirelessly transmitted packets that each repeater had received at a received signal strength higher than any other repeater. [0036] In one embodiment, the repeaters are grouped and the switch handles each group of repeaters separately. Even so, if a mobile station moves to a location in which a different repeater in a different group is associated with the mobile station, any data buffered by the switch may be forwarded to the mobile device through the new repeater using a single data transfer within the switch.

[0037] In the following description, numerous details are set forth in order to provide a thorough understanding of the present invention. It will be apparent, however, to one skilled in the art, that the present invention may be practiced without these specific details. In other instances, well-known structures and devices are shown in block diagram form, rather than in detail, in order to avoid obscuring the present invention.

[0038] Some portions of the detailed descriptions which follow are presented in terms of algorithms and symbolic representations of operations on data bits within a computer memory. These algorithmic descriptions and representations are the means used by those skilled in the data processing arts to most effectively convey the substance of their work to others skilled in the art. An algorithm is here, and generally, conceived to be a self-consistent sequence of steps leading to a desired result. The steps are those requiring physical manipulations of physical quantities. Usually, though not necessarily, these quantities take the form of electrical or magnetic signals capable of being stored, transferred, combined, compared, and otherwise manipulated. It has proven convenient at times, principally for reasons of common usage, to refer to these signals as bits, values, elements, symbols, characters, terms, numbers, or the like.

[0039] It should be borne in mind, however, that all of these and similar terms are to be associated with the appropriate physical quantities and are merely convenient labels applied to these quantities. Unless specifically stated otherwise as apparent from the following discussion, it is appreciated that throughout the description, discussions utilizing terms such as "processing" or "computing" or "calculating" or "determining" or "displaying" or the like, refer to the action and processes of a computer system, or similar electronic computing device, that manipulates and transforms data represented as physical (electronic) quantities within the computer system's registers and memories into other data similarly represented as physical quantities within the computer system memories or registers or other such information storage, transmission or display devices.

[0040] The present invention also relates to apparatus for performing the operations herein. This apparatus may be specially constructed for the required purposes, or it may comprise a general purpose computer selectively activated or reconfigured by a computer program stored in the computer. Such a computer program may be stored in a computer readable storage medium, such as, but is not limited to, any type of disk including floppy disks, optical disks, CD-ROMs, and magnetic-optical disks, read-only memories (ROMs), random access memories (RAMs), EPROMs, EEPROMs, magnetic or optical cards, or any type of media suitable for storing electronic instructions, and each coupled to a computer system bus.

[0041] The algorithms and displays presented herein are not inherently related to any particular computer or other apparatus. Various general purpose systems may be used with programs in accordance with the teachings herein, or it may prove convenient to construct more specialized apparatus to perform the required method steps. The required structure for a variety of these systems will appear from the description below. In addition, the present invention is not described with reference to any particular programming language. It will be appreciated that a variety of programming languages may be used to implement the teachings of the invention as described herein. [0042] A machine-readable medium includes any mechanism for storing or transmitting information in a form readable by a machine (e.g., a computer). For example, a machine-readable medium includes read only memory ("ROM"); random access memory ("RAM"); magnetic disk storage media; optical storage media; flash memory devices; electrical, optical, acoustical or other form of propagated signals (e.g., carrier waves, infrared signals, digital signals, etc.); etc.

Exemplary Network Architecture

[0043] Figure 3 illustrates one embodiment of a network architecture.

Referring to Figure 3, a LAN backbone 102 interfaces a number of desktops 103_-103n to Internet 101. Note that the present invention does not require that a LAN backbone be included. All that is necessary is that there be a communication mechanism that is capable of receiving packets from other devices and /or sending packets to other devices.

[0044] Similar to Figure 1, LAN backbone 102 includes firewall 102A, corporate server 102B and Ethernet switch 102C. However, in contrast to Figure 1, LAN backbone 102 also includes switch 301 which interfaces to repeaters 302_-302₃. Although only three repeaters are shown, alternative embodiments may utilize any number of repeaters with a minimum of one. In one embodiment, switch 301 is coupled to repeaters 302₁-302₃ via a wired connection, such as cabling. In one embodiment, the wired connection may comprise CAT5 cabling.

[0045] Each of the repeaters 302ι-302₃ receives wireless communications from devices (e.g., mobile stations such as, for example, a mobile phone, a cellular phone, a cordless phone, a headset, a voice-enabled mobile station, a laptop computer system, a personal digital assistant, a computer-data-enabled mobile station, a speakerphone, video game controller, a DVD controller, a stereo controller, a TV controller, etc.) in the coverage areas of the repeaters. In one embodiment, these wireless communications are performed according to the 802.11 protocol. That is, each of the mobile stations in each of cells 310_- 310_n exchanges packets with the repeaters 302ι-302₃ using the 802.11 protocol. [0046] In one embodiment, switch 301 includes 802.11 MAC protocol software that allows switch 301 to communicate with repeaters 302ι-302₃. Different from the prior art, many of the 802.11 MAC functionality typically associated with the access points, as described above in the Background section, are taken out of the repeaters 302ι-302_n and centralized in switch 301. More specifically, the MAC layer is split to enable transfer of messages over wiring (e.g., CAT5 cabling). As such, repeaters 302ι-302₃ and switch 301 are interfaced at the inside the 802.11 MAC layer as described below. [0047] In one embodiment, switch 301 includes one or more Ethernet connectors (e.g., external Ethernet connector) so that a computer system, such as desktop computer system 303, or other device, has an Ethernet connection to LAN backbone 102 via switch 301. Similarly, in one embodiment, one or more of repeaters 302_-302₃ includes an Ethernet connector to enable a device (e.g., computer system, such as desktop computer system 304) to gain access, via a repeater, such as repeater 302₃, to switch 301 and the rest of the communication system. In such a case, the wiring coupling switch 301 to repeaters 302_-302₃ may combine 802.11 information including management and control (as opposed to solely data) information with traditional Ethernet packets on the same wiring (e.g., CAT5).

Distributed Receiver Diversity Approach

[0048] The network architecture described above allows for overlapping coverage between cells supported by the repeaters. This overlapping coverage allows for receiver diversity.

[0049] The packets from the mobile stations in each of the cells are broadcast and may be received by multiple repeaters. By allowing multiple repeaters to receive packets from one of the mobile stations, collisions and dropped packets may be reduced or avoided. For example, if a collision occurs or if a packet is dropped by one of the repeaters, then a particular packet can still be received by other repeaters. In this manner, the use of repeaters described herein provides for higher reliability.

[0050] In an embodiment in which mobile stations exchange packets with repeaters using the 802.11 protocol, each packet from a mobile station includes an Ethernet MAC address, which is embedded in the packet. Each packet may be received by one or more repeaters. Each repeater that receives a packet from a mobile station without errors (i.e., cleanly) determines the received signal strength of the packet in a manner well-known in the art. The received signal strength is converted into an indication, such as a received signal strength indicator (RSSI). The repeater forwards the packet, along with the RSSI. In one embodiment, the repeater encapsulates the packet into an Ethernet packet with the RSSI in a header and forwards the Ethernet packet to switch 301. In one embodiment, the RSSI is specified in a value from 1 to 127. These 128 discrete values can be mapped to dB signal strength values based on the particular implementation being used. Thus, all packets received from mobile stations by a repeater without errors are forwarded to switch 301. Switch 301 knows which repeater sent the packet(s) because it is received on its preassigned port.

[0051] In one embodiment, the fact that a particular repeater received a packet without errors is communicated to all other repeaters. In one embodiment, this is accomplished by having the repeater send each encapsulated packet and its RSSI as a broadcast packet to switch 301. This broadcast packet is similar to those broadcast packets used in Ethernet and includes a special broadcast address, which is recognized by switch 301. In another embodiment, only the header of the packet, which includes the RSSI and uniquely identifies the packet, is encapsulated and sent as a broadcast packet to the other repeaters. In this case, the data portion of the packet is not forwarded.

[0052] In response to receiving the broadcast packet with the specific broadcast address, switch 301 broadcasts the packet on all of the other ports used for communication between switch 301 and the other repeaters. [0053] In one embodiment, upon receiving a packet without error from a particular mobile station, the repeater sets a timer within which it is to receive packets received by other repeaters that are duplicates to the packet it has already received. When the timer expires, the repeater examines the RSSI of the packet it received (without error) with the RSSI values of duplicate packets received by other repeaters. Based on that information, the repeater determines if it is to send the acknowledgement packet. Thus, if the time expires without receiving a duplicate packet, the repeater sends the acknowledgement. If the timer expires and the repeater receives a duplicate packet, thereafter, it is treated as a new packet. To avoid this, the timer time out value is set to handle the worst case time delay that a repeater may face in receiving duplicate packets.

[0054] Note that switch 301 forwards each packet received from repeaters (note duplicates) to the rest of the communication system (e.g., LAN backbone, other mobile stations, the Internet, etc.). In one embodiment, this occurs after de-duplication of packets so that only one copy of each packet is forwarded.

[0055] Once the broadcast packets have been received, all the repeaters know what packets were received cleanly by the others and at what RSSI the packets were received by the other repeaters. Thereafter, each repeater selects the packet with the highest RSSI and determines the repeater that received it. In other words, each repeater performs a comparison on the received signal strength of the packets it received that were also received by one or more other repeaters. For each of the packets that a repeater receives at a power level higher than any of the other repeaters that received that packet, that repeater sends an acknowledgement back to the mobile station acknowledging that the packet was received without errors. This prevents all the repeaters that receive the packet cleanly from sending multiple acknowledgements to mobile station. [0056] In one embodiment, if two repeaters have the same receive signal strength for a packet, the repeater with the lower port number (the port number by which switch 301 is coupled to the repeater) is the repeater that is elected to send the acknowledgement to the mobile station. In this manner, only one repeater is selected to send the acknowledgement to the mobile station and, thus, the receiver diversity is handled in the network architecture in a distributed fashion. In one embodiment, to enable the repeaters to determine which is to send the acknowledgement in case of a packet received with the same received signal strength by multiple repeaters, each packet includes identification information, such as its switch port number, to enable the determination of which has the lowest port number. Note, in an alternative embodiment, the repeater with the highest port number may be the one to send the acknowledgement or other pre-assigned priority information may be used by the repeaters in such situations.

[0057] Figure 4A is a flow diagram of one embodiment of a receiver diversity process performed by a repeater. The process is performed by processing logic that may comprise hardware (circuitry, dedicated logic, etc.), software (such as is run on a general purpose computer system or a dedicated machine), or a combination of both.

[0058] Referring to Figure 4 A, processing logic initially receives a 802.11 packet (processing block 401). In response to the 802.11 packet, processing logic determines the received signal strength (e.g., RSSI) (processing block 402). In one embodiment, this processing logic comprises a hardware mechanism, such as a radio frequency (RF) device (e.g., integrated circuit (e.g., RF IC 1002 in Figure 10)) in the repeater. In such a case, the RF device sends the RSSI to a baseband processor in the repeater.

[0059] Thereafter, processing logic encapsulates 802.11 packet and RSSI in an Ethernet packet (processing block 403) and sends the Ethernet packet to the switch (processing block 404). In one embodiment, a baseband processor (e.g., baseband processor 1001 in Figure 10) performs the encapsulation and sends the Ethernet packet to the switch.

[0060] Later in time, processing logic receives one or more packets from the switch that are duplicates of the 802.11 packet. These duplicate packets are transmitted by other repeaters and encapsulated by those repeaters, along with their RSSIs (processing block 405). Processing logic in the repeater compares RSSIs for the duplicate packets (processing block 406). In one embodiment, a baseband processor (e.g., baseband processor 1001 in Figure 10) performs the comparison. If the repeater determines it received the 802.11 packet with the highest RSSI, then processing logic sends the acknowledgment packet to the mobile station (processing block 407). [0061] Figure 4B is a flow diagram of one embodiment of a receiver diversity processing performed by a switch. The process is performed by processing logic that may comprise hardware (circuitry, dedicated logic, etc.), software (such as is run on a general purpose computer system or a dedicated machine), or a combination of both.

[0062] Referring to Figure 4B, processing logic initially receives a packet from a repeater (processing block 411). In response to the packet, processing logic determines that the packet is to be sent to the other repeaters and re- broadcasts the received packet to other repeaters (processing block 412). Then processing logic sends only one copy of the packet to the rest of the network (processing block 413).

Token-based Receiver Diversity Approach

[0063] Note that the above receiver diversity procedure is particularly useful when gigabit or faster Ethernet communication exists between switch 301 and repeaters 302_-302_n. However, if such is not the case, another technique for receiver diversity may be utilized. For example, a token-based receiver diversity procedure may be used. In this case, switch 301 has a token for every mobile station on the 802.11 network and it gives the token to one of the repeaters. In other words, switch 301 pre-assigns the token before a packet is even transmitted by a mobile station. The repeater stores the token in a table that lists all mobile stations for which it has a token. The repeater with the token sends the acknowledgement packet to the mobile stations listed in the table when those mobile stations send packets that are received by the repeater. Therefore, a comparison of received signal strengths for duplicate packets is not necessary. Note that this token based mechanism, if the repeater with the token does not receive a packet cleanly, but another repeater does, that packet will be forwarded to the switch and not acknowledged to the mobile client. However, the switch moves the token before a subsequent packet is sent by mobile station. Therefore, this will only occur for one packet. [0064] In one embodiment, switch 301 includes a database with a listing of mobile stations and repeater numbers corresponding to the repeater that has been designated to acknowledge packets received from the mobile station and, thus, has the token. The table may also include additional information describing the repeater itself.

[0065] Since switch 301 receives all packets and their received signal strength, switch 301 can determine the closest repeater to a particular mobile station. If the repeater determined to be closest to the particular mobile station is different than the one previously identified as closest, then switch 301 moves the token to a new repeater, i.e. the one that is closer to the mobile station. The token may be moved on a packet-by-packet basis or every predetermined number of the packets (e.g., 10 packets, 100 packets, etc.). [0066] Switch 301 may employ a timer to indicate the time during which duplicate packets may be received in much the same manner the timer is used by the repeaters in the distributed approach described above. [0067] Figure 4C is a process for managing repeaters using a token- based mechanism. The process is performed by processing logic that may comprise hardware (circuitry, dedicated logic, etc.), software (such as is run on a general purpose computer system or a dedicated machine), or a combination of both.

[0068] Referring to Figure 4C, processing logic first determines the location of mobile stations with respect to repeaters (processing block 451). Processing logic then assigns a token for each of the mobile stations to one of the repeaters (processing block 452) and stores an indication of the repeater assigned to each mobile station (processing block 453). This information is stored in a table in memory. This table is referred to herein as an access list. In one embodiment, this table includes a listing of mobile stations and an indication of which repeater and /or switch port number is assigned to the mobile station. The table may be the same data structure used for location tracking described below. [0069] In one embodiment, the switch assigns a token by sending an

Add Token command to the repeater, which causes the repeater to add a new mobile station to its table of mobile devices that the repeater supports. This command includes the MAC address of the mobile station. [0070] Subsequently, processing logic periodically tests whether the repeater assigned the token for a particular mobile station is still the closest repeater to that mobile station (processing block 454). If so, then the processing is complete. If not, then processing logic moves the token to the closest repeater (processing block 455) and updates the table (e.g., the access list) to reflect the new repeater that is closest to the mobile station (processing block 456). Processing logic also updates the switch port to reflect the new repeater for use when sending packets to the mobile station from the switch. [0071] In one embodiment, the switch moves the token by sending a

Delete Token command to the repeater that currently has it, causing the repeater to delete the token (and assorted MAC Address) from its list of supported mobile stations, and by sending an Add Token command to the repeater that is currently closest to the mobile station. [0072] Figure 4D is one embodiment of a token-based process for handling packets. The process is performed by processing logic that may comprise hardware (circuitry, dedicated logic, etc.), software (such as is run on a general purpose computer system or a dedicated machine), or a combination of both.

[0073] Referring to Figure 4D, processing logic receives a token from the switch (processing block 470) and stores the token in a table stored in a repeater memory that indicates all the mobile stations for which the repeater has a token (processing block 471).

[0074] Subsequently, when processing logic receives a packet from mobile station (processing block 472), processing logic compares the MAC address of the 802.11 packet from the mobile station with the address in the table (processing block 473). At this time, processing logic tests whether the MAC address of a packet equals an address in the table (processing block 474). If so, processing logic provides an acknowledgment (ACK) packet to the mobile station (processing block 475). If not, processing logic ignores the packet.

[0075] Note that since all repeaters communicate the fact that they received a packet from a mobile station along with the received signal strength to switch 301, switch 301 is able to determine the coverage area of the transmission of the mobile station. In one embodiment, each packet received by the switch 301 from the repeaters terminates in a network processor in switch 301 (e.g., network processor 1206 of Figure 12), which determines the coverage area because it has access to the RSSI values. By determining the coverage area of the transmission, switch 301 is able to track the location of a particular device.

Downstream Communication Scheduling

[0076] For communications in the reverse direction (e.g., in the downstream direction), in one embodiment, the repeater transmitters are scheduled to reduce collisions. This scheduling is useful because repeaters can be close enough to interfere with one another. Because of this, switch 301 schedules the transmissions to prevent the collisions when the repeaters are actually transmitting.

[0077] For example, if a packet is destined for a particular IP address, then switch 301 performs an address translation to translate, for example, the

IP address into an Ethernet MAC address. Switch 301 uses the Ethernet MAC address to search in a location tracking database to determine which repeater is closest to the mobile station having the Ethernet MAC address. Once the repeater is identified by switch 301, then switch 301 knows the switch port on which the packet should be sent so that it is sent to the repeater listed in the location tracking database (for forwarding by the repeater to the mobile station).

[0078] Once the repeater (and the port number) has been identified, switch 301 checks whether an interference problem would be created if the packet is sent by switch 301 to the mobile station at that time. An interference problem would be created if there are other transmissions that would be occurring when the packet is forwarded onto its destination mobile station. If no interference problem would exist, switch 301 sends the packet through the identified port to the repeater most recently determined to be closest to the mobile station. However, if an interference problem would be created by sending the packet immediately, then switch 301 delays sending the packet through the identified port to the repeater most recently determined to be closest to the mobile station.

[0079] In one embodiment, to determine if an interference problem would exist if a packet is sent immediately upon determining the switch port number on which the packet is to be sent, switch 301 maintains and uses two databases. One of the databases indicates which of the repeaters interfere with each other during their transmissions. This database is examined for every downstream packet that is to be sent and switch 301 schedules the transmission of downstream packets so that repeaters that interfere with each other when they transmit at the same time do not transmit at the same time. The other database is a listing of mobile stations and the corresponding set of repeaters that last received the transmissions. If two mobile stations have overlapping sets, then it is possible for their acknowledgement packets to interfere when they simultaneously receive non-interfering data packets from different repeaters. Because mobile stations send acknowledge packets upon receiving downstream packets, there is a possibility that mobile stations will interfere with each other when sending their acknowledgement packets. Switch 301 takes this information into account during scheduling and schedules downstream packets to the mobile stations to reduce the occurrence of mobile stations interfering with other when sending acknowledgment packets. [0080] The information in these two databases may be collected by sending out test packets to the WLAN to determine which repeaters and mobile devices cause the interference described above. Location - Tracking by Received Signal Strength (RSSI)

[0081] Figure 5A illustrates one technique for location tracking by RSSI.

Referring to Figure 5 A, switch 301 obtains the RSSI for each packet received by the repeaters and may have multiple RSSI values for a packet when that packet is received by two or more different repeaters. More specifically, a mobile station communicates with two (or more) repeaters and one repeater is going to have a stronger received signal strength than the other for the same packet. Based on this information, switch 301 is able to determine that a mobile station is closer to one repeater than the other. By continually monitoring the received signal strength, switch 301 can track the movement of a mobile station with respect to the repeaters.

[0082] Figure 5B is a flow diagram of one embodiment of a process for performing location tracking by a switch. The process is performed by processing logic that may comprise hardware (circuitry, dedicated logic, etc.), software (such as is run on a general purpose computer system or a dedicated machine), or a combination of both. In one embodiment, the processing logic comprises a network processor in the switch (e.g., network processor 1206 of Figure 12).

[0083] Referring to Figure 5B, processing logic compares the RSSI for the duplicate packets received by different repeaters from a mobile station (processing block 550) and tests whether the repeater with the highest RSSI for the packet is the repeater listed as closest to the mobile station in a location tracking table (e.g., database) (processing block 551). If not, processing logic updates the table to indicate that the repeater that received the packet with the highest RSSI is the closest repeater (processing block 552). Processing logic also switches port assignment for the mobile station to the new repeater. [0084] In one embodiment, the location tracking table may include a listing of mobile stations and their individually assigned repeaters. The location tracking table may also be referred to herein as the active station list. This table may also include, or include instead of the assigned repeater, an indication of the switch port by which the switch is to communicate with the repeater assigned to each mobile station.

Mobility Supported By Routing

[0085] Figure 6A illustrates mobility supported by routing. Referring to

Figure 6A, the dotted arrow path for communication from switch 301 to mobile station 601 through repeater 302₂ is the original communication path with the network. As the mobile station 601 moves, a routing handoff occurs so that communication occurs over the solid arrowed path. In order to accomplish this handoff, switch 301 reroutes the packet to a different port. For example, if the first communication path illustrated as the dotted line arrow was on port 1, switch 301 may switch the packet to port 5, the port that associated with the communication path through repeater 302o. Thus, mobility is supported by simply moving a packet to a different port of switch 301 that is assigned to a different repeater. In such a situation, the mobility provisions of the 802.11 protocol may be ignored.

[0086] In one embodiment, switch 301 determines that a particular mobile station is closer to a different repeater (by monitoring the received signal strength of duplicate packets). As described above, switch 301 maintains a table (e.g., database, active station list, etc.) of all mobile stations in the 802.11 network and includes an indication of the repeater closest to each mobile station. Switch 301 performs port-based routing and may use the table in the same manner an IP routing table is used. Switch 301 has an Ethernet port for each repeater. When switch 301 determines that a mobile station is closer to a repeater that is different than the one listed in the database (based on the received signal strength of duplicate packets among multiple repeaters), then switch 301 updates the database. Thereafter, if a packet is received by switch 301 for that mobile station, switch 301 merely sends it out on the Ethernet port assigned to the repeater that was most recently determined to be the closest to that mobile station. Multi-Switch System

[0087] Figure 7 illustrates one embodiment of a multi-switch system.

Referring to Figure 7, the network architecture includes switches 701 and 702 are communicably coupled to server 712. In one embodiment, server 712 is part of a LAN backbone through which access to the Internet and incorporates other resources made. Alternatively, server 712 may act as an interface to another portion of the communication system. Each of switches 701 and 702 is coupled to one or more repeaters in the same manner as described above with respect to Figure 3. In still another embodiment, server 712 may exist within one of, or both, switches 701 and 702.

Protocol Architecture

[0088] Figure 8 illustrates one embodiment of a protocol architecture.

Referring to Figure 8, switch 801 is shown having a network layer 801 A and a MAC layer 801B. In one embodiment, the network layer 801A comprises a TCP/IP network layer. MAC sublayer 801B communicates with a MAC sublayer of each of repeaters 802_-802N. Thus, in contrast to the prior art in which the 802.11 MAC layer is completely within the access point, the 802.11 MAC layer is split between switch 301 and repeaters 802_-802N, and the MAC sublayer of the repeaters performs much less functionality than the MAC sublayer of the access points described above.

[0089] In one embodiment, the repeater MAC sublayer is responsible for performing portions of the 802.11 protocol including handling CSMA/CA, DIFS/EIFS interframe spacing (IFS) timing, SIFS timing and control, beacon frames (during transmit only), generating acknowledgement (of ACK) frames (during transmit only) on data packets received, such as 802.11 data frames and generating CTS (clear-to-send) frames in response to RTS (request-to-send) frames. The repeater MAC sublayer may also respond to the resetting of internal network allocation vectors (NAVs) which are embedded into (e.g., RTS and CTS frames). Each of the above repeater MAC functions may be implemented in a manner that is well-known is the art. [0090] In addition to the MAC sublayer, each of repeaters 802_-802N includes an 802.11 physical layer or other wireless physical layer. [0091] The switch MAC sublayer is responsible for handling multiple frame types during reception from the repeaters. In one embodiment, the MAC frame types the switch is capable of handling include an association request, reassociation request, probe request, ATIM, disassociation, authentication, deauthentication, PS-Pol, CTS (updates NAV in repeaters), ACK (in response to data frames), data and Null.

[0092] The switch MAC frame types that are accommodated during transmission include an association response, a reassociation response, probe response, AΗM, disassociation, deauthentication, PS-Pole, data, Null and RTS (updates NAV in repeater). It should be noted that the MAC frame types that the switch accommodates during receive and transmit are well known in the arts and part of the 802.11 standard. Each of the above switch MAC functions may be implemented in a manner that is well-known is the art [0093] Figure 10 illustrates one embodiment of a hardware architecture for a repeater. Referring to Figure 10, an RF chip 1002 receives and transmits RF transmissions using antenna 1003. In one embodiment, RF chip 1002 comprises a standard 802.11 RF chip. In one embodiment, antenna 1003 comprises a dual-diversity antenna. Communications received by RF chip 1002 are forwarded on to baseband processor 1001, which is a digital chip that is described in further detail below. Similarly, transmissions to be sent are received by RF chip 1002 from baseband processor 1001. [0094] Baseband processor 1001 is a digital chip that performs the reduced MAC functions as described above. The repeater also includes a port for coupling to switch, port 1007. Baseband processor 1001 handles communication with switch 301 using this port. In one embodiment, this port also transfers information through the port at lOOMb/s bits per second. Port 107 may also provide power to baseband processor 1001. [0095] A desktop port 1006 may be included to allow desktop or other systems to plug into the repeater. Also, in one embodiment, an LEDs 1005, such as an activity LED, power LED, and /or link LED, may be included in the repeater as well.

[0096] Figure 11 is a block diagram of one embodiment of the baseband processor of a repeater. Baseband processor 1001 includes a repeater MAC and control unit 1105 that interfaces with RF chip 1002 using a protocol. In one embodiment, the interface comprises a TCP/IP layer and an 802.11 MAC sublayer. The repeater MAC /control unit 1105 is coupled to switch 1103. In one embodiment, MAC /control unit 1105 communicates with switch 1103 using a TCP/IP layer and an 802.11 MAC sublayer tunneled inside Ethernet packets. Switch 1103 is also coupled to MAC/PHY layer unit 1104 which interfaces the baseband processor to desktop port 1006. Switch 1103 is also coupled to the activity /power /link LEDs 1005. Similarly switch 1103 is coupled to the MAC/physical layer unit 1001 that interfaces the rest of the components on baseband processor 1001 to switch port 1007 via switch 1103. Also coupled to switch port 1007 is a power distribution unit 1102. In one embodiment, power distribution unit obtains power from the CAT5 wiring and provides it to the rest of baseband processor 1001.

[0097] Figure 12 is a block diagram of one embodiment of a switch.

Referring to Figure 12, the switch includes one or more ports 1201 to repeaters 1201. Although 12 are shown, any number may be included. Ports 1201 are coupled to a switching processor 1202. In one embodiment, switching processor 1202 switches 13 ports of gigabit Ethernet and allows broadcast packets to be received on one port and broadcast on the others without involving the rest of the switch. In one embodiment, switching processor 1202 comprises a Broadcom BRCM 5633 gigabit switching processor. [0098] HyperTransport controller 1203 is coupled to switching processor

1202 and provides a gigabit ethernet interface to the rest of the switch architecture. In one embodiment, the HyperTransport controller 1203 includes a diagnostic porthole 1204 and another ethernet port 1205 for use, for example, coupled to a corporate LAN. [0099] In one embodiment, HyperTransport controller 1203 comprises a

Galaileo HyperTransport controller sold by Marvell. [00100] A network processor 1206 is coupled to HyperTransport controller 1203 and performs the majority of the functions of the switch, including the receiver diversity functions and location-tracking functions described above, with the exception of the rebroadcast of the broadcast packets received by the switch, which is handled by switching processor 1202. In one embodiment, network processor 1206 is coupled to a boot memory 1209, a DRAM 1207 and one or more LED's 1208. In one embodiment, network processor 1206 comprises a PMC-Sierra RM9000X2 sold by PMC-Sierra, boot memory 1209 comprises an MB boot flash AMD AM29LV640D boot flash memory and DRAM 1207 comprises 64MB synchronous DRAM (SDRAM). [00101] In one embodiment, the network processor 1206 includes a PCI interface to a processor 1210. Processor 1210 may host certain applications, such as, for example, firewall applications. Processor 1210 may perform these functions with the use of hard disk 1211, DRAM 1213 and console port 1211. Console port 1211 may provide access to a monitor or keyboard or other peripheral device. In one embodiment, processor 1210 comprises a pentium processor manufactured by Intel Corporation of Santa Clara, California. [00102] In one embodiment, network processor 1206 executes software instructions, which performs the 802.11 MAC layer. Network processor 1206 may also execute a wireless LAN configuration module to configure the wireless LAN network, a priority traffic administration (e.g., traffic shaping) module, a management software (e.g., Cisco IOS), a security protocol (e.g., 802.1x) module, and a VPN/firewall module. Processor 1210 executes a location tracking module to perform the location tracking. Processor 1210 may also execute one or more of the following software modules: clustering /HA, RADIUS/ DHCP, session mobility, third party applications, XML Web services, user administration software, and network management software.

An Exemplary MAC Software Architecture [00103] Figure 13 is one embodiment of a distributed MAC architecture.

The 802.11 MAC layer is distributed between the switch and a number of the repeaters connected to the switch. On one side, the MAC is terminated on the switch and on the other side the MAC is terminated on the stations. Thus, in this way, the distributed architecture is "one to many" relationship. [00104] The MAC sublayer on the repeater is engaged in performing hard real time functions related to the time synchronization (BEACON, PROBE request /response processing), receiving and transmitting 802.11 frames, including acknowledgment of the received frames. [00105] The MAC sublayer on the switch is centralized and controls multiple repeaters. In one embodiment, the MAC sublayer on the switch includes centralized management of the mobile stations and handles mobile stations in power save mode.

[00106] In one embodiment, the switch runs multiple instances of the

MAC sublayer on the switch. In this manner, the switch may support multiple, separate logical groupings of repeaters on the switch. By being able to run multiple instances of the MAC sublayer of the switch, the architecture offers very flexible configuration of the wireless communication system and allows at least the following benefits. First, tuning of the size of the RF coverage per logical grouping of repeaters. Second, the roaming of the stations is easy to control. Third, the management of mobile stations in power save mode is centralized. That is, the frames for the mobile stations in power save mode are buffered in the MAC sublayer on the switch and can be exchanged between other instances of the MAC sublayer on the same switch (between MAC instances) when the mobile station in power save mode is roaming. [00107] Referring to Figure 13, each of the units may be implemented in hardware, software, or a combination of both. Data_SAP unit 1301 exchanges messages with the LLC layer, conveying MSDUs from and to the LLC layer. Fragmentation unit 1302 performs fragmentation of outgoing MPDUs and MMPDUs. In one embodiment, since the sending of the fragmented PDU by repeater has some timing constraints, the fragmented PDUs between the switch and the repeater are transferred in one tunneling protocol message. The tunneling protocol covers this case by putting a number of fragments is in the tunneling protocol header. Power save unit 1303 performs power save device management, including TIM (Traffic Indication Map) management, in which TIM are sent to the repeaters periodically. It is a Tunneling protocol procedure. The repeaters use the updated TIM to construct a BEACON frame and buffering of unicast MPDUs for mobile stations in power save mode. In one embodiment, the switch maintains buffered unicast PDUs for all mobile stations in power save mode. Broadcasts and multicast PDUs are not buffered at the switch and are sent to the repeaters to be sent out immediately after any beacon containing a TIM element with a DTIM count field with a value of 0. Power save unit 1303 also performs PS-Poll request and response handling [00108] Routing unit 1305 routes data frames to MAC Data SAP unit 1301 and management inbound frames to management_SAP unit 1309. De- fragmentation unit 1304 performs de-fragmentation of inbound frames. Management SAP unit 1309 includes an interface to MIB unit 1308 and MLME service unit 1307. MLME services unit 1307 handles the incoming associate and re-associate frames, as well as disassociate requests, and processes authentication and de-authenticate requests and generates authentication and de-authenticate response frames.

[00109] MIB management unit 1308 performs get and set functions to get and set parameters of the repeater, and reset functions to reset all the parameters of a repeater and return the parameters to default values. The MIB variables located on the repeater are managed using a tunneling protocol. [00110] With respect to block tunneling protocol layer 1306, both MPDUs and MMPDUs frames between the switch and repeater are transferred by the tunneling protocol. In one embodiment, the 802.11 frames are encapsulated into Ethernet frames. The tunneling protocol header is placed after the fourteen bytes of the Ethernet header. This protocol transfers both data and management frames as well as special defined tunneling protocol control messages. [00111] On the repeater, transmit unit 1311 transfers frames from MAC to

PHY transmitter, generates FCS, inserts timestamps in the beacons and probe responses, performs DCF timing (SIFS, DIFS, EIFS), handles ACK, RTS, CTS, and performs a back-off procedure.

[00112] Receive unit 1312 transfers frames from PHY to MAC, receives the MPDUs from the PHY, calculating and checking the FCS value (Frames with valid FCS, length and protocol version are sent for receive filtering). Receive unit also filters valid received frames by destination address, and Bssld for group destination addresses, as well as handles ACK, CTS and RTS. Other functions include detection of duplicated unicast frames, updating the NAV using Duration/ID value from 802.11 frames, maintenance of the channel state based on both physical and virtual carrier sense, time slot reference generation, and providing Busy, Idle & Slot signals to Transmission. [00113] Synchronization unit 1313 processes the MLME start request in which it starts a new BSS and set all parameters for BEACON frame. Synchronization unit 1313 generates Beacon frames periodically and handles Probe request and response frames.

[00114] Repeater management unit 1314 relays all MIB set/get requests, start requests, reset requests, request /confirm characteristic commands to a proper block on the repeater.

[00115] With respect to block tunneling protocol 1 layer 1310, both

MPDUs and MMPDUs frames between the switch and repeater are transferred by the tunneling protocol. The frames are encapsulated into the Ethernet frames and the tunneling protocol header is placed after the fourteen bytes of the Ethernet header. This protocol transfers both data and management frames as well as special defined tunneling protocol control messages.

An Exemplary Switch Software Architecture

[00116] The switch contains the switching and management planes.

Figure 14 illustrates one embodiment of the switching plane. Referring to Figure 14, the switching plane contains the switch MAC sublayer (i.e., the upper MAC), a switch management entity (SwME) and a switching layer. The switching layer interfaces with the Ethernet drivers and performs switching function. The Ethernet drivers are connected to the 10/100 BT ports of the switch (PORT1 to PORT24 ) or connected to another Ethernet switch with its uplink connected to the Gigabit interface on the switch. The simulator may also be connected to the any of these ports. In one embodiment, in order to support this kind of abstraction, the tunneling protocol header contains the number of the Ethernet port handling the repeater.

Data Traffic Procedures

[00117] Figures 15-18 illustrate the communication network and exemplary data traffic process. Referring to Figure 15, switch 1501 is shown coupled to router 1502 and repeaters 1-3, via ports 1-3. Stations (ST A) 1-4 are mobile stations that communicate wirelessly with the repeaters 1-3. Router 1502 is also shown coupled to computer system 1503.

[00118] Figure 16 illustrates an exemplary process for transferring data traffic from a mobile station to a desktop. Referring to Figure 16, repeater 1602 receives the one or more 802.11 data frames (packets) and encapsulates each received 802.11 data frame into one or more Ethernet packets, adding an Ethernet frame header and a tunneling protocol header to each Ethernet packet. Thereafter, repeater 1602 sends the Ethernet frames (packets) to switch MAC sublayer 1603 on the switch. At the switch, switch MAC sublayer 1603 processes the Ethernet frames by stripping off the 802.11 MAC header and tunneling protocol headers and switches Ether frames (packets) with encapsulated IP packets to the proper switch port. Switch MAC sublayer 1603 sends the Ethernet frames (packets) to router 1604 (backbone). Router 1604 routes each Ethernet frame to a destination, such as, for example, computer system 605.

[00119] Figure 17 illustrates an exemplary process for transferring data traffic between two mobile stations. In this case, the destination address is another mobile station address and the switch MAC sublayer processes both the 802.11 and tunneling protocol headers and switches the packet to the proper port. Referring to Figure 17, a first station, station 1701, sends 802.11 data frames to a first repeater, repeater 1702. Repeater 1702 receives the 802.11 data frame and encapsulates the 802.11 frames into Ethernet frames, including adding an Ethernet frame header and tunneling protocol header to each 802.11 frame. Repeater 1702 sends the encapsulated 802.11 data frames to switch MAC sublayer 1703. Switch MAC sublayer 1703 processes the 802.11 and tunneling headers and switches Ethernet frames to the repeater (repeater 1704 in this example) handling the destination station (station 1705 in this example). Switch MAC sublayer 1703 encapsulates the 802.11 data frames into Ethernet frames and sends them to repeater 1704. Repeater 1704 receives the encapsulated 802.11 data frames and sends the 802.11 data frames to station 1705.

[00120] Figure 18 illustrates an exemplary process for transferring data traffic from a desktop to a mobile station. Referring to Figure 18, computer system 1806 encapsulates IP packets into Ethernet frames. For the first IP packet destined to a mobile station, the router starts an ARP procedure in order to obtain the corresponding MAC address. Router 1805 sends an ARP request to switch MAC sublayer 1804 to request the MAC for this IP broadcast. Switch MAC sublayer 1804 encapsulates the ARP request into an 802.11 packet and then encapsulates this packet into an Ethernet packet, essentially creating a new Ethernet frame with an embedded 802.11 MAC header and tunneling protocol header. Switch MAC sublayer 1804 broadcasts this packet to all repeaters, repeaters 1802-1803 in this example. The mobile station, station 1801, with the IP address contained in the ARP request sends an ARP response with its MAC address. Repeater 1802 receives the ARP response and encapsulates the 802.11 frames into Ethernet frames, adding an Ethernet frame header and tunneling protocol header. Repeater 1802 sends the encapsulated ARP response to switch MAC sublayer 1804, which strips off the 802.11 MAC header and switches the Ethernet frame with encapsulated ARP response packet to the backbone port. [00121] After this procedure, the router takes the station MAC address from the ARP response and routes all IP packets for this mobile station as described above. Since the switch MAC sublayer has the configuration information about MAC and IP addresses, the ARP response could come from the MAC.

Management Procedures

[00122] There are a number of management procedures supported by the distributed MAC Architecture. In one embodiment, these include starting up the switch, resetting the MAC, starting a new BSS, synchronization, authentication, and de-authentication, association, disassociation and reassociation.

[00123] With respect to starting up the switch, the switch is started by the switch management entity (SwME). To configure and start the switch and the repeaters, the SwME issues commands to the switch MAC sublayer on the switch. The commands intended for the repeaters are transferred using the tunneling protocol. Layers of the tunneling protocol are running on the switch and the repeaters.

[00124] With respect to MAC reset, the switch and repeaters cooperate to perform a reset of the MAC. Since the MAC is distributed between the switch and repeaters, the reset process is modified to support this architecture. In one embodiment, the switch management entity sends a reset request to each of the repeaters as part of a tunneling protocol process and receives a reset response indicating if the reset was successful. The reset process may set the MAC to initial conditions, clearing all internal variables to the default values. MIB attributes may be reset to their implementation-dependent default values. [00125] With respect to the start process, the switch management entity requests that the MAC entity start a new BSS. The switch management entity generates the request to start an infrastructure BSS (with the MAC entity acting as an access point) and sends it to all MAC entities where the switch is acting as a multiple access point. Each repeater responds with an indication as to whether the start process was successful.

[00126] With respect to synchronization, the synchronization process determines the characteristics of the available BSSs and allows for synchronizing the timing of a mobile station with a specified BSS (switch MAC entity). In one embodiment, the synchronization process begins with an instance of the switch MAC sublayer generating a beacon frame, which is encapsulated and sent to the repeaters periodically. The repeater updates the timestamp of the beacon frame before sending the beacon frame in the air. Based on the beacon frame, the mobile station synchronizes its timers. [00127] The switch management entity also causes authentication to establish a relationship between a station MAC sublayer and the instances of the switch MAC sublayers. In one embodiment, a mobile station is authenticated if its MAC address is in the access list on the switch. Similarly, de-authentication is supported to invalidate an authentication relationship with a switch MAC entity. In one embodiment, de-authentication is initiated by the mobile station. In this case, the instance of the switch MAC sublayer on the switch associated with the repeater assigned to the mobile station updates the station state as maintained by the switch. The results of de-authentication is that the state of the mobile station is listed in the switch as unauthenticated and unassociated.

Association

[00128] Data frames for a mobile station re forwarded from the repeater that has the token for the mobile station. If a repeater without the token receives the data frames, it forwards only a short frame with the RSSI (in the tunneling protocol header) to the switch. The switch keeps track of the RSSI for the mobile station. If the repeater without the token has better reception and if the repeater with the token has "high" error ra te, the switch has to re-assign the token. The RSSI and token are part of the tunneling protocol header. Token re- assignment is a part of the Tunneling protocol. The token assignment occurs within the association process.

[00129] Figure 19 is a data flow diagram of one embodiment of an association and token assignment process. Referring to Figure 19, an association request is generated by a mobile station and sent by the mobile station, via the mobile station MAC. Repeater 2 has the token for the mobile station. Therefore, repeater 2 encapsulates the association request, along with is RSSI and BSSID, into an Ethernet packet and sends the encapsulated packet to the switch. Repeater 1, which does not have the token for the mobile station, forwards a short frame with the RSSI in the tunneling protocol header. [00130] The switch takes the RSSIs for the two identical frames and determines which one is stronger. Based on which is stronger, the switch either allows the repeater that has the token and station MAC for the mobile station to keep them (e.g., repeater 2) or reassigns them to the repeater with the higher RSSI (e.g., repeater 1). In either case, the switch sends an association response encapsulated in an Ethernet packet with the token and association ID to the repeater, which de-encapsulates it and forwards it to the mobile station, via the mobile station MAC.

Re-association

[00131] The following procedure describes how a mobile station becomes re-associated with another switch MAC entity (logical access point). Figure 20 is a block diagram of two MAC sublayer instances in a switch. Referring to Figure 20, two (or more) instances of the switch MAC sublayer run on the switch (offering the access points (APs) inside the same switch). Each instance has its own BSSID (the MAC address of the MAC instance). Both MAC instances are managed by the same switch management entity (SwME). The SwME manages these as multiple access points (APs) inside the switch. In one embodiment, communication between MAC instances is through the SwME. Both MAC instances as well as the switch management entity (SwME) reside on the same switch. Communication between the MAC instances can be direct or through the SwME. In one embodiment, the SwME has knowledge of all MAC instances and is involved in this communication. Thus, the switch acts as a distribution system containing multiple switch MAC sublayer instances (multiple logical access points) in which roaming is centralized in the switch. [00132] In one embodiment, the association request from the mobile station is encapsulated and sent by the repeater to the switch. The association request with the BSSID of the first MAC sublayer instance is sent from the second MAC sublayer instance through the SwME to the first MAC sublayer instance. As a result, the first MAC sublayer instance generates a response representing that mobile station has been already associated with the first MAC sublayer instance. Using this process, the station does not have to go again through authentication procedure and it can be automatically associated with the second MAC sublayer instance. When the second MAC sublayer instance receives the response, it associates the station. Thus, when the station roamed, the handover procedure is performed in the switch. Therefore, the switch acts as a complete distribution system with multiple logical access points.

[00133] As described above, when a station roams between two MAC sublayer instances (logical access points) inside one distribution system, there is only one repeater controlled by one MAC sublayer instance. In one embodiment, a mobile station can roam from one repeater to another repeater controlled by the same MAC sublayer instance (logical access point) without a need to associate again, and only the token re-assignment procedure described herein has to be performed. In one embodiment, the station is not aware of the token re-assignment procedure.

[00134] If a mobile station moves from one repeater belonging to one logical access point (one MAC sublayer instance) to a second repeater belonging to a second logical access point (second MAC sublayer instance), the station has to be re-associated and the token re-assignment procedure has to be performed. The handover procedure is performed in the switch. Again, the station is not aware of any token assignment procedures. Note that mobile stations are associated with switch MAC sublayers instances not with a repeater. If a station is controlled by a repeater, the repeater has a token for that station. All repeaters controlled by a particular MAC sublayer instance are associated with a station if the station is associated with that MAC sublayer instance, and only one repeater has a token for that station.

[00135] A user can configure the switch to have any number of MAC instances. This may be configured using a parameter. Also configurable is which repeater belongs to MAC instance. For example, if the switch has 64 ports, it can be configured to act as 8 access points (8 upper MAC instances running concurrently), and 8 repeaters per access point (one upper MAC sublayer controlling 8 repeaters).

[00136] Figure 21 is a data flow diagram of one embodiment of a reassociation process. Referring to Figure 21, a mobile station SME generates a re-association request and sends it to a repeater, repeater 4 in this case, along with its BSSID via the mobile station MAC. It knows that it needs to make a reassociation request because it has received a BEACON frame with different BSSID (i.e., a different MAC instance), meaning that it roamed. The repeater receives the re-association request, encapsulates the packets of the reassociation request with the RSSI into an Ethernet packet, and sends the Ethernet packet to the instance of the switch MAC sublayer associated with the repeater. In response thereto, the instance of the switch MAC sublayer generates an indication to the switch management entity indicating that a reassociation request has been made.

In response to the indication, the switch management entity causes a new AID (association id) to be assigned to the mobile station, a token for the mobile station to be assigned to a new repeater, and the previous token assignment to be deleted. In one embodiment, the association identifier (AID) is a number (value between 0 and 2007) assigned to a mobile station by the switch or an access point during the association procedure. It is a 802.11 standard defined parameter. After the station is associated, it will insert the AID in every message. More specifically, the switch management entity updates the entry for the mobile station in the access list, including setting the new access point address to the address of the instance of the switch MAC sublayer associated with the repeater. The switch management entity also assigns a token and an association ID.

[00137] The switch management entity sends a delete token command to the instance of the switch MAC layer associated with the repeater previously assigned to the mobile station, which the instance of the switch MAC layer forwards to the repeater (repeater 3 in this case).

[00138] The instance of the switch MAC sublayer (upper MAC 2 in this case) associated with the repeater that forwarded the re-associate request (repeater 4 in this case) sends a re-associate response frame to the repeater with the token, association ID, and an indication that the re-association was successful. The repeater de-encapsulates the packet, keeps the mobile station MAC token, and forwards the de-encapsulated re-associate response frame to the mobile station MAC with the association ID and the successful indication.

Disassociation

[00139] A mobile station may request disassociation with a specified peer

MAC entity that is acting as an access point. The mobile station may request this due to inactivity, because a switch is unable to handle all currently associated mobile stations, etc.

[00140] Figure 22 is a flow diagram on one embodiment of a disassociation process. This can happen if the station wants to disassociate for the following reason - "Disassociated due to inactivity".

[00141] Referring to Figure 22, a disassociation request is generated by the SME on the mobile station and sent by the mobile station MAC as a disassociate request frame with the BSSID (i.e., the instance identifiers). The

BSSID is a basic service set identifier representing the MAC address of an upper MAC instance. Each repeaters that receives the disassociate request frame without errors encapsulates i t with its RSSI and forwards it to the switch, regardless of whether it has the token for the mobile station. In response to the receiving the disassociate request frame, the switch MAC determines whether the mobile station is in the access list and changes the state of the mobile station in the access list to authenticated and unassociated, removes all parameters from the access list entry for the mobile station, and deletes the token and association ID. In one embodiment, the access list is dynamically created hash table containing a records for all authenticated stations, in which each record contains a station MAC address, association identifier, BSSID, a station state, and a repeater port number which has station token. In other words, on the switch MAC, the state of the mobile station is updated and its AID is deleted. The switch then sends a disassociate response frame encapsulated in an Ethernet frame to the repeater having the token. Embedded in the tunneling protocol header of the frame is a tunneling protocol command to delete the token, which causes the repeater having the token to delete the token. Thereafter, the repeater that deleted the token sends the de-encapsulated disassociate response frame to the MAC of the mobile station with an indication that disassociation was successful.

[00142] In one embodiment, this process can be initiated by the switch management entity. This can happen if the switch decides to disassociate the mobile station because of inactivity or because a switch is unable to handle all currently associated mobile stations.

[00143] Whereas many alterations and modifications of the present invention will no doubt become apparent to a person of ordinary skill in the art after having read the foregoing description, it is to be understood that any particular embodiment shown and described by way of illustration is in no way intended to be considered limiting. Therefore, references to details of various embodiments are not intended to limit the scope of the claims which in themselves recite only those features regarded as essential to the invention.

Claims

CLAIMSI claim:

1. A method comprising: running two or more instances of a switch MAC sublayer on a switch; and managing the two or more instances of the switch MAC sublayer as multiple logical access points inside the switch.

2. The method defined in Claim 1 further comprising re- associating a mobile station with a first instance of the switch MAC sublayer after the mobile station moves from a first repeater belonging to a second instance of the switch MAC sublayer to a second repeater belonging to the first instance of the switch MAC sublayer.

3. The method defined in Claim 2 further comprising sending an association request from the first instance of the switch MAC sublayer to the second instance of the switch MAC sublayer.

4. The method defined in Claim 3 wherein the association request is sent through an entity on the switch.

5. The method defined in Claim 4 wherein the entity comprises a management entity.

6. The method defined in Claim 2 further comprising performing token reassignment.

7. The method defined in Claim 1 further comprising communicating between at least two of the two or more instances of the switch MAC sublayer.

8. The method defined in Claim 7 wherein communication is directly between the at least two of the two or more instances of the switch MAC sublayer.

9. The method defined in Claim 7 wherein communication is through a management entity on the switch.

10. The method defined in Claim 1 wherein at least one of the two or more instances of the switch are associated and communicate with a repeater.

11. An apparatus comprising: means for running two or more instances of a switch MAC sublayer on a switch; and means for managing the two or more instances of the switch MAC sublayer as multiple logical access points inside the switch.

12. The apparatus defined in Claim 11 further comprising means for re-associating a mobile station with a first instance of the switch MAC sublayer after the mobile station moves from a first repeater belonging to a second instance of the switch MAC sublayer to a second repeater belonging to the first instance of the switch MAC sublayer.

13. The apparatus defined in Claim 12 further comprising means for sending an association request from the first instance of the switch MAC sublayer to the second instance of the switch MAC sublayer.

14. The apparatus defined in Claim 13 wherein the association request is sent through an entity on the switch.

15. The apparatus defined in Claim 14 wherein the entity comprises a management entity.

16. The apparatus defined in Claim 12 further comprising means for performing token reassignment.

17. The apparatus defined in Claim 11 further comprising means for communicating between at least two of the two or more instances of the switch MAC sublayer.

18. The apparatus defined in Claim 17 wherein communication is directly between the at least two of the two or more instances of the switch MAC sublayer.

19. The apparatus defined in Claim 17 wherein communication is through a management entity on the switch.

20. The apparatus defined in Claim 11 wherein at least one of the two or more instances of the switch are associated and communicate with a repeater.

21. A method comprising: a switch acting as a distribution system containing a plurality of switch wireless MAC sublayer instances.

22. An architecture comprising: a first repeater having a first instance of repeater MAC sublayer functionality; a second repeater having a second instance of repeater MAC sublayer functionality; and a switch coupled to the first and second repeaters and having first and second instances of switch MAC sublayer functionality, wherein the first instance of repeater MAC sublayer functionality and the first instance of switch MAC sublayer functionality cooperate to execute a MAC layer in a distributed fashion and the second instance of repeater MAC sublayer functionality and the second instance of switch MAC sublayer functionality cooperate to execute a MAC layer in a distributed fashion.

23. The architecture defined in Claim 22 wherein the switch further comprises an entity to manage the first and second instances of the switch MAC sublayer.

24. The architecture defined in Claim 23 wherein communication between the first and second instances of the switch MAC sublayer occurs through the entity.

25. The architecture defined in Claim 24 wherein the entity comprises a management entity for the switch.

26. The architecture defined in Claim 22 wherein the first and second repeaters are coupled to first and second ports on the switch, the first and second ports being different.

27. The architecture defined in Claim 26 wherein the first and second repeaters are coupled to first and second ports on the switch via a wired connection.

28. The architecture defined in Claim 27 wherein the wired connection comprises an Ethernet connection.

29. An architecture comprising: at least one mobile station to transmit and receive packets; a plurality of repeaters to receive packets from the at least one mobile device, each of the plurality of repeaters having a repeater MAC sublayer; and a switch communicably coupled to the one or more repeaters, the switch having a configurable number of instances of a switch MAC sublayer and configuring at least two repeaters of the plurality of repeaters to cooperate with different instances of the switch MAC sublayer.

30. A method comprising: a mobile station, associated with a first repeater controlled by a first MAC instance, sending a request to a second repeater, controlled by a second MAC instance, to be associated with the second MAC instance; the second repeater sending the request to a switch MAC sublayer; the switch assigning a new association identifier to the mobile station to associate the mobile station with the second instance; the switch assigning the token for the mobile station to the new repeater; and the switch deleting the token assignment to the first repeater.

31. The method defined in Claim 30 further comprising the switch disassociating the mobile station from the first MAC instance.

32. The method defined in Claim 30 further comprising sending a delete token command to the first repeater.

33. The method defined in Claim 32 wherein sending a delete token command to the first repeater is performed by the switch.

34. The method defined in Claim 31 further comprising updating a list used by the switch to identify associations between MAC instances and mobile stations to reflect that the second repeater has a token for the mobile station.

35. The method defined in Claim 31 further comprising the switch causing packets buffered for the mobile station by an instance of a switch MAC sublayer associated with the first repeater to be forwarded to an instance of the switch MAC sublayer associated with the second repeater for forwarding to the mobile station via the second repeater.

36. The method defined in Claim 31 wherein the switch forwards the buffered packets through a management entity.

37. The method defined in Claim 31 further comprising the second repeater identifying itself to the switch using an identifier in a packet.

38. The method defined in Claim 37 wherein the identifier is a tunneling ID.

39. An apparatus comprising: a switch; a first repeater; a second repeater; a mobile station, associated with the first repeater controlled by a first MAC instance, the mobile station including means for sending a request to the second repeater, controlled by a second MAC instance, to be associated with the second MAC instance; the second repeater having means for sending the request to a switch MAC sublayer; the switch having: means for assigning a new association identifier to the mobile station to associate the mobile station with the second instance, means for assigning the token for the mobile station to the new repeater, and means for deleting the token assignment to the first repeater.

40. The apparatus defined in Claim 39 further comprising means for disassociating the mobile station from the first MAC instance.

41. The apparatus defined in Claim 39 further comprising means for sending a delete token command to the first repeater.

42. The apparatus defined in Claim 41 wherein means for sending a delete token command to the first repeater is performed by the switch.

43. The apparatus defined in Claim 40 further comprising means for updating a list used by the switch to identify associations between MAC instances and mobile stations to reflect that the second repeater has a token for the mobile station.

44. The apparatus defined in Claim 40 further comprising means for causing packets buffered for the mobile station by an instance of a switch MAC sublayer associated with the first repeater to be forwarded to an instance of the switch MAC sublayer associated with the second repeater for forwarding to the mobile station via the second repeater.

45. The apparatus defined in Claim 40 wherein the switch forwards the buffered packets through a management entity.

46. The apparatus defined in Claim 40 further comprising the second repeater means for identifying itself to the switch using an identifier in a packet.

47. The apparatus defined in Claim 46 wherein the identifier is a tunneling ID.