Wireless channel allocation in a base station processor

The use of wireless network infrastructure is increasing, which allows computer devices to communicate with wired networks such as the Internet via wireless media. In wireless data networks, multiple local computer devices, such as PCs (personal computers), are supported via wireless subscriber access The subscriber access unit provides a wireless wireless link to the base station processor. The base station processor is also connected to an internet gateway that provides a connection to a wired network. Similar to the cellular telephone network, the base station processor allocates multiple Radio channels on demand to allow transmission and reception of messages to and from subscriber units. Radio channels are assigned to messages sent and received to and from subscriber units on behalf of local computer equipment.

For a typical base station processor, a radio channel shared with a subscriber unit is a scarce resource. Messages are often queued until a channel is available. In addition, wired networks generally use a variety of methods to detect the speed at which These schemes alleviate congestion by slowing down the transmission of messages, thereby reducing throughput and avoiding overloading the recipient. Such a scheme interprets the message queue in the base station processor as congestion in the wired network, resulting in reduced. throughput. In particular, the protocols used in wired networks are not suitable for efficient communication over wireless connections.

TCP / IP networks use congestion control schemes such as slow start, congestion avoidance, fast retransmission, and fast recovery. According to the slow start method specified in Internet RFC 2581, an alarm message (ack) is regarded as a reply message to each message If ack is not received in a timely manner, the number of additional messages sent will decrease, reducing throughput. This does not mean that the message queue in the base station processor represents congestion in the base station processor, but rather the propagation delay inherent in the wireless network. However, this propagation delay is interpreted as congestion by wired line protocols such as TCP / IP.

<p> Therefore, by anticipating the arrival of reply messages and reserving available channels to transmit messages through base station processors, wireless network throughput can be reduced to that of wired network protocols such as slow start. It is beneficial to provide methods and devices that are not degraded. by the congestion control function. </p>

<p> The present invention provides a system and method for allocating radio channels in a radio communication system to support message transmission between a subscriber and a base station processor. The latency is determined by the timing of the expected reply message from the response node to the outgoing message. The waiting time manager in the base station processor calculates the waiting time and stores the calculated value in the allocation table. The scheduler reserves one channel available at the end of the latency indicated in the quota table. prior to receiving the reply message. Just before the end of the wait, a reply message is received and the scheduler allocates the channel specified in the allocation table. Send and receive reply messages with applicable subscribers using the reserved channel.</p><p> The latency manager calculates the latency using various transmission parameters specified by the wired network protocol. For example, in a TCP / IP network, the transmission parameters used to calculate latency include window size, space available in the window, average message size Includes, message type, number of messages received in the session, maximum number of unprocessed acks, and other transmission parameters. </p><p> In drawings, the same reference numerals refer to the same parts in different drawings. The invention shown in the accompanying drawings. In drawings, the same reference numerals refer to the same parts in different drawings. The Drawings Are Not Necessarily In Size And The Emphasis Is On Showing The Principles Of The Invention. </p>

<figref num="1"> FIG. 6 Is A Block Diagram Of A Communication System Suitable For Performing Radio Channel Allocation As Defined Herein. </figref><figref num="2"> Indicates A Base Station processor communicating with multiple subscriber access units. </figref><figref num="3a"> The message transmission in the system of Fig. 2 is shown. </figref><figref num="3b">The Channel Allocation Table Corresponding To The Message In Figure 3A Is Shown. </figref><figref num="4"> The Flow Chart Of The Channel Allocation Defined In This Specification Is Shown. </figref><figref num="5a"> Figure 1 Shows The Retrieval Of A Web Page Using The System. </figref><figref num="5b"> The Channel Allocation Table Corresponding To The Message In Figure 5A Is Shown. </figref><figref num="5c"> The Timing Chart Corresponding To The Channel Allocation Table In Figure 5B Is Shown. </figref><figref num="6"> Shown Is A Subscriber Profile Table For The Channel Allocations Defined Herein. </figref>

FIG. 1 is a block diagram of a communication system 10 that performs channel allocation in a wireless network as defined herein. The communication system includes a local computer device such as PC 12, a subscriber access unit 14, a base station processor 16, and an Internet gateway 18. The PC 12 communicates with the subscriber unit 14 via a wired connection 20. The subscriber unit 14 communicates with the base station processor 16 via the wireless connection 26. The base station processor communicates with the Internet gateway 18 via a wired link 24. The Internet gateway 18 is adapted for communication over public access networks such as the Internet.

Therefore, the PC 12 can access the network server 18 (any remote entity located on the Internet or other networks) via a combination of wired lines 20, 24 and wireless connections 26 provided ... Wired connections 20 and 24 are generally supported by protocols Wireless Connection 26 is a protocol described in the pending US patent application "Dynamic Frame Sizing for Multi-Channel Transmission," issued September 2, 1999 as PCT Application No. WO 99/44341. Supported by such protocols. Generally, the PC 12 provides the Internet Protocol (IP) packet to the subscriber 14 via a wired connection 20 (eg, an Ethernet® connection). The subscriber 14 removes the framing of the IP packet and transfers the data in the IP packet to the base station processor 16 via the wireless connection 26 according to the wireless link protocol. The base station processor 16 extracts wireless connection frames and sends the frames in IP packet format to the Internet gateway 18 via the wired connection 24. Therefore, the subscriber 14 and the base station processor 16 are considered to be the "end points" of the radio connection 26.

FIG. 2 shows the base station processor 16 in detail. The base station processor 16 communicates with a plurality of subscribers 14a to 14d. An additional subscriber unit 14 (x) may be provided. The subscriber communicates with the base station processor via radio channels 22a-22j. An additional channel 22 (x) may be provided. As mentioned above, channel 22 is used to send and receive messages to and from subscriber 14. The scheduler 28 allocates channel 22 on demand and assigns available channels to messages transmitted between subscriber 14 and base station processor 16.

Although channel 22 is unidirectional between subscriber 14 and switch 16, multiple channels can be assigned to messages originating from or addressed to specific subscriber 14. In the illustrated example, channel 22a is assigned to carry a message from base station processor 16 to subscriber 14b. Further, the channel 22b is assigned and the message from the subscriber 14c is received by the base station processor 16, while the channel 22c is assigned and the message is transmitted to the subscriber 14c. In addition, channels 22d and 22e are assigned to send messages to subscriber 14d, while channels 22f are assigned to receive messages from subscriber 14d. As mentioned above, the scheduler 28 generally assigns a channel to a subscriber immediately and responds to a channel request for sending and receiving a message with the subscriber 14.

Invoke message traffic on one channel using two dedicated channels that are common to all subscribers 14 .Subscriber 14 uses common access channel 30 to request a channel from base station processor 16. The common paging channel 32 is used to notify subscriber 14 that a channel is assigned. The subscriber 14 then sends the message to the PC 12 or the base station processor 16 depending on the direction in which the message is transmitted.

The base station processor 16 also includes a time control manager 34 that determines the latency delay and an allocation table 36, both of which are described in detail below. As mentioned above, in general message transmission, there is a large amount of latency delay between For example, radio propagation delay occurs in message transmission from base station 16 to subscriber 14 (FIG. 1). Network propagation delays occur when messages are transmitted over the Internet or other public access networks. There are other latency delays, which will be described below. Protocols such as TCP / IP usually expect a reply message, generally ack, to be returned in response to a message transmitted to the response node. According to the present invention as defined herein, Channel allocation means that a channel is assigned to a message transmitted from a sender in any transmission direction. For example, the reply message is sent back to the sender by the node that receives the message. Therefore, the channel allocation reservation for the reply message is made in advance when the message is sent and received to and from the subscriber 14. the channel allocation reservation for the reply message is made in advance when the message is sent and received to and from the subscriber 14. the channel allocation reservation for the reply message is made in advance when the message is sent and received to and from the subscriber 14.

Figures 3a and 3b show the detailed configuration of base station processor 16 including latency manager 34, allocation table 36, and scheduler 28. The latency manager 34 is a process that calculates the latency delay associated with the reply message sent by the response node 40. The allocation table 36 shows the entries 38a, 38b for each of the allocated channels 22b, 22c, and the associated latency T. ₀ And T ₀ + T _L It is a memory structure that stores and. The scheduler 28 reads the allocation table 36 and the latency information to determine the channel allocation for the expected message.

In general message transmission, the PC 12 sends a connection request message to the response node 40, as indicated by arrow 42. Message 42 is time T ₀ Will be sent at. Therefore, in the allocation table 36, the time T ₀ Entry 38a for channel 22b assigned to subscriber 14c is written in. When message 42 is received via channel 22b, latency manager 34 examines message 42. The latency manager 34 determines that the type of message 42 is a TCP / IP connection request, Predicting An Ack Thereby As A Reply Message.

The Latency Manager 34 Determines The Latency Elapsed Before The Reply Message Is Received By The Base Station Processor 16. For Example, The Wait Time Manager 34 Is DerutaT ₁Determine that the ISP (Internet Service Provider) delay 44, indicated by, will occur between Internet Gateway 18 and Internet 50. In addition, ΔT ₂ The network propagation delay 46, indicated by, occurs when message 42 is transmitted over the Internet 50, ΔT. ₃ The response node delay 48, indicated by, determines that it will occur when the response node 40 processes message 42 and sends a return message. Therefore, the waiting time T _L 52 is calculated by the latency manager, T _L = ΔT ₁ + ΔT ₂ + ΔT ₃Will be. The latency manager then writes entry 38b to allocation table 36 to indicate that after the latency, a reply message 54 from response node 40 to subscriber 14c is expected. This causes channel 22c to time T ₀ + T _L Is assigned to subscriber 14c. Reply message 54 is sent by response node 40 and time T ₀ + T _L Is received by the base station processor 16. According to the allocation table 36, the scheduler 28 allocates the channel 22c and transmits the reply message 54 to the subscriber 14c.

In another embodiment, the channel is reserved as a bulk pool in the quota table and is not assigned to a particular subscriber until the reply message is actually received.

As an example above, the latency manager 34 has a latency T based on Message Type And The The Corresponding Expected Reply Message. _LTo calculate. Many protocols, including the TCP / IP protocol, specify not only the reply message, but also other transmission parameters. Therefore, the method of determining the latency delay depends on multiple factors, depending on the protocol used. In the TCP / IP protocol, such factors include, for example, window size, space available in the window, average message size, number of unprocessed acks, message type, number of messages received in the session, And transmission parameters such as the maximum number of unprocessed ackes For example, TCP / IP uses the sliding window performance enhancements defined in Internet RFC1323. Such features are used in conjunction with transmission parameters to improve performance via the base station processor as defined herein .

TCP / IP networks operate according to the sliding window protocol to achieve reliable stream delivery while maximizing bandwidth. According to this protocol, both endpoints of a TCP / IP connection negotiate an acceptable window size. The window size represents the maximum number of bytes that a When the transmitting unit receives an acknowledgment for the first packet in the window, it slides the window and sends the next packet. Windows are commonly referred to for the maximum number of unacknowledged packets. .

For message 42 sent in the example of Figure 3a, the latency manager inspects TCP / IP packets in a non-destructive manner to determine the type of message. Examine another aspect of the TCP / IP packet listed above to get the transmission parameters and use In the following examples of FIGS. 4 and 5a-5c, the latency manager 34 further includes a subscriber profile table 56 that stores the transmission parameters corresponding to each of the subscribers 14.

According to the flowchart shown in FIG. 4 and the system diagram of FIG. 3a, the message is received by the base station processor 16 as shown in step 100. The latency manager 34 checks the TCP / IP packet information as shown in step 102 As shown in step 106, the corresponding transmission parameters are retrieved. As shown in step 108, the transmission parameters are updated to reflect. Perform a search in the subscriber profile table to find the entry corresponding to the subscriber, as shown in step 104. At step 110, it is determined whether a reply message to supplement the message is expected. If the reply message is unpredictable, the message is sent, as shown in step 120, and the control loop returns to step 100 and remains in that state until the next message is received, As shown in step 122. If the reply message is expected, the latency manager 34 calculates the latency 52 using the subscriber's transmission parameters updated in step 108, as shown in step 112. As shown in step 114, a new entry corresponding to the calculated latency 52 is stored in the allocation table 36. The message is then sent to the response node 40, as shown in step 116. The control loop returns to step 118 and remains in that state until the next message is received. The control loop returns to step 118 and remains in that state until the next message is received. The control loop returns to step 118 and remains in that state until the next message is received.

5a-5c show in detail another embodiment of the message transmission sequence of FIG. 3b. Connection request 42 is time T ₀ Is transmitted from PC12. The latency manager 34 inspects the packet information and determines the subscriber 14d. The The latency manager looks up the transmission parameters for subscriber 14d in the subscriber profile table 56 and updates the parameters accordingly to match the new packet information. The waiting time manager 34 determines that the connection alert message 54 is returned as a reply message. 34 has a latency ΔT as a result of the updated transmission parameters. ₁ , ΔT ₂ , ΔT ₃ Calculate and wait T _A T _{A Notifies the scheduler 28 to assign channel 22d to subscriber 14d.}₁ = ΔT ₂ + ΔT ₃ + ΔT _{Calculate as. The latency manager 34 stores entry 58 in allocation table 36 and time T, as shown by entry 68 in timing chart 86.} A

The response node 40 then sends a reply message 54. The base station processor 16 receives the response message 54, and the waiting time manager 34 inspects the packet information of the response message 54. The latency manager looks up the transmission parameters for subscriber 14d in The subscriber profile table 56 and updates entry 58 accordingly. The waiting time manager 34 determines that the type of reply message 54 is a connectable email and decides to send the request message as a reply message from the PC.

When the reply message is sent to subscriber 14d, the latency is calculated as follows: Radio propagation time ΔT ₄Represents the latency associated with transmission over the wireless connection 26 between the base station processor 16 and the subscriber 14d. Subscriber response time ΔT ₅Represents the waiting time associated with transmission between the subscriber 14d and the PC 12 over the wired 20. Therefore, the latency manager 34 uses the subscriber profile table to ΔT. ₄ + ΔT ₅ Waiting time from ΔT _B To calculate. Write the corresponding entry 60 to allocation table 36 and time T as shown by entry 70 in timing chart 86. _B Notifies the scheduler to allocate channel 22f to subscriber 14d.

After receiving ack54, PC12 sends an HTTP get message 78. Search the subscriber profile table 56 for the corresponding transmission parameter and update the parameter accordingly to match message 78. As a result of the updated transmission parameters, the latency manager 34 determines the possibility that The HTTP get ack80 and the HTTP data message 82 will be sent as reply messages at the same time. This causes the wait manager to T _C = ΔT ₁ + ΔT ₂ + ΔT ₃ From waiting time T _C And write two entries to the allocation table 36. Time T, as indicated by entries 72 and 74 in timing chart 86 _C In, for subscriber 14d, entry 62 allocates channel 22d and entry 64 allocates channel 22e.

When an HTTP data message is received on switch 16, the latency manager 34 determines that the HTTP data ack84 is a reply message and writes entry 66, as shown by entry 76 in timing chart 86. T _D = ΔT ₄ + ΔT ₅ Assign channel 22f in.

FIG. 6 shows an example of the subscriber profile table 56. Each entry 86 stores a transmission parameter 88 corresponding to a message received by a particular subscriber 14. Such parameters determine the window size, the space available in the window, the average message size , the number of outstanding acks, the type of message, the number of messages received in the session, and the maximum number of outstanding ackes. Including. You can specify other parameters defined by the TCP / IP protocol or other protocols such as those used by base station processors.

Those skilled in the art will appreciate that the programs that define the behaviors and methods defined herein can be applied to base station processors in many forms. These forms include a) information permanently stored on non-writable storage media such as ROM devices, b) information modifiable and stored on magnetic and optical media, c) Communicating using baseband or broadband signaling methods, such as in electronic networks such as the Internet or telephone modem lines. Information, including, but not limited to, information transferred to a computer via a medium. These operations and methods can be achieved with software that can be executed from memory by the processor. As an alternative, these operations and methods include, for example, application specific integrated circuits (ASICs), state machines (states). It is achieved using hardware components such as machine), controllers or other hardware components or devices, or using a combination of hardware and software components.

Although the present invention has been illustrated and described in detail in preferred embodiments, those skilled in the art will make various modifications to the shape or details without departing from the spirit and scope of the invention, which is limited to the claims. understood that it is possible. Therefore, the present invention shall not be limited except as limited by the claims.