KR101158912B1 - Stall avoidance method using a window in hsdpa system - Google Patents

Abstract

Updating the next-expected transmission sequence number (NET) or the receiver window position to ensure that the NET always falls within the receiver window range to prevent unnecessary delays in delivering data blocks in order to avoid stall conditions and achieve high speed data transmission capabilities for a high-speed downlink packet access (HSDPA) system.

Description

STAD AVOIDANCE METHOD USING A WINDOW IN HSDPA SYSTEM}

1 is a diagram illustrating a network structure of a UMTS to which the prior art and the present invention are applied.

2 is a diagram illustrating a radio interface protocol structure between a terminal and a UTRAN based on the 3GPP radio access network standard;

3 is a MAC layer structure of the terminal side used in the HSDPA system.

4 is a diagram illustrating an example of a process of transmitting / receiving a MAC-hs PDU in an HSDPA system.

5 is a flowchart showing a deadlock avoidance method using a conventional window.

6 is a flowchart illustrating a deadlock avoidance method using a window according to the present invention.

7 (A) to 7 (C) are diagrams illustrating an operation when the NET becomes smaller than the reception window starting point as the reception window moves due to the reception of a new PDU.

8 (A) to 8 (D) show the operation when all PDUs in the reception window are successfully received so that the NET becomes larger than the reception window endpoint.

The present invention relates to a high speed downlink packet access (HSDPA) of a UMTS wireless mobile communication system, and more particularly, to a deadlock avoidance method in which a terminal controls a reordering buffer by using a receiver number and a receiving window.

UMTS (Universal Mobile Telecommunications System) is a third generation mobile communication system that has evolved from the European standard Global System for Mobile Communications (GSM) system.It is based on GSM Core Network and Wideband Code Division Multiple Access (WCDMA) access technologies. The aim is to provide an improved mobile communication service.

1 is a diagram illustrating a network structure of a UMTS to which the prior art and the present invention are applied.

As shown in FIG. 1, a UMTS system is composed of a user equipment (UE), a UTMS radio access network (UTRAN), and a core network (CN). The UTRAN consists of one or more Radio Network Sub-systems (RNS), each of which is a Radio Network Controller (RN) and one or more base stations (Node B) managed by the RNC. It consists of.

The Node B is managed by the RNC, and receives the information transmitted from the physical layer of the terminal in the uplink, and transmits data to the terminal in the downlink, and serves as an access point of the UTRAN for the terminal.

The RNC is in charge of allocating and managing radio resources and serves as an access point with the CN. In the RNC, an RNC managing a dedicated resource for a specific terminal is called a SRNC (Serving RNC), and an RNC managing a common resource for a plurality of terminals in a cell is called a CRNC (Controlling RNC). In addition, all RNCs that the terminal passes through except the SRNC when the terminal moves are called DRNC (Drift RNC).

The interface between the RNC and the CN is called Iu interface, and the interface between the SRNC and the DRNC is called Iub. In addition, the interface between the RNC and Node B is called an Iub interface. Each interface provides control information or a data transmission service through a transport bearer. For example, a bearer provided by the Iub interface is called an Iub transmission bearer, and the Iub transmission bearer provides control information or data transmission service between the RNC and the Node B.

FIG. 2 shows a structure of a radio interface protocol between a terminal and a UTRAN (UMTS Terrestrial Radio Access Network) based on the 3GPP radio access network standard.

The air interface protocol of FIG. 2 consists of a physical layer, a data link layer, and a network layer horizontally, and is vertically divided into a user plane for transmitting data information and a control plane for transmitting control signals. The user plane is an area in which user traffic information is transmitted, such as voice or IP packet transmission, and the control plane is an area in which control information is transmitted, such as network interface or call maintenance and management. The protocol layers of FIG. 2 are based on the lower three layers of the Open System Interconnection (OSI) reference model, which are well known in communication systems, and include L1 (first layer), L2 (second layer), and L3 (first layer). Three layers).

The physical layer (PHY), which is the first layer, provides an information transfer service to a higher layer by using various wireless transmission technologies, and a transport channel with a medium access control layer on a higher layer. Connected via Channel. Thus, data between the medium access control layer and the physical layer is moved through the transport channel.

Medium Access Control (MAC) belonging to the second layer provides a reassignment service of MAC parameters for allocating and reallocating radio resources. The MAC is connected to a radio link control layer, which is an upper layer, through a logical channel, and various logical channels are provided according to the type of information to be transmitted.

In general, a control channel is used to transmit control plane information, and a traffic channel is used to transmit information of a user plane. MAC is divided into a MAC-b sublayer, a MAC-d sublayer, a MAC-c / sh sublayer, and a MAC-hs sublayer according to the type of a transport channel managed. The MAC-b sublayer manages a broadcast channel (BCH), which is a transport channel for broadcasting system information, and the MAC-c / sh sublayer manages a forward access channel (FACH) or a downlink shared with other terminals. Manage common transport channels such as Shared Channel. In the UTRAN, the MAC-c / sh sublayer is located in the CRNC and manages channels shared by all terminals in a cell, so that there is one for each cell. In addition, there is one MAC-c / sh sublayer in each terminal. The MAC-d sublayer is responsible for managing a dedicated channel (DCH), which is a dedicated transport channel for a specific terminal.

Therefore, the MAC-d sublayer of the UTRAN is located in the SRNC that manages the corresponding terminal, and there is one MAC-d sublayer in each terminal. The MAC-hs sublayer performs functions related to high speed data packet access (HSDPA), such as packet scheduling or HARQ (Hybrid ARQ) operation.

The Radio Link Control (RLC) layer supports reliable data transmission and performs segmentation and concatenation of RLC Service Data Units (SDUs) from lower layers. have. The RLC SDU delivered from the upper layer is sized according to the processing capacity in the RLC layer, and then header information is added to the MAC layer in the form of a Protocol Data Unit (PDU). In the RLC layer, there is an RLC buffer for storing RLC SDUs or RLC PDUs from above.

The Broadcast / Multicast Control (BMC) layer is located above the RLC layer and is a terminal located in a specific cell (s) by scheduling a cell broadcast message (CB message) delivered from a core network. (UE) to perform the function of broadcasting.

The Packet Data Convergence Protocol (PDCP) layer is located above the RLC layer, allowing data transmitted over network protocols such as IPv4 or IPv6 to be efficiently transmitted over a relatively low bandwidth wireless interface. . To this end, the PDCP layer performs header compression, a function of reducing unnecessary control information used in a wired network. That is, the PDCP layer transmits only the necessary information in the header portion of the data, thereby increasing the transmission efficiency of the radio section.

The radio resource control (RRC) layer located at the bottom of the third layer is defined only in the control plane, and controls transport channels and physical channels in connection with setting, resetting, and releasing radio bearers (RBs). In charge of. In this case, the RB refers to a service provided by the second layer for data transmission between the UE and the UTRAN. In general, the RB is configured to define characteristics of a protocol layer and a channel necessary to provide a specific service. The process of setting specific parameters and operation methods. In addition, when the RRC layer of a specific terminal and the RRC layer of the UTRAN is connected to send and receive RRC messages with each other, the terminal is in the RRC connected state (Connected state), when the terminal is not connected to the idle state ( Idle state).

High Speed Downlink Packet Access (HSDPA) system supports up to 10Mbps downlink based on W-CDMA and provides shorter packet transmission latency and improved capacity.

Hereinafter, the operation of the terminal for the HSDPA will be described.

3 is a MAC layer structure of the terminal side used in the HSDPA system.

As shown in FIG. 3, the MAC layer on the terminal side is divided into three sublayers: a MAC-d sublayer, a MAC-c / sh sublayer, and a MAC-hs sublayer. Referring to FIG. 3, let's find out how the data from the physical layer (PHY) is transferred to the RLC layer in the MAC layer.                         

The MAC-hs PDU delivered to MAC-hs through the HS-DSCH (High Speed Downlink Shared Channel), which is a transport channel for HSDPA, is first stored in one of several HARQ processes in the HARQ block. In this case, which process is stored may be known from the HARQ process identifier included in the downlink control signal.

If there is an error in the MAC-hs PDU, the HARQ process transmits NACK information to the UTRAN to request retransmission of the MAC-hs PDU. If there is no error, the HARQ process forwards it to the reordering buffer and sends it to the UTRAN. Send ACK information. The reordering buffer exists for each priority, and the HARQ process delivers the MAC-hs PDU to the reordering buffer by using a queue identifier (Queue Id) included in the MAC-hs PDU. The big feature of the reorder buffer is that it supports in-sequence delivery, so that MAC-hs PDUs are sequentially delivered based on a transmission sequence number (TSN), if the corresponding MAC-hs If the MAC-hs PDU before the PDU has not been received, the received MAC-hs PDU is stored in the reordering buffer, and when all the previous MAC-hs PDUs are received, the MAC-hs PDUs are transferred to the upper part together with the corresponding MAC-hs PDUs.

The serial number (TSN) is 6 bits and modulo operation is performed. In general, since several HARQ processes operate, the reordering buffer receives MAC-hs PDUs out-of-sequence. Therefore, in order to deliver MAC-hs PDUs in-sequence to higher layers, a reorder buffer is necessary.

When MAC-hs PDUs in the reorder buffer are delivered to the upper level, they are first delivered to the disassembly block. The decomposition block serves to segment MAC-hs PDUs in which a plurality of MAC-d PDUs are combined into a plurality of MAC-d PDUs. Thereafter, the decomposition block transfers the divided MAC-d PDUs to the MAC-d sublayer, and the transport channel multiplexing block of the MAC-d sublayer that has received the MAC-d PDU is a logical channel identifier included in each MAC-d PDU. After the C / T field is recognized, the MAC-d PDU is transferred to the RLC through the logical channel by controlling the channel switching block.

4 is a diagram illustrating an example of a process of transmitting and receiving a MAC-hs PDU in an HSDPA system. Although the MAC-d PDU is actually stored in the transmission buffer, it is illustrated that MAC-hs PDUs (= one or several MAC-d PDUs) are stored for convenience of description, and 8 HARQ processes are assumed. In addition, although the size of each MAC-hs PDU may be different from each other, the MAC-hs PDUs are shown in the same size because it is shown only conceptually.

4 is a process of transmitting to the receiving side when the MAC-hs PDU is TSN = 13 to TSN = 22 in the transmission buffer.

First, the TSN is transferred from the low MAC-hs PDU to the empty HARQ process. In FIG. 4, the MAC-hs PDU having TSN = 13 is delivered to HARQ process # 1, and the MAC-hs PDU having TSN = 14 is delivered to HARQ process # 8. In other words, it is not related to the TSN and the HARQ process number and is simply transferred to an empty process.

When the HARQ process receives a random MAC-hs PDU, it transmits it to the receiver for a specific transmission time interval (TTI) and stores it for retransmission in the future. Only one MAC-hs PDU can be transmitted during any one TTI, so only one HARQ process is active. The HARQ process that transmits the MAC-hs PDU informs the receiver of its process number through a downlink control signal. Here, the downlink control signal is transmitted through a channel different from the MAC-hs PDU.

The reason why the HARQ processes of the transmitting side and the receiving side are matched is that a stop-and-wait ARQ scheme is used for each HARQ process. That is, the HARQ process 1 that transmits the MAC-hs PDU having TSN = 13 does not transmit another MAC-hs PDU until the transmission of the MAC-hs PDU is successful. If the MAC-hs PDU is not received during the predetermined TTI, the first HARQ process of the receiver transmits NACK information to the transmitter through an uplink control signal. On the contrary, when the MAC-hs PDU is normally received, the HARQ process transmits ACK information to the sender and transfers the MAC-hs PDU to the reordering buffer indicated by the queue ID. That is, since the reordering buffer exists for each priority, the HARQ process sees the queue identifier included in the MAC-hs PDU and delivers the MAC-hs PDU to the corresponding reordering buffer. The MAC-hs PDU delivered to the reordering buffer is immediately transferred to the upper level if all previous MAC-hs PDUs are transferred to the upper level, but is temporarily stored in the reordering buffer if the previous MAC-hs PDU is not transferred to the upper level. That is, since the reordering buffer can deliver MAC-hs PDUs only in-sequence, MAC-hs PDUs that are not delivered are stored in the reordering buffer.

In FIG. 4, if a MAC-hs PDU having TSN = 14 is received while a MAC-hs PDU having TSN = 13 is received, a MAC-hs PDU having TSN = 13 may be received. Until it is stored in the reorder buffer. Thereafter, when a MAC-hs PDU having TSN = 13 is received, TSN = 13 and TSN = 14 are transferred upward. When the MAC-hs PDU is delivered to a higher level, as described above, the MAC-hs PDU is divided and transmitted in units of MAC-d PDUs.

However, in the process of transmitting data through the radio channel, some MAC-hs PDUs may not be properly transmitted from the UTRAN to the UE despite several retransmission attempts. This lack of transmission of a specific MAC-hs PDU for a long time reduces the transmission efficiency of the HSDPA system. In other words, the HSDPA system was developed for high-speed data communication. If a large number of MAC-hs PDUs are left in the MAC-hs reordering buffer for a long time due to one MAC-hs PDU not received, the entire data transmission is performed. It will reduce the efficiency. This is contrary to the intention of introducing the HSDPA system.

In order to prevent the stall phenomenon of the MAC-hs PDU, HSDPA introduces a stall avoidance method using a window. Before describing the deadlock avoidance method using the conventional window, let's first describe the environment variables used in the reorder buffer.

The receiver number (next_expected_TSN: NET) is a value immediately after the TSN of the last received PDU among MAC-hs PDUs sequentially received. That is, NET indicates the TSN value of the first MAC-hs PDU to be received sequentially. The NET is updated whenever a PDU with a TSN of NET (TSN = NET) is received and has an initial value of zero.

The receiving window endpoint RcvWindow_UpperEdge means the largest TSN value in the receiving window of the reordering buffer. When the MAC-hs PDU arrives at the receiving side for the first time, the receiving window endpoint is set to the highest value among the TSNs of the received PDUs, and the initial value of the receiving window endpoint is 63.

The receiving window starting point (RcvWindow_LowerEdge) means the smallest TSN value in the receiving window of the reordering buffer. The reception window start point is a value obtained by subtracting the reception window size from the reception window end point plus one.

The Receive Window defines TSN values of MAC-hs PDUs that can be received without changing the position of the Receive Window. The receive window includes TSNs from the receive window start point to the receive window end point.

Receive Window Size (Receiver Window Size, RECEIVE_WINDOW_SIZE) means the size of the receiving window and is set by the upper end of the MAC entity.

Next, a deadlock avoidance method using a conventional window will be described.

In a deadlock avoidance method using a conventional window, the receiving side moves the receiving window when a MAC-hs PDU having a TSN larger than the receiving window endpoint is received. The receiving side no longer waits for the MAC-hs PDUs that have not yet been received among the MAC-hs PDUs corresponding to the TSN lower than the starting point of the updated receiving window according to the movement of the receiving window. The stored MAC-hs PDUs are delivered to the upper end to avoid deadlocks of the MAC-hs PDUs.

The deadlock avoidance method using the conventional window will be described in detail with reference to FIG. 5 as follows.

The receiving side receives the MAC-hs PDU whose TSN is SN (= any number) (S10), and compares the received SN value with the receiving window range (S11). As a result of the comparison, if the SN value is within the reception window range, the receiving side compares the SN value and NET and checks whether the MAC-hs PDU corresponding to the SN has been previously received (S12).

If the SN is smaller than NET or the MAC-hs PDU corresponding to the SN value has been previously received, the receiver deletes the received MAC-hs PDU (S13). On the other hand, if the SN is greater than or equal to NET and the MAC-hs PDU corresponding to the SN value has never been previously received, the receiver stores the corresponding MAC-hs PDU at the location indicated by the SN value in the reordering buffer (S14).

In step S11, if the SN value is outside the reception window, the receiving side stores the received MAC-hs PDU at the point indicated by the SN value in the reordering buffer, i.e., the upper end of the receiving window endpoint, and then stores the receiving window endpoint at the corresponding SN value. (S15, S16). Thereafter, the receiving side transfers MAC-hs PDUs whose TSN value is smaller than the reception window start point to the partition block among MAC-hs PDUs stored in the reordering buffer (S17) and updates the NET to the reception window starting point (S18).

When the processes S14 and S18 are performed, the receiving side converts the MAC-hs PDU whose TSN value is NET from the PDUs stored in the reordering buffer to the MAC-hs PDU immediately before the first MAC-hs PDU not received as a decomposition block. Transfer (S19). Here, the first MAC-hs PDU not received refers to the MAC-hs PDU with the smallest TSN among the MAC-hs PDUs whose TSN is equal to or larger than NET and not yet received. When the process (S19) is finished, the receiving side updates the NET to the TSN of the first MAC-hs PDU not received (S20).

In the deadlock avoidance method using the conventional window, the receiving side sets the NET value to 0 and the receiving window end point to 63 in the initial initialization step. Therefore, for example, if the size of the receiving window is 32, the receiving window is initially determined from the value of the TSN is 32 to the value of the TSN is 63 according to the definition of the receiving window.

The transmitter sets the TSN value of the MAC-hs PDU to be transmitted first to 0, and uses TS, 1,2,3 .. If no loss occurs in the wireless space, the first PDU arriving at the receiver will be a MAC-hs PDU with a TSN value of zero. However, since the MAC-hs PDU is located outside the defined receiving window, the receiving side advances the window according to the TSN of the MAC-hs PDU.

As a result, the reception window reset according to the process illustrated in FIG. 5 is set from a value of 33 to a value of 0 for TSN. In addition, despite the fact that the MAC-hs PDU expected by the receiver afterwards is a MAC-hs PDU whose TSN is 1, the NET value is set to 33.

Also, even more problematic is that although the received MAC-hs PDU is a MAC-hs PDU that can be immediately delivered to the upper layer, the MAC-hs PDU is not delivered to the upper decomposition block and accumulated in the rearrange buffer so that unnecessary delay of delivery is required. This will occur. In this case, all MAC-hs PDUs having a TSN lower than the reception window starting point, that is, MAC-hs PDUs having a TSN value of 33 and MAC-hs PDUs having a TSN value of 63, are received in order for the MAC-hs PDU to be transmitted to the upper end. The receiving window is moved by the MAC-hs PDU received or later received, so that the MAC-hs PDU having a TSN value of 0 must be located smaller than the starting point of the receiving window. This situation also occurs when the MAC-hs PDUs with low TSN values, such as 1 or 2, arrive at the receiver at the beginning of the receive window operation. Therefore, MAC-hs PDUs that are sequentially received and immediately delivered to the upper level stay in the reordering buffer more than necessary, causing unnecessary propagation delay conditions.

This unnecessary propagation delay does not only occur at the initial stage of HSDPA operation. If all MAC-hs PDUs within the receive window range are received normally, the receive window does not move and NET will be updated to the receive window endpoint + 1. In these cases, when NET is pointing to the receiving window endpoint + 1, if the MAC-hs PDU with TSN value of the receiving window endpoint + 1 arrives at the receiving side, the receiving side will receive the receiving window because the MAC-hs PDU is outside the receiving window. Newly adjust the range of, update the NET value to the reception window starting point, and store the received MAC-hs PDU in the reorder buffer. However, although the MAC-hs PDUs are sequentially received and can be immediately delivered to the upper stage, since the MAC-hs PDUs follow the operation of the conventional receiving window, the MAC-hs PDUs are not transferred to the decomposition block but are stored in the reordering buffer, thereby unnecessary transfer delay. The situation arises.

This delay condition occurs when the NET value is located outside the receiving window. In addition, although the transmission delay in the initial operation of the receiving side is NET as 0, the situation occurs because the receiving window is set to a value of 32 to 63 with a TSN value of 32, so that the NET value is located outside the receiving window. to be.

Whenever the situation where the NET value is located outside the receiving window occurs, the receiving side can send the received MAC-hs PDU to the upper stage immediately, but it stores it in the reordering buffer to delay delivery of the corresponding MAC-hs PDU. do. This propagation delay may cause higher level errors or significantly degrade the quality of service.

 Accordingly, an object of the present invention is to ensure the high speed data transmission capability of the HSDPA by preventing the unnecessary propagation delay of the MAC-hs PDU by ensuring that the receiver number is always located inside the reception window.

In order to achieve the above object, in a high speed downlink packet access (HSDPA) system, a deadlock avoidance method according to an aspect of the present invention comprises the steps of: receiving a protocol data unit (PDU); Processing the received PDU according to environment variables and serial numbers such as a reception window and a receiver number, and updating the environment variables; Checking whether the receiver number is present in the reception window; Updating the receiver number or the receiving window so that the receiver number can be included in the receiving window when the receiver number exists outside the receiving window.

Preferably, the updating step is characterized in that the starting point of the reception window to match the receiver number.

Preferably, the updating step is characterized in that the end point of the reception window to match the receiver number.

In order to achieve the above object, in the HSDPA (High Speed Downlink Packet Access) system for receiving and processing data using a receiver number and a receiving window, the deadlock avoidance method according to another aspect of the present invention, the protocol data Receiving a unit (PDU); Updating the reception window with reference to the serial number of the received PDU; Checking whether the receiver number is smaller than the updated reception window start point; And updating the receiver number so that the receiver number is included in the reception window when the receiver number is smaller than the starting point of the reception window.

Preferably, the receiver number is updated to match the start point of the reception window.

In order to achieve the above object, in the HSDPA (High Speed Downlink Packet Access) system for receiving and processing data using the receiver number and the receiving window, a deadlock avoidance method according to another aspect of the present invention, the protocol Receiving a data unit (PDU); Updating the receiver number by referring to the serial number of the received PDU; Checking whether the updated receiver number is greater than the receiving window endpoint; And updating the reception window so that the reception number is included in the reception window when the receiver number is larger than the reception window endpoint.

Preferably, the receiver number is updated to match the reception window endpoint.

The present invention is implemented in a W-CDMA mobile communication system. However, the present invention can also be applied to communication systems operating according to other standards. Hereinafter, preferred embodiments of the present invention will be described in detail.

The present invention ensures that the NET value is always located inside the reception window on the receiving side (terminal), thereby preventing unnecessary transmission delay of MAC-hs PDUs that may occur when using the conventional technology, thereby improving the high-speed data transmission capability of HSDPA. I want to secure it.

To this end, in the present invention, the initial range of the reception window is adjusted so that the NET is located in the reception window from the beginning of the protocol operation. It suggests how to re-update NET or receiving window to be in value.

Specifically, in the present invention, the initial value of the reception window endpoint is changed to 0, such as NET, not 63, so that NET is within the reception window range from the beginning of the protocol operation. If the NET becomes smaller than the receiving window starting point due to the update of the receiving window, the NET is renewed to the TSN value of the receiving window starting point.If the NET becomes larger than the receiving window end point due to the updating of the NET, the receiving window end point is reset to NET. Suggest ways to update.

As described above, the present invention provides an example of a method of matching the receiving window endpoint with the NET when the NET becomes larger than the receiving window endpoint due to an initial protocol operation or an update of the NET. It should be noted that the receiving window endpoint does not necessarily have to match the NET. Instead, the receiving window should be updated to an appropriate location so that the NET falls within the receiving window range. That is, in the present invention, the reception window end point is set to NET only as an example, the reception window start point may be set to NET, and other values within the reception window range may be set to NET.

6 is a flowchart illustrating a deadlock avoidance method using a window according to the present invention.

The receiving side receives the MAC-hs PDU whose TSN is SN (= any number) (S20), and compares the received SN value with the receiving window range (S21). As a result of the comparison, if the SN value is within the reception window range, the receiver compares the SN value with NET and checks whether the MAC-hs PDU corresponding to the SN has been previously received (S22).                     

If the SN is smaller than NET or the MAC-hs PDU corresponding to the SN value has been previously received, the receiver deletes the received MAC-hs PDU (S23). On the other hand, if the SN is greater than or equal to NET and the MAC-hs PDU corresponding to the SN value has not been previously received, the receiver stores the corresponding MAC-hs PDU at the location indicated by the SN value in the reordering buffer (S24).

If the SN value is outside the receiving window in step S21, the receiving side stores the received MAC-hs PDU in the reordering buffer at the point indicated by the SN value, that is, above the receiving window endpoint, and stores the receiving window endpoint as the corresponding SN value. (S25, S26). Thereafter, the receiving side transfers MAC-hs PDUs whose TSN value is smaller than the reception window start point from the reordering buffer to the split block among the MAC-hs PDUs stored in the reordering buffer (S27). Then, the receiving side compares the NET with the receiving window starting point, and updates the NET to the receiving window starting point only when the NET is smaller than the receiving window starting point (S28, S29).

7 illustrates an operation in which NET becomes smaller than the reception window starting point as the reception window moves due to the reception of a new PDU. In this case, it is assumed that the size of the reception window is 5. FIG. 7 illustrates an example of receiving a MAC-hs PDU having a TSN of 10 while a MAC-hs PDU having a NET of 4 and a MAC-hs PDU having a 7 is received.

When the MAC-hs PDU having the TSN of 10 is received in the situation as shown in FIG. 7A, the PDU having the TSN of 10 is located outside the range of the receiving window, so that the movement of the receiving window is not shown as shown in FIG. Happens. As the receiving window moves, the MAC-hs PDU (TSN = 5) having a TSN smaller than the receiving window starting point, which is located outside the range of the receiving window, is transferred to the upper end. Since the current NET is smaller than the reception window start point, that is, 6, the NET is updated to 6 as shown in Fig. 7C.

Once the processes S24 and S28 or S29 are performed, the receiving end performs the MAC-hs PDU whose TSN value is NET from the PDUs stored in the reordering buffer to the MAC-hs PDU immediately before the first MAC-hs PDU not received. Transfer to the decomposition block (S30). Here, the first MAC-hs PDU not received refers to the MAC-hs PDU with the smallest TSN among the MAC-hs PDUs whose TSN is equal to or larger than NET and not yet received.

When the process (S30) is finished, the receiving side updates the NET to the TSN of the first MAC-hs PDU that has not been received (S31), and then compares the updated NET with the receiving window endpoint, where the NET is larger than the receiving window endpoint. By only updating the receiving window, the receiving window end point is set to NET (S32, S33).

8 is an operation when all PDUs in the reception window are successfully received so that the NET becomes larger than the reception window endpoint. Again, it is assumed that the size of the receiving window is five. FIG. 8 illustrates an example in which a MAC-hs PDU having a TSN of 4 is received in a situation where NET is 4 and a MAC-hs PDU having a TSN of 5, 6, 7, or 8 is received.

When the MAC-hs PDU having the TSN of 4 is received in the situation as shown in FIG. 8 (A), the MAC-hs PDU having the TSN of 4 is placed at the 4 position in the reorder buffer as shown in FIG. From the MAC-hs PDU of 4 to the MAC-hs PDU of 8 immediately before the first MAC-hs PDU (TSN = 9) not received, the corresponding TSN is transferred to the decomposition block. NET is then updated to 9. Therefore, when NET becomes larger than the receiving window end point as shown in FIG. 8 (C), the end point of the receiving window is updated to NET to move the receiving window as shown in FIG. 8 (D).                     

When the receiving window is updated in step S33, the receiving window endpoint may be updated to correspond to NET, but other parts of the receiving window may be updated to correspond to NET.

As described above, when using the HSDPA using the conventional technology, even if the terminal can directly transfer the received MAC-hs PDU to the upper end, the terminal stored in the reordering buffer for a long time, causing unnecessary data transfer delay. Therefore, when updating the NET value and the reception window and processing the MAC-hs PDU according to the deadlock avoidance method according to the present invention, the receiving side (terminal) prevents unnecessary data transmission delay and data transmission error, thereby preventing high-speed data transmission. There is an effect that can be made possible.

In addition, the present invention has been described with reference to the embodiments shown in the drawings, which are merely exemplary, and those skilled in the art will understand that various modifications and equivalent other embodiments are possible therefrom. . Therefore, the true technical protection scope of the present invention will be defined by the technical spirit of the appended claims.

Claims (49)

A method of processing data blocks on a wireless communication system, Receiving data blocks associated with a sequence number; Determining whether the serial numbers of the received data blocks are outside a receiver window; Comparing a next-expected transmission sequence number (NET) that is next expected when it is determined that the serial numbers of the received data blocks are outside the reception window; And Setting the NET to be within the updated receive window if the NET is identified as being out of range of the updated receive window as a result of the comparison. . 2. The method of claim 1, wherein the NET is a transmission sequence number (TSN) of the last received data block of the last order. 3. The method of claim 2, wherein the TSN is an identifier for a transmission serial number on a High Speed Downlink Shared Channel (HS-DSCH). 3. The method of claim 2, wherein the TSN is used for reordering purposes to support sequential transmissions to higher stages. 2. The method of claim 1, wherein if the NET is greater than the updated receive window endpoint, set the NET as the updated receive window endpoint. 2. The method of claim 1, wherein if the NET is greater than the updated receive window endpoint, the NET is set to a suitable location within the updated receive window range. . 2. The data of a wireless communication system according to claim 1, wherein if the NET is larger than the updated receiving window, the data advances any suitable place in the receiving window while the NET is within the updated receiving window range. How to process blocks. 2. The method of claim 1, wherein if the NET is less than the updated receive window start point, set the NET as the updated receive window start point. 2. The radio of claim 1, wherein if NET is less than the updated receive window, then NET = receive window upper edge value-receive window size + 1 wireless. A method of processing data blocks on a communication system. 2. The method of claim 1, wherein said NET is a variable that is processed by a receiving end. 2. The method of claim 1, wherein the NET is updated to TSN = NET when delivered to the disassembly entity of the data block. 2. The method of claim 1, wherein the initial value of NET is zero. The method of claim 1, wherein the data blocks are Medium Access Control-high speed (MACH) protocol data units (PDUs). 2. The method of claim 1, wherein said steps are performed for a high-speed downlink packet access (HSDPA) system. 2. The method of claim 1, wherein said steps are performed for avoidance stall conditions. 2. The method of claim 1, wherein if the received data block associated with the serial number was previously received, the received data block is discarded. An apparatus for processing data blocks on a wireless communication system, A receiving module for receiving data blocks associated with a sequence number; A determination module for determining whether a serial number of the received data blocks is outside a receiver window; A next-expected transmission sequence number (NET), which is next expected when the determination module determines that the serial number of the received data blocks is outside the reception window, and the range of the updated reception window; A comparison module for comparing; And And a setting module for setting the NET to be within the updated receiving window if the NET is found out of the range of the updated receiving window as a result of the comparison by the comparing module. Apparatus for processing data blocks on a network. 18. The method of claim 17, wherein the NET processed by the modules is a TSN following a transmission sequence number (TSN) of the last sequence data block received. Device. 19. The apparatus of claim 18, wherein the TSN is an identifier for a transmission serial number on a High Speed Downlink Shared Channel (HS-DSCH). 19. The apparatus of claim 18, wherein the TSN is used for rearrangement purposes to support sequential transmission to higher stages. 18. The system of claim 17, wherein if the NET processed by the modules is greater than the updated receive window endpoint, set the NET as the updated receive window endpoint. Device for processing. 18. The method of claim 17, wherein if the NET is greater than the updated receive window endpoint, the NET is set to a suitable location within the updated receive window range. Device. 18. The system of claim 17, wherein if the NET is greater than the updated receive window, the data advances in a suitable location in the receive window while the NET is within the updated receive window range. Device for processing blocks. 18. The apparatus of claim 17, wherein if the NET is less than the updated receive window start point, set the NET as the updated receive window start point. 18. The method of claim 17, wherein if the NET is less than the updated receive window, the NET = receive window upper edge value-receive window size + 1 wireless. Apparatus for processing data blocks on a communication system. 18. The apparatus of claim 17, wherein the NET is a variable number processed by a receiving end. 18. The apparatus of claim 17, wherein the NET is updated to TSN = NET when delivered to the disassembly entity of the data block. 18. The apparatus of claim 17, wherein the initial value of NET is zero. 18. The apparatus of claim 17, wherein the data blocks are Medium Access Control-high Speed (MAC-hs) Protocol Data Units (PDUs). 18. The apparatus of claim 17, wherein the modules operate for a high-speed downlink packet access (HSDPA) system. 18. The apparatus of claim 17, wherein the modules operate for deadlock stall conditions. 18. The apparatus of claim 17, wherein if the received data block associated with the serial number was previously received, the received data block is discarded. 18. The apparatus of claim 17, wherein the modules are part of a MAC entity. 18. The apparatus of claim 17, wherein the modules are part of a MAC-hs entity. 18. The apparatus of claim 17, wherein the modules operate within a mobile station (MS). 18. The apparatus of claim 17, wherein the modules are part of a wireless handset. 18. The apparatus of claim 17, wherein the modules are part of a network. 18. The apparatus of claim 17, wherein the modules are part of a base station. 18. The apparatus of claim 17, wherein the modules are part of a Node B. 18. The apparatus of claim 17, wherein the modules are part of a Radio Network Controller (RNC). A method of processing data blocks on a wireless communication system, A first step of receiving a Protocol Data Unit (PDU) having a sequence number (SN); In a second step of comparing the serial number with a receiver window range, If the serial number is not within the receiving window range, Storing the PDU at a location within a buffer defined by the SN, Update the reception window so that an end value of the reception window is SN; Deliver all PDUs with SN less than the start of the receive window to a disassembly block, and Next verify that the next-expected transmission sequence number (NET) is less than the start of the receive window, If small, update the NET to be equal to the starting point of the receive window and proceed to the third step, If it is not small, proceeding to the third step; A third step of delivering specific PDUs stored in the buffer in a decomposition block from the PDU with the NET to the PDU immediately preceding the PDU that has not yet been received; And And a fourth step of updating the NET to be equal to the SN of the first PDU that has not yet received the NET. 42. The method of claim 41, wherein after the fourth step, verify that the updated NET is greater than the end value of the receive window, and if so, set the NET equal to the end value of the receive window; Repeating the process from the first step or completing the process or repeating the process. 43. The method of claim 42, wherein the NET is a TSN following a transmission sequence number (TSN) of the last sequence data unit received. 44. The method of claim 43, wherein the TSN is an identifier for a transmission serial number on a High Speed Downlink Shared Channel (HS-DSCH). 45. The method of claim 44, wherein said TSN is used for rearrangement purposes to support sequential transmission to a higher level. 45. The method of claim 44, wherein if the NET is greater than the receive window endpoint, set the NET to a suitable location within the receive window range. 45. The system of claim 44, wherein if the NET is greater than the updated receive window, data is moved over a suitable location in the receive window while the NET is within the updated receive window range. How to process blocks. 45. The method of claim 44, wherein the NET is updated to TSN = NET when delivered to the disassembly entity of the data unit. 45. The method of claim 44, wherein the initial value of NET is zero.

Images