Académique Documents
Professionnel Documents
Culture Documents
Table of Contents
Table of Contents
Chapter 1 GPRS Fundamentals .......................................................................................................1 1.1 GPRS Overview.....................................................................................................................1 1.2 Evolution of GPRS Standards and Services .........................................................................1 1.3 Comparison Between GPRS and HSCSD ............................................................................2 1.4 EDGE Overview.....................................................................................................................3 1.5 Advantages and Disadvantages of the GPRS.......................................................................4 Chapter 2 GPRS Network Architecture ...........................................................................................6 2.1 Overall GPRS Structure.........................................................................................................6 2.2 Logical System Architecture of the GPRS .............................................................................7 2.3 Major Network Entities of GPRS............................................................................................7 Chapter 3 GPRS Protocol Layers ..................................................................................................11 3.1 GPRS Data Transmission Plane .........................................................................................11 3.2 GPRS Signaling Plane.........................................................................................................12 3.3 GPRS Network Interface Protocols .....................................................................................15 3.3.1 Um Interface..............................................................................................................15 3.3.2 Gb Interface...............................................................................................................19 3.3.3 Gs Interface...............................................................................................................21 3.3.4 Gn/Gp Interface.........................................................................................................21 3.3.5 Gi Interface................................................................................................................23 3.3.6 Gr Interface ...............................................................................................................23 3.3.7 Gd Interface...............................................................................................................23 3.3.8 Gc Interface...............................................................................................................23 3.3.9 Gf Interface................................................................................................................23 Chapter 4 GPRS Radio Subsystem ...............................................................................................24 4.1 GPRS Radio Interface Channels .........................................................................................24 4.2 Channel Coding ...................................................................................................................26 1.1.1 Channel Coding of GPRS PDTCH............................................................................26 4.2.2 Channel Coding of EGPRS PDTCH .........................................................................28 4.2.3 Channel Coding for PACCH, PBCCH, PAGCH, PPCH, PNCH and PTCCH/D .......35 4.2.4 Channel Coding for the PRACH................................................................................35 4.3 Media Access Control Mode................................................................................................36 4.4 Multislot Capability of MS ....................................................................................................36 4.4.1 Multislot Configuration...............................................................................................36 4.4.2 MS Classes for Multislot Capability...........................................................................36 4.5 Power Control ......................................................................................................................39 4.6 Paging Handling...................................................................................................................39 4.6.1 Packet Paging ...........................................................................................................39 4.6.2 Paging Co-ordination ................................................................................................40 4.6.3 Network Operation Modes ........................................................................................40 4.7 Packet Access Modes .........................................................................................................41 4.8 GPRS Cell Selection and Reselection.................................................................................42 4.8.1 Relationship Between GPRS Cell Selection and GSM Cell Selection......................42 4.8.2 Relationship Between GPRS Cell Reselection and GSM Cell Reselection .............42 4.8.3 Network Control Modes.............................................................................................42 Chapter 5 GPRS Contents and Quality .........................................................................................44 5.1 Bearer Services ...................................................................................................................44 5.2 GPRS Supplementary Services...........................................................................................45 5.3 Applications of GPRS Services ...........................................................................................45 5.4 Relations Between GPRS Network and Circuit Switching Service .....................................46 5.5 GPRS Service Quality .........................................................................................................47
Confidential Information of Huawei. No Spreading without Permission
Table of Contents
Chapter 6 GPRS Numbering Plan and Functions ........................................................................51 1.1 IMSI......................................................................................................................................51 6.2 P-TMSI.................................................................................................................................52 6.3 NSAPI/TLLI ..........................................................................................................................52 6.4 PDP Address and Type .......................................................................................................53 6.5 Tunnel Identifier (TID)..........................................................................................................53 6.6 Routing Area Identifier (RAI)................................................................................................53 6.7 Cell Identifier ........................................................................................................................54 6.8 GSN Address and Numbering .............................................................................................54 6.9 Access Point Name (APN)...................................................................................................54 Chapter 7 GPRS Entity Information Storage ................................................................................55 7.1 HLR......................................................................................................................................55 7.2 MS........................................................................................................................................56 7.3 GGSN ..................................................................................................................................56 7.4 SGSN...................................................................................................................................57 Chapter 8 GPRS Mobility Management Flow................................................................................59 8.1 Overview ..............................................................................................................................59 8.2 MM Status and MM Context ................................................................................................59 8.3 GPRS Attach/Detach ...........................................................................................................62 8.3.1 GPRS Attach .............................................................................................................62 8.3.2 GPRS Detach............................................................................................................62 8.4 GPRS Location Management Function ...............................................................................63 8.4.1 Cell Updating Procedure ...........................................................................................63 8.4.2 Routing Area Updating Procedure ............................................................................64 8.4.3 Periodical RA/LA Updating Procedure ......................................................................64 8.4.4 User Data Management Procedure ..........................................................................64 8.4.5 MS Class Mark Processing Function ........................................................................64 8.5 Security Management..........................................................................................................65 8.5.1 GPRS Authentication and Encryption .......................................................................65 8.5.2 P-TMSI Reallocation .................................................................................................65 8.5.3 User Data and GMM/SM Signaling Privacy ..............................................................65 Chapter 9 GPRS PDU Transmission..............................................................................................67 Appendix Frame Relay...................................................................................................................69 A.1 Frame Relay Concept ........................................................................................................69 A.2 Frame Relay Structure .......................................................................................................70 A.3 Frame Relay Working Principle..........................................................................................70 A.4 Congestion Control.............................................................................................................71 A.5 Frame Relay Technical Feature .........................................................................................72 A.6 FR Application on GPRS Gb Interface...............................................................................73
ii
Error! Style not defined.Error! Style not defined. Phase 3 07.60 user interworking 08.14 Gb layer-1 08.16 Gb-layer network service 08.18 BSSGP and Gb interface 09.16 Gb layer-2 09.18 Gb layer-3 09.60 Gn&Gp interface 09.61: External network interworking Major revised standards physical layer 08.58&08.60: Abis interface and TRAU frame structure change 09.02: Add the Gr and Gd protocols 11.10: TBR-19 MS test 11.2X BSS test 11.11 SIM 12.XX O&M
The GPRS, as a stepping-stone to 3G, would be developed in the following two phases after it is commercially available, according to the ETSIs proposal. Phase 1: enable GSM subscribers to access to data communication applications such as e-mail, and Internet using their mobile phones. Phase 2: EDGE GPRS (E-GPRS for short). In global mobile communication market, domestic mobile carriers have tested the waters of deploying GPRS multimedia services so that subscribers are accessible to a variety of banking functions including stock trading and bank transfer with their mobile phones. On December 21 2000, China Mobile Communications Corporation announces the formal launch of the construction of GPRS network known as Monternet in Beijing. To date, China Mobile has completed the two-phase project of the GPRS and made the GPRS commercially available in most cities of China.
Data rate
devices, software upgrade is involved. Connection duration, and number of channels occupied Circuit-switched-based; easy-to-plan and design for wireless and network
Amount of data transferred; connection duration and QoS. Lack of experience on wireless side; complex network planning after the increase of data traffic
The HSCSD is a type of service that enhances the radio interface data transmission rate by multiplexing several full-rate Traffic Channels (TCH). At present, the rate of the MSC switching matrix is 64 kb/s. Therefore the incoming switching rate must be less than 64 kb/s to avoid large alteration of the MSC. After the GSM network introduces the HSCSD service, the supported user data rate can be up to 38.4 kb/s (4 timeslots), 57.6 kb/s (Four timeslots; 14.4 kb/s channel coding) or 57.6 kb/s (six timeslotsTransparent data transmission service). The HSCSD is applicable to the realtime services, for example, video conference, while the GPRS the burst data service. As a type of circuit-switched data service, though the HSCSD supports the radio resource negotiation and adjustment (non-transparent transmission service) on the radio interface, yet a timeslot must be occupied even if there is no data transmission. When the data traffic increases, new BTSs or a large amount of radio channels shall be configured. With respect to the GPRS, the MS only requests the radio resource before transmitting data; in other time, the MS with the PDP context active does not request any radio resource. The network needs to judge the MS contention on the uplink link. Several MSs may share the radio resource of the same timeslot. The reuse of uplink resources may change along with the variation of the USF. On the downlink channel, the queuing mechanism is adopted so that several MSs can share the downlink resources, differentiated with the TFIs, of multiple timeslots. In respect of the network construction, the GPRS entails larger network alteration compared with the HSCSD, but the GPRS occupies the minimum erlang, minimizes the BTS investment and provides services even if no frequencies and cells are added. The operator can dynamically allocate the radio channels between the voice and data services based on the traffic load and actual requirements. Especially when idle channels are in the idle and burst state as a result of the setup, release and blocking of the circuits-switched calls, they can be utilized by the GPRS instead of the HSCSD. The HSCSD requires no change of hardware device except for the rate adaptation devices. but the GPRS necessitates the configuration of the SGSN and GGSN as well as software upgrade of the network devices such as HLR and so on. From the developments point of view, the GPRS network structure paves the way for the rd construction of the 3 generation mobile communication networks. The GPRS network will be primarily adopted as the 3G core network in the first phase.
The EDGE includes the Enhanced General Packet Radio Service (EGPRS) and ECSD. The compatibility and inheritability with the original GSM network are taken into full consideration when the EDGE is introduced to the GSM network, and therefore both ECSD and EGPRS have little impact on the core network. As an enhancement of the GPRS, the EGPRS has the following improvements over the GPRS:
1) 2)
Adopt the 8PSK modulation at the RF layer to greatly enhance the rate of a single channel. Modify the RLC/MAC at the link layer and define consummate link control algorithm.
The ECSD is introduced as an enhancement of the HSCSD. To provide the service of 57.6 kb/s, the HSCSD needs to bind four TSs, while ECSD only needs to bind two. The MSC is generally based on the circuit switching of 64 kb/s, but the rate design of the ECSD breaks through the 64 kb/s. The ECSD shall address the internal 64kb/s transmission problem of BSS through the frame numbering and reorganization at the receiving end.
4)
Though the GPRS dramatically enhances the spectrum utilization in comparison with the existing non-voice data service, yet it still cannot get rid of the following disadvantages:
Actual transmission rate is lower than the theoretical one: To reach the theoretical transmission rate of 171.2 Kbps, a subscriber shall occupy the whole 8 TSs without any error protection program. In practice, it is impossible for a single GPRS subscriber to occupy all TSs. In addition, there are constraints on the TS support capability of the GPRS terminals. Therefore, the theoretical maximum rate needs re-proving by taking account of the practical environmental constraints. The terminal does not support the wireless termination function. After a subscriber confirms the volume-based charging for the service contents when enabling the GPRS, the subscriber has to pay for undesired spam contents. Whether the GPRS terminal supports the wireless termination threatens the application and market exploration of the GPRS. The modulation is not optimal. The GPRS adopts the GMSK modulation mode. The EDGE employs a new modulation mode eight-phase-shift keying (8 PSK), and allows higher bit rate on the radio interface. The 8 PSK modulation is also used in the UMTS. Transmission delay: The GPRS packet switching technology transmits data in different directions but to reach the same destination, so the data of one or several packets may be lost during the radio link transmission.
5) 6) 7)
Introduce 3 major components to the GSM network: Serving GPRS Supporting Node (SGSN). Gateway GPRS Support Node (GGSN). Packet Control Unit (PCU). Perform software upgrade of related components of the GSM network.
BSC PCU
As shown in the above figure, the portable computer connects to the GPRS cellular phone through serial or radio mode. The GPRS cellular phone communicates with the BTS. Different from the circuit-switched data calls, the GPRS packets are transmitted from the BTS to the SGSN instead of being transmitted to the voice network through the MSC. The SGSN communicates with the GGSN. The GGSN handles the packet data before transmitting them to the destination network, for example, the Internet or X.25 network. Upon receiving the IP packets from the Internet with the MS address, the GGSN forwards them to the SGSN which then transmits the packets to the MS.
Table 1-2 lists the interfaces defined in the GPRS network architecture.
Table 2-1 List of interfaces defined in the GPRS network architecture Interface R Gb Gc Gd Gi Gn Gp Gr Gs Gf Um Description The reference point between the Mobile Terminal (MT) (for example, mobile phone) and the Terminal Equipment (TE) (for example, the portable computer). The interface between the SGSN and BSS. The interface between the GGSN and HLR. The interface between SMS and GMSC; the interface between SMS-IWMSC and SGSN The interface between the GPRS and external packet data The interface between SGSNs and between SGSN and GGSN in the PLMN. The interface between GSNs of different PLMNs. The interface between the SGSN and HLR. The interface between the SGSN and MSC/VLR. The interface between the SGSN and EIR. The interface between MS and GPRS network side
8)
GPRS MS
Confidential Information of Huawei. No Spreading without Permission
The GPRS MS consists of the TE and MT. The MS is actually an integrated MT after the TE functions are integrated into the MT. TE The TE, used to transmit and receive the packet data of the end user, refers to the computer operated and used by the end user. The TE can either be a stand-alone desktop computer, or integrated with the handset MT. In a sense, the GPRS network provides all functions for the sake of establishing a path between the TE and external data network to transmit packet data. MT The MT on the one hand communicates with the TE and on the other hand communicates with the BTS over the air interface. The MT can establish a logical link to the SGSN. The MT of the GPRS must be configured with the GPRS functional software to enable the GPRS. From the perspective of the TE, the MT acts as a modem for TE in the GPRS network. The functions of both MT and TE can be integrated to one physical device. MS The MS can be regarded as the device that integrates the functions of both MT and TE. It can either be an independent entity or two entities (TE + MT). The MS can be classified into the following three categories based on the capabilities of the MS and network: Class-A GPRS MS: The Class-A MSs can attach to the GSM and GPRS network simultaneously, activate and receive system messages from two systems, and implement Packet Switched Service (PS) and Circuit Switched Service (CS) concurrently. Class-B GPRS MS: The Class-B MSs are similar to Class A MSs with the exception that Class-B MSs will not support simultaneous traffic. If there is a circuit-switched call incoming to a Class-B MS, the MSC/VLR sends a Suspend message to the SGSN. Upon receiving the Suspend message, the SGSN suspends (temporarily terminates) the GPRS connection. After the circuit switching, the MSC/VLR then sends a Restore message to the SGSN to restore the GPRS connection. Class-C GPRS MS: The Class-C GPRS MSs cannot attach to the GPRS and GSM networks concurrently, and they only support manual switching between the PS and CS.
9)
As a processing unit added on the BSS side, the PCU implements the PS processing on the BSS side and management of packet radio channel resources. Currently the PCU networking structure includes the following three types: A. Integrated into the BTS; B. Integrated into the BSC; C. Independently configured, as shown in Figure 2-3. Huawei GPRS adopts the type C networking mode.
CCU
BSC PCU Gb
GSN C
a lot of bills. The CG is thus introduced to combine and pre-process bills before they are sent to the billing center so as to relieve the load on the billing center. In addition, NEs such as SGSN and GGSN do not have to interface with the billing center after the CG is configured.
10
SNDCP LLC RLC MAC GSM RF Um RLC MAC GSM RF BSSGP Network Service L1bis Gb
MS
BSS
SGSN
GGSN
1)
2)
3) 4)
GSM RF: The physical layer, the RF interfaces, enables data transmission over Um interface, while the LLC provides various logical channels for Um interface. The carrier bandwidth of the GSM Um interface is 200kHz, and a carrier is divided into 8 physical channels. RLC/MAC: Provides RLC and MAC functions. The RLC layer supports the acknowledged and unacknowledged transmission between the MS and BSS, and provides a reliable link independent of the radio solution. The MAC layer defines and allocates the GPRS logical channels of the Um interface so that they can be shared among MSs. The MAC also maps the LLC frames into the physical channel of the GSM. The RLC/MAC is standardized in the GSM04.60. SNDCP: Implements such functions as segmentation and compression of subscriber data. The SNDCP is defined in the GSM04.65. LLC: Provides end-to-end reliable error-free logical data links. Based on the High-level Data Link Control, the LLC provides highly reliable encrypted logical links. The LLC builds the LLC address and frame field on the SNDC data unit from the SNDC layer to generate the complete LLC frame. In addition, the LLC can implement point-to-multipoint addressing and data frame retransmission control, and support several types of QoS delay registration. The LLC is standardized in the GSM04.64.
11
5)
Base Station System Application GPRS Protocol (BSSGP) layer: Contains the functions of the network layer and partial functions of the transport layer, and interprets the routing and QoS information. The BSSGP is standardized in the GSM08.18. 6) Network Service: The data link layer protocol adopts the frame relay mode. The NS is standardized in the GSM08.16. 7) L1: Physical layer. 8) L2: Data link layer protocol. The common Ethernet protocols can be adopted. 9) IP: Network layer protocol, used for routing of subscriber data and control signaling. 10) UDP/TCP: Transport layer protocol. The UDP/TCP is used to set up the end-to-end reliable link. The connection-oriented TCP features the protection and traffic control functions to ensure accurate data transmission. As the non-connection-oriented protocol, the UDP provides no error recovery capability and only acts as the transmitter/receiver of datagram without concerning whether packets are correctly received. 11) GPRS Tunnel Protocol (GTP): The GTP transmits the packet data by utilizing the tunnel established between GSNs. The GTP is standardized in the GSM09.60.
MS-SGSN-GGSN
GSN-GSN
GGSN-HLR
12
Figure 3-3 Signaling plane between SGSN and HLR, EIR, and SMS-GMSC/ SMS-IWMSC
13
14
Defined in GSM0364
Physical RF MS Um
Physical RF Network
I. Physical layer
The physical layer consists of the physical RF and physical link sub-layers. The physical RF layer modulates and demodulates the physical waveform. It modulates the bit sequence received at the physical link layer into waveform, or demodulates the received waveform into the bit sequence required at the physical link layer. Defined by the GSM05 series specifications, the physical RF layer contains the following contents: Carrier frequency features and GSM channel structure; modulation mode of transmitting waveform and data rate of GSM channel; features and requirements of the transmitter and receiver. The physical link layer provides the information transmission services on the physical channel between the MS and network. Forward Error Correction (FEC) coding; detecting and correcting transmitted code words and providing indication of error code words; block interleaving; performing quadrature interleaving on the four consecutive burst TDMA frames. Radio channel measurement: Includes receive signal quality and level, measurement time advance, and physical link layer congestion detection. Wireless management: Includes cell selection and reselection, power control of transmitter, and battery power management, for example, the Discontinuous Reception (DRX) process.
1)
MAC layer
The MAC layer defines the process that several MSs share the transmission media (that is, PDCH). It also provides the MS contention arbitration and conflict avoidance, detection and recovery methods on the uplink. The contention arbitration is not required for the downlink transmission from network to several MSs. The MAC layer functions also allow a single MS to concurrently use several physical channels.
Confidential Information of Huawei. No Spreading without Permission
15
The MAC layer of the GPRS provides the following functions: Provide highly efficient data and signaling multiplexing on the uplink and downlink, and leave the multiplexing control to the network side. On the downlink, the multiplexing is controlled based on the scheduling mechanism; on the uplink, the multiplexing is controlled by allocating media to a single user. For the mobile-initiated channel access, the MAC layer performs contention arbitration for channel access attempts, including conflict detection and recovery. For the mobile-terminated channel access, the MAC layer allocate resources by the sequential access attempts. Priority handling
2)
RLC layer
The RLC functions define the process of selectively re-transmitting unsuccessfully transmitted RLC data blocks. The RLC/MAC layer provides the non-acknowledged and acknowledged operation modes. The RLC layer implements the assembly and disassembly of the LLC-PDU packets, and transmits data between peer layers over the sliding window protocol by adopting the acknowledged or non-acknowledged mode. The size of the RLC sliding window is 64. Huawei PCU supports the acknowledged and non-acknowledged modes of the RLC layer. It can specify the RLC modes of the uplink and downlink data transmission based on the MS requests and downlink LLC-PDU packet type respectively. If the acknowledged mode is adopted, each transmitted data block of the Temporary Block Flow (TBF) must be acknowledged by the peer; otherwise re-transmission is required. The TBF is released after all data are transmitted and acknowledged by the peer. If the non-acknowledged mode is adopted, the transmitted data blocks do not have to be acknowledged by the peer, and the lost or incorrectly transmitted data blocks are replaced with the fill bits. The TBF is released after the data transmission is complete.
3)
The radio block is the basic unit for radio transmission and allocation of radio resources. The RLC/MAC block consists of the MAC header, and RLC data block (or RLC/MAC control block) and generally contains four normal bursts. Each radio block consists of four consecutive TDMA frames. The transmission data and control information have different radio block structures, as shown in the following figure:
Radio block
MAC header RLC header RLC data
The control block is uniformly called the RLC/MAC control block because it contains the resource allocation information (handled at the MAC layer) and protocol ACK/NACK information (handled at the RLC layer).
16
3. LLC layer
LLC: Transport layer protocol. Based on the High-level Data Link Control, the LLC provides highly reliable encrypted logical links. The LLC builds the LLC address and frame field on the SNDC data unit from the SNDC layer to generate the complete LLC frame. In addition, the LLC can implement point-to-multipoint addressing and data frame retransmission control, and support several types of QoS delay registration. The LLC is standardized in the GSM04.64. Figure 3-10 shows the function model of the LLC layer.
GPRS Mobility Management Layer 3
LLGMM LLGMM LL3 LL5 LL9 LL11 LLSMS
SNDCP
SMS
LLC layer
Logical Link Entity SAPI=7
Multiplex Procedure
LLC layer
GRR BSSGP
MS
SGSN
BSSGP
BSSGP layer
The layer-3 users can adopt the SubNetwork Dependent Convergence Protocol (SNDCP), GMM/SM and SMS services. The LLC provides logical links for these services. The LLC frame structure is shown as follows:
17
Information field
8 PD
7 C/R
6 X
5 X
1 SAPI 0001 0011 0101 0111 1001 1011 Corresponding service GPRS Mobility Management (GMM) Subscriber data 1 Subscriber data 2 Short Message Service (SMS) Subscriber data 3 Subscriber data 4 SAP name GMM QoS1 QoS2 SMS QoS3 QoS4
SAPI
The PD (protocol indication bit) indicates whether current frame is an LLC frame or invalid frame. The C/R (command/response bit) indicates whether current frame is a command or response frame. The Service Access Point Identity (SAPI) contains 4 bits and 16 values. Currently only 6 values are adopted. The above figure shows the services in relation to the 6 values. The RLC Data Transmission Performance Measurement and LLC Data Transmission Performance Measurement in Huawei GPRS traffic measurement reflect the transmission features of the LLC layer.
4. SNDCP
The SNDCP is located between the network layer and LLC layer. It supports various network layers which share the same SNDCP. Therefore, the multivariate data from different data sources can pass the LLC layer. The SNDC implements the following functions: Map the SNDC primitive from the network layer to the LLC primitive of the LLC layer, or vice versa. Multiplex the N-PDUs from one or several NSAPIs into one LLC SAPI by adopting the multichannel technology. Compress the redundant control information and subscriber data. Segmentation and reassembling. Figure 3-12 shows the transmission platform of the SNDCP and LLC layers.
18
The SNDCP adopts the services provided by the LLC layer to multiplex the to-be-transmitted data from different sources. The Network layer Service Access Point Identifier (NSAPI) is the index of the PDP context. The PDP employs the services provided by the SNDCP layer. The PDP of the same type may have several PDP contexts and NSAPIs. Several different PDPs may adopt the same NSAPI, as shown in Figure 3-13.
3.3.2 Gb Interface
The Gb interface (Gb interface is the interface between the SGSN and PCU in Huawei GPRS network) is used to implement packet data transmission, mobility management and session management between the SGSN and the BSS/MS. The Gb interface is mandatory for the GPRS networking.
1)
The several physical layer configurations and protocols defined in GSM 08.14 are available here. The physical resources shall be configured through the Operation and Maintenance (O&M) process.
2)
The Frame Relay (FR) sub-layer of the Gb interface belongs to the NS Sub-Network Service protocol. The FR module enables the interworking of sub-network so that the PCU may connect to the SGSN through point-to-point connection or the frame relay
Confidential Information of Huawei. No Spreading without Permission
19
network. The point-to-point connection refers to the direct connection between the PCU and SGSN. Generally the PCU acts as the DTE and the SGSN the DCE. You may flexibly set the network features of the PCU and SGSN. Huawei PCU supports the above two connection modes. The link layer protocol of the Gb interface is based on the FR and defined in the GSM 08.16. Establish a FR virtual circuit between the SGSN and BSS, which is to be multiplexed by the LLC PDU from multiple subscribers. This virtual circuit may be multi-hop and traverse the network consisting of FR switching nodes. The frame relay is used for signaling and data transmission.
3)
The NS here particularly refers to the network service control part of the NS protocol. The NS layer protocol implements such functions as NS Service Data Unit (SDU) data transmission, NS-VC link management, load sharing of subscriber data and network congestion status indication and network status report. NS SDU data transmission All messages transmitted over the Gb interface are sent at the NS layer in the form of virtual circuit. The normal running of the NS layer guarantees the stable running of the upper layer protocols. In normal cases, the NS layer ensures the sequence of the NS SDUs transmitted through the Link Selection Parameters (LSP); in exceptional cases (for example, load sharing), the sequence cannot be well ensured. NS-VC status management The NS-VC status management involves such operations as resetting, blocking, unblocking and testing the NS-VC. If the BSS or SGSN wants to stop certain NS-VC, it sends a BLOCK message to the peer entity to block the NS-VC and switches the service on the NS-VC to other NS-VCs. If the BSS or SGSN wants to unblock certain NS-VC, it sends an UNBLOCK message to the peer entity to unblock the NS-VC, re-shares the load among services at the NS layer and informs the NS subscribers (for example, BSSGP layer) of the transport capability of the new NS layer. The status of either a new NS-VC established between peer NSs or a NS-VC reset upon the system failure is "Blocked and Activated. If the BSS or SGSN wants to detect whether the end-to-end communication on certain NS-VC exists, it can send a test message to the peer to test the connection. The test operation cannot be performed upon successful reset, and test messages are periodically re-transmitted. Load sharing of subscriber data. One of the most important functions of the NS layer is to perform load sharing of the subscriber data. When upper layer subscribers transmit data to the NS layer, the system allocates an LSP for each subscriber and encapsulates it to the data packet. The NS layer ensures the sequence of subscriber data transmission based on the LSPs. The NS layer selects one or several available NS-VCs to transmit the subscriber data packets based on the LSP and BVCI so that the load is shared among all unblocked NS-VCs of the same NSE. Congestion status indication Upon detecting the lower layer link failure or congestion, the NS layer notifies the NS layer subscribers through the congestion indication and status message, and at the same time informs them of the transmission capability of the NS layer so that the subscribers can handle accordingly.
4)
BSSGP layer
The BSSGP provides radio-specific data, QoS and selection information to satisfy the requirements of data transmission between the BSS and SGSN. In the BSS, it is used as the interface between the LLC frame and RLC/MAC block; in the SGSN, it is used as the interface between the RLC/MAC information and LLC frame. The BSSGP has a one-to-one relationship between the SGSN and BSS. That is, if a SGSN handles several BSSs, the SGSN must have a BSSGP in relation to each BSS.
20
Though distributed on both sides of the Gb interface, the BSSGP has asymmetrical functions on two sides of the Gb interface. The BSSGP implements the following functions: Signaling message and subscriber data transmission. Traffic control of downlink data. Blocking and unblocking of the BVC. Dynamic configuration and management of the BVC. Error detection of interface messages. The BSSGP contains the following basic procedures: Uplink and downlink data transmission procedure. Paging procedure. Radio access capability notification procedure. Radio access capability request and response procedure. Radio status procedure. Suspension and restoration procedure. FLUSH_LL (Logic Link) procedure. Traffic control procedure. Blocking and unblocking of the PTP BVC. Reset procedure of the BVC. Tracing procedure.
3.3.3 Gs Interface
As the interface between the SGSN and MSC/VLR, the Gs interface adopts the SS7 to carry the BSSAP+. The SGSN implements mobility management of the MS through the cooperation between the Gs interface and MSC, including such operations as joint Attach/Detach and update of joint routing area/location area. The SGSN also receives the CS paging information from the MSC and transmits it to the MS through the PCU. If the Gs interface is not introduced, the paging coordination and update of joint location area/routing area will be unavailable, and this hinders the improvement of connection rate and decrease of signaling load.
The GTP (core protocol of Gn/Gp interface) is adopted between the GSNs in the GPRS backbone network. The Gn refers to the interface between the SGSNs and between SGSN and GGSN in the same PLMN. The Gp refers to the interface used between GSNs of different PLMNs. The Border Gateway and firewall are added. The BG routing protocol is provided through the BG to implement the communication between GSNs of different PLMNs. The subscriber data and signaling between GSNs in the GPRS backbone network are transmitted by adopting the GTP. The GTP is standardized in the GSM09.60. The GTP signaling platform implements the GTP signaling processing, including session establishment, modification and deletion as well as tunnel maintenance. The GTP data transmission platform implements the GPRS tunnel encapsulation/ decapsulation and forwarding of packet data. Figure 3-14 shows the GTP message format: The first 20 bytes are the header.
21
Version: Protocol version bit. PT: Protocol type bit, including GTP and GTP. Spare bit: Set to 111 currently. N-PDU sequence number of the SNN and SNDCP. For the signaling message: SNN is 0; the SNN of the N-PDU transmitting end is 255, and that at the receiving end is omitted. For data N-PDU: If the SNN is set to 1, the GTP header contains SNDCP N-PDU SN; if the SNN is set to 0, the N-PDU will be transmitted in non-acknowledged mode at the LLC layer, and the N-PDU SN shall be set to 255. Message Type: Indicates whether the signaling message or data N-PDU tails the GTP header. For signaling message: Set based on the signaling message type (path management signaling message, tunnel management signaling message, location management signaling message and mobility management signaling message). For subscriber data N-PDU: Set it to 255. Length: Refers to the number of bytes (excluding header) of the GTP signaling or subscriber data packets. Sequence number: Refers to the incremental sequence number of the signaling messages and tunnel transmitted N-PDUs. Flow label: Refers to the flow flag. The flow label is not used in the path management and location management messages, and is thus set to 0; in the tunnel management and mobility management messages, the flow label is set in the signaling request message to indicate a GTP flow, exclusive of the established PDP and SGSN context request messages. In the data message, the flow label is used to identify the N-PDU flow. It is established and updated by the recipient in the context and selected in the case of SGSN change. TID: Refers to the tunnel ID. In the signaling message, the TID of path management, location management and mobility management messages is set to 0; in the tunnel management message, the TID indicates the destination GSN of the MM and PDP context. In the data messages, the TID indicates the tunnel where the N-PDU is located.
22
Information Elements /N-PDU The signaling message consists of the GTP header, followed by information elements. The data message prefixes a GTP header to the data N-PDU and encapsulates the message into the G-PDU so as to add subscriber-specific information, such as the IMSI, NSAPI and session-related flow label.
2)
The GTP signaling messages are transmitted over the UDP/IP. The subscriber packet data can be transmitted over the UDP/IP connectionless path or TCP/IP connection-oriented path. In addition, the GTP-based IP networking technology is adopted to encapsulate the IP addresses of the source and destination GSNs.
3.3.5 Gi Interface
The Gi interface refers to the interface between the GPRS and external PDN. The GPRS interconnects with various public packet networks such as Internet or ISDN through the Gi interface, on which such operations as protocol encapsulation/de-capsulation, address translation (for example, translating IP address of private network into that of public network), user access authentication and authorization shall be performed.
3.3.6 Gr Interface
As the interface between the SGSN and HLR, the Gs interface adopts the SS7 to carry the MAP+. The SGSN obtains the MS-related data from the HLR through the Gr interface. The HLR stores the GPRS subscriber data and routing information. In the case of update of inter-SGSN routing area, the SGSN will update related location information in the HLR. In the case of any data change, the HLR will also inform the SGSN to handle accordingly.
3.3.7 Gd Interface
The Gd refers to the interface between the SGSN and Short Message Service Gateway MSC (SMS-GMSC)/Short Message Service - InterWorking MSC (SMS-IWMSC). The SGSN receives short messages over the Gd interface and forwards them to the MS. The SMS of the GPRS is implemented through the coordination among the SGSN, SMS-GMSC, SMS-IWMSC and Short Message Center (SMC) over the Gd interface. If the Gd interface is not provided, the Class-C MSs cannot receive/transmit short messages after they attach to the GPRS network.
3.3.8 Gc Interface
As the interface between the GGSN and HLR, the Gc interface is used by the GGSN to request current SGSN address information of the subscriber from the HLR by using the IMSI when the network initiates service request to the MS. In mobile data service, this interface is used when the network initiates service request to the MS.
3.3.9 Gf Interface
As the interface between the SGSN and EIR, the Gf interface is used to authenticate the IMEI of the MS.
Confidential Information of Huawei. No Spreading without Permission
23
24
2)
PBCCH+PCCCH+PDTCH+PACCH+PTCCH PCCCH+PDTCH+PACCH+PTCCH PDTCH+PACCH+PTCCH PBCCH+PCCCH Where, PCCCH=PPCH+PRACH+PAGCH+PNCH If the PBCCH is required in the cell, the first combination is adopted, and only one such channel combination is allowed in a cell. If there is a large number of MSs in a cell, and the PCCCH is quite busy, one or several channels of the second combination can be configured. The presence of the first combination is the prerequisite for the existence of the second combination in a cell. The third combination is used for uplink and downlink packet data transmission. One or several channels in such a combination can be configured in a cell. The GPRS PCU supports all the above channel combinations. The channels in the third combination can be divided into fixed PDCH and dynamic PDCH. The fixed PDCH is used to transmit the GPRS packet data and cannot be preempted by the circuit switching service. The dynamic PDCH can be dynamically switched between the TCH and PDTCH based on the service requirements. The TCH is used during system initialization and may switch to the PDCH when there is a demand for packet switching service, or vice versa.
3)
The GPRS packet channel adopts the structure of 52 multiframes, and each packet channel contains 52 multiframes. Each four frames form a radio block. Therefore, a radio channel consists of 12 radio blocks and 4 idle frames, as shown in Figure 4-1.
T= Frames used for PTCCH; X= Idle frames Figure 4-1 PDCH multiframe structure
PBCCH: The PBCCH can be mapped onto such radio blocks as B0, B3, B6 and B9, with the number subject to the busy/idle degree of the broadcast channels. The mapping is performed based on the above sequence. PCCCH: The PAGCH and PPCH can be mapped onto any radio block (except the radio block occupied by the PBCCH) of the downlink channel. The PRACH is mapped onto the uplink frame in relation to the radio blocks occupied by PBCCH, PAGCH and PPCH. PDTCH: The PDTCH can be mapped onto any radio block to transmit packet data. PACCH: The PACCH can be mapped onto any radio block to transmit air interface radio signaling. PTCCH: The 12 and 38 uplink frames of each 52 multiframes constitute a uplink PTCCH and those of two neighboring 52 multiframes a downlink PTCCH.
th th
4)
Temporary Block Flow (TBF): A TBF is a physical connection used by the two RR peer entities (MS and BSS) to support the unidirectional transfer of LLC PDUs on packet
Confidential Information of Huawei. No Spreading without Permission
25
data physical channels. The TBF consists of the RLC/MAC block carrying one or several LLC PDUs, and only exists during data transmission. Temporary Flow Identity (TFI): Refers to the flag of the TBF. The TFI is used to differentiate the data flows when they share the same resources. One TFI is allocated for one TBF (each radio block contains one TFI), and it shall be unique among all TBFs in the same direction of the PDCH occupied by this TBF. The same TFI is allowed in the same direction of other PDCHs or in the reverse direction of current PDCH. The TBF is uniquely identified through the TFI and data transmission direction. The TFI contains 5 bits, with the value ranging from 0 to 31. Uplink State Flag (USF): The Uplink State Flag (USF) is used on PDCH channel(s) to allow multiplexing of uplink radio blocks (generally 4 consecutive burst pulses) from different mobile stations. The USF is transmitted in all downlink radio blocks to indicate the user of the next uplink radio block in the same timeslot. The USF contains three bits to indicate eight states. It can be used for multiplexing of uplink radio blocks. That is, eight MSs can be multiplexed in the same timeslot on the uplink channel through the USF, and the network dynamically adjust the uplink radio resources allocated to certain MS by changing the value of the USF. But on the PCCH, the value of the USF can only be 111 to indicate that related uplink radio blocks contain the PRACH.
Generally, the GPRS networks currently activated all support CS-1 and CS-2.
26
the RLC data block can adopt these four coding schemes, but the radio blocks adopting CS-1 does not contain the reserved part. For the radio blocks carrying RLC/MAC Control blocks, all control channels except PTCCH/U and PRACH adopt CS-1. The first step of the coding procedure is to add a Block Check Sequence (BCS) for error detection. For CS-1 - CS-3, the second step consists of pre-coding USF (except for CS-1), adding four tail bits and a convolutional coding for error correction that is punctured to give the desired coding rate. For CS-4 there is no coding for error correction.
1)
CS-1 coding
CS-1 is the same coding scheme as specified for SDCCH. Add 40 BCS bits to 184 information bits (including 3 USF bits) through the FIRE code, and then add four tail bits to constitute 228 bits. Then the 228 bits, after 1/2 convolutional coding, becomes 456 bits.
2)
CS-2 coding
Add 16 BCS bits used to detect errors in 271 information bits (including 3 USF bits), perform pre-coding for the 3 USF bits to get 6 bits, and add 4 tail bits to constitute 294 bits. Then the 294 bits, after the 1/2 convolutional coding, becomes 588 bits. Puncture 132 bits from the 588 bits to output 456 bits.
3)
CS-3 coding
Add 16 BCS bits used to detect errors in 315 information bits (including 3 USF bits), perform pre-coding for the 3 USF bits to get 6 bits, and add 4 tail bits to constitute 338 bits. Then the 338 bits, after the 1/2 convolutional coding, becomes 676 bits. Puncture 220 bits from the 676 bits to output 456 bits. Figure 4-2 shows the channel coding from CS-1 to CS-3.
puncturing
456 bits
Figure 4-2 Radio channel coding of CS-1 to CS-3
4)
CS-4 coding
Add 16 BCS bits used to detect errors in 431 information bits (including 3 USF bits), perform pre-coding for the 3 USF bits to generate 12 bits, and finally output 456 bits directly without performing convolutional coding. Figure 4-3 shows the channel coding of CS-4.
27
456 bits
Figure 4-3 Radio channel coding of CS-4
28
MCS-3 Family A 37 octets 37 octets MCS-6 MCS-9 MCS-3 34+3 octets Family A padding 34+3 octets 37 octets 37 octets
MCS-8
MCS-2 Family B 28 octets 28 octets MCS-5 MCS-7 MCS-1 Family C 22 octets 22 octets 28 octets 28 octets
MCS-4
Figure 4-4 General description of the EGPRS modulation and coding scheme
To allow incremental redundancy, the header part of the radio block is independently coded from the data part. Three different header formats are used, one for MCS-7, MCS-8 and MCS-9, one for MCS-5 and MCS-6, and one for MCS-1 to MCS-4. The first two formats are for 8-PSK modes, the difference being in the number Sequence Numbers carried (2 for MCS-7, -8 and -9; 1 for MCS-5 and -6). The third format is common to all GMSK modes. The header is always interleaved over four bursts. The following figures show the coding and truncating procedures in all modulation and coding schemes.
29
Rate 1/3 convolutional coding 36 bits 135 bits puncturing SB = 8 36 bits 124 bits 612 bits P1 1836 bits puncturing
612 bits P2
612 bits P3
612 bits P1
612 bits P2
612 bits P3
1392 bits
Figure 4-5 MCS-9 coding and truncating process; un-encoded 8-PSK; two RLC blocks per 20ms
3 bits USF
45 bits
564 bits TB
Rate 1/3 convolutional coding 36 bits 135 bits puncturing SB = 8 36 bits 124 bits 612 bits P1 1692 bits puncturing
612 bits P2
612 bits P3
612 bits P1
612 bits P2
612 bits P3
1392 bits
Figure 4-6 MCS-8 coding and truncating process; 8-PSK at 0.92 data rate; two RLC blocks per 20ms
30
Rate 1/3 convolutional coding 36 bits 135 bits puncturing SB = 8 36 bits 124 bits 612 bits P1 1404 bits puncturing
612 bits P2
612 bits P3
612 bits P1
612 bits P2
612 bits P3
1392 bits
Figure 4-7 MCS-7 coding and truncating process; 8-PSK at 0.76 data rate; two RLC blocks per 20ms
3 bits USF
33 bits
RLC/MAC HCS E Hdr.
Rate 1/3 convolutional coding 36 bits 99 bits +1 bit SB = 8 36 bits 100 bits 1836 bits puncturing P1 1248 bits P2 1248 bits
1392 bits
Figure 4-8 MCS-6 coding and truncating process; 8-PSK at 0.49 data rate; one RLC block per 20ms
31
Rate 1/3 convolutional coding 36 bits 99 bits +1 bit SB = 8 36 bits 100 bits 1404 bits puncturing P1 1248 bits P2 1248 bits
1392 bits
Figure 4-9 MCS-5 coding and truncating process; 8-PSK at 0.37 data rate; one RLC block per 20ms
3 bits USF
36 bits
RLC/MAC HCS E Hdr.
Rate 1/3 convolutional coding 12 bits 108 bits puncturing SB = 12 12 bits 68 bits 372 bits P1 1116 bits puncturing
372 bits P2
372 bits P3
464 bits
Figure 4-10 MCS-4 coding and truncating process; un-encoded GMSK; one RLC block per 20ms
32
Rate 1/3 convolutional coding 12 bits 108 bits puncturing SB = 12 12 bits 68 bits 372 bits P1 948 bits puncturing
372 bits P2
372 bits P3
464 bits
Figure 4-11 MCS-3 coding and truncating process; GMSK at 0.85 data rate; one RLC block per 20ms
3 bits USF
36 bits
RLC/MAC HCS E Hdr.
Rate 1/3 convolutional coding 12 bits 108 bits puncturing SB = 12 12 bits 68 bits 372 bits P1 672 bits puncturing
372 bits P2
464 bits
Figure 4-12 MCS-2 coding and truncating process; GMSK at 0.66 data rate; one RLC block per 20ms
33
Rate 1/3 convolutional coding 12 bits 108 bits puncturing SB = 12 12 bits 68 bits 372 bits P1 588 bits puncturing
372 bits P2
464 bits
Figure 4-13 MCS-1 coding and truncating process; GMSK at 0.53 data rate; one RLC block per 20ms
The USF has 8 states, which are represented by a binary 3 bit field in the MAC header. The USF is encoded to 12 symbols similarly to GPRS, (12 bits for GMSK modes and 36 bits for 8PSK modes). The Final Block Indicator (FBI) bit and extended (E) bit are encoded with the data part. The first step of the coding procedure is to add a Block Check Sequence (BCS) for error detection. The second step consists of adding six tail bits (TB) and a 1/3 rate convolutional coding for error correction that is punctured to give the desired coding rate. Each MCS uses different truncating schemes, which is represented by Pi, to give the desired coding rate. In both 8-PSK and GMSK modes, the stealing bits (SBs) in blocks denote block header formats. In 8-PSK mode, 8 SBs are used to denote four types of block header formats. In GMSK mode, 12 SBs are used to denote two types of block header formats, where the first 8 bits denotes CS-4. The details of the EGPRS coding schemes are shown in Table 4-2.
Table 4-2 Coding parameters for the EGPRS coding schemes RLC blocks per radio block (20ms) Raw data within one radio block Header code rate Data rate kb/s
Tail payload
Modulation
Code rate
Scheme
Family
BCS
HCS
8PSK
2 2 2 1
GMSK
1 1 1 1
2x592 2x544 2x448 592 544+48 448 352 296 272+24 224
A A B A B C A B
8 2x12 12 2x6 6
59.2 54.4 44.8 29.6 27.2 22.4 17.6 14.8 13.6 11.2
34
Error! Style not defined.Error! Style not defined. RLC blocks per radio block (20ms)
Tail payload
Modulation
Code rate
MCS-1
MCS-1 to MCS-4 apply the GMSK modulation mode, but their data rates differ from those of the CS-1 and CS-4. This data rate variation is specially designed for the link adaptation control algorithm of the EGPRS. The data in the GPRS can only be re-transmitted in the original coding mode, so it may never succeed in the case of radio transmission environment degradation. To address the problem, the EGPRS coding mode is designed to split a data block that originally uses a coding mode with a higher data rate into two data blocks using a coding mode with a lower data rate. For example, a RLC block using MCS-9 can be divided into two RLC blocks using MCS-6 during re-transmission. In view of the poor performance of MCS-9 in adverse radio transmission environment, the MCS-8 is designed with some protection capabilities and smaller valid data. MSC-8 and MCS-9 belong to the same family. When switching to MCS-3 or MCS-6, 3 or 6 padding octets shall be added to the data octets, respectively. According to the specification, all control logical channels in both EGPRS and GPRS adopt CS-1.
4.2.4 Channel Coding for PACCH, PBCCH, PAGCH, PPCH, PNCH and PTCCH/D
The channel coding for the PACCH, PBCCH, PAGCH, PPCH, PNCH, and PTCCH/D is corresponding to the coding scheme CS-1. The channel coding for the PTCCH/U is identical to PRACH.
Scheme
Family
BCS
HCS
0.53
0.53
176
8.8
35
them are 0), in turn, to the 11 information bits to get 21 bits. Use the 21 bits to perform 1/2 convolutional coding and get 42 bits. Puncture 6 bits (with serial numbers: 0, 2, 5, 37, 39 and 41) from the 42 bits to constitute 36 bits.
36
a maximum of four timeslot bundling is supported in uplink and downlink. The Class-A MSs at other multislot capability classes (1929) demonstrate a complicated physical layer design, so they probably will not be launched into market recently. Class-B MSs support the concurrent sending and receiving function, and generally their multislot capability is from class 1318. They support up to eight timeslot bundling in uplink and downlink. Class-A MSs are required to be able to transmit and receive uplink and downlink data at the same time, and be configured with two transceivers, so it adds difficulty in radio frequency design and enlarges the MS size. It is estimated that they may not be launched into market recently. The following table lists the multislot classes supported by the MS:
Table 4-3 Multislot classes supported by the MS Multislot class 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 1 2 2 3 2 3 3 4 3 4 4 4 3 4 5 6 7 8 6 6 6 6 6 8 8 8 8 8 8 Maximum number of slots Rx 1 1 2 1 2 2 3 1 2 2 3 4 3 4 5 6 7 8 2 3 4 4 6 2 3 4 4 6 8 Tx Sum 2 3 3 4 4 4 4 5 5 5 5 5 NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA Tta 3 3 3 3 3 3 3 3 3 3 3 2 NA NA NA NA NA NA 3 3 3 2 2 3 3 3 2 2 2 2 2 2 1 1 1 1 1 1 1 1 1 a) a) a) a) a) 0 b) b) b) b) b) b) b) b) b) b) b) Minimum number of slots Ttb 4 3 3 3 3 3 3 2 2 2 2 2 3 3 3 2 1 0 2 2 2 2 2 2 2 2 2 2 2 Tra 2 1 1 1 1 1 1 1 1 1 1 1 a) a) a) a) 0 0 c) c) c) c) c) c) c) c) c) c) c) Trb 1 1 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 1 1 1 1 1 1 1 1 1 1 1 Type
b) 1: with frequency hopping or change from Rx to Tx; 0: without frequency hopping and no change from Rx to Tx c) 1: with frequency hopping or change from Rx to Tx; 0: without frequency hopping and no change from Rx to Tx The meanings of multislot configuration parameters are as follows: Rx
Confidential Information of Huawei. No Spreading without Permission
37
Rx is the maximum number of downlink timeslots that can actually be used by an MS per TDMA frame, that is, the maximum number of timeslots used to receive data. The MS must be able to support the receiving timeslots of all the integer values from 0Rx (depending on the services supported by the MS). The receiving timeslots can be inconsecutive. For Class-A MS, several receiving timeslots must be allocated within a window at the size of Rx in a TDMA frame, in which the transmitting timeslots cannot be arranged among the receiving timeslots. Tx Tx is the maximum number of uplink timeslots that can actually be used by an MS per TDMA frame, that is, the maximum number of timeslots used to transmit data. The MS must be able to support the transmitting timeslots of all the integer values from 0Tx (depending on the services supported by the MS). The sending timeslots can be inconsecutive. For Class-A MS, several receiving timeslots must be allocated within a window at the size of Tx in a TDMA frame, in which the receiving timeslots cannot be arranged among the transmitting timeslots. Sum Sum is the total number of timeslots that can actually be used by an MS for uplink and downlink data transmission per TDMA frame. First the MS shall be subject to the restrictions on Tx and Rx. In addition, the MS must be able to support all combinations of integer values of Rx and Tx where 1 Rx + Tx Sum (depending on the services supported by the MS). For example, if Rx = 3, Tx = 3, and SUM = 4, it indicates that the MS uses three timeslots to receive data and at most one timeslot to send data. If two timeslots are used for data receiving, a maximum of two timeslots can be used for data sending. A maximum of four timeslots can be used concurrently. Sum is not applicable to all classes. Tta Tta relates to the time needed for the MS to perform adjacent cell signal level measurement and get ready to transmit. For Class-A MS, Tta is the minimum number of timeslots between the last transmitting or receiving timeslot and the next transmitting timeslot. During the period, the MS performs related measurement. For Class-B MS, Tta is not used. For CS multislot configuration, Tta is not used. Ttb Ttb relates to the time needed for the MS to get ready to transmit. Ttb is only used to select service instead of performing adjacent cell signal level measurement. For Class-A MS, Ttb is the minimum number of timeslots between the last receiving timeslot and the next transmitting timeslot, or between the last transmitting timeslot and the next transmitting timeslot. During the period, the frequency is changed. For Class-B MS, Ttb is the minimum number of timeslots after the transmission of the last burst in the last TDMA frame and before the transmission of the first burst in the next TDMA frame. Tra Tra relates to the time needed for the MS to perform adjacent cell signal level measurement and get ready to receive.
38
For Class-A MS, Tra is the minimum number of timeslots between the last transmitting or receiving timeslot and the next receiving timeslot. During the period, the MS performs related measurement. For Class-B MS, Trb is the minimum number of timeslots after the transmission of the last burst in the last TDMA frame and before the receiving of the first burst in the next TDMA frame. Trb Trb relates to the time needed for the MS to get ready to receive. Trb is only used to select service instead of performing adjacent cell signal level measurement. For Class-A MS, Trb is the minimum number of timeslots between the last transmitting timeslot and the next receiving timeslot. It stands for the minimum number of timeslots between the last receiving timeslot and the next receiving timeslot. During the period, the frequency is changed. For Class-B MS, Trb is the minimum number of timeslots after the receiving of the last burst in the last TDMA frame and before the receiving of the first burst in the next TDMA frame.
39
interface. After processing the response forwarded from the PCU, the SGSN starts to transmit downlink data.
1)
The network sends CS paging messages to the MS attached to the GPRS network on the GPRS paging channel or GPRS traffic channel. The MS only needs to monitor one paging channel. When allocated with a PDCH, the MS will receive the CS paging messages on the PDCH.
2)
The network sends CS paging messages to the MS attached to GPRS network on the CCCH, which can also be used for packet paging. The MS only needs to monitor one the CCCH paging channel. Even if the MS has been allocated with a PDCH, the CS paging messages are also received on the CCCH paging channel.
3)
The network sends CS paging messages to the MS attached to the GPRS network on the CCCH paging channel. The network can send GPRS paging messages on a PDCH (if configured in the cell) or CCCH paging channel. If a cell is configured with a PDCH,
40
the MS needs to monitor two paging channels to receive both CS and PC paging messages at the same time, because the network does not perform paging coordination. When the system is configured with the Gs interface, set the MS to operate in mode I; when the system is not configured with the Gs interface or PCCCH, set the MS to operate in mode II; when the system is configured with a PCCCH instead of Gs interface, set the MS to operate in mode III.
Table 4-4 Paging modes in different network operation modes Network operation mode CS paging channel PS paging channel Paging mode
PPCH
PPCH
PCH
PCH
PACCH
The Gs interface shall be used. For the MS attached to the GPRS network, if it is in the packet idle mode, the CS and PS paging messages will be transmitted on the same channel, so the MS only needs to monitor one paging channel; if it is in the packet transfer mode, the circuit paging can be transmitted on the PACCH. When the MS is in packet idle mode, all paging messages shall be transmitted on the PCH and the MS only needs to monitor the PCH; when the MS is in packet transfer mode, the circuit paging messages shall also be transmitted on the PCH. The circuit paging messages are only transmitted on the PCH. The packet paging messages can be transmitted on the PPCH (If a cell is configured with a PCCCH, the MS needs to concurrently monitor the PCH and PPCH.) or PCH.
II
PCH
PCH
III
PCH
PPCH
Note: The cells in the same routing area shall be configured with the same network operation mode.
41
In the short access and one phrase access mode, the MS is allocated with radio resources (such as TFI, dynamically allocated USF or fixedly allocated radio block bit list) once for all. In the two phrase access mode, the MS is allocated with one radio block in the first time, on which the MS transmits packet resource request messages. In the second time, the MS is allocated with resources again (including TFI, USF or radio block bit list), on which the MS starts to transmit data. As an 8 or 11 bit access burst, a packet channel request carries a little information. A packet resource request is a RLC/MAC signaling packet adopting the CS-1 and carries comparatively more information (including the TLLI and multislot capability of the MS and the radio priority), so it is of benefit for allocating appropriate resources for the MS. The PCU supports all the access types and treats the paging response, cell update and mobility management access types as two phrase access.
4.8.1 Relationship Between GPRS Cell Selection and GSM Cell Selection
In idle mode, the GPRS MS only execute the cell selection process. The cell selection procedure of GPRS is the same with that of GSM network. Before the allocation of the GPRS dedicated channel, the GPRS MS always uses the GSM signaling resources.
4.8.2 Relationship Between GPRS Cell Reselection and GSM Cell Reselection
The network does not provide the PBCCH and PCCCH. When the NC0 mode is adopted, the cell reselection procedure of GPRS is the same with that of GSM network. When an MS is in the GPRS Standby or Ready state, the MS executes the cell reselection. Only when the Class-A MS is in the CS mode, the network will select cells according to the handover process. After the CS is released, the MS starts to reselect cells. As a supplement of the GSM cell reselection algorithm, new algorithms C31 and C32 (used only when the system is configured with the PBCCH) in GPRS are proposed. If the PBCCH does not exist in the cell of the MS, the MS will monitor the system messages broadcasted by the BCCH and adopt the C1/C2 in CS mode to perform cell reselection.
42
4) 5) 6)
NC0: Refers to the general MS control mode. In this mode, the MS automatically reselects cells. NC1: Refers to the MS control mode with measurement reports. In this mode, the MS transmits measurement reports to network side and automatically reselects cells. NC2: Refers to the network control mode. In this mode, the MS transmits measurement reports to network side but does not automatically reselect cells.
NC1 and NC2 are only applicable when the MS is in Ready state. NC0 is applicable when the MS is in Standby state. Note: In current GPRS phase, only NC0 is applicable.
43
1)
The PTP service provides the transmission of one or more packets between two subscribes. The PTP service is started by the sender and received by the receiver. The PTP service includes point-to-point connectionless service (PTP-CLNS) and point-to-point connection service (PTP-CONS). The PTP-CLNS belongs to the datagram service and is applied to the burst non-interactive services. It is supported by the Connectionless Network Protocol (CLNP) and Internet Protocol (IP). The PTP-CONS is applied to the burst events and interactive application services. It is supported by the Connection Network Protocol (CONP) such as the X.25 protocol.
2)
With the PTM service, a message can be transmitted to multiple subscribers. The GPRS PTM service enables a subscriber to send a message to multiple subscribers who have the simple service request. The PTM service falls into the following types: Point-to-multiple broadcasting (PTM-M) service: With this service, a message can be sent to all subscribers in a certain area. It is a unidirectional service and it cannot ensure that all subscribers can receive the message. The time for providing the packet data and the QoS is determined based on the negotiation between the GPRS operator and the PTM-M provider. The retransmission is determined based on the forehand negotiated plan. Point-to-multiple group broadcasting (PTM-G) service: With this service, a message can be sent to a specific subscriber group in a certain area. It can be unidirectional, bidirectional, or multi-directional. This service is suitable for the mobile data communication, that is, providing bidirectional communication for data subscribers in a corporation. It is widely used in the dispatch management, taxi dispatch management, corporation secret information management, and special news broadcasting. IP multipoint broadcasting (IP-M) service: It is part of the IP series service. With this service, a message can be transmitted among the IP-M participants. An IP-M subscriber can be a fixed-line or mobile IP subscriber. The service area is not defined. An IP-M subscriber can be a subscriber in the PLMN or a group of subscribers in the Internet. Addition to providing the PTP and PTM services, the GPRS network provides various telecom services such as E-MAIL, WAP, stock information, and portable office.
44
1) 2) 3)
Call forwarding unconditional (CFU) and call forwarding on MS unreachable (CFNR) services Closed user group (CUG) service Advice of charge information (AoCI) and advice of charge charging (AoCC) services
Currently, the special service provided by the GPRS network to the subscriber is the barring of GPRS interworking profiles (BGIWP) service. With this service, the interworking profiles are barred, thus preventing subscribers from accessing the external data networks.
1)
Messaging service
With the messaging service, versatile contents can be provided to the mobile subscribers, such as stock price, sports news, whether forecast, flight information, news headlines, entertainment, and transportation information.
2)
Chatting service
Currently, Internet chatting group is much more popular. Under the coordination with Internet, the GPRS network enables mobile subscribers to join the internet chatting group.
3) 4)
With this service, file sharing and remote coordination become much easier. People in different places can work with the same shared files.
5)
The non-voice mobile service can be used to assign new tasks to employees of outgoing in business and keep contact with them.
6)
Through extending the enterprise E-mail system on the PC of the employee office, the employee can keep contact with the office. In this case, the E-mail sent to the PC can be forwarded to the mobile terminal, thus expanding the applications of the E-mail.
7)
The internet E-mail can be converted to a gateway service which cannot be stored by messages or the mailbox service that can store any information. In the situation of gateway charging the service, the radio mail platform can convert information from SMTP to SMS and then sends the information to the SMSC.
8) 9)
Still images such as photos, figures, post cards, greeting cards, and lecture paper can be sent and received over the mobile network.
45
2)
Network operation modes Mode I: In this mode, the network sends the CS message to the mobile station on the GPRS paging channel. This means that the MS monitors one PCCCH only. The MS can receive and exchange paging messages on this channel. Mode II: In this mode, the network sends the CS and packet messages to the MS on the CCCH paging channel. This means that the MS monitors one CCCH only. Even the MS has been allocated a PCCCH, the packet message can still be broadcasted on the CCCH. Mode III: In this mode, the CS message is sent on the CCCH. The packet message can be sent either on the CCCH or on the PCCCH. This means the MS simultaneously monitors these two paging channels in order to receive both the CS and packet messages.
When there is the Gs interface in the network, the network works in mode I. all GPRS MS CS calls are sent by the MSC to the SGSN. After that, the SGSN universally initiates the paging. Huawei GPRS system supports the Gs interface. When there is no Gs interface, the network works in mode II or mode III. In mode II, the network does not allocate the PCCCH. In mode III, if there is the PCCCH in the network, the GPRS paging is broadcasted on this channel.
3)
When there is the Gs interface in the network, the relation is set up between the MSC and the SGSN for coordinating the CS and packet services. Three types of GPRS terminals are used under the coordination with the three network operation modes to implement the relations between the CS and packet services. Combined GPRS/IMSI attach and detach function To implement the GPRS/IMSI attach function, the MS sends a request to the SGSN, the SGSN notifies the MSC/VLR, and then the MSC/VLR stores the SGSN address. After the MS implements the IMSI attach function, it also implements the GPRS attach function, thus saving the radio resources. Combined route area/location update function When an MS changes its route/location, the MS sends a request to the SGSN, the SGSN sends the location update request to the VLR, and then the VLR knows the MS location change. Circuit switching function implemented by GPRS
46
When receiving a paging or a message from the MS, the MSC/VLR sends a paging request to the SGSN, and then the SGSN sends the CS paging through the GPRS paging channel and the packet data channel. The MS responds the paging through the normal CS channel.
1)
2)
Throughput: It refers to the number of the packet data units (PDU) successfully transmitted in a specified reference time duration in one direction. The Q933 protocol defines the time duration as a second and PDU as the packet data units in one domain. Successfully" indicates that there is no error in the FCS. Transmission delay: It refers to the transmission time measured between two edge points of a specified reference point, that is, the difference of the time when the first bit of a PDU passes the left edge point and the time when the last bit of the PDU passes the right edge point.
When being registered, the subscriber subscribes the default QoS script. Each PDP context has an independent QoS script associated with it. When the PDP context is activated, the MS negotiates with the network on the QoS script. The MS can ask for the QoS that is different from the subscribed QoS. In addition, the QoS class is determined when the subscriber is registered. The QoS service parameters include service precedence class, reliability class, delay class, and throughput class. The following table lists the service parameters that are required to be registered when providing the different GPRS services and the registration method. Generally, negotiation method is adopted.
Table 5-1 QoS service parameter registration list Service Parameter Service precedence (priority) Reliability Throughput Simultaneous use class Delay Security management / encryption Interworking profile Password Active / De-active Geographical area Address 2, 3, 4 2, 4 2, 3, 4 2 2, 4 1 2 3 N/A PTP-CLNS PTP-CONS
Note: Network determined User determined per subscription and registration User determined per request Negotiable
47
1. Service precedence (priority) class In normal case, relevant service precedence class is kept. In case of resource shortage or network congestion, the network determines which packets should be discarded based on the service precedence. The GPRS service defines three precedence classes, that is, high precedence, normal precedence, and low high precedence. 2. Reliability class The QoS context reliability parameters indicate the transmission features required by subscriber applications. They define the loss possibility, retransmission possibility, non-sequential reaching possibility, and error possibility of the service data units. The subscriber application based on the X.25 protocol requires higher reliability class. Low reliability class results in errors. The reliability is classified into three classes in the ETSI0260 protocol, as listed in Table 5-2.
Table 5-2 Data reliability class Reliability class 1 10 Lost SDU probability
-9
Example of application characteristics. Error sensitive, no error correction capability, limited error tolerance capability. Error sensitive, limited error correction capability, good error tolerance capability. Not error sensitive, error correction capability and/or very good error tolerance capability.
10
-4
10
-5
10
-5
10
-6
10
-2
10
-5
10
-5
10
-2
The reliability is classified into five classes in the 3GPP TS 24.008 protocol. The coding mode is listed in Table 5-3.
Table 5-3 Data reliability class Value 000 001 010 011 110 100 101 111 1 2 3 4 5 Reliability class Description In the direction from MS to network, refer to signatured Reliability, in the direction from network to MS, reserved. Acknowledged GTP, LLC, and RLC; Protected data Unacknowledged GTP; Acknowledged LLC and RLC, Protected data Unacknowledged GTP and LLC; Acknowledged RLC, Protected data Unacknowledged GTP, LLC, and RLC, Protected data Unacknowledged GTP, LLC, and RLC, Unprotected data Reserved
3. Delay class When bearing data service, the GPRS network does not adopt the store-and-forward mode. In this case, the transmission delay of the packet data is restricted by the network transmission technology. The QoS-defined delay refers to the maximum mean delay and the maximum delay of 95% packets when the end-to-end packet data passes the GPRS network. The GPRS-defined delay is classified into four (14) classes.
48
Classes 13 indicate that reliable performance can be provided. Class 4 indicates that the minimum performance can be ensured. The network must support at least class 4 (Best Effort). Currently, most networks support class 4. The coding and the meaning in the 3GPP TS 24.008 protocol are listed in Table 5-4.
Table 5-4 Delay class 128-byte SDU Value 001 010 011 100 000 Other 4 (most effective value) Delay class 1 (measured value) 2 (measured value) 3 (measured value) 4 (most effective value) Mean delay (S) 95% delay (S) 1024-byte SDU Mean delay (S) 95% (S)
<0.5 <1.5 <2 <7 <5 <25 <15 <75 <50 <250 <75 <375 Not defined In the direction from the MS to the network, it indicates the subscribed delay class; in the direction from the network to the MS, this value is reserved. Not defined
4. Throughput class The throughput class includes the peak throughput class and the mean throughput class. The peak throughput class defines the maximum data rate for transmitting each PDP context in the network. The peak throughput can be classified into nine classes. It refers to the number of transmitted bytes measured every minute on the R (interface between the MS and the TE) and the Gi reference points. The coding and the meaning in the 3GPP TS 24.008 protocol are listed in Table 5-5.
Table 5-5 Peak throughput class Value 0001 0010 0011 0100 0101 0110 0111 1000 1001 0000 Other Peak throughput class 1 2 3 4 5 6 7 8 9 1 Peak throughput (byte/s) Up to 1000 (8kbit/s) Up to 2000 (16kbit/s) Up to 4000 (32kbit/s) Up to 8000 (64kbit/s) Up to 16000 (128kbit/s) Up to 32000 (256kbit/s) Up to 64000 (512kbit/s) Up to 128000 (1024kbit/s) Up to 256000 (2048kbit/s) In the direction from the MS to the network, it indicates the subscribed peak throughput class; in the direction from the network to the MS, this value is reserved. Up to 1000 (8kbit/s)
The mean throughput class defines the expected mean transmission rate in the remaining time after the PDP context is activated when the data is transmitted in the GPRS network. For the convenience of charging, even the network is able to provide more throughputs, it restricts the throughput to a certain class. The mean throughput is classified into 19 classes. It refers to the number of transmitted bytes measured every hour on the R and the Gi reference points. The time measured includes the idle time (no data transmission). The most effective value indicates the negotiable throughput class determined based on the MS requirement and the network
49
available resource. The coding and the meaning in the 3GPP TS 24.008 protocol are listed in Table 5-6.
Table 5-6 Mean throughput class Value 00001 00010 00011 00100 00101 00110 00111 01000 01001 01010 01011 01100 01101 01110 01111 10000 10001 10010 1110 11111 00000 Other 31 Peak throughput class 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 Reserved 31 Peak throughput (byte/s) 100 (~0.22bit/s) 200 (~0.44bit/s) 500 (~1.11bit/s) 1000 (~2.2bit/s) 2000 (~4.4bit/s) 5000 (~11.1bit/s) 10000 (~22bit/s) 20000 (~44bit/s) 50000 (~111bit/s) 100000 (~0.22kbit/s) 200000 (~0.44kbit/s) 500000 (~1.11kbit/s) 1000000 (~2.2kbit/s) 2000000 (~4.4kbit/s) 5000000 (~11.1kbit/s) 10000000 (~22kbit/s) 20000000 (~44kbit/s) 50000000 (~111kbit/s) __ Best effort. (best effective throughput) In the direction from the MS to the network, it indicates the subscribed peak throughput class; in the direction from the network to the MS, this value is reserved. Best effort. (best effective throughput)
50
X.25 TLLI SGSN address SGSN No. TID=IMSI+ NSAPI GGSN address GGSN No. IP
6.1 IMSI
It is the same as the original GSM subscriber, all GPRS subscribers (except for anonymous access subscribers) must have their own IMSIs. Note: Anonymous access: A special MS can access the network in anonymous mode without the IMSI/IMEI authentication and encryption. All fee of an anonymous access is paid by the called party. The network
51
operators decide whether to support the anonymous access service. Currently, China GSM network does not support the reverse charging service.
6.2 P-TMSI
P-TMSI: When the MS is attached to the GPRS network, the SGSN allocates a temporary P-TMSI to the MS for the packet call. The SGSN must combine the P-TMSI and the IMSI. It can be regarded that an MS is allocated with two identities.
6.3 NSAPI/TLLI
The network layer service access point identifier/temporary link level identity (NSAPI/TLLI) is used for routing in the network layer in a paired mode. To an MS, an NSAPI/TLLI is unique in a routing area (RA). The NSAPI indicates the address of the SubNetwork Dependent Convergence Protocol (SNDCP) level in the PDP application layer, as shown in Figure 6-2.
Signalling
SM S
SN CP D
N PI SA
LLC
Figure 6-2 SNDCP-LLC layer structure
Either the X.25 or the IP protocol has its own NSAPI. Temporary link level identity (TLLI): It uniquely identifies a logical link between the MS and the SGSN in an RA. In an RA, the TLLI maps the IMSI one to one. It is directly generated by the SGSN or generated based on the P-TMSI. If the network allocates a new P-TMSI, the TLLI is also updated. A TLLI contains 32 bits and bit 0 is the least-significant bit. A TLLI can be generated in the following mode: local, foreign, random, and auxiliary. The MS and the SGSN judges which mode a TLLI belongs to based on the TLLI institute.
Table 6-1 TLLI coding mode 31 1 1 0 0 0 30 1 0 1 1 1 29 T T 1 1 1 28 T T 1 1 0 27 T T 1 0 X 26~0 T T R A X TLLI type Local TLLI Foreign TLLI Random TLLI Auxiliary TLLI Reserved
52
Note: In the above table, T indicates that the TLLI is generated from the P-TMSI, A" from SGSN, R from random mode, and X indicates reserved.
1) 2)
It is generated from the P-TMSI and valid only in its associated RA. It is derived from the P-TMSI and another RA. It is reported to the SGSN when the RA of the MS is updated.
3) 4)
If an MS has no valid P-TMSI (new access MS), a random TLLI is provided by itself. It is chosen by the SGSN to provide the identity for an anonymous access MS.
53
1) 2)
APN network identifier: This identifier is mandatory and allocated by the network operator to the ISP or company. It is identification with the same function as an Internet domain name. APN operator identifier: This identifier is optional in the form of xxx.yyy.gprs (such as MNC.MCC.gprs) and used to indicate the homed network.
The APN network identifier is stored in the HLR as subscriber subscription data. When initiating the packet service, a subscriber can provide the APN to the SGSN for the SGSN to choose the GGSN to be accessed and for the GGSN to judge the external network to be accessed. In addition, the HLR can store a wildcard. In this case, the subscriber or the SGSN can choose an APN that is not stored in the HLR. A subscriber can choose different GGSNs based on the APNs. That is to say, one subscriber can activate multiple PDP contexts, and each context is associated with one APN. The purpose of the subscriber choosing different APNs is to choose the external networks based on different GGSNs.
54
Table 7-1 lists the GPRS subscription data stored in the HLR.
Table 7-1 HLR data storage list Field IMSI MSISDN SGSN Number SGSN Address SMS Parameters Description
IMSI is the main reference key. The basic MSISDN of the MS. The SS7 number of the SGSN currently serving this MS. The IP address of the SGSN currently serving this MS. SMS-related parameters, e.g., operator-determined barring. Indicates that the MM and PDP contexts of the MS are deleted MS Purged for GPRS from the SGSN. Indicates that the MS is not reachable through an SGSN, and that MNRG the MS is marked as not reachable for GPRS at the SGSN and possibly at the GGSN. The GSN number and optional IP address pair related to the GGSN that shall be contacted when activity from the MS is detected and MNRG is set. The GSN number shall be either the GGSN-list number of the GGSN or the protocol-converting GSN as described in the subclauses "MAP-based GGSN- HLR Signaling" and "GTP and MAP-based GGSN- HLR Signaling". Each IMSI contains zero or more of the following PDP context subscription records: PDP Context Identifier Index of the PDP context. PDP Type PDP type, e.g., X.25, PPP, or IP. PDP address, e.g., an X.121 address. This field shall be empty if PDP Address dynamic addressing is allowed. A label according to DNS naming conventions describing the Access Point Name access point to the external packet data network. The quality of service profile subscribed. QoS ProfileSubscribed is QoS Profile Subscribed the default level if a particular QoS profile is not requested. Confidential Information of Huawei. No Spreading without Permission
55
Specifies whether the MS is allowed to use the APN in the domain of the HPLMN only, or additionally the APN in the domain of the VPLMN.
7.2 MS
Table 7-2 MS data storage list Field IMSI MM State P-TMSI P-TMSI Signature Routing Area Cell Identity Kc CKSN Ciphering algorithm Classmark DRX Parameters SIM X Description
International Mobile Subscriber Identity. Mobility management state, IDLE, STANDBY, or READY. X Packet Temporary Mobile Subscriber Identity. X A signature used for identification checking purposes. X Current routing area. Current cell. X Currently used ciphering key. X Ciphering key sequence number of Kc. Selected ciphering algorithm. MS classmark. Discontinuous reception parameters. The RLC/MAC radio priority level for uplink SMS Radio Priority SMS transmission. Each MM context contains zero or more of the following PDP contexts: PDP Type PDP type, e.g., X.25, PPP, or IP. PDP Address PDP address, e.g., an X.121 address. PDP State Packet data protocol state, INACTIVE or ACTIVE. Specifies whether the MS is allowed to use a dynamic Dynamic Address Allowed address. APN Requested The APN requested. NSAPI Network layer Service Access Point Identifier. TI Transaction Identifier. QoS Profile Requested The quality of service profile requested. QoS Profile Negotiated The quality of service profile negotiated. The RLC/MAC radio priority level for uplink user data Radio Priority transmission. SNDCP sequence number of the next uplink N-PDU to be Send N-PDU Number sent to the SGSN. SNDCP sequence number of the next downlink N-PDU Receive N-PDU Number expected from the SGSN.
7.3 GGSN
Table 7-3 GGSN data storage list Field IMSI NSAPI MSISDN PDP Type PDP Address Dynamic Address APN in Use QoS Profile Negotiated SGSN Address Description International Mobile Subscriber Identity. Network layer Service Access Point Identifier. The basic MSISDN of the MS. PDP type, e.g., X.25, PPP, or IP. PDP address, e.g., an X.121 address. Indicates whether PDP Address is static or dynamic. The APN Network Identifier currently used. The quality of service profile negotiated. The IP address of the SGSN currently serving this MS.
56
OMQ000001 GPRS Fundamentals ISSUE 1.0 Field MNRG Recovery SND SNU Charging Id Reordering Required
Error! Style not defined.Error! Style not defined. Description Indicates whether the MS is marked as not reachable for GPRS at the HLR. Indicates if the SGSN is performing database recovery. GTP sequence number of the next downlink N-PDU to be sent to the SGSN. GTP sequence number of the next uplink N-PDU to be received from the SGSN. Charging identifier, use to identify charging records generated by SGSN and GGSN. Specifies whether the GGSN shall reorder N-PDUs received from the SGSN.
7.4 SGSN
Table 7-4 SGSN data storage list Field IMSI MM State P-TMSI P-TMSI Signature IMEI MSISDN Routing Area Description
IMSI is the main reference key. Mobility management state, IDLE, STANDBY, or READY. Packet Temporary Mobile Subscriber Identity. A signature used for identification checking purposes. International Mobile Equipment Identity The basic MSISDN of the MS. Current routing area. Current cell in READY state, last known cell in STANDBY or Cell Identity IDLE state. Time elapsed since the last LLC PDU was received from the Cell Identity Age MS at the SGSN. VLR Number The VLR number of the MSC/VLR currently serving this MS. The IP address of the new SGSN where buffered and not New SGSN Address sent N-PDUs should be forwarded to. Authentication Triplets Authentication and ciphering parameters. Kc Currently used ciphering key. CKSN Ciphering key sequence number of Kc. Ciphering algorithm Selected ciphering algorithm. Radio Access Classmark MS radio access capabilities. SGSN Classmark MS network capabilities. DRX Parameters Discontinuous reception parameters. Indicates whether activity from the MS shall be reported to MNRG the HLR. Indicates whether activity from the MS shall be reported to NGAF the MSC/VLR. Indicates whether paging for GPRS and non-GPRS PPF services can be initiated. SMS Parameters SMS-related parameters, e.g., operator-determined barring. Recovery Indicates if HLR or VLR is performing database recovery. The RLC/MAC radio priority level for uplink SMS Radio Priority SMS transmission. Each MM context contains zero or more of the following PDP contexts: PDP Context Identifier Index of the PDP context. PDP State Packet data protocol state, INACTIVE or ACTIVE. PDP Type PDP type, e.g., X.25, PPP, or IP. PDP Address PDP address, e.g., an X.121 address. APN Subscribed The APN received from the HLR. APN in Use The APN currently used. NSAPI Network layer Service Access Point Identifier. TI Transaction Identifier. Confidential Information of Huawei. No Spreading without Permission
57
OMQ000001 GPRS Fundamentals ISSUE 1.0 Field GGSN Address in Use VPLMN Address Allowed QoS Profile Subscribed QoS Profile Requested QoS Profile Negotiated Radio Priority Send N-PDU Number Receive N-PDU Number SND SNU Charging Id Reordering Required
Error! Style not defined.Error! Style not defined. Description The IP address of the GGSN currently used. Specifies whether the MS is allowed to use the APN in the domain of the HPLMN only, or additionally the APN in the domain of the VPLMN. The quality of service profile subscribed. The quality of service profile requested. The quality of service profile negotiated. The RLC/MAC radio priority level for uplink user data transmission. SNDCP sequence number of the next downlink N-PDU to be sent to the MS. SNDCP sequence number of the next uplink N-PDU expected from the MS. GTP sequence number of the next downlink N-PDU to be sent to the MS. GTP sequence number of the next uplink N-PDU to be sent to the GGSN. Charging identifier, used to identify charging records generated by SGSN and GGSN. Specifies whether the SGSN shall reorder N-PDUs before delivering the N-PDUs to the MS.
58
1) 2)
The GMM specific function includes: GPRS attach/detach, cell updating, and RA updating. The GMM security function includes: GPRS authentication and encryption, P-TMSI re-allocation, user data, and GMM/SM signaling confidentiality.
1)
DILE status
In GPRS IDLE status, the subscriber is not attached to the GPRS mobility management. The MM context in the MS and the SGSN does not contain the valid location and routing information of the subscriber. In this case, the mobility management cannot be implemented. Under this condition, data cannot be transmitted between the SGSN and the subscriber. The GPRS MS is regarded unreachable. To set up the MM context between the MS and the SGSN, the MS must be attached to the GPRS mobility management.
2)
STANDBY status
In STANDBY status, the MS is attached to the GPRS mobility management and the MM context is set up between the MS and the SGSN. The MS can receive the PS paging and the CS paging through the SGSN. However, the data cannot be transmitted between the MS and the SGSN. In this status, the MS can perform the GPRS RA updating and GPRS cell selection and reselection. When moving to a new RA, the MS implements the mobility management to notify the SGSN. When moving from one cell to another in the same RA, the MS
Confidential Information of Huawei. No Spreading without Permission
59
does not notify the SGSN. Therefore, the SGSN MM context contains the GPRS RAI location information only. In STANDBY status, the MS can activate or deactivate the PDP context. Before sending or receiving the data, the MS must activate a PDP context. When the MS is in STANDBY status, the paging is implemented in the RA level. When receiving a paging of the PDP or the PTM-G data but the MS is in STANDBY status, the SGSN checks whether the paging proceed flag (PPF) is reset. If yes, the SGSN sends the paging request message to the RA where the MS resides; If not, the SGSN does not send the request. If the MS answers the paging, its MM context changes to READY status. After the SGSN receives the paging response message, its MM context changes to READY status. Similarly, if the MS sends the data or signaling, its MM context changes to READY status. After the SGSN receives the data or signaling, its MM context changes to READY status. When the MS or the network await-order timer expires, the SGSN implements the implicit detachment to change the MM identifier of the MS to IDLE status. After that, the SGSN removes the MM of the MS and the PDP context.
3)
READY status
In READY status, the SGSN implements the mobility management on the MS in a specific cell level. Through the mobility management process, the MS provides its current cell information to the network. The selection and reselection of the GPRS cell can be implemented by the MD or by the SGSN. The cell global identification (CGI: containing RAC and LAC) is contained in the BSSGP header of the packet sent by the MS. In this status, the MS can send or receive the PDP PDU and the SGSN can initiate non-GPRS paging to the MS. The data sent by the SGSN is forwarded by the BSS to the GPRS cell where the MS resides. In READY status, the MS can activate or deactivate the PDP context. When the MS is in READY status, the paging is implemented in the cell level. During the packet data transmission, the MM context is in READY status. When the data transmission is complete, the MM context is still in READY status. READY status is controlled by a ready timer (T3314). If the timer expires, the MM context changes from READY status to STANDBY status. When the MS initiates GPRS detach, the MM context changes from READY status to IDLE status.
60
IDLE
IDLE
GPRS Attach
GPRS Detach
GPRS Attach
READY
READY
PDU transmission
PDU reception
STANDBY
STANDBY
MM State Model of MS
1)
GPRS attach: The MS requires accessing the GPRS service and sets up a logical link to the SGSN. The MM context is created in the MS and the SGSN respectively.
2)
From STANDBY to IDLE Implicit detach: The MM context in the SGSN returns to IDLE status and the PDP context returns to INACTIVE status. The MM and PDP contexts in the SGSN and the PDP context in the MS will be removed. Location removal: After receiving a MAP location removal message from the HLR, the SGSN removes the MM and PDP contexts. From STANDBY to READY PDU sending: The MS sends an LLC PDU message to the SGSN as the response to the paging. PDU receiving: The SGSN receives an LLC PDU sent by an MS.
3)
61
4)
From READY to STANDBY Ready timer timeout: The MM contexts in the MS and the SGSN return to STANDBY status. Forced to STANDY status: Before the ready timer expires, the SGSN sends a message, indicating the MM context to return to STANDBY status. Abnormal RLC: To avoid radio interface transmission failure or irrevocable radio transmission faults, the MM context of the SGSN returns to STANDBY status. From READY to IDLE GPRS detach: The MS or the network requires that the MM context returns to IDLE status, and the PDP context returns to INACTIVE status. The SGSN removes the MM and PDP contexts. The PDP context in the SGSN is removed. Location removal: After receiving a MAP location removal message from the HLR, the SGSN removes the MM and PDP contexts.
5)
62
When an MS that is in READY status moves from one cell to another in the same RA, the MS initiates the cell updating procedure. The cell updating procedure is described as follows: The MS sends a free-type upstream LLC frame that contains the MS ID to the SGSN to initiate the cell updating procedure. After receiving the LLC frame, the BSS adds the CGI (RAC+LAC) of the new cell to the header of the BSSGP frame and then sends the frame to the SGSN. After receiving the BSSGP frame, the SGSN saves the CGI of the new cell to the MM context of the MS. The services sent to the MS are directly transmitted to this cell.
2)
During packet data transmission, if the MS finds another more suitable adjacent cell through signal measurement or cell selection parameters broadcasted on the PBCCH/BCCH, it stops receiving the system messages from the previous cell but starts to receiving the system messages from the new cell. Then the MS enters this new cell and sends a CELL UPDATE message to the SGSN. This message is transparently transmitted to the PCU. When the SGSN receives the CELL UPDATE message and identifies that the MS is receiving the downstream packets, it sends a PURGE message (containing the BVCIs of both the previous and new cells and the TLLI of the MS) to the PCU to notify the PCU that the MS moves from one cell to another. The PCU finds the previous cell through the BVCI of the previous cell. After that, the PCU exports or transfers the TLLI-related LLC frames that are not transmitted or not confirmed from the previous cell to the transmission queue of the new cell. After that, the PCU reallocates resources for the MS in the new cell. The new TBF stream is set up for the MS in the new cell and then the data transmission is started. Note: If the cell updating procedure is implemented in different BSSs, the PCU removes the TLLI-related LLC frames from the previous cell. If the data transmission is implemented in LLC confirm mode, the LLC-PDUs removed by the PCU will be retransmitted. If the cell reselection procedure is implemented in LLC non-confirm mode, the LLC-PDUs removed by the PCU will be discarded.
63
64
The SGSN regards the radio class mark as an information field and provides it to the BSS in each downstream BSSGP PDU. The SGSN stores the SGSN class mark and transmits it to the new SGSN. To improve the efficiency, the initial access stage is advanced in the specifications for the BSS to directly obtain the simple radio access class mark from the MS. In this case, the BSS does not need to obtain the entire radio access class mark from the SGSN, thus quickly implementing transmission triggering for the MS. The simplified class mark can be contained in the initial random access message or in the first upstream radio block.
65
66
relay
SNDCP GTP UDP / TCP IP Network service L1bis Gb L2 L1 Gn
GTP
LLC
Relay
LLC BSSGP
UDP / TCP IP L2 L1 Gi
MS
BSS
SGSN
GGSN
This section describes how the PDU is transmitted from one side of the GPRS network to the other side and how the relevant protocol is implemented when the PDU is transmitted over the interface. Take the case when a PC sends an E-mail to a GPRS MS for example. The application layer (the PC here) generates an IP packet and sends it to the GGSN through the external network (IP network or X.25 network). When the IP packet reaches the GGSN, it is called network packet data unit (N-PDU). The N-PDU will be transmitted orderly on the layers of the protocol stack after being added a header at each layer. After a GTP header is added to the N-PDU on the GTP layer, the N-PDU becomes the G-PDU. The GTP is the interface protocol bearing the N-PDU between the GSNs. On the signaling platform, it specifies the channel management and control. On the data transmission platform, it is used to transmit the user packet data through the established tunnel between the GSNs. The tunnel is defined by the GTP header. The TID of the GTP header indicates which tunnel the N-PDU belongs to. The receiving GSN identifies the MM and PDP contexts through this TID. Thus, the PDU is multiplexed between the GSNs through the GTP. The N-PDU that contains the GTP header is transmitted to the transport layer. In the GPRS specifications, the UDP/IP is used for transmitting the GTP signaling and the tunnel established on the UDP/IP connectionless path or the TCP/IP connection-oriented path is used for transmitting PDUs. To be specific, the UDP/TCP layer adds the UDP/TCP header to the N-PDU. The header contains the port address, flow control, and error prevention information. The IP layer adds the IP header to the N-PDU. The header contains the sending and receiving GSN addresses, and routing information. In addition, the IP layer performs the segmentation on the N-PDU to meet the maximum transmission unit (MTU) restriction of the IP layer. After being added with the headers of every layer, the N-PDU is transmitted to the SGSN by physical circuits through the Gn interface. When receiving the N-PDU, the protocol stack of the SGSN removes the headers layer by layer and then sends the N-PDU from the trunk layer to the SubNetwork Dependent Convergence Protocol (SNDCP) layer. At this layer, the data is compressed and segmented to improve the channel utilization to meet the transmission requirement of
Confidential Information of Huawei. No Spreading without Permission
67
the MTU of 1520 bytes at the NS layer of the Gb interface FR network. The SN-PDU is classified into the connection-oriented SN-DATA PAU and connectionless SN-UNIDATA PDU formats based on the transmission type (confirmation mode and non-confirmation mode). The segmentation and reassembly of the two formats are different. In addition, the SNDCP implements the multiplexing (multiple PDUs use one NSAPI) and segmenting (one PDU uses multiple NSAPIs) of different types of PDUs. After being added the NSDCP header, the N-PDU is sent to the LLC layer. The LLC layer compresses the SN-DATA PDU or SN-UNIDATA PDU and then adds a header to the PDU to generate the LLC frame. The LLC frame containing the SN-PDU is called LLC block. The LLC layer provides a highly reliable encrypted logical link between the MS and the SGSN, which is uniquely identified by the TLLI. The LLC frame header contains the control information unit, frame check sequence (FCS), and SAPI. The SAPI indicates the frame associated with a certain PDP context. This frame can be GMM, SMS, or SNDCP service. The BSSGP interface is under the LLC layer, providing routing information for the NS layer. It notifies the LLC block through which route the LLC block can access the FR physical layer. The BSSGP also provides control parameters for the retransmission of the radio interface RLC/MAC. The PDU is transmitted to the BSS through the Gb interface of the FR network. After receiving the PDU, the BSS transfers all BSSGP information to the RLC layer. The RLC layer segments the LLC block into smaller RLC block, that is, the TBFs. The TBF exists only in data transmission. Each TBF is uniquely identified by a TFI allocated. The RLC block header added to the N-PDU contains the TFI and BSN. The LLC information unit length of the RLC block is related to the coding scheme of the radio interface, that is, 22 (CS1), 32 (CS2), 38 (CS3), and 52 (CS4). Under the RLC layer is the MAC layer. The MAC layer provides upstream and downstream signaling and data multiplexing. It determines the contention and precedence among the channel access attempts initiated by the MSs. The N-PDU is added with the MAC header and then transmitted to the MS through the radio interface physical network. The MS orderly removes all headers added to the PDU based on the MS protocol and obtains the complete IP application layer data, that is, the E-mail sent by the PC.
68
The PVC is the fixed virtual circuit established between the subscribers for information transmission and exchange. In SVC mode, no fixed circuit is established between the terminal users. When there is a request for data transmission, the subscriber initiates the request for establishing the virtual circuit. When the data transmission is complete, the request of clearing the virtual circuit is initiated. Similar to the subscriber line of the telephony network, the network establishes the relevant virtual circuit through the calling request. After the communication is complete, the virtual circuit is released through the signaling. Virtual circuits are connection-oriented.
69
Information 1~4096byte
Information (I)
|________Trailer_______| 2 byte
FCS
2 byte
Address (A)
1byte
Ending flag (F)
Flag (F): Delimits the beginning and end of the frame. The value of this field is represented as the 8-bit binary number 01111110. The flag is classified into starting flag and ending flag. Data link connection identifier (DLCI): Indicates the virtual connection of the bearing path on the subscriber network interface or network interface. The length of a DLCI is 10, 16, or 23 bits. In the address of two bytes, the length of a DLCI address is 10 bits, ranging 01023. Command/Response (C/R): It is not used currently and can be set to any value. Extended address (EA): It can be set to 0 or 1. The value 0 means another address byte follows the current address byte. The value 1 means this is the last byte of the address field. To compatible with the ISDN, channel D can only adopt the 2-byte format. The following table lists the bits complying with the 2-byte format.
8 7 DLCI 6 5 DLCI FECN BECN 4 3 2 C/R DE 1 EA(0) EA(1)
Forward explicit congestion notify (FECN): This bit is used to notify the end user of congestion for the purpose of preventing data loss. Backward explicit congestion notify (BECN): This bit is used to notify the source user there is congestion in the opposite direction from the one the frame is traveling. Discard eligibility (DE): When the network is congested, the frame may be discarded for bandwidth processing. When it is set to 1, it indicates discard; when it is set to 0, it indicates not discard (for frames with higher precedence).
70
the direction for the frame based on the DLCI in the routing table. The FR virtual circuit is an end-to-end logical link made up by multiple DLCI logical connections. When user data information is capsulated into the frame and sent to the network node equipment. The node equipment analyses the DLCI in the frame and then queries the PVC routing table to find the DLCI of the next PVC, thus accurately sending the frame to the next node equipment. The FR network subscriber interface supports up to 1024 virtual circuits. The range of the DLCIs available to subscribers is 161007. The following figure illustrates the FR principle with one frame of a series of them that are sent from the local terminal to the network. Suppose that there is an FR terminal connected to port X and the DLCI=a before FR switching. When the switch begins to receive the frame, it checks the 2-byte FCS in the trailer to see whether the FCS is correct, length is proper, and CLCI=a is allocated. If any information is incorrect, the frame is discarded. If all information is correct, the switch queries the routing table and finds that the frame with DLCI=a received from port X should be sent from port Y. It changes the DLCI from a to b, that is, DLCI=b. In this case, the FCS must be re-calculated before the frame is sent. From the call setup, the routing tables of all switches that the frame is traveling must contain the entries as shown in the following figure. During the transmission of the frame until it reaches the destination, the DLCI of the frame varies based on different links. The figure illustrates one direction only. It is the same with the other direction. The transmission on two directions is independent from each other and can be configured with different pass rate.
RF principle
The FR network is composed of FRADs and FR switching devices (FRSDs). The FR standard defines the following applications: interface protocol, network signaling protocol, and network services.
71
1)
Congestion notification
Explicit congestion notification: The congestion is detected by checking whether the FECN or BECN bit is set to 1. After detecting the congestion, the system adjusts the window size to prevent the situation. To be specific, when the FR node detects that the cache or processor is overloaded, it set the FECN bit to 1. After detecting that the rate of FECN=1 is over the threshold value, the receiving higher-layer protocol reduces the size of the receiving window to reduce the frame sending rate. On the other side, when detecting that the cache or processor is overloaded, the FR node sets the BECN to 1 to notify the sending party to reduce the frame sending rate. Implicit congestion notification: When detecting that the frame is discarded or the delay is rather long, the user terminal sends the notification for the congestion. In Huawei Gb interface configuration, the PCU and SGSN set the DE, FECN, and BECN to 1.
2)
The three broadband network control parameters applied by subscribers when accessing the FR network are: CIR, Bc, and Be. Consent information rate (CIR): It indicates the ensured transmission rate (bit/s) at a time interval Tc (measuring period) when the network operation is normal. Consent burst (Bc): It indicates the allowed maximum information volume (bit) at a time interval Tc (measuring period). Exceed burst (Be): It indicates the allowed maximum information volume (bit) that exceeds Bc at a time interval Tc (measuring period). Tc = Bc/CIR The FR monitors the information volume on the virtual circuit at each time interval (Tc). It determines whether a new access request is approved based on the transmission rate and network remaining bandwidth.
3)
Traffic volume forced restriction Within Tc, when user data transmission volume B <= Bc, the system permits the transmission. Within Tc, when Bc < user data transmission volume B <= (Bc + Be) and the network is not encountered with heavy congestion, the system permits the transmission. If the network is heavily congested, the frames with DE=1 are discarded. Within Tc, when user data transmission volume B > (Bc + Be), the system discards the frames.
The FR forcedly restricts the traffic volume. The restriction principles are as follows:
72
4)
communication delay. In the FR network, end user terminals are used to implement the flow control and error correction. The length of an FR frame is longer than an IP packet, which can contain up to 1600 bytes.
MS
BSS
Figure 9-4 GPRS Gb interface protocol
SGSN
The FR is applied to the network service (NS) layer that employs the Gb interface protocol. The NS layer is composed of two parts: NS subnetwork part and NS control part. The former one defines layer-2 protocol and adopts FR currently and ATM in the future. The latter one uses the service of the former one for communication. The NS layer is responsible for the NS PDU communication between the SGSN and the PCU. It has the following functions:
1) 2) 3)
NS PDU transmission. Network congestion indication: The NS subnetwork part (FR) can implement the congestion restore control. Status indication: It is used to notify NS users about NS-related events, such as transmission performance change.
To provide end-to-end communication between the BSS and the SGSN regardless of the structure of the Gb interface, network service virtual connection (NS-VC) concept is introduced. The NS-VC is the end-to-end virtual connection between the BSS and the SGSN. For the FR network, the NS-VC is the PVC. Each NS-VC is identified by NS-VC identification (NS-VCI). An NS-VCI uniquely identifies an NS-VC. The NS-PDU is transmitted on the NS-VC. The NS-VC is the virtual connection between the NS control entities. The NS-PDU is encapsulated into the NS control PDU and then the NS control PDU is encapsulated into the NS subnetwork PDU. In the SGSN, multiple NS entities (NSE) are defined. Each PCU maps an NES. One NSE manages a group of NSVCIs as a specific node.
73