Are LTE and mobile WiMAX really 4G networks? A look at ITU-R IMT Advanced attributes

There have been several recent articles comparing and contrasting LTE vs. mobile WiMAX as potential 4G network technologies. Indeed, SPRINT and Clearwire have been marketing their WiMAX offering as a 4G service. But is it really 4G? Is LTE a 4G technology? What about ultra-wideband?

It turns out that no broadband wireless technology can legitimately claim to be 4G, because the ITU has not completed its 4G specifications yet (4G is known as IMT Advanced in the ITU-R). Hence, there is no benchmark document(s) for which to evaluate any of the proposed 4G technologies. Several important ITU documents and presentations are listed in the References at the bottom of this article.

What are the attributes of 4G?

The ITU has stated that 4G will be a "completely new, fully IP-based integrated system of systems and network of networks- achieved after convergence of wired and wireless networks." This is in sharp contrast to current 3G networks, which are circuit-switched based, with an overlay of data carrying capabilities (EVDO for CDMA and HSDPA for GSM). 4G networks will be entirely IP packet switched. Mobile voice will be carried as a stream of IP packets–VoIP over the equivalent MAC sub-layer.

According to the ITU, 4G networks will be capable of providing 100 Mbps and 1 Gbps downstream rates, in outdoor (mobile) and indoor (fixed access) environments, respectively. 4G networks will have end-to-end QoS and high security, offer any kind of services at any time as per user requirements, anywhere with seamless interoperability, always on, affordable cost, one billing and fully personalized.

Here are the key features of 4G/IMT-Advanced (as per ITU-R document: BACKGROUND ON IMT-ADVANCED, 7 March 2008- see References below):

• a high degree of commonality of functionality worldwide while retaining the flexibility to support a wide range of services and applications in a cost efficient manner;
• compatibility of services within IMT and with fixed networks;
• capability of interworking with other radio access systems;
• high quality mobile services;
• user equipment suitable for worldwide use;
• user-friendly applications, services and equipment;
• worldwide roaming capability; and,
• enhanced peak data rates to support advanced services and applications (100 Mbit/s for high and 1 Gbit/s for low mobility were established as targets for research).
• These features enable IMT-Advanced to address evolving user needs, which implies a network that will evolve and change over time.

The 4G Radio technologies will include:

• Orthogonal Frequency-Division Multiple Access (OFDMA)- a highly efficient multi-carrier modulation scheme, and
• Multiple Input Multiple Output (MIMO)- a multi-antenna system that minimizes data errors and optimizes speed.

How do LTE and mobile WiMAX stack up against the 4G/IMT-Advanced Objectives?

Both LTE and WiMAX are based on OFDMA and MIMO technologies and both are all IP networks with QOS and (some) security. 3GPP’s initial LTE specification, due in March 2009, is that organization’s choice for a 4G network. As noted in an earlier post today, IEEE 802.16 TG m is chartered with ammending the IEEE 802.16 WirelessMAN-OFDMA specification to meet IMT-Advanced requirements, while offering continuing support for legacy WirelessMAN-OFDMA equipment. This will provide an UPGRADE path for existing mobile WiMAX networks based on IEEE 802.16-2005 to a 4G-like network.

But it remains to be seen if either of these networks will qualify as 4G/ IMT-Advanced networks when they are initially rolled out. It will take a long time to validate their attributes and operational performance against the numerous IMT Advanced requirements. Let’s not take all the hype too seriously at this time.

LTE vs. WiMAX: The 4G Wireless War

Remember when 3G was the future of wireless data? It’s not even universally available in the U.S. yet, and the race is already well underway to replace it. WiMAX, the 4G network technology that counts Sprint and Intel among its boosters, has a head start. But it’s losing ground to Long Term Evolution (LTE).

LTE’s promise of high-speed, two-way wireless data promises an “all-IP” mode of communications in which voice calls are handled via VoIP. It’s also designed to handle video well, and to permit roaming through multiple systems–from cellular to Wi-Fi and satellite.

LTE is considered by many to be the obvious successor to current-generation 3G technologies, based on WCDMA, HSDPA, HSUPA and HSPA, in part because it updates UMTS technology to provide significantly faster data rates for both uploading and downloading, while preserving backwards compatibility with existing handsets based on older standards. Verizon Wireless, has already said that it will support LTE as its 4G technology of choice, abandoning its current CDMA based network.

Speed, theoretically superior to WiMAX, would give LTE an edge for bandwidth-hungry applications such as live TV and video downloads. LTE handsets are also expected to embrace automatic roaming to non-cellular systems, such as Wi-Fi and satellite.

It’s true that WiMAX, unlike LTE, is available today–but it’s only in the early stages of rollout. (Sprint-backed Clearwire, the only company to roll out WiMAX in the U.. to date, offers service only in scattered areas in sixteen states.) Analysts express doubts that phone manufacturers, networking companies, app developers, operators, and carriers will ever make WiMAX a popular replacement for 2G or 2.75G facilities and services.

Still, WiMAX may endure–Clearwire has vowed to build a nationwide network. But the leisurely pace of its rollout indicates extra caution about the necessary investments. And Clearwire is controlled by Sprint, widely considered the weakest of the major U.S wireless carriers.

Whether they bet on LTE, WiMAX, or some combination of the two, major carriers, hardware companies, and other telecommunications players cannot postpone decisions about their 4G plans–even though it’s not yet clear how the competing technologies will sort themselves out. Investing mammoth amounts of money on building out what may be a temporary technology is high risk–especially during the worst economic crisis since the Great Depression–but they can’t leave the market open to their competitors.

The matter of superiority, WiMAX vs. LTE, is mind-boggling to industry observers, even if it might not be to a genius, or to electrical and wireless engineers. Innovation advocates might see LTE as a natural evolution of technology. Yet some technology writers have described it as unusual, in the logical sequence of technological advancement. At least, the adoption of LTE shows that the best decision, in the acceleration of wireless-connectivity technology, is not to wait for the economic recession to hit rock bottom or reverse.

The CTIA Wireless trade show in Las Vegas last month made the industry’s 4G road map a bit clearer. Most players, including Motorola and Verizon, said that they would go straight to LTE without touching WiMAX. Nokia, went further: According to a Financial Times report, Anssi Vanjoki, Nokia’s head of sales and manufacturing, compared WiMAX’s prospects to those of Betamax.

Worldwide, LTE’s prospects look promising. Some observers say that China will go directly to LTE, bypassing WiMAX. Major Chinese telecommunications players, including China Mobile and Huawei, are believed to be working hard to step up to LTE in a year or two.

My home, Pakistan, would also benefit from LTE. Currently, almost all the mobile operators, including the formerly state-owned landline monopoly Pakistan Telecommunication Company (PTCL), have flooded the consumer market with phones, cameras, music players, and USB modems that use a form of connectivity that’s similar to WiMAX but slower. These devices in Pakistan offer Internet connectivity of 300-kbs. Companies such as Wateen Telecom of the United Arab Emirates have tried to offer WiMAX, but without much success; but hardly succeeded; PTCL has tried a package of cellular connectivity, satellite TV, and broadband Internet that is also far from a success so far. China Mobile is one of the five major mobile operators in Pakistan, and other Chinese companies such as ZTE and Huawei are major players, so Pakistan’s 4G future will likely mirror that of China.

Countries such as Sweden and Finland, which are small but well-developed and technology-rich can benefit from this transitional period of wireless technologies, during which 3G, WiMAX, and LTE will coexists. Examples could be Sweden, with rich file-sharing experience, and Finland, with Nokia having early experimentation on real time interactive videos. Next in line are rapidly developing countries, including China, India, and Pakistan.

The U.S., a traditional leader in innovation and technological advancement, may struggle to adopt 4G as rapidly as other countries. Why? One reason is the difficulty of ramping up LTE during a period of recession. Another is the indecisiveness of U.S. industry heavyweights about next-generation standards. But even if the U.S.’s 4G future is somewhat murky, wireless connectivity is bound to evolve towards higher speed, great traffic capacity and more reliable connections.

802.11n and 4G

IEEE 802.11n is a proposed amendment to the IEEE 802.11-2007 wireless networking standard to significantly improve network throughput over previous standards, such as 802.11b and 802.11g, with a significant increase in raw (PHY) data rate from 54 Mbit/s to a maximum of 600 Mbit/s. Most devices today support a PHY rate of 300 Mbit/s, with the use of 2 Spatial Streams at 40 MHz. Depending on the environment, this may translate into a user throughput (TCP/IP) of 100 Mbit/s.

According to the book "WI-Fi, Bluetooth, Zigbee and Wimax":

802.11n is the 4th generation of wireless lan technology.

• First generation (IEEE 802.11) since 1997 (WLAN/1G)
• Second generation (IEEE 802.11b) since 1998 (WLAN/2G)
• Third generation (802.11a/g) since 2000 (WLAN/3G)
• Fourth generation (IEEE 802.11n) (WLAN/4G)

The distinguishing features of 802.11n are:

• Very high throughput (some hundreds of Mbps)
• Long distances at high data rates (equivalent to IEEE 802.11b at 500 Mbps)
• Use of robust technologies (e.g. multiple-input multiple-output [MIMO]and space time coding).

In the N option, the real data throughput is estimated to reach a theoretical 540 Mbps (which may require an even higher raw data rate at the physical layer), and should be up to 100 times faster than IEEE 802.11b, and well over ten times faster than IEEE 802.11a or IEEE 802.11g. IEEE 802.11n will probably offer a better operating distance than current networks. IEEE 802.11n builds upon previous IEEE 802.11 standards by adding MIMO. MIMO uses multiple transmitter and receiver antennae to allowfor increased data throughput through spatial multiplexing and increased range by exploiting the spatial diversity and powerful coding schemes. The N system is strongly based on the IEEE 802.11e QoS specification to improve bandwidth performance. The system supports basebands width of 20 or 40MHz.

Note that there is 802.11n PHY and 802.11n MAC that will be required to acheive 540Mbps.

To achieve maximum throughput a pure 802.11n 5 GHz network is recommended. The 5 GHz band has substantial capacity due to many non-overlapping radio channels and less radio interference as compared to the 2.4 GHz band. An all-802.11n network may be impractical, however, as existing laptops generally have 802.11b/g radios which must be replaced if they are to operate on the network. Consequently, it may be more practical to operate a mixed 802.11b/g/n network until 802.11n hardware becomes more prevalent. In a mixed-mode system, it’s generally best to utilize a dual-radio access point and place the 802.11b/g traffic on the 2.4 GHz radio and the 802.11n traffic on the 5 GHz radio.

A lot of phones are coming with inbuilt WiFi (or 802.11 a/b/g) and this WiFi is a must on Laptops or they wont sell. The main difference in 802.11n, compared to previous generation of 802.11 is that there is a presence of MIMO. 802.11 family uses OFDM which is the same technology being adopted by LTE. The new LTE handsets will have advantage of easily integrating this 802.11n technology and the same antennas can be reused. In fact the same is applicable for WiMAX as it supports MIMO and OFDM. Ofcourse we will have problems if they are using quite different frequencies as the antennas ore optimised to range of frequencies, this is something that has to be seen.

UMTS Interfaces

Many new protocols have been developed for the four new interfaces specified in UMTS: Uu, Iub, Iur, and Iu. This tutorial is organized by the protocols and shows their usage in the interfaces. That means protocols will be described individually. Only the references to the interfaces are indicated. Interface specific explanations of the protocols are, however, not included. Before we review the individual interface protocols, we introduce the UMTS general protocol model.

General Protocol Model [3G TS 25.401]
UTRAN interface consists of a set of horizontal and vertical layers (see Figure 9). The UTRAN requirements are addressed in the horizontal radio network layer across different types of control and user planes. Control planes are used to control a link or a connection; user planes are used to transparently transmit user data from the higher layers. Standard transmission issues, which are independent of UTRAN requirements, are applied in the horizontal transport network layer.

Figure 9. UTRAN Interface—General Protocol Model

Five major protocol blocks are shown in Figure 9:

• Signaling bearers are used to transmit higher layers’ signaling and control information. They are set up by O&M activities.
• Data bearers are the frame protocols used to transport user data (data streams). The transport network–control plane (TN–CP) sets them up.
• Application protocols are used to provide UMTS–specific signaling and control within UTRAN, such as to set up bearers in the radio network layer.
• Data streams contain the user data that is transparently transmitted between the network elements. User data is comprised of the subscriber’s personal data and mobility management information that are exchanged between the peer entities MSC and UE.
• Access link control application part (ALCAP) protocol layers are provided in the TN–CP. They react to the radio network layer’s demands to set up, maintain, and release data bearers. The primary objective of introducing the TN–CP was to totally separate the selection of the data bearer technology from the control plane (where the UTRAN–specific application protocols are located). The TN–CP is present in the Iu–CS, Iur, and Iub interfaces. In the remaining interfaces where there is no ALCAP signaling, preconfigured data bearers are activated.

Application Protocols
Application protocols are Layer-3 protocols that are defined to perform UTRAN–specific signaling and control. A complete UTRAN and UE control plane protocol architecture is illustrated in Figure 10. UTRAN–specific control protocols exist in each of the four interfaces.

Figure 10. Iu RANAP Protocol Architecture

Figure 11. Application Protocols

Iu: Radio Access Network Application Part (RANAP) [3G TS 25.413]

This protocol layer provides UTRAN–specific signaling and control over the Iu (see Figure 11). The following is a subset of the RANAP functions:

• Overall radio access bearer (RAB) management, which includes the RAB’s setup, maintenance, and release
• Management of Iu connections
• Transport of nonaccess stratum (NAS) information between the UE and the CN; for example, NAS contains the mobility management signaling and broadcast information.
• Exchanging UE location information between the RNC and CN
• Paging requests from the CN to the UE
• Overload and general error situation handling

Iur: Radio Network Sublayer Application Part (RNSAP) [3G TS 25.423]

UTRAN–specific signaling and control over this interface contains the following:

• Management of radio links, physical links, and common transport channel resources
• Paging
• SRNC relocation
• Measurements of dedicated resources

Figure 12. Iur RNSAP Protocol Architecture

Iub: Node B Application Part (NBAP) [3G TS 25.433]

UTRAN specific signaling and control in the Iub includes the following (see Figure 13):

• Management of common channels, common resources, and radio links
• Configuration management, such as cell configuration management
• Measurement handling and control
• Synchronization (TDD)
• Reporting of error situations

Uu: Radio Resource Control (RRC) [3G TS 25.331]

This layer handles the control plane signaling over the Uu between the UE and the UTRAN (see also Figure 13). Some of the functions offered by the RRC include the following:

• Broadcasting information
• Management of connections between the UE and the UTRAN, which include their establishment, maintenance, and release
• Management of the radio bearers, which include their establishment, maintenance, release, and the corresponding connection mobility
• Ciphering control
• Outer loop power control
• Message integrity protection
• Timing advance in the TDD mode
• UE measurement report evaluation
• Paging and notifying

(Note: The RRCs also perform local inter-layer control services, which are not discussed in this document.)

Two modes of operation are defined for the UE—the idle mode and the dedicated mode. In the idle mode the peer entity of the UE’s RRC is at the Node B, while in the dedicated mode it is at the SRNC. The dedicated mode is shown in Figure 10.

Higher-layer protocols to perform signaling and control tasks are found on top of the RRC. The mobility management (MM) and call control (CC) are defined in the existing GSM specifications. Even though MM and CC occur between the UE and the CN and are therefore not part of UTRAN specific signaling (see Figure 15), they demand basic support from the transfer service, which is offered by duplication avoidance (see 3G TS 23.110). This layer is responsible for in-sequence transfer and priority handling of messages. It belongs to UTRAN, even though its peer entities are located in the UE and CN.

Figure 13. Uu and Iub RRC Protocol Architecture

Transport Network Layer: Specific Layer-3 Signaling and Control Protocols
Two types of layer-3 signaling protocols are found in the transport network layer:

1. Iu, Iur: Signaling Connection Control Part (SCCP) [ITU-T Q.711–Q. 716] This provides connectionless and connection-oriented services. On a connection-oriented link, it separates each mobile unit and is responsible for the establishment of a connection-oriented link for each and every one of them.
2. Iu–CS, Iur, Iub: ALCAP [ITU–T Q.2630.1, Q.2150.1, and Q.2150.2]. Layer-3 signaling is needed to set up the bearers to transmit data via the user plane. This function is the responsibility of the ALCAP, which is applied to dynamically establish, maintain, release, and control ATM adaptation layer (AAL)–2 connections. ALCAP also has the ability to link the connection control to another higher layer control protocol. This and additional capabilities were specified in ITU–T Q.2630.1. Because of the protocol layer specified in Q.2630.1, a converter is needed to correspond with underlying sublayers of the protocol stack. These converters are called (generically) signaling transport converter (STC). Two converters are defined and applied in UTRAN:
   • Iu–CS, Iur: AAL–2 STC on message transfer part (MTP) level 3 (broadband) for Q.2140 (MTP3b) [Q.2150.1]
   • Iub: AAL–2 STC on service-specific connection-oriented protocol (SSCOP) [Q.2150.2]

Transport Network Layer Specific Transmission Technologies
Now that we have a circuit-switched and packet-switched domain in the CN and a growing market for packet-switched network solutions, a new RAN must be open to both types of traffic in the long run. That network must also transmit the Layer-3 signaling and control information. ATM was selected as the Layer-2 technology, but higher-layer protocols used in the transport network layer demonstrate the UMTS openness to a pure IP solution.

Iu, Iur, Iub: ATM [ITU-T I.361]
Broadband communication will play an important role with UMTS. Not only voice but also multimedia applications such as videoconferencing, exploring the Internet, and document sharing are anticipated. We need a data link technology that can handle both circuit-switched and packet-switched traffic as well as isochronous and asynchronous traffic. In UMTS (Release ’99), ATM was selected to perform this task.

An ATM network is composed of ATM nodes and links. The user data is organized and transmitted in each link with a stream of ATM cells. AALs are defined to enable different types of services with corresponding traffic behavior. Two of these are applied in UTRAN:

1. Iu–CS, Iur, Iub: AAL–2 [ITU-T I.363.2]—With AAL–2, isochronous connections with variable bit rate and minimal delay in a connection-oriented mode are supported. This layer was designed to provide real-time service with variable data rates, such as video. Except for the Iu–PS interface, AAL–2 is always used to carry the user data streams.
2. Iu–PS, Iur, Iub: AAL–5 [ITU-T I.363.5]—With AAL–5, isochronous connections with variable bit rate in a connection-oriented mode are supported. This layer is used for Internet protocol (IP) local-area network (LAN) emulation, and signaling. In UTRAN, AAL–5 is used to carry the packet-switched user traffic in the Iu–PS-interface and the signaling and control data throughout.

In order to carry signaling and control data, the AAL–5 has to be enhanced. Here, UTRAN offers both a classical ATM solution and an IP–based approach:

1. Signaling AAL and MTP3b—To make signaling AAL (SAAL) available in place of the AAL–5 service-specific convergence sublayer (SSCS), the SSCOP, which provides a reliable data transfer service, and the service-specific coordination function (SSCF), which acts as coordination unit, are defined.
2. Iu, Iur, Iub: SSCOP [ITU–T Q.2110]—The SSCOP is located on top of the AAL. It is a common connection-oriented protocol that provides a reliable data transfer between peer entities. Its capabilities include the transfer of higher-layer data with sequence integrity, flow control, connection maintenance in case of a longer data transfer break, error correction by protocol control information, error correction by retransmission, error reporting to layer management, status report, and more.

Two versions of the SSCF are defined: one for signaling at the user-to-network interface (UNI), and one for signaling at the network to node interface (NNI):

1. Iub: SSCF for at the UNI (SSCF) [ITU–T Q.2130]—The SSCF–UNI receives Layer-3 signaling and maps it to the SSCOP and visa versa. The SSCF–UNI performs coordination between the higher and lower layers. Within UTRAN, it is applied in Iub with the NBAP and ALCAP on top of the SSCF–UNI.
2. Iu, Iur: SSCF at the NNI (SSCF-NNI) [ITU–T Q.2140]—The SSCF-NNI receives the SS7 signaling of a Layer 3 and maps it to the SSCOP, and visa versa. The SSCF-NNI performs coordination between the higher and the lower layers. Within UTRAN, MTP3b has the higher Layer 3, which requires service from the SSCOP-NNI.

Figure 14. Iu–PS Protocol Architecture

Originally the SS7 protocol layer, SCCP relies on the services offered by MTP, so the Layer-3 part of the MTP must face the SCCP layer:

Iu, Iur: MTP3b [ITU–T Q.2210]—Signaling links must be controlled in level 3 for: message routing, discrimination and distribution (for point-to-point link only), signaling link management, load sharing, etc. The specific functions and messages for these are defined by the MTP3b, which requires the SSCF–NNI to provide its service.

The Layer-3 signaling and control data can also be handled by an enhanced IP stack using a tunneling function (see Figure 12). Tunneling is also applied for packet-switched user data over the Iu–PS interface (see Figure 14).

• IP over ATM
  • lu-PS, Iur: IP [IETF RFC 791, 2460, 1483, 2225], user datagram protocol (UDP) [IETF RFC 768] The IP can be encapsulated and then transmitted via an ATM connection, a process which is described in the RFC 1483 and RFC 2225. Both IP version 4 (IPv4) and IP version 6 (IPv6) are supported. IP is actually a Layer-3 protocol. UDP is applied on top of the unreliable Layer-4 protocol. The objective is to open this signaling link to future pure IP network solutions.

In order to tunnel SCCP or ALCAP signaling information, two protocols are applied:

• Iu–PS and Iur: Simple Control Transmission Protocol (SCTP) [IETF SCTP]—This protocol layer allows the transmission of signaling protocols over IP networks. Its tasks are comparable with MTP3b. On Iu–CS, SS7 must be tunneled between the CN and the RNC. The plan is that this is to be done with the Iu–PS and Iur [IETF M3UA].

The following does the tunneling of packet-switched user data:

• Iu–PS: GPRS tunneling protocol (GTP) [3G TS 29.060]—The GTP provides signaling through GTP–control (GTP–C) and data transfer through GTP–user (GTP–U) procedures. Only the latter is applied in the Iu–PS interface because the control function is handled by the RANAP protocol. The GTP–U is used to tunnel user data between the SGSN and the RNC.

Figure 15. UMTS Air Interface Uu

Iu, Iur, Iub: The Physical Layers [3G TS 25.411]
The physical layer defines the access to the transmission media, the physical and electrical properties, and how to activate and deactivate a connection. It offers to the higher-layer physical service access points to support the transmission of a uniform bit stream. A huge set of physical-layer solutions is allowed in UTRAN, including ETSI synchronous transport module (STM)–1 (155 Mbps) and STM–4 (622 Mbps); synchronous optical network (SONET) synchronous transport signal (STS)–3c (155 Mbps) and STS–12c (622 Mbps); ITU STS–1 (51 Mbps) and STM–0 (51 Mbps); E-1 (2 Mbps), E-2 (8 Mbps), and E-3 (34 Mbps); T-1 (1.5 Mbps) and T-3 (45 Mbps); and J-1 (1.5 Mbps) and J-2 (6.3 Mbps).

With the above protocol layers, the interfaces Iu, Iur, and Iur are fully described. There is only the air interface left for a more detailed analysis:

The Air Interface Uu [3G TS 25.301]
The air interface solution is usually a major cause for dispute when specifying a new RAN. Figure 15 shows the realization of the lower parts of the protocol stack in the UE. As can be seen, a physical layer, data link layer, and network layer (the part for the RRC) have been specified.

The physical layer is responsible for the transmission of data over the air interface. The FDD and TDD W–CDMA solutions have been specified in UMTS Rel. ’99. The data link layer contains four sublayers:

• Medium Access Control (MAC) [3G TS 25.321]—The MAC layer is located on top of the physical layer. Logical channels are used for communication with the higher layers. A set of logical channels is defined to transmit each specific type of information. Therefore, a logical channel determines the kind of information it uses. The exchange of information with the physical layer is realized with transport channels. They describe how data is to be transmitted over the air interface and with what characteristics. The MAC layer is responsible for more than mapping the logical channels into the physical ones. It is also used for priority handling of UEs and the data flows of a UE, traffic monitoring, ciphering, multiplexing, and more.
• Radio Link Control (RLC) [3G TS 25.322]—This is responsible for acknowledged or unacknowledged data transfer, establishment of RLC connections, transparent data transfer, quality of service (QoS) settings, unrecoverable error notification, ciphering, etc. There is one RLC connection per radio bearer.

The two remaining Layer-2 protocols are used only in the user plane:

• Packet Data Convergence Protocol (PDCP) [3G TS 25.323]—This is responsible for the transmission and reception of radio network layer protocol data units (PDUs). Within UMTS, several different network layer protocols are supported to transparently transmit protocols. At the moment, IPv4 and IPv6 are supported, but UMTS must be open to other protocols without forcing the modification of UTRAN protocols. This transparent transmission is one task of PDCP; another is to increase channel efficiency (by protocol header compression, for example).
• Broadcast/Multicast Control (BMC) [3G TS 25.324]—This offers broadcast/multicast services in the user plane. For instance, it stores SMS CB messages and transmits them to the UE.

UMTS Network Architecture

UMTS (Rel. ’99) incorporates enhanced GSM Phase 2+ Core Networks with GPRS and CAMEL. This enables network operators to enjoy the improved cost-efficiency of UMTS while protecting their 2G investments and reducing the risks of implementation.

In UMTS release 1 (Rel. '99), a new radio access network UMTS terrestrial radio access network (UTRAN) is introduced. UTRAN, the UMTS radio access network (RAN), is connected via the Iu to the GSM Phase 2+ core network (CN). The Iu is the UTRAN interface between the radio network controller (RNC) and CN; the UTRAN interface between RNC and the packet-switched domain of the CN (Iu–PS) is used for PS data and the UTRAN interface between RNC and the circuit-switched domain of the CN (Iu–CS) is used for CS data.

"GSM–only" mobile stations (MSs) will be connected to the network via the GSM air (radio) interface (Um). UMTS/GSM dual-mode user equipment (UE) will be connected to the network via UMTS air (radio) interface (Uu) at very high data rates (up to almost 2 Mbps). Outside the UMTS service area, UMTS/GSM UE will be connected to the network at reduced data rates via the Um.

Maximum data rates are 115 kbps for CS data by HSCSD, 171 kbps for PS data by GPRS, and 553 kbps by EDGE. Handover between UMTS and GSM is supported, and handover between UMTS and other 3G systems (e.g., multicarrier CDMA [MC–CDMA]) will be supported to achieve true worldwide access.

Figure 3. Transmission Rate

The public land mobile network (PLMN) described in UMTS Rel. ’99 incorporates three major categories of network elements:

• GSM Phase 1/2 core network elements: mobile services switching center (MSC), visitor location register (VLR), home location register (HLR), authentication center (AC), and equipment identity register (EIR)
• GSM Phase 2+ enhancements: GPRS (serving GPRS support node [SGSN] and gateway GPRS support node [GGSN]) and CAMEL (CAMEL service environment [CSE])
• UMTS specific modifications and enhancements, particularly UTRAN

Network Elements from GSM Phase 1/2
The GSM Phase 1/2 PLMN consists of three subsystems: the base station subsystem (BSS), the network and switching subsystem (NSS), and the operations support system (OSS). The BSS consists of the functional units: base station controller (BSC), base transceiver station (BTS) and transcoder and rate adapter unit (TRAU). The NSS consists of the functional units: MSC, VLR, HLR, EIR, and the AC. The MSC provides functions such as switching, signaling, paging, and inter–MSC handover. The OSS consists of operation and maintenance centers (OMCs), which are used for remote and centralized operation, administration, and maintenance (OAM) tasks.

Figure 4. UMTS Phase 1 Network

Network Elements from GSM Phase 2+

GPRS
The most important evolutionary step of GSM toward UMTS is GPRS. GPRS introduces PS into the GSM CN and allows direct access to packet data networks (PDNs). This enables high–data rate PS transmission well beyond the 64 kbps limit of ISDN through the GSM CN, a necessity for UMTS data transmission rates of up to 2 Mbps. GPRS prepares and optimizes the CN for high–data rate PS transmission, as does UMTS with UTRAN over the RAN. Thus, GPRS is a prerequisite for the UMTS introduction.

Two functional units extend the GSM NSS architecture for GPRS PS services: the GGSN and the SGSN. The GGSN has functions comparable to a gateway MSC (GMSC). The SGSN resides at the same hierarchical level as a visited MSC (VMSC)/VLR and therefore performs comparable functions such as routing and mobility management.

CAMEL
CAMEL enables worldwide access to operator-specific IN applications such as prepaid, call screening, and supervision. CAMEL is the primary GSM Phase 2+ enhancement for the introduction of the UMTS virtual home environment (VHE) concept. VHE is a platform for flexible service definition (collection of service creation tools) that enables the operator to modify or enhance existing services and/or define new services. Furthermore, VHE enables worldwide access to these operator-specific services in every GSM and UMTS PLMN and introduces location-based services (by interaction with GSM/UMTS mobility management). A CSE and a new common control signaling system 7 (SS7) (CCS7) protocol, the CAMEL application part (CAP), are required on the CN to introduce CAMEL.

Network Elements from UMTS Phase 1
As mentioned above, UMTS differs from GSM Phase 2+ mostly in the new principles for air interface transmission (W–CDMA instead of time division multiple access [TDMA]/frequency division multiple access [FDMA]). Therefore, a new RAN called UTRAN must be introduced with UMTS. Only minor modifications, such as allocation of the transcoder (TC) function for speech compression to the CN, are needed in the CN to accommodate the change. The TC function is used together with an interworking function (IWF) for protocol conversion between the A and the Iu–CS interfaces.

UTRAN
The UMTS standard can be seen as an extension of existing networks. Two new network elements are introduced in UTRAN, RNC, and Node B. UTRAN is subdivided into individual radio network systems (RNSs), where each RNS is controlled by an RNC. The RNC is connected to a set of Node B elements, each of which can serve one or several cells.

Figure 5. UMTS Phase 1: UTRAN

Existing network elements, such as MSC, SGSN, and HLR, can be extended to adopt the UMTS requirements, but RNC, Node B, and the handsets must be completely new designs. RNC will become the replacement for BSC, and Node B fulfills nearly the same functionality as BTS. GSM and GPRS networks will be extended, and new services will be integrated into an overall network that contains both existing interfaces such as A, Gb, and Abis, and new interfaces that include Iu, UTRAN interface between Node B and RNC (Iub), and UTRAN interface between two RNCs (Iur). UMTS defines four new open interfaces:

• Uu: UE to Node B (UTRA, the UMTS W–CDMA air interface
• Iu: RNC to GSM Phase 2+ CN interface (MSC/VLR or SGSN)
  • Iu-CS for circuit-switched data
  • Iu-PS for packet-switched data
• Iub: RNC to Node B interface
• Iur: RNC to RNC interface, not comparable to any interface in GSM

The Iu, Iub, and Iur interfaces are based on ATM transmission principles.

The RNC enables autonomous radio resource management (RRM) by UTRAN. It performs the same functions as the GSM BSC, providing central control for the RNS elements (RNC and Node Bs).

The RNC handles protocol exchanges between Iu, Iur, and Iub interfaces and is responsible for centralized operation and maintenance (O&M) of the entire RNS with access to the OSS. Because the interfaces are ATM–based, the RNC switches ATM cells between them. The user’s circuit-switched and packet-switched data coming from Iu–CS and Iu–PS interfaces are multiplexed together for multimedia transmission via Iur, Iub, and Uu interfaces to and from the UE.

The RNC uses the Iur interface, which has no equivalent in GSM BSS, to autonomously handle 100 percent of the RRM, eliminating that burden from the CN. Serving control functions such as admission, RRC connection to the UE, congestion and handover/macro diversity are managed entirely by a single serving RNC (SRNC).

If another RNC is involved in the active connection through an inter–RNC soft handover, it is declared a drift RNC (DRNC). The DRNC is only responsible for the allocation of code resources. A reallocation of the SRNC functionality to the former DRNC is possible (serving radio network subsystem [SRNS] relocation). The term controlling RNC (CRNC) is used to define the RNC that controls the logical resources of its UTRAN access points.

Figure 6. RNC Functions

Node B
Node B is the physical unit for radio transmission/reception with cells. Depending on sectoring (omni/sector cells), one or more cells may be served by a Node B. A single Node B can support both FDD and TDD modes, and it can be co-located with a GSM BTS to reduce implementation costs. Node B connects with the UE via the W–CDMA Uu radio interface and with the RNC via the Iub asynchronous transfer mode (ATM)–based interface. Node B is the ATM termination point.

The main task of Node B is the conversion of data to and from the Uu radio interface, including forward error correction (FEC), rate adaptation, W–CDMA spreading/despreading, and quadrature phase shift keying (QPSK) modulation on the air interface. It measures quality and strength of the connection and determines the frame error rate (FER), transmitting these data to the RNC as a measurement report for handover and macro diversity combining. The Node B is also responsible for the FDD softer handover. This micro diversity combining is carried out independently, eliminating the need for additional transmission capacity in the Iub.

The Node B also participates in power control, as it enables the UE to adjust its power using downlink (DL) transmission power control (TPC) commands via the inner-loop power control on the basis of uplink (UL) TPC information. The predefined values for inner-loop power control are derived from the RNC via outer-loop power control.

Figure 7. Node B Overview

UMTS UE
The UMTS UE is based on the same principles as the GSM MS—the separation between mobile equipment (ME) and the UMTS subscriber identity module (SIM) card (USIM). Figure 8 shows the user equipment functions. The UE is the counterpart to the various network elements in many functions and procedures.

Figure 8. UE Functions

3G/UMTS

3G – History:

First generation wireless, or 1G, refers to analog networks introduced in the mid-1980s. Examples include Advanced Mobile Phone Service (AMPS) used in North America and Total Access Communications System (TACS) used in the UK. As mobile communications grew in popularity, networks often became overloaded, resulting in busy signals and dropped calls. The solution was second-generation wireless, or 2G, which emerged in the early 1990s. 2G technologies were digital and offered the much-needed capacity that 1G analog systems did not afford. Several technologies were widely used:

• TDMA (IS-54 and IS-136)
• GSM (a TDMA based technology)
• CDMA IS-95 or cdmaOne (a CDMA based technology)

However, these 2G technologies are incompatible with each other. Thus, mobile service subscribers were still often limited to using their phones in a single country or region.

In an effort to standardize future digital wireless communications and make global roaming with a single handset possible, the ITU established a single standard for wireless networks in 1999. Called IMT-2000, which is commonly referred to today as 3G, the initiative set forth the requirements (mentioned above) for the third generation of wireless networks.

3G – The Standard:

3G stands for third-generation wireless technology and networks. The concept of a single standard evolved into a family of five 3G wireless standards. Of those five, the most widely accepted are CDMA2000, WCDMA (UMTS) and TD-SCDMA. According to the ITU and IMT-2000, a wireless standard must meet minimum bit-rate requirements to be considered 3G:

• 2 Mbps in fixed or in-building environments
• 384 Kbps in pedestrian or urban environments
• 144 Kbps in wide area mobile environments
• Variable data rates in large geographic area systems (satellite)

In addition to providing faster bit rates and greater capacity over previous-generation technologies, 3G standards excel by effectively:

• Delivering mobile data
• Offering greater network capacity
• Operating with existing second-generation technologies
• Enabling rich data applications such as VoIP, video telephony, mobile multimedia, interactive gaming and more

3G Today:
Today, WCDMA (Wideband CDMA) and CDMA2000 are by far the dominant standards in terms of current commercial services, operator deployment plans and vendor support. Launched commercially by wireless operators in 2000, CDMA2000 1X was the world’s first operational 3G technology, capable of transmitting data faster than most dial-up services. Today, more than 190 million people enjoy the benefits of CDMA2000 1X, which provides enhanced data capacity compared to all 2G technologies.

Also known as UMTS (Universal Mobile Telecommunications System), WCDMA (Wideband CDMA) is the 3G standard chosen by most GSM/GPRS wireless network operators wanting to evolve their systems to 3G network technology. WCDMA offers enhanced voice and data capacity and peak data rates faster than most dial-up services and average rates consistently greater than GSM/GPRS (Global System for Mobile communications/General Packet Radio Service) and EDGE (Enhanced Data for GSM Evolution). As of February 2006, more than 51 million subscribers were using WCDMA for their mobile voice and data needs.

Architecture:

Generation Wireless

The Next Generation

Vendors and carriers are working to develop and deploy the next generation of wireless systems, often referred to as 2.5G, which is packet-based and increases data communication speeds to as high as 384 Kbps.

Upgrading will involve a newer radio network for modified air interface, cell planning, and modifications to the core and backbone network. However, 2.5G systems can use many existing infrastructure elements. CDMA-based carriers say they'll have a less expensive migration path than GSM/ TDMA-based carriers. For newer CDMA equipment, CDMA-based carriers may have only to change channel cards in their base stations and upgrade the network software, while GSM/ TDMA-based networks may require close to an entire network overlay.

3G technology would join the different 2G wireless systems into a global system providing data rates of about 2 Mbps. CDMA has emerged as the multiple access scheme of choice for 3G. The proposed 3G evolution path for TDMA-based systems, including GSM, is W-CDMA (Wideband CDMA), a standard proposed by Ericsson, while CDMA systems will evolve to CDMA 2000 systems. W-CDMA will incorporate an airlink that uses a 5-MHz-wide carrier to enable systems to support speeds of up to 2 Mbps; CDMA 2000 will combine three 1.25-MHz carriers to accomplish its rates.

In March 1999, Ericsson and Qualcomm ended a two-year patent dispute over W-CDMA by entering into cross-licensing agreements. This settlement was reached to foster the development of a single 3G CDMA-based standard. Today, almost all wireless equipment manufacturers have signed patent licenses with Qualcomm for CDMA products, including those that incorporate W-CDMA (see "Getting to 3G: Migration Path for Existing 2G and 2.5G Systems").

Many analysts expect the next-generation technology's cost and subscribers' lack of enthusiasm to keep it from becoming widespread in the United States until 2003 or later. However, many prominent service providers, like AT&T Wireless, Cingular, Sprint PCS and Verizon, have developed migration plans. These companies plan to deploy 2.5G networks this year and migrate to 3G networks in the next few years. AT&T had a TDMA-based network. After forming an alliance with NTT DoCoMo, however, AT&T began migrating to a GSM-based network, en route to deploying a GPRS (General Packet Radio Service) network. A few 2.5G deployments can be found in China, Europe, Japan, Korea and a few U.S. cities.

How soon migration to 3G will actually happen is a big question. The answer depends on the maturity of the technology and the appeal of mobile data applications, as well as the costs of implementation. It also depends on how well subscribers respond to 2.5G offerings. There is also the issue of frequency spectrum availability, which is an acute problem in the United States. The FCC plans to hold auctions for 3G licenses in September 2002. NTT DoCoMo expects to launch some 3G services in the coming months.

GPRS

GPRS was introduced as a packet-switched intermediate step to transport high-speed data efficiently over GSM- and TDMA-based networks. GPRS uses eight time slots in the 200-KHz channel and can support IP-based packet data speeds up to about 115 Kbps.

The two main additional infrastructure elements here are the SGSN (Serving GPRS Service Node) and the GGSN (Gateway GPRS Service Node). GPRS packetizes the user data and transports it using an IP backbone, with the GGSN acting as the gateway between the GPRS network and other packet-based networks, like the Internet. These GGSNs also connect to other GPRS networks to provide GPRS roaming.

The SGSN can be considered a mobile switching center. It enables mobility management by keeping track of all the mobile stations on the network, and it provides mobile-data packet routing to and from an SGSN service area.

GPRS uses the GSM network to look up the location-register databases to obtain subscriber-profile data. Enabling GPRS on a GSM network will also require a new air interface for packet-switched traffic.

With GPRS offering speeds between 14.4 Kbps and 115 Kbps, it should allow for better wireless Internet access. However, achieving the theoretical maximum GPRS data transmission speed of about 170 Kbps would require a single user to take over all eight time slots without any error protection.

Another standard, called EDGE (Enhanced Data Rates for Global Evolution), has been specified to improve the throughput per time slot in GPRS to support data rates of up to 384 Kbps using the same 200-KHz TDMA carrier.

CDMA Moves to 3G

Next-generation CDMA networks will come in two phases. The first is CDMA 2000 1X, and the second is CDMA 2000 3X. The "1X" and "3X" refer to the number of 1.25-MHz-wide air interface channels used.

CDMA 2000 1X would enable CDMA systems to improve data performance by providing IP-based packet data speeds of about 144 Kbps in the 1.25-MHz channel. Through modulation improvements and better power control, these systems would more than double the capacity of the earlier IS-95 systems.

CDMA 2000 1X uses a PDSN (Packet Data Serving Node) as the packet data gateway and the Mobile IP protocol to allow mobility management between the cellular network and a packet data network (see "Cellular Technology Comparison").

The newer generation of wireless technologies represents one of the biggest opportunities for equipment vendors and carriers to provide both businesses and users with value-added, location-independent services while opening up new sources of revenue.

Blesson Mathews is a research associate with the Center for Emerging Network Technologies at Syracuse University.