第二代移动通信系统向第三代移动通信系统 演进 徐国鑫 Xuguoxin@bupt.edu.cn 北京邮电大学无线通信新技术研究室

1。 移动通信的发展 2。无线网络的演进 3。核心网的演进 4。业务的演进

数据速率最高达2Mb/s，数据在线连接宽带数据业务 移动通信各代典型系统特点 典型代表 技术 特性 第一代 AMPS 小区制蜂窝系统 模拟话音 第二代 GSM 数字蜂窝（TDMA） 数字话音，数据速率9.6Kb/S 第二代半 GPRS 通用分组数字蜂窝 数据速率115Kb/s,数据在线连接 第三代 W-CDMA 宽带码分多址，实现宽带多媒体业务 数据速率最高达2Mb/s，数据在线连接宽带数据业务 后三代 …

Electronic Conference 移动数据业务发展方向 Information Services SMS Chat Room E-Mail m-stock trading m-banking m-cash Picture Mail Route planning Multi-player Games Mobile Office Schedule Management Work flow Management Electronic Conference File Sharing Radio Music Video Mail Video streaming Web Interactive TV TV Conference 低速数据 高速数据 WAP GPRS 3G 99 Q1 99 Q4 00 Q4 01 Q4

频率因素比较—频率适用性

中国频率资源分配: 1900 1920 1980 2010 2025 60 MHz 20 MHz 15 MHz FDD (上行) TDD 卫星 空 30 MHz 2110 2170 1885 85 MHz 双工间隔 190 MHz 空中接口 模式 频段 射频带宽 可用性 TD-SCDMA TDD 20 + 15 MHz 1.6 MHz 现在 W-CDMA FDD 60 + 60 MHz 5 MHz 现在 ? TD-SCDMA TDD 15 + 85 MHz 1.6 MHz 现在

多址方式 FDMA FDMA/TDMA CDMA Frequency Power Time Power Frequency Time

第二代 系统空中接口-GSM 微蜂窝小区结构 工作频段 ：800/1900或900M/1800M 数字化技术---语音信号数字化 新的调制方式---GMSK、QPSK等 FDMA/TDMA 频谱利用率高，系统容量大 便于实现通信安全保密 信道宽度：200khz 每个载波的用户数 8个

第二代系统空中接口-CDMA 用户的接入方式采用码分多址（CDMA） 软容量、软切换，系统容量大 抗多径衰落 可运用话音激活、分集接收等先进技术 频率800/1900 信号带宽 1.25MHz 每个信道的用户数 约为20个 频道划分方式：FDMA+CDMA

3G 第三代数字蜂窝移动通信系统 3G的主要特点： 支持移动多媒体业务 宽带CDMA技术 高频谱效率 FDMA/TDMA/CDMA 从电路交换到分组交换 从媒体(media) 到 多媒体(Multi-media) 高保密性 全球范围无缝漫游系统 微蜂窝结构 3G 主流技术： WCDMA cdma2000 TD-SCDMA

移动网的主要元素 核心网 无线网络 移动网 移动用户

网络需求 保证业务的传输 对于高速业务 对于大量用户 对低速用户 高吞吐率 灵活性 可靠性 高效传输 低成本网络

业务需求 2G 到 3G平滑过渡 明显的3G网络收益 在3G引入中的地域补充 提供2G的主要业务 高速多媒体传输

Network Architecture for Both 2G and 3G Services 3G在2G网络上重叠 在两个网上共享用户信息 2G RAN 2G Core Network 移动用户 HLR (User Profile) 3G RAN 3G Core Network (3GPP R99)

For migration from 2G to 3G 双网服务 在引入阶段，在已存在的2G网络上做3G业务补充 用户必须激活起所期望得到服务的网络。 HLR 管理激活状态 HLR Case:B Activation State → GSM ACTIVE IMT INACTIVE Case:A　Activation State → IMT ACTIVE GSM INACTIVE IMT ACTIVE PDC ACTIVE (3) Interrogation of ACTIVATION STATE (2)　Incoming Call GMSC (4)　Call termination to IMT (Case: A) (4)’　Call termination to PDC (Case: B) MSC/VLR MSC PDC Service Activation (1)　IMT Service Activation (1)’　PDC Service Activation PDC IMT IMT Service Activation Call terminating MSISDN: 090-1234-5678 CASE A CASE B IMT-2000 PDC

市场驱动力 社会发展趋势 市场趋势 技术趋势 语音和数据的集成 需要大量带宽的媒体 基于Internet的业务 新的增值应用 用户归属的竞争 在智能终端上的易用廉价服务 移动应用 社会发展趋势 通信全球化 逐渐增长的Internet 和email应用 远程办公 开发的标准和户操作性 更好的频谱利用率 更大的系统容量 更小、更轻，更便宜 技术趋势

Emerging Challenges for the Operator More 2G spectrum Falling barriers & market dereg International Competition Aggressive competition Innovative service offerings Maintaining profit margins Voice: A commodity service Players Upcoming 3G New 3G players Improving Cellular Technology Increasing capacity Maintaining profit margins Existing wireless operators have to look beyond traditional voice services in order to maintain and grow margins Loss of voice premiums New sources of revenue essential to build a business case for the development of new technology

Market Progression Toward Wireless Traditional Data Frame relay ATM TCP/IP Ethernet (wired) Wireless 3G Cellular/PCS evolving to UMTS Fixed wireless (LMDS/MMDS) Wireless LAN (802.11) Traditional PSTN POTS ISDN xDSL Traditional Wireless Voice TDMA CDMA GSM IS-41/GSM-MAP

Wireless Applications Vertical Business Mobile Office Personal Finance Entertainment Information & Travel Shopping Friends & Family Route Planning

Third Generation (ITU) Requirements High Speed Packet & Circuit Data 144 kbps vehicular 384 kbps pedestrian 2 Mbps indoor Asymmetric & Symmetric Transmission Enhanced Voice Communications Entertainment E-Commerce Information Transaction Travel Multi-Media Applications High Spectral Efficiency (“The wireless network must grow into (integrate with) the larger communications environment to prepare for tomorrow’s customers.”) Voice and data are converging over wireless as well as wired networks. In just a few years, forecasters predict wireless will become the preferred access technology, with as many wireless phones in the world as wired ones. And wireless users increasingly expect better data service. Yet current mobile wireless networks don't have the bandwidth for heavy data usage. So service providers, equipment vendors, and standards bodies around the world are developing the third generation of mobile wireless technology, which will carry data at rates well into the kilobit range and even into the megabit range for fixed, in-building use. Lucent, which has complete product lines for first (analog) and second (present-day digital) generation networks, is deeply involved in developing third-generation, or 3-G, wireless technology for all three major air interfaces – global system for mobile communications (GSM), time-division multiple access (TDMA), and code-division multiple access (CDMA). (too data focused?) Global Roaming 3G  3G 2G  3G Multi-Environment Operation Vehicular,Outdoor to Indoor, Indoor Office,Satellite

Wireless Technologies Map

Operational 3G Network

Air Interface Evolution to 3G IMT-2000 Capable Systems Existing Spectrum New Spectrum UMTS / W-CDMA GSM GSM+ w/GPRS (Rate 1&2) GSM++ / IS136 HS EDGE/GPRS (Rate 3&4) TDMA (IS-136) TDMA IS-136+ cdma2000 (5MHz) CDMA (IS95) CDMA (ANSI-95-B) Lucent provides an evolution path from all major 2G air interfaces and is the only vendor committed to supporting both TDMA and CDMA-based 3G standards. cdma2000 (1.25MHz)

Air Interface Evolution (Based on GSM Infrastructure)

2G vs. 3G Wireless Migration

3G Trials UMTS cdma2000 TDMA UWC-136HS/EDGE Sprint Bell Atlantic IDO / DDI UMTS NTT DoCoMo Vodafone- Air Touch Omnitel TDMA UWC-136HS/EDGE AT&T Wireless

Wireless Applications Environment External Databases Voice & Fax Mail server Post-paid Billing Personal Portal Pages Pre-paid Billing Email Server Customer care VPN Service Applications AnyPath Unified Messaging Voice Browser Internet Message Management Network Management E-commerce SMS-C WAP Gateway CBC USSD Server Internet LAN/Intranet Telephone Applications & Content Location Database SMS CB USSD CSD Packet IP/ATM Core Network MSC SDH/Optical GGSN/SGSN Mobile Access Micro browser

Packet Network Roadmap All IP Network Feature Server 3G Packet RAN Voice & Data Packet Core Network Packet Core/ Hybrid Packet? PSTN Gateway PSTN Wireless Multimedia: Integrated Multimedia via 3G Packet RAN Common Feature Servers for Wireless and Wireline Common Core Network for TDMA & UMTS Common VoP Network for TDMA, cdma2000 & UMTS Circuit RAN Voice PSTN PSTN Gateway Packet RAN Data Packet Core Network Circuit Core Wireless Packet Data: Add EDGE/GPRS Data Services (TDMA) Add cdma2000/M-IP Data Services (CDMA) Bypass PSTN with Packet IP/ATM Core Network (Voice Tandem) Circuit RAN Voice & Data PSTN Wireless Circuit Voice and Low Speed Circuit Data RAN - Radio Access Network

IP is Evolving to Meet the Need IP Extensions to Support Cellular Attributes Passive Connectivity Mobility Management Handoff Mechanisms Quality of Service Billing Mechanisms Ubiquitous Scalability

Introduction for 3G Mobile Communication System 1) Why 3G? 2) 2G to 3G Evolution 3) Comparison between 2G , 2.5G & 3G. 4) Standardizing Committee for 3G. 5) Two major proposals for 3G mobile communication. 6) Comparison between W-CDMA & cdma2000. 7) Harmonization efforts for global systems.

3) Comparison between 2G , 2.5G & 3G

6) Comparison between W-CDMA & cdma2000

BC (Before Cellular) Single Channel Radio Telephone Trunked Radio High power transmitter on a high spot Each user has own channel range of 50 miles Trunked Radio “Trunking efficiency” by sharing channels More spectrum efficient Cellular like Trunked radio, share the channels Low power= small coverage area Low power = frequency reuse More expensive infrastructure

Data - What Are Consumer Needs? Affordable Terminal Service “Attractive” terminal Human factors input / output Battery life Form factor Responsiveness Useful data (“walled garden”) Quick response Sorted response (time critical / location critical)

Applications 2G Wireless Networks 2.5G Wireless Networks Games with limited graphics Video-Games Audio/Video Clip Download Streaming Audio/Video Still Picture Web Surfing Banking Email with attachments SMS/ e-mail Information Services File Transfer Mobile Intranet Access Video Conferencing GSM GPRS UMTS Courtesy Don Gibson, Philips, Santa Clara, CA

Data Services- Middleware Preferred language should be XML compliant

Data Services- Business issues Usefulness of information Metcalfe’s law: value of network increase as the exponent of connected nodes: V~ KN Corollary: value of a wireless network is inversely proportional to the number of incompatible standards: V ~ 1/S High speed data Few will pay for high speed video (5 to 10 x voice call) Many will pay for useful, responsive service- high speed allows for responsiveness

SMS Very Popular, Except in US “Rule of thumb” at 30% pen usage explodes US penetration is about 35% But we have 5 standard - 10% max on any one Metcalfe’s law in reverse European environment calling party pays - no cost for incoming Internet cost per minute SMS is a way of making money go father Postulation - SMS would be explosive in North America if the internet was established as the common interface

2G offered customer’s clear value Smaller phone - in your pocket / purse Longer battery life - stop worrying about it Better MMI on the phone More affordable service - $30 bucks per month Maybe poor voice, but better coverage Less dropped calls - worked better

What is 3 G supposed to offer? Original Definition Common World Standard Higher Speed Data 384 kp/s for full area coverage 2 Mb/s for local area coverage Isochronous Services (voice / video) supported Data Packet, “always on” Service supported Reality Family of standards at least two bands, thanks to North America

Reasons for 3G Business drivers to move forward Need for more voice spectrum in some regions of the world Hope that data will become a major new source of revenue Technology push 3 G data business case is unclear In some markets, might be competitively priced to wireline, but not in US 3 G needs a data killer application, as voice revenue will drop Market must find “low payload” services with high utility Tie in with location and voice technology Carriers must aggressively price the service - “learning curve” Thus 3 G data business can be proven by 2.5 G data business

Operators’ 2.5G / 3G networks will not meet expectations Expected data rate [kbps] 9-14 100-300 384-2000 Actual data rate [kbps] 9-14 10-30 30-150

It’s all about Integration and Balance Nokia RIM Palm VII PIM Visor Web Access Messaging Voice ? For starters, let’s take a look at VisorPhone, our first step toward achieving a more fully integrated product. VisorPhone is optimized for two of the four key functions described earlier: PIM and voice. It SUPPORTS messaging and web access, but it is not really optimized for them. It does voice and PIM very well, but the other functions are not as optimized and work but could be done better if there were less trade-offs. Other examples of of this include the RIMM Blackberry product which is great at e-mail but lacks in PIM functionality, or the Palm VII which is great at quick Web access but lacks in messaging. In the case of VisorPhone, it runs on a voice network, not a data network, and as such can’t have always-on connection, so it is not ideal for e-mail. What we and everyone else in the industry has done so far is to optimize products for some features, and do the best we can on the others, recognizing inherent limitations.

Mobile Multimedia Services

General architecture of the IP-based IMT network platform Integration Applications AP1 AP2 AP3 APn Middleware Service support sub-layer Location Accounting/billing Media conversion Distribution Basic network management sub-layer RRM MM C/SM Security QoS ISDN/ PSTN IP-based transport NW IP IP IP IP Radio Internet IP IP Radio IP Radio General architecture of the IP-based IMT network platform

GSM GPRS UMTS 2G 与3G网络结构 PDN Other PLMN MSC SGSN GMSC BSC BTS Abis AUC VLR MSC SGSN GMSC BSC BTS Abis BSS AUC HLR EIR PSTN E D C F H A GSM GGSN PDN Gb Gp Gc Gf Gn Gr Gs GPRS Other PLMN RNC NodeB Iub RNS Iur Iu-CS Iu-PS UMTS Gi PLMN: Public land mobile network UMTS:Universal Mobile Telecommunication System

3GPP R4网络结构 R99 R4 SCP HLR SCP HLR MSC MSC MSC Server MSC Server MGW CAP Over TDM MAP Over TDM CAP Over TDM/IP MAP Over TDM/IP TUP/ISUP MSC MSC MSC Server MSC Server TDM TUP/ISUP/BICC H.248 H.248 TDM/ATM/IP MGW MGW RAN RAN RAN RAN RAN RAN R99 R4 注：R99与R4的分组域无变化，图中未画出。 承载独立呼叫控制 bearer independent call control BICC

3GPP R5网络结构图

Call State Control Function (CSCF) Home Subscriber Server (HSS) Transport Signalling Gateway Function (T-SWG) Roaming Signalling Gateway Function (R-SGW) Composite Gateway Media Gateway Control Function (MGCF) Media Gateway Function (MGW) Multimedia Resource Function (MRF) MSC Server Gateway MSC Server

WCDMA的切换 1、硬切换 移动台在不同频率或不同系统（GSM）之间的切换 2、软切换 移动台在同一频率下的不同基站之间的切换 3、更软切换 移动台在同一基站的不同扇区之间的切换

软切换与更软切换 更软切换： 在BTS内部完成 基站和移动台都按照RAKE 接收机方式接收 软切换： 网络需要一条独立的传输链路 UL/DL分集方式不同 MS:MRC RAKE合并 RNC:帧选择合并

WCDMA 压缩模式与切换 测量间隙 正常帧 正常帧 • 在压缩模式下需要更多发射功率=>影响WCDMA系统的覆盖 IS-95 GSM 频间测量的用途 用于频间和系统间切换 无频间测量， 很难利用多载波 用于切换 由于TX和RX的不连续，测量很容易 频间测量方法 压缩模式 测量间隙 压缩模式 正常帧 正常帧 • 在压缩模式下需要更多发射功率=>影响WCDMA系统的覆盖 • 在压缩模式帧中不能进行功率控制=> 需要更高的Eb/N0 => 影响 WCDMA 容量

向3G网络演进的策略 网络演进需考虑的问题 主要方案 平滑性和阶段性 保护用户权益 保护运营者投资 不能频繁改动网络 增强网络方案 新技术之间有一定间隔 无线技术的选择 核心网络兼容性 经济性 标准化程度 频率问题 试验及执照 主要方案 增强网络方案 保证业务的连续性 叠加网络方案 建立全新核心网和平台 全IP网络演进

二代向三代的过渡策略 两个问题 网络兼容与过渡 终端的兼容 第三代系统必然要在一段时间内与第二代系统相共存 双模甚至多模终端 ITU建议采用“软件无线电”技术。这种技术是采用空间软件下载和自适应技术来实现传统用硬件实现的功能，目前看来，实现这种技术仍存在困难。

向第三代演进的策略 策略1： 首先采用增强的第二代网络，延续一段时间。第三代的无线接入网通过适配功能单元接入到增强的第二代核心网；在较晚的时间再采用第三代的核心网络。第三代核心网络与第二代网络间采用互通功能单元进行互通，而逐步淘汰第二代核心网。

向第三代演进的策略 策略2： 从开始阶段就建设第三代的核心网络。第二代与第三代之间采用互通功能单元互通，逐步将第二代的无线接入网直联到第三代的核心网络，而淘汰第二代核心网。

两种演进策略的大致比较

R99 network architecture (3GPP TS 23.002) BSS RNS (RAN) CN A IuPS MS Gs Gn Gr Gc C D G F Gf Gi IuCS Gb SGSN Iur Iubis RNC NB Abis BTS BSC MSC VLR B GGSN GMSC HLR AuC EIR PSTN IP Networks Cu SIM-ME i/f or ME USIM SIM Um Iuu Gp E

2G network reuse CN With R99 the 2G CN can be reused and 2G subscribers will be migrated gradually to 3G Major CN evolution will appear with R4 and R5 (all IP) Operators need to have clear roadmaps and upgrade paths and shall only invest in a new CN when 3G traffic really takes off RAN Most of the 2G sites can be reused In urban areas where physical space for equipment is limited existing 2G cabinets shall be swapped for higher capacity / smaller footprint 2G cabinets to allow new 3G cabinets or for new 2G / 3G cabinets In urban areas antennas shall be swapped for wide band or dual band antennas

Network sharing - Technical solutions overview Site sharing Antenna sharing Grid sharing RAN sharing (NB and RNC) Full MVNO Geographical network sharing (national roaming) Technology sharing (UMTS/EGPRS/GPRS)

Network sharing - Site sharing MSC/VLR SGSN GGSN HLR GMSC MSC/VLR SGSN GGSN HLR GMSC RNC NB RNC NB NB

Comments on site sharing Site sharing is a “good practise” well developed for 2G (co-location) and generally allowed and even encouraged by regulators for 3G to limit planning and environmental impact An estimation states that for the UK 60% of the sites have at least two 2G tenants Site sharing always needs to be considered on the top of other higher level of network sharing Site sharing (co-location) ratio shall increase naturally with the number of operators

Network sharing - Antenna sharing MSC/VLR SGSN GGSN HLR GMSC MSC/VLR SGSN GGSN HLR GMSC RNC NB RNC NB NB

Shared antenna configuration examples UMTS single band antenna 1 x 2 Couplers (3dB loss ~ 50% interference load) UMTS A UMTS B UMTS/GSM1800 wide band antenna « Two in One» UMTS A GSM1800 A UMTS B 2 x2 Couplers (6 dB loss ~ 75% interference load) UMTS/GSM1800 wide band antenna UMTS A UMTS B GSM1800 A .

Network sharing - Grid sharing MSC/VLR SGSN GGSN HLR GMSC MSC/VLR SGSN GGSN HLR GMSC RNC RNC NB

Network sharing - RAN sharing MSC/VLR SGSN GGSN HLR GMSC MSC/VLR SGSN GGSN HLR GMSC RNC NB

Comments on RAN sharing RAN Sharing is standardised by 3GPP Some vendors have already announced the availability of equipment for 2002 but not all of them even if they claim that they can be ready in few months The number of supported NB is a limiting factor for the dimensioning of RNC and therefore the shared RNC solution shall be carefully analysed even in low capacity areas

...any questions are welcome! Thank you! ...any questions are welcome! 徐国鑫 xuguoxin@bupt.edu.cn 无线新技术研究室