Unity Eludes 3G Mobile Phone Standards

Communications is the buzzword that will ring in the new millennium. Among its many aliases are wired, global roaming, and wireless networking.

The large annual growth of mobile-phone usage is greatly changing the way we live. Congregating in Times Square to see the ball drop midnight on New Year’s eve may no longer happen. Some industry watchers predict that instead families will be informed of the new year’s arrival with nanosecond precision by colorful and intriguingly shaped personal information appliances.

In the United States, mobile phones are a phenomenon waiting to happen. Because this country has been well-served by the POTS, ISDN, and other LAN/WAN wired services in the past, mobile-phone sales were driven more by need than fashion. Certainly, salespeople and busy managers on the go aren’t the only ones that have mobile phones today, but shoppers using personal phones in U.S. grocery stores still is less prevalent than in Europe.

According to Ken Woo, assistant to the president and CEO at AT&T Wireless Services, about 5 million of the world’s 16 million CDMA IS-95 phones are being used in the United States.¹ Half of the 9 million AT&T Wireless customers use IS-136 TDMA phones and half continue with analog phones. They are not alone. Approximately 100 million wireless analog 1G phones still are in use worldwide.

By comparison, GSM, which started in Europe, now has reached more than 100 million subscribers in 120 countries, and its use is growing.2 Nevertheless, the greater spectrum efficiency of the new wideband CDMA and American and Japanese support are reasons that it is forecast to eventually overtake GSM in numbers of subscribers.3

One reason for so much recent press coverage of the lengthy deliberation over new 3G phone standards, such as wideband CDMA, is the tantalizing size of the market for mobile phones. In 1996, more mobile phones than PCs were shipped. The large projected growth is partially responsible for the need to develop a new standard.

A multitude of new services and applications which wide bandwidth will supply also need a new standard. For example, the marketing effort to differentiate 3G from 2G phones is concentrated on mobile e-mail and internet and stationary video conferencing and multimedia. Although there is some customer need for these capabilities today, network operators are counting on the latent demand for such services to be awakened by their availability.

Because the bulk of the phone market will be consumer-driven, another goal of a common, worldwide 3G standard is further cost reduction through very large-scale production. For phone users who really have to roam globally today, handsets for the Iridium satellite systems now are being sold for about $3,000 each. 3G phones won’t be commercially attractive, in spite of their added features, if they are much more expensive than current 2G models—about $100 to $200.

So where is the push for new 3G systems coming from? Not the customer.

“The phone manufacturer is the key driver at present,” according to Chris Foreman, marketing manager at Racal Instruments. “The network operator wishes to retain his customers and ensure optimum voice quality and call drop rates and tries to offer the latest features. The user is only interested in the capability to roam freely and maybe send data quickly and cheaply.”

The Clash of the Standards

IMT-2000, the ultimate 3G standard envisioned by the ITU, requires full coverage and high mobility for 144 kb/s (vehicular), full coverage and low mobility for 384 kb/s (pedestrian), limited stationary coverage for 2 Mb/s, high-spectrum efficiency compared to existing systems, and high flexibility.

Two of the favored CDMA 3G proposals are cdma2000 and W-CDMA. Wideband CDMA proposals from ARIB, ETSI, and TTA have been grouped together for the purposes of this comparison and referred to as W-CDMA. Cdma2000 refers to wideband CDMA proposals from TIA (TR45.5) and TTA. The term wideband CDMA is used throughout the article when a specific proposal is not referenced.

There is also a high-speed TDMA proposal, UWC-136, which is compared in Table 1 with the two CDMA 3G proposals.

All three schemes provide backwards compatibility with the GSM/MAP (SS7) network, but the air-interfaces are not compatible with each other. Both types of wideband CDMA proposals also include several advanced properties:

Multirate service.

Packet data.

Complex spreading.

A coherent uplink using a user-dedicated pilot.

Additional pilot channel in the downlink for beamforming.

Seamless interfrequency handover.

Fast power control in the downlink.

Optional multiuser detection.

Underlying any technical advantages one scheme may have over another is the fact that many U.S. operators of 2G networks are just beginning to profit from their original investments. As a result, there is a strong motivation for the 3G system(s) finally adopted to be evolutionary advances upon the present services—not something totally new.

Operators are developing 2½G features that don’t change the installed base too radically. For example, higher data rates can be achieved by using several IS-95 CDMA or GSM channels concurrently.

For GSM, this method of working is called HSCSD and now a standard for the first products that appeared in 1998. HSCSD provides data rates from 9.6 kb/s to 115.2 kb/s when using one to eight time slots, respectively. For IS-95 CDMA, the 1998 IS-95B version of the specification allows concatenation of up to eight codes for the transmission of higher bit rates up to 115.2 kb/s.

Other elements of 3G systems also can be viewed as enhancements to present technologies. A good example is the CA3D scheme used in the base stations proposed for wideband CDMA. Simulations show that six-antenna CA3D can increase cell capacity by a factor of 4.2 over a single antenna receiver where links are limited by interference. Antenna array diversity allows the base station to sort out multipath interference to make use of the additional power arriving at different times from the main signal.

Another base-station improvement is interference cancellation. It’s not practical to try to remove the interference caused by all other users, but subtractive cancellation can be used to remove the effect of the strongest interference source.

Wideband-CDMA error correction also has been developed beyond 2G to include sophisticated encoding techniques. Because 3G will carry both voice and data, it’s necessary to handle high and low bit-rate signals. A number of encoding/error correcting methods are used singly or in combination to achieve multirate transmission, high bandwidth, and orthogonality. These schemes take lots of MIPs to implement and change both the base and mobile stations significantly.

Finally, because the information bandwidth is increased, which allows a greater number of users within an RF channel, the crest factor of the resulting wideband-CDMA RF envelope will be higher than that of a conventional IS-95 CDMA system. A higher crest factor and a more stringent ACP specification for wideband CDMA systems require better transmitter power amplifier linearity.

If it all sounds technically complex, that’s because it is. On the other hand, technical strengths are probably not the main reason that one IMT-2000 proposal will win over the others, if indeed, a single one does.

In their book, Wideband CDMA for Third Generation Mobile Communications, the authors concluded, “In general, we can state that any of the proposed technologies, TDMA, CDMA, OFDM, or hybrid CDMA/TDMA could be used as an IMT-2000 air interface. The technical differences are not that large…[and] many of the noticed weaknesses of the previous schemes can be overcome by further development.

“For example, in the FRAMES comparison, CDMA was noted to be better suited for moderately varying bit rates rather than for packet data. Since then, however, a new packet access mode for wideband CDMA…has been developed. [Similarly], in FRAMES, no antenna diversity was used in the downlink, and thus, it was concluded that the downlink limits capacity. However, different solutions such as antenna diversity in the mobile station, base-station transmit diversity, and interference cancellation in the mobile station are being discussed as a solution for this problem.”⁵

The Plight of the 2½G Test Engineer

Mobile-phone system developers have access to technology that will allow them to improve their version of present 2G service to evolve toward the eventual IMT-2000 standard. Also, 2G phone production will not stop soon. So, the mobile-phone development and test environment will remain a multiple-standard one for some time. As shown in Figure 1, many more standards emerged at the beginning of this 2½G period which may ultimately coalesce into one or a few true IMT-2000 standards.

When we asked leading test and measurement companies “What’s an engineer to do?”, the answers fell into two categories. The first focused on a particular feature of a new or proposed wireless standard such as bandwidth.

From a power measuring point of view, Gurpreet Kohli, vice president of sales and marketing at Boonton Electronics said, “Present test equipment will need to increase the video bandwidth of the sensing elements. To accurately measure power in wideband-CDMA systems using modulation methods that result in pseudorandom or noise-like signals, the measurement of infrequent power peaks over long time periods must be conducted. Peak power meters that can analyze wideband-CDMA signals must have statistical analysis capability.”

Also commenting on the difficulties of measuring the power of wideband-CDMA signals was Steve Reyes, marketing manager at Giga-tronics. “The power meter must have enough amplitude bandwidth to track the power envelope of the signal over a wide dynamic range. The sampling rate of our latest instrument has increased from a few kilohertz to a random sampling rate of 2.5 to 5 MHz. The random sampling rate eliminates aliasing effects when undersampling a modulated signal with coherent properties.”

Other companies emphasized flexibility, especially during the 2½G period until the details of the IMT-2000 standard are determined. “The test specifications, performance requirements, and test methods or procedures have not been completed or defined,” according to Frank Palmer, a 3G applications specialist at Hewlett-Packard. “This means that test equipment should have as much flexibility as possible built into address potential changes in these areas. For example, the proposed chip rates have been 3.6864 and 4.096 Mc/s, but recently there has been discussion of 3.84 Mc/s.”

Similar concerns were echoed by Steve Stanton, product marketing manager for the Tektronix Wireless Communications Test Product Line. “Signal stimulus must provide the correct RF envelope, which requires that symbol rates, filtering, and orthogonal coding be correct for the standard to be tested. Some wideband-CDMA base stations may operate with multiple carriers, and amplifiers must be tested under real-world operating conditions.

“Wideband CDMA uses orthogonal codes of varying length, while cdmaOne [IS-95] uses 64-bit Walsh codes,” he continued. “These different codes will affect the RF envelope of the input signals to amplifiers, and the signal generators used for component test must accurately create this new signal. During the period when standards are not yet finalized, it will be necessary to have great flexibility in test equipment to provide future proofing.”

Unless you are actually developing 3G phones, you shouldn’t be too concerned about testing them right now. Instead, direct your attention to 2½G products which may have less technical support initially and are already shipping. You can obtain official details of the new 2G enhancements from the relevant standards associations listed in the glossary.

Remember that 2G, 2½G, and 3G equipment with their different CDMA and TDMA standards will coexist for several years. The fact that the three main IMT-2000 proposals do not have compatible air-interfaces only strengthens the argument for choosing very flexible test equipment.

References

1. “Tektronix Wireless Test Backgrounder: The Path To Third-Generation Mobile Systems,” Tektronix, p. 3.

2. Ojanperä, T. and Prasad, R., Wideband CDMA for Third Generation Mobile Communications, Artech House, Boston, 1998, pp. 5, 9.

3. “CDMA Wireless Business Opportunities,” Datacomm Research, 1998.

4. Ojanperä, T., and Prasad, R., op cit., p. 12.

5. Ibid., pp. 398, 399.

Glossary

1G first generation

2G second generation

2½G transition period between 2G and 3G

3G third generation

8-PSK 8-phase, phase shift keying modulation scheme

ACP adjacent channel power

ACPR adjacent channel power ratio

AMPS Advanced Mobile Phone Service

ANSI American National Standards Institute

ARIB Association for Radio Industry and Business (Japan)

ARQ automatic repeat request

CA3D coherent adaptive antenna array diversity

CDF cumulative distribution function

CDMA code division multiple access

cdma2000 wideband CDMA as developed in the United States, backwards compatible with cdmaOne

cdmaOne original American CDMA IS-95 system

DS direct spreading

DSS direct sequence spread spectrum

DTCH dedicated traffic channel

EDGE enhanced data rates for GSM/global Evolution

EIA Electronic Industries Association

ETSI European Telecommunications Standards Institute

FRAMES 1995 research project aimed at defining an UMTS air interface proposal

GMSK Gaussian filtered minimum shift keying

GPS/MAP global positioning system/mobile application part

GSM global system for mobile communications

HSCSD high-speed circuit switched data

IMT-2000 The final, worldwide mobile telephone standard

IS-136 U.S. TDMA standard with digital control channel, backwards compatible with AMPS, also called D-AMPS

IS-136+ modulation enhancements to IS-136

IS-136HS wider carrier bandwidth enhancements to IS-136; 200 kHz and 1.6 MHz; adopted EDGE proposals

IS-41 Wireless Intelligent Network (WIN) standard supported by UWCC

IS-54 U.S. TDMA standard with analog control channel, backwards compatible with AMPS

IS-95/B U.S. CDMA system, backwards compatible with AMPS

ISDN integrated services digital network

ITU International Telecommunications Union

Mc/s mega chips/second

MIPs million instructions per second

OFDM orthogonal frequency division multiplexing

PDC personal digital cellular, Japan only, some similarities to IS-54

POTS plain old telephone service

QPSK quadrature phase shift keying

SS7 signaling system 7

TDMA time division multiple access

TIA Telecommunications Industry Association, formerly part of EIA

TTA Telecommunications Technology Association (Korea)

UMTS universal mobile telecommunications system

UTRA UMTS terrestrial radio access

UWC-136 Universal Wireless Communications-136, extended version of IS-136 proposed by UWCC

UWCC Universal Wireless Communications Consortium

Video Bandwidth Refers to the amplitude bandwidth of an RF sensor required to track the modulation envelope.

W-CDMA wideband CDMA proposals from ARIB, ETSI, and TTA

Real-Time Spectrum Analyzer

The Model 3066 DC to 3 GHz Spectrum Analyzer continuously captures a 5-MHz wide band of frequencies during successive very short frame periods. By positioning the frequency band to coincide with a CDMA or GSM channel, all bursts and modulation can be captured and displayed. Both time-domain and frequency mask triggering are provided. Displays include spectragrams, spectrum dB amplitude vs frequency, waterfall, modulation vector diagram, and modulation analysis. Separate time and frequency samples are stored in memory and can be recalled to show events ahead of and after the trigger. $43,500. Tektronix, (800) 426-2200, ext. 1085.

Arb With Five Types of Noise

The 100 MS/s Model 395 Universal Waveform Generator provides 16 types of synthesized waveforms, including sines to 40 MHz, squares to 50 MHz, triangles to 10 MHz, and user-defined arbitrary signals. Output is up to 10 Vp-p into a 50-W load. The five programmable noise functions produce analog white noise to 10-MHz bandwidth, digital noise to 50-MHz bandwidth, comb function, signal-plus-noise, and signal-plus-comb. White and digital noise have programmable sequence length. These signal-generation capabilities, combined with flexible triggering, modulation, and gating facilities, allow CDMA and GSM waveforms to be simulated for device and subassembly testing. From $3,995. Wavetek Wandel & Goltermann, (800) 223-9885.

Integrated CDMA Tester

The Rohde & Schwarz CDMA Test System TS 8180 comprises the CDMA Radiocommunication Tester CMD80, two signal generators, a power supply, and the interface for CDMA phones. The system was developed in close cooperation with and its software is licensed by QUALCOMM. It features fully and semiautomatic tests of CDMA phones according to IS-98-A and J-STD-018, including the AMPS standard. The system components are integrated into a 19″ rack. As well as production testing, the TS 8180 can be used to develop or verify CDMA terminal equipment. Call company for price. Tektronix, (800)-426-2200.

RF Power Meter

The 4530 Series RF Power Meter makes peak and CW power measurements with up to 20-MHz video bandwidth, depending upon the sensor used. Statistical features include histograms and CDFs for analysis of complex signals such as CDMA and HDTV. The effective sampling rate is up to 50 MS/s for repetitive signals and 1.25 MS/s maximum for single-shot capture. Measurement of average, maximum, and minimum power; the peak-to-average power ratio; and RF voltage and display of power vs time also are provided. GPIB and RS-232-C ports are standard; a second channel is optional. Single-channel 4531A: $3,650; dual-channel 4532A: $5,650. Boonton Electronics, (973) 386-9696.

Fast-Sampling RF Power Meter

The single- or dual-channel 8650A Series Universal Power Meters feature a 20-MS/s maximum uniform sampling rate for CW signals and a 2.5- to 5-MS/s random sampling rate for modulated signals. Random sampling minimizes aliasing effects caused by undersampling complex signals such as 3G wideband CDMA. Statistical power measurement analysis, automatic burst measurement, time gating, and peak hold functions are provided. A range of high-frequency RF diode sensors supports power measurements up to 40 GHz with a 10-MHz video bandwidth and an 80-dB dynamic range. Single-channel: $3,310; dual-channel: $4,870. Giga-tronics, (800) 726-4442.

Automatic CDMA Test System

TASKIT

/CDMA TestPilot™ Software turns a stand-alone base-station emulator into an automatic test system capable of CDMA handset performance characterization. The Windows 95/98/NT-compatible package translates test specifications to instrument settings, configures the test instruments, performs automatic test execution, and logs results. TestPilot includes drivers that support the HP 8924C Mobile Test Set and Tektronix’s CMD80 CDMA Radiocommunication Tester. Standard, predefined test suites are provided, but it also is possible to control advanced test configurations using a pair of base-station emulators for soft handoff testing or an external spectrum analyzer for mobile transmitter measurements. From $8,950. Telecom Analysis Systems, (732) 544-8700.

cdma2000

W-CDMA

UWC-136

Supports high data rates