The coming generation of digital personal communications systems (PCS) is about to change the face of RF communications test equipment. The changes will be more than skin deep, because the digital nature of these new communications systems requires new tests as well as new test equipment.
PCS will operate at 1,900 MHz in the United States and utilize digital modulation techniques and access technologies such as time division multiple access (TDMA) or code division multiple access (CDMA). TDMA is the modulation scheme used in the Global System Mobile (GSM) communications service in operation throughout Europe. TDMA also is used for Japan’s Pacific digital cellular system and personal handy phone system service, and other countries world-wide.
CDMA is not currently in full operation in a public network. However, it has been used in many trial networks and has been selected for future use in some countries, including Korea and Thailand. The two digital modulation schemes will compete for dominance in the U.S. PCS market.
While both TDMA and CDMA are digital modulation techniques, they differ in the way calls are separated. In TDMA, the user is assigned a time slot and frequency channel. In CDMA, all users occupy the same wideband spectrum at all times during a conversation, with the individual signals identified by a unique code and spreading sequence.
PCS is the logical next step in sophistication from an 800- MHz cellular telephone service. Unlike cellular, which got its start and continues to thrive primarily on revenues from business users, PCS 1900 will offer low-cost equipment and services to virtually anyone. The services will include voice, fax, e-mail and data. The bidding process for PCS licenses began in the United States in December 1994.
PCS Demands on Test Equipment
Like all new digital communications systems, PCS at 1,900 MHz (frequently called DCS 1900) conforms to the Open Systems Interconnection (OSI) model prepared by the International Standards Organization. This model consists of seven layers, each representing a portion of the overall network. The physical layer (Layer 1) is the most fundamental element of the model and applies in this case to the actual physical medium between base stations and mobiles.
Layer 1 not only fully describes the frequencies, but also the modulation type, power levels, power ramping and the slot frame structure implicit in a TDMA system. CDMA has other considerations, since it is a spread spectrum technique and differs considerably from TDMA.
Layer 2 structures the messages to match the physical constraints specified in Layer 1. Layer 3, the network layer, manages all calling and related activity of the network, including call management, mobility management and radio resource management.
Several characteristics of digital communications systems impact the equipment used to measure their performance:
Signaling and traffic information is purely digital and, in the case of DCS 1900, assigned to time slots. In analog systems, the information is assigned solely on the basis of frequency channel.
The phase and amplitude of the carrier are continuously varied over time. Conventional signal generators, modulation analyzers and power meters cannot effectively generate and simulate this type of signal.
DCS 1900 requires strict time synchronization between the base and mobile stations. To make effective measurements, test equipment also must synchronize in the same way.
TDMA systems have a transmit dynamic range of greater than 70 dB. When coupled with the need to sample at a multiple of the symbol rate, this stretches the limits of most A/D converter technologies.
Frequency hopping, a characteristic of GSM-derived systems such as DCS 1900, requires test equipment that analyzes or generates a frequency-hopping signal.
Speech is digitized as 64 kbit/s PCM and then compressed to 13 kbits/s in the case of DCS 1900. This makes single-tone testing unacceptable because it does not adequately approximate speech.
CDMA imposes additional constraints on test equipment. It uses a spreading sequence to distribute the RF energy over a 1.25-MHz bandwidth. It is not possible to use conventional RF power meters and spectrum analyzers to measure the total in-band power or interference to other users.
Specific Test Considerations
Signal generators designed to test DCS 1900 and other candidate systems must convert a bit sequence into a corresponding modulated RF carrier. Conventional signal generators cannot do this because there is no direct way to control the phase of the carrier.
Newer instruments have I/Q modulators that accomplish this task. Modulation bandwidth and switching times must be sufficient for the latest digital systems (Figure 1).
Another RF evaluation function becoming more important is simulation of multipath, or fading, conditions. More and more RF path simulation systems imitate propagation conditions over various types of terrain. These new instruments often factor-in vehicle speed within the simulation, making accurate performance evaluations more meaningful (Figure 2).
Receiver sensitivity is no less important in digital systems than it has always been in analog communications. One method of testing receiver sensitivity that does not require the mobile terminal to be opened is the loop-back technique. In this scenario, the test equipment sends a data stream on its modulated carrier to the mobile station and commands the mobile stations to remodulate and retransmit the demodulated data back to the test set.
The test set can then compare the data stream received from the mobile with the original pattern. The signal level of the modulated RF carrier produced by the test equipment is reduced until the error rate exceeds an acceptable level.
The RF level at which this occurs is then a measure of the receiver sensitivity of the mobile UUT. This replaces signal-to-noise and distortion tests that have been the criteria for analog mobile radio receiver sensitivity. Digital radios are evaluated for their performance in one or more of these areas:
Bit error rate (BER)—the percentage of bad bits in the entire transmission.
Frame-erasure rate—the percentage of bad frames.
Residual bit-error rate—the BER measured only in “good” frames.
Some traditional measurement routines are retained, such as co-channel and adjacent-channel rejection, blocking, and spurious response. TDMA systems also require measurement of adjacent time-slot rejection. Transmitter tests include measurement of mean RF carrier power during a signal burst, power-time profile of a burst, phase and frequency errors, and spectra caused by modulation and switching transients.
Examining the Next Two Layers
For the test equipment to examine the data in a wireless system, it must synchronize to the timing of Layer 1 and the message structuring of Layer 2. If possible, this should be done without need for an external time slot or frame triggers.
If trigger information is provided externally, timing offsets caused by external influences may cause errors in the evaluation of physical parameters and make it impossible to interpret data. It requires the test equipment to generate and interpret broadcast control channels, training sequences and other signaling information. The test equipment also must map logical channels onto the physical channels of time and frequency.
In Layer 3, familiar terms found in analog networks appear. Registration, call setup, call clearing, intracell and intercell channel switching, RF power-level control and queuing are all found in Layer 3.
Test equipment designed to address Layer 3 requires substantial amounts of processing power, memory and software to completely simulate the communications environment. That is, in essence, what a test set for a mobile station must do—convince the mobile that it is connected to a functioning network. Errors must be detected, logged and stored. Most likely as DCS 1900 systems begin appearing, test equipment that combines both RF parameter and Layer 3 protocol testing will be introduced.
A New Measurement Environment
Without some scientific leap, coming generations of wireless communications services will all use digital modulation techniques of some kind. There is little doubt that new digital modulation schemes will use the full spectrum and provide the highest levels of signal quality over the most varied signal conditions.
Each system will require sophisticated equipment to test the fundamental properties of the network that are contained in Layers 1, 2 and 3 of the OSI model. This equipment will combine RF parametric testing in the time, frequency and modulation domains with Layer 2 and Layer 3 protocol analysis and will be easier to use than its less-sophisticated analog predecessors.
About the Author
Bob Buxton is Product Marketing Manager of RF and wireless test products at Tektronix. He was awarded an M.Sc. degree at the University College of London and is a Chartered Engineer and member of the Institution of Electrical Engineers. As an internationally noted lecturer, Mr. Buxton has addressed audiences on the topics of digital modulation techniques and GSM and other digital wireless standards. Tektronix, Inc., P.O. Box 500, Beaverton, OR 97077, (503) 627-5757.
Copyright 1995 Nelson Publishing Inc.
May 1995
|
|