Essential tests for Bluetooth products

Every Bluetooth product has to be passed as suitable for use under a rigorous qualification program. Wilfried Tiwald of Rohde & Schwarz describes the company's TS8960 test system

An essential precondition for the success of Bluetooth technology is the interoperability and undisturbed function of different products. To ensure this, the Special Interest Group (SIG) has defined a qualification program that has to be passed by every product to be marketed under the Bluetooth label.

Bluetooth operates in the license-free ISM (industrial scientific medical) band at 2.4GHz. A data rate of I Mbit/s makes this standard ideal for wireless data communication in many applications.

The TS8960 test system has been designed for RF measurements in line with the Bluetooth qualification program and for in-depth measurements in development and quality assurance. According to the specification, two test modes are available for RF measurements, the Tx test mode and the loopback test mode. In the Tx test mode, the EUT sends a settable data pattern; in the loopback test mode, the data pattern sent by the test system to the FUT is returned and is thus available for evaluation.

In line with the test specification, the TS8960 offers 16 standard test programs for measuring the RF characteristics of an EUT. First up are the transmitter tests to determine the output power of the EUT, the quality of the RF output spectrum, and modulation parameters such as frequency deviation, frequency accuracy and drift

Dirty transmitter

The TS8960 also provides a 'dirty transmitter' test. Since Bluetooth units are normally stationary, there are no fading effects. However, the signals used for measuring the receiver sensitivity are not completely undisturbed. The Bluetooth specification defines a dirty transmitter, the signal of which features a drift for many parameters such as frequency, symbol duration, deviation and modulation index. The dirty transmitter is realised in the TS8960 by the interplay of several components. The Bluetooth signalling unit produces the symbol drift, and the SMIG03 varies the deviation, frequency and modulation index.

Other receiver tests include the measurement of carrier/interference performance, blocking performance and intermodulation performance.

The specification defines an "out-of-band spurious emissions" transceiver test. According to the European standard, spurious emissions are searched for only in a frequency range of up to 12.75GHz, but the American standard prescribes measurements up to 25GHz.

Other tests can be performed in addition to the measurements defined by the test specification. For example, the sensitivity test can be used to determine the current receiver sensitivity. For this measurement, the test signal level can be reduced until the bit error rate just exceeds the 0.1% defined by the specification. The level at which the BEB is still within the 0.1% limit corresponds to the receiver sensitivity. Another example is the free receiver measurement which permits user-configured receiver tests to be performed.

The instruments in the test system are controlled by a system controller via the IEEE bus and an Ethernet link. The signalling unit in the system is responsible for call setup, test mode selection and the BER measurement. It contains an RF front-end for transmission and reception. Two Vector Signal Generators SM1003 are provided in addition. One produces the useful signal, the other the bluetooth-modulated interference signal for receiver tests. A third generator provides the unmodulated interference signal for measuring blocking and intermodulation suppression. The spectrum analyser measures the power, the modulation spectrum and the spurious emissions

Minimal errors

All signals are amplified and filtered in the signal switching and conditioning unit (SSCU), which contains relays, directional couplers, combiners, amplifiers and several filters. Since accurate level values are required for RF measurements, the TS8960 performs fully automatic path compensation to minimise measurement errors. For this purpose, two RF probes are installed at essential test points in the SSCU. Channel A of the Dual-Channel Power Meter NRVD is used for monitoring the transmit level of the simulator, and channel B for monitoring the received levels. Any level errors that occur in the current measurements are automatically corrected with the aid of stored reference values.

A Rubidium frequency standard supplies the highly accurate reference frequency. The system can optionally be equipped with power supply units for the EUTSs.

The standard model of the Test System TS 8960 is designed for fully automatic measurements of conducted interference up to 12.75 GHz. Options are available for measurements up to 25 GHz and for measurements of radiated spurious emissions.

All parameters can be varied. All test cases can also be performed with parameters that differ from those in the test specification. For example, the level and frequencies of the test signal and noise signal generators can be varied in a wide range for receiver tests. Packet types and even signalling parameters can also be changed.

An easy to use graphical interface ensures straightforward operation of the TS8960. Menus are available for equipment under test, test cases, path compensation, self-test and utilities.

In the EUT menu, the characteristics of the EUT, such as power class, Bluetooth address, signalling information, supported data packets, antenna gain, and so on, must be entered. Subsequently, one or more test cases can be selected and executed in the test case menu. When the self-test menu is selected, a comprehensive self test is performed.

After a test has been completed, the system documents all the main measurement results in a comprehensive test report and in plots. Zoom and marker functions allow fast and accurate evaluation of graphics.

Rohde & Schwarz
h122@industrialnetworking.co.uk