[an error occurred while processing this directive]

here. Requires Adobe Acrobat Reader,

free download.]

Bluetooth wireless technology is an open specification for a wireless personal area network that provides limited-range, wireless communications for voice and data transmissions between information appliances. It is a promising solution for portable devices and for interconnecting computer and communications devices via a radio link, rather than with cumbersome cables. The Bluetooth module includes a radio transceiver that can be built into various host devices with a need to communicate with other devices.

Mainstream applications include mobile phones, laptop PCs and personal digital assistants (PDAs). All connections are real-time, fast and secure.

Operating in the 2.4 GHz industrial/scientific/medical (ISM) band, interference can occur between Bluetooth and other devices such as wireless local area networks (LAN) services and appliances such as microwave ovens.

To help Bluetooth operate successfully in the crowded environment of the ISM band, modulation is based on frequency-shift keying (FSK). FSK modulation is used because it has reduced susceptibility to interference. However, Bluetooth goes a step further by employing a spread-spectrum frequency-hopping scheme.

Operating at 1.6 khops/sec in a pseudorandom sequence, the transmitter alternates among 79 carrier frequencies over the range 2.4 to 2.4835 GHz, with channels spaced 1 MHz apart. This eliminates the need to perform rigorous radio planning and improves link reliability in the presence of other ISM spectrum users.

Governed by the pseudorandom generator in the master device, the signal sent to the slave hops from channel to channel. The signal dwells at each channel frequency for a 675 ms interval in what is called a ‘time slot’. The master transmits during even-numbered time slots and is the slave during odd-numbered time slots. Voice bits and data bits are transmitted in packets. These packets may straddle one, three, or five of the 675 ms time slots.

In single-slot mode, the slots are frequency hopped in accordance with the pseudorandom sequence governed by the master (see Figure 1). Notice that in multi-slot mode, the signal dwells at the same frequency for two intervals (2 × 675 ms) and then does a ‘catch up’, returning to the governing pseudorandom sequence.

A Bluetooth radio can be implemented with a number of system architectures, from direct frequency-modulated, voltage-controlled oscillator/analog discriminator to IQ modulated/digital demodulator designs. Figure 2 shows a typical block diagram of a Bluetooth device in which the transceiver employs an IQ modulator/demodulator. This figure depicts the RF transmitter, the receiver and the controller/processor.

Typically, designs employ a single local oscillator — such as the voltage-controlled oscillator/phase lock loop (VCO/PLL) shown in this figure — that switches between receive and transmit functions. The transmitter up-converts the baseband information arriving at the digital-to-analog converter(s) (DAC) to the frequency-modulated carrier. Frequency hopping and bursting are performed at this level.

Likewise, the receiver downconverts and demodulates the RF signal and feeds the baseband signal to the Gaussian FSK modem.

The scheme relies on GFSK using two closely-spaced frequencies to represent a ‘0’ or ‘1’. The peak allowable frequency deviation from the carrier center frequency is 175 kHz. Unlike standard FSK, GFSK employs a Gaussian filter prior to modulation and transmission. This filter rounds each transmitted pulse, thereby reducing spectral width and enabling the signal to fit more efficiently into a narrower frequency band.

The term BT = 0.5, with reference to Bluetooth modulation, refers to the bandwidth-time product, a design parameter that dictates the time spread of the frequency-shaping pulse. It is the product of the 3 dB baseband bandwidth of the Gaussian filter and the bit duration. A value of 0.5 results in a relatively small pulse width and thereby reduces the amount of intersymbol interference.

As for data throughput, the base rate is 1M symbols/s. The effective data throughput is somewhat lower, however, because of protocol overhead. The asynchronous channel supports an asymmetric link of 721 kb/s, maximum, in either direction while enabling 57.6kb/s in the return direction. The alternative is a 432.6 kb/s symmetric link.

With regard to power and depending on the class of service, the transmitter output ranges from 0.25 to 100 mW. The maximum permissible level is Class 1 at 100 mW (+20 dBm). At this power level, the operating range is approximately 100 meters.

The Bluetooth device can operate in different modes: normal, transmitter (TX) test and loopback test (illustrations of the loopback and transmitter mode appear in Figure 3). Transmitter Mode is the normal mode (the same mode in which standard Bluetooth communication occurs). In this mode, if a test system is programmed to function as a master and sends POLL packets to the Bluetooth device (slave), the device will confirm the reception of these packets by sending back a NULL packet. (The description of POLL and NULL packets can be found in readily availble documents).

In this mode the test equipment is able to examine test packets and determine characteristics such as initial carrier frequency tolerance.

The loopback test mode is where the Bluetooth device (slave) is asked to decode the packets sent by the test system (master) and respond by sending back the payload using the same packet type. The bit error rate test requires such loopback capability. The tester sends a packet such as a PN9 sequence. The device under test will then sends back those same packets. The tester compares what is received with what was sent. If they are all the same, the bit error rate is zero. If 1% of the bits are wrong, the BER is 1%.

Different set-ups may be used for Bluetooth transmitter tests, depending on whether one is testing an entire Bluetooth device, an RF transmitter or an RF component of the transmitter.

One way to test the transmitter performance of a full Bluetooth device is to use a Bluetooth test set. The test set and device under test (DUT) form a piconet where the tester acts as master and the DUT acts as slave. The test set establishes a link with the device in either the normal or test mode using the standard Bluetooth protocol. With the device in test mode, the test set will acquire complete control of DUT operation.

For instance, the test set can put the device into loopback mode or transmit mode, disable frequency hopping and ask the device to transmit at specific frequencies as required by the Bluetooth RF test specification.

Three other types of transmitter measurement set-ups are illustrated in Figure 4. These three set-ups require the use of a signal analyzer, which could be a spectrum analyzer or a vector signal analyzer. Additional equipment includes signal generators and possibly a power meter, power supply, oscilloscope and network analyzer.

Set-up 1 is an example of a set-up to test the transmitter performance of a full Bluetooth module, while set-up 2 is used for testing only a Bluetooth transmitter. Set-up 3 is for testing RF components of a transmitter.

Set-up 1 differs from the set-up of Figure 3 in that there is no Bluetooth communication established between the device and the test equipment, so the test equipment doesn't have any control of the DUT operation. For this set-up, a special internal test facilities utility must be implemented in the device. This utility must have the ability to instruct the device to transmit the packets it receives. This will enable a Bluetooth signal from the signal generator to be fed to the Bluetooth device receiver and then be looped back through its own transmitter for analysis.

In set-up 2, the utility must have the capability to control the type of transmission — frequency hopping on or off, different types of packets, etc. — to provide the right conditions to test the Bluetooth transmitter.

Set-up 3 can be used in testing the amplifier of a Bluetooth transmitter, as well as in a variety of other tests.

If a direct cable connection is not possible between the Bluetooth device and the measurement equipment, a suitable coupling device, such as an antenna, will be necessary. The path loss between antennas must be accounted for in the calculations. This can be evaluated using a network analyzer.

This article has discussed the tests required for Bluetooth devices in a highly abbreviated fashion. Given the enormous growth expected in the near future for Bluetooth-enabled devices, regulatory standards are adapting and changing to meet the safety and technical challenges of new Bluetooth features and products. Developers should consult the latest standard information to ensure proper compliance.

1. “Bluetooth RF Measurement Fundamentals,” Agilent Application Note 1333-1.

2. “Specification of the Bluetooth System — Version 1.1, Volume 1 “Core,” February 22, 2002 — Bluetooth SIG.

3. “Mobile Communications,” Jochen H. Schiller, Addison-Wesley, 2000.

For additional information, go to the Agilent Website at www.get.agilent.com.find/bluetooth/. For full information on the tests required, refer to the Bluetooth RF Test Specification, prepared by the Bluetooth SIG at www.Bluetooth.org. Synopses of the specific tests discussed in this article are available from Agilent's Web site under “Transmitters Measurements,” “Transceiver Tests” and Receiver Measurements.”