4336 views|0 replies

52

Posts

0

Resources
The OP
 

Bluetooth transceiver IC test [Copy link]

The first official version of the Bluetooth specification, version 1.0, was released in July 1999. Since then, many manufacturers have launched cost-effective integrated circuit chips that support Bluetooth products. As Bluetooth products become more and more popular, manufacturers need to complete a large amount of testing work at a lower cost. This article focuses on the various test parameters defined in the Bluetooth technical specification for Bluetooth RF front-end transceivers.

Today, there is hardly any electronic engineer who has not heard of the concept of "Bluetooth". The word comes from the Danish King Harald Blaatand in the 10th century AD, who established a communication system in Norway and Denmark to connect his subjects. Bluetooth technology was developed to enable personal digital assistants (PDAs), mobile phone peripherals and other mobile computing devices to communicate without using expensive dedicated cables. For this reason, Bluetooth is also called "Personal Area Network (PAN)". The most basic requirements for Bluetooth products are low price, high reliability, low energy consumption and limited working range.

Initially, Bluetooth was defined as short-range communication (10 to 15 meters) using the globally applicable 2.4GHz ISM band. However, recent improvements by chip manufacturers have pushed Bluetooth technology far beyond its original design level. Some OEM manufacturers hope to use Bluetooth technology in 20 to 30-meter office environments and 100-meter open environments. They expect to use Bluetooth as a network connection technology to enable laptop users to access the local area network through a wireless access point.

Bluetooth technology consists of four main parts: application software, Bluetooth stack, hardware and antenna. This article focuses on the hardware and RF front-end transceiver, and introduces the various test parameters defined in the Bluetooth technology specification.

Bluetooth transceiver

(Figure 1) Typical RF Bluetooth block diagram

The test requirements for an integrated RF transceiver can be illustrated by a typical RF Bluetooth block diagram (Figure 1).

◆Bluetooth transmitter Bluetooth wireless signal is modulated by Gaussian frequency shift keying (GFSK). After the transmission data (Tx) is filtered by a Gaussian filter, the output of the filter is used to modulate the VCO frequency. According to the logic level of the serial input data stream, the VCO frequency will deviate from its center frequency to the positive and negative ends. The offset determines the modulation index of the transmitter. The modulated signal is amplified and then transmitted by the antenna.

The Bluetooth radio operates in half-duplex mode, with an RF multiplexer switch (located before the antenna) connecting the antenna to transmit or receive mode.

◆ The Bluetooth receiver is similar to the receiving part of the device. The GFSK signal transmitted from another Bluetooth device is also received by the antenna. During this period, the switch is connected to the low noise amplifier (LNA) to amplify the received signal (Rx). The next mixer converts the received signal down to the IF frequency (usually a few MHz). During this step, the PLL/VCO part used for transmission is used as the receiver down-conversion local oscillator to demodulate the IF signal and recover the data.

Spread Spectrum


(Figure 2) The FHSS system is used when the carrier is modulated with the original data and transmitted using a frequency range consistent with the frequency hopping code of each link endpoint.

A unique feature of Bluetooth wireless communication is its use of spread spectrum technology, which was originally developed for military applications where wireless data transmission must be secure and reliable. Traditionally narrowband applications consume more power and stay on one frequency for a long time, making the spectrum easily detectable; after the transmitter power is distributed (spread) over a larger bandwidth, the signal now looks more like random noise, which is equivalent to sacrificing bandwidth efficiency in exchange for reliability and security. Due to the lower power density, these systems have less interference to other signal receivers, and even if there is a signal loss band, the data can be recovered at other frequencies, thereby enhancing the ability to resist interference and noise. The two most important forms of spread spectrum are frequency hopping (FHSS) and direct sequence (DSSS). FHSS systems are used when the carrier is modulated with the original data and transmitted using a frequency range consistent with the frequency hopping code of each link endpoint (Figure 2). In this way, data lost due to interference on one frequency can be transmitted on another frequency. The spreading code generator in FHSS directly modulates the carrier frequency using GFSK modulation technology.

GFSK modulation

GFSK is a modulation technique that linearly modifies the carrier frequency of a portion of the carrier period by data within the duration of a bit. The frequency change rate is a function of the data rate, and the frequency change size is a function of the data amplitude. The relationship between them is expressed by the modulation index β.

The FSK signal modulation index β is calculated by the following formula:

β=2Δf/fi

Here fi is the data frequency in Hertz (typically 1 MHz for Bluetooth) and Δf is the frequency deviation of the carrier.

If 140kHz is selected as the Bluetooth carrier frequency deviation, then:

β=2Δf/fi=280kHz/1MHz=0.28 and

FFSK=Acos(2πfc(t)+0.28π∫m(t)dt)

Here A is the amplitude of the digital data, and m(t) is the digital data that lasts for one bit time and has a DC level of ±1.

A voltage-controlled oscillator with a sensitivity of 140kHz/V can be used as an FSK modulator, with β=0.28. The input data stream is usually passed through a limiter to ensure that the circuit frequency difference is 140kHz.

The carrier frequency deviation (transmit mode) depends on the amplitude of the input data stream, and vice versa. The data amplitude of the demodulated carrier is a function of the carrier deviation (receive mode), which is very important for the system bit error rate (BER).

BER is a function of the power of each transmitted bit relative to the noise contained in each bit. The relationship between them is expressed as Eb/No, that is, the power-to-noise ratio of each bit. Eb/No can be improved by reducing receiver noise or increasing the transmit power, or by increasing the power of each transmitted bit. Increasing the carrier frequency deviation can increase the power of each transmitted bit, thereby increasing Eb/No and reducing the bit error rate; but its negative impact is that increasing the frequency deviation will increase the bandwidth and reduce the number of channels in the system.

Effective communication requires a minimum bit error rate, and the Bluetooth technical specification stipulates that the BER is 0.1% at 72dBm, that is, there is 1 error in every 1,000 bits of data stream. The consistency specification requires that the measured sensitivity (as BER) exceed 1.6 million bits on three frequencies. Since this test uses a standard single-slot (DH1) data packet and takes at least 25 seconds, in order to save time, even if the number of frequencies is reduced in actual applications, only a small number of bits are measured.

Bluetooth transceiver test specifications

The Bluetooth standard stipulates the following requirements for RF carrier modulation data:

Modulation: Gaussian frequency shift keying (GFSK) Gaussian filter: 0.5 Output power: 0dBm and +20dBm Data rate: 1Mb/s Channel bandwidth: 1MHz Frequency deviation (Δf): 140kHz~175kHz (modulation index 0.28~0.35) Bit error rate (BER) sensitivity: 0.1% @ -72dBm

Bluetooth defines 1mW or 0dBm as the nominal system, with a peak transmit power of no more than 20dBm, designed for short-range operation without interfering with other wireless systems, using Gaussian filtered frequency shift keying (GFSK) modulation in a 1MHz bandwidth (carrier spacing of 1MHz). The United States and Europe (except France and Spain) have 79 1MHz channels, while France, Spain and Japan have only 23 1MHz channels in the 2.4GHz range.

Figure 3 shows the packet timing protocol using DH1, DH2 and DH5 packets for 7 consecutive time intervals. Since the transmit and receive packets are the same length, DH1 has a symmetrical link, and the transceiver transmits data in even time intervals and receives data in odd time intervals. DH3 uses 3 time intervals and DH5 uses 5. The payload of DH3 and DH5 packets is longer, and since the overhead of the protocol is fixed (access code + header), it can provide higher data throughput.

For the transmitter, the following are some of the more important test parameters:

Modulated carrier power -20dB bandwidth Carrier frequency tolerance Transmitter frequency offset Modulation index Transmitter settling time Transmitter adjacent channel leakage power


Figure 4 shows a typical test setup when the transceiver is in transmit mode. The transceiver is set to test the -20dB bandwidth parameter, powered by the Bluetooth device power supply, and the PLL/VCO encodes the required digital pattern in the above Bluetooth channel (2.4-2.5GHz). After the device completes encoding in the channel, there will be a pre-defined wait time for the VCO to set to the encoded carrier frequency.

The PLL is then put into an open loop so that modulation is performed and pseudo-random bit sequence (PRBS) data is provided to the Tx data pin. An RF receiver is connected to the Bluetooth transmitter antenna using an RF port. After a predefined time in the digital pattern, the RF receiver is triggered to start receiving the transmitted signal and perform a fast Fourier transform on the digital samples received by the microwave receiver. For the captured signal, the power of the carrier frequency and the frequency portion on both sides of the GFSK signal is searched, that is, 20dB is subtracted from the coded carrier frequency, and then the bandwidth is calculated.

Similar to the transmit part, some important receive test parameters include:

Receiver sensitivity Co-channel interference Adjacent channel interference Intra-modulation Maximum input level

Figure 4 is also used to set up the transceiver when testing the receiver sensitivity parameter (receive mode). The parameters for the interference signal test are similar to the sensitivity test, but the former also contains an interfering modulation signal. The co-channel interference, adjacent channel interference and inter-modulation test setups require an additional RF modulation source using a two-tone RF synthesizer.

In order to generate a modulated carrier for the sensitivity test, a PRBS data stream needs to be created and stored in the test program array. The IF carrier is modulated with a random bit stream using a mathematical GFSK modulator. The IF modulated signal comes from the VHF waveform generator and is up-converted to the Rx test frequency by the RF source. The signal is then connected to the device under test through the RF port.

The device under test demodulates the modulated RF signal and sends the bit stream to the Rx data pin, which is then collected by a digital sampling instrument. The collected data is compared with the original PRBS data and used during modulation to calculate the bit error rate (BER). The receiver sensitivity test will measure the BER at different input powers.

As the production of Bluetooth ICs continues to increase, testing may become a very costly process in the manufacturing process. The most economical way to meet the requirements is to complete various tests in a very short time. Automatic test equipment (ATE) is the best choice for providing economical testing solutions. The test instrument must have a high-quality front-end design and sufficient flexibility to meet various test requirements. The RF signal source setting time of the tester should be shorter than that of the device under test, and the digital subsystem must be advanced to meet the strict digital performance requirements of mixed-signal RF devices. In addition, the test instrument should also have a high-speed, high-resolution DSP device to obtain signals from Bluetooth devices and a powerful DSP engine to meet the processing requirements of each test.

By: Nelson Lee TK
Test Technical Supervisor
ST Assembly Test Services Ltd

This post is from RF/Wirelessly
 

Guess Your Favourite
Find a datasheet?

EEWorld Datasheet Technical Support

快速回复 返回顶部 Return list