Bluetooth EDR Technology and Applications

JasonYoo

Bluetooth EDR Technology and Applications [Copy link]

Bluetooth wireless technology is a short-range communication system that provides voice and data connections between different information appliances. The maximum data rate originally designed was 1 Mb/s, but after the addition of the Enhanced Data Rate (EDR) feature, the Bluetooth core specification has evolved to support maximum data rates of 2 Mb/s and 3 Mb/s. The Bluetooth Special Interest Group (SIG; Bluetooth Wireless Technology Alliance), which is responsible for promoting and promoting Bluetooth wireless technology, has defined test specifications for RF layer compliance testing, including EDR mode. In addition to briefly introducing the market demand that has driven the evolution of the core specification from v1.0 to v1.2 and even v2.0+EDR, this article will also explore the Bluetooth EDR standard in depth, including new test cases for performing preliminary testing of transmitter and receiver designs.

Bluetooth System Overview The Bluetooth system operates in the unlicensed 2.4 GHz Industrial, Scientific and Medical (ISM) radio band. Low-power RF transmissions provide communication between different devices with a transmission distance of up to 10 to 100 meters. Through the Bluetooth system, up to 8 devices can form an ad hoc network without the need for formal wireless infrastructure. To mitigate interference and attenuation, Bluetooth wireless technology uses frequency hopping spread spectrum (FHSS) operation. FHSS also facilitates multi-access and coexistence of Bluetooth systems with other types of wireless systems. The basic frequency hopping pattern is a pseudo-random sequence of 79 channel frequencies in the ISM band. After the Bluetooth system adopts adaptive frequency hopping (AFH), it can avoid channels with known interference and greatly improve performance. The nominal value of the frequency hopping rate is 1600 times per second. The Bluetooth system provides point-to-point or point-to-multipoint connections. Two or more devices that share the same physical channel form a special network, called a piconet. In addition to one device acting as the master, a piconet can contain up to seven other devices, called slaves. All devices in a piconet are synchronized to a common clock reference and the frequency hopping pattern provided by the master. Bluetooth devices can operate in two or more overlapping piconet, forming a so-called scatternet. Figure 1 shows the network topology of a scatternet consisting of two different piconet. In this figure, a device is the master in one of the piconet, but a slave in the other. A device cannot be the master of more than one piconet at the same time, as this would mean that the different piconets are synchronized, but the specification specifies that each piconet must operate independently using a different frequency hopping pattern and master clock. 　 Figure 1: A scatternet network topology consisting of two piconet networks sharing the same device. 　 The physical channel of a Bluetooth system is subdivided into time slots and transmitted using time division duplex (TDD) technology. The master transmits in even-numbered time slots and the slave transmits in odd-numbered time slots. The length of the time slot is a function of the frequency hopping rate and is nominally 625 μs. Data is transmitted between the master and slave devices in packets contained within the time slot. A device can use one, three, or five consecutive time slots to accommodate a packet put together by the master. The packet contains an access code, a header, a guard band, and a payload. The payload contains user data, which must be modulated on the RF carrier using one of several modulation schemes, such as GFSK as specified in Bluetooth v1.0 and v1.2 and π/4-DQPSK or 8DPSK as proposed in the core specification v2.0+EDR. In a Bluetooth system channel, the physical link is formed between the master and active slave devices in the Bluetooth piconet. Physical links include synchronous connection-oriented (SCO) links for voice and asynchronous connectionless links (ACL) for data. SCO and extended SCO (eSCO) links are point-to-point links between the master and slave devices and can be considered circuit-switched connections. ACL links provide packet-switched connections between the master and all active slave devices in the piconet. The master device controls all traffic in the piconet, allocates capacity to different SCO links, and handles the polling mode of ACL links. The core system protocols of the Bluetooth system include radio (RF), link control (LC), link manager (LM), and logical link control and adaptation (L2CAP). RF is the first layer (layer 1) in the protocol stack.

Evolution of Bluetooth v1.0 to v1.2 A major and necessary change in the Bluetooth core specification v1.2 was the adoption of Adaptive Frequency Hopping (AFH). One of the reasons for this change was the coexistence issues between the Bluetooth system and 802.11b/g WLAN. The two wireless systems share the same frequency range in the 2.4 GHz ISM band, with the WLAN system using direct sequence spread spectrum (DSSS) or OFDM technology with channels up to 22 MHz wide and the Bluetooth system using FHSS technology on 79 channels spaced 1 MHz apart. When both systems are present, there is a 28% chance that the two devices will collide. Using AFH specified in v1.2, the Bluetooth system can detect interference, such as WLAN signals, and avoid frequency channels that are known to be subject to interference. When necessary, the system will reduce the number of available channels to 20. Another major and necessary improvement in v1.2 results in faster connection speeds. Connection times of up to 4 to 5 seconds were common with radios based on the v1.0 specification, but v1.2 reduced connection times to less than 0.5 seconds after changes to the inquiry and paging operations. Faster connections not only provide a more pleasant user experience, but also reduce manufacturing test time and reduce the overall cost of the product. Other improvements include enhanced features that improve link quality and more efficient process control. For example, v1.2 adds the eSCO logical transport link, which is a SCO link that retransmits when errors occur in the data stream. The original Bluetooth core system's SCO support uses a fixed-size payload to transmit data synchronously at a fixed rate at a fixed time interval, while the eSCO link provides a more flexible combination of packet types and selectable packet data content and time slot cycles, and improves data transmission speed by supporting a variety of bit rates. The original supported voice transmission rate was 64 kb/s, but the v1.2 specification has increased the transmission rate to 288 kb/s after adding several new packets.

Evolution of Bluetooth v1.2 to v2.0+EDR By adding EDR features to v1.2 and updating bugs found in that version, the Bluetooth SIG defined the latest core system specification v2.0+EDR. This specification has all the features of v1.2 and implements two new modulation modes in the payload portion of the packet. These EDR packet types provide maximum data rates of 2 Mb/s and 3 Mb/s. The maximum data rates are able to exceed the 1 Mb/s basic rate by using phase shift keying (PSK) to modulate the RF carrier, increasing the number of bits per symbol by 2 to 3 times. The 2 Mb/s EDR packet uses π/4-DQPSK modulation and the 3 Mb/s EDR packet uses 8DPSK modulation. For more information on EDR modulation modes, refer to the Basic Rate and EDR Packet Format section of this application note. π/4-DQPSK modulation is a mandatory feature of all v2.0+EDR compliant radios, while 8DPSK modulation types are optional. To maintain backward compatibility with v1.2, all Bluetooth v2.0+EDR compliant radios require a mandatory mode called Basic Rate. Basic Rate uses GFSK modulation throughout the packet to achieve a maximum data rate of 1 Mb/s, as specified in the older core specifications. It is important to note that the spectrum occupied by all three modulation types is roughly the same, as both Basic Rate and EDR packet types maintain a 1 Ms/s symbol rate. The occupied bandwidth is slightly higher when EDR modulation is used, as it uses a root-raised cosine filter instead of the narrower Gaussian filter implemented in the Basic Rate packet. The FCC relaxed the –20 dB occupied bandwidth requirement from 1.0 MHz to 1.5 MHz, allowing Bluetooth EDR radios to be used in the 2.4 GHz ISM band.

Bluetooth EDR Market Drivers The Bluetooth system is designed to provide short-range wireless connections between portable, personal handheld devices. Devices in a Bluetooth piconet form a natural personal area network (PAN) around the user, and any Bluetooth device that enters the piconet can connect to any other device in the PAN. In addition, Bluetooth devices can access local area networks (LANs) or wide area networks (WANs) through network-capable personal gateway devices. For example, a Bluetooth-enabled PC can access the Internet through a Bluetooth mobile phone connected to a WAN data service. As Bluetooth wireless technology becomes more and more popular in a variety of consumer products, providing higher data rates and longer battery life has become the only way to expand new applications. For example, consumer demand for short-range wireless connections has evolved from running a single application to running multiple applications in the same PAN. High-bandwidth applications, such as stereo speakers running simultaneously with wireless input/output devices (mice, keyboards, and printers), as well as other multimedia and gaming applications, have a pressing need for Bluetooth systems. With the introduction of EDR, multiple applications can more efficiently utilize available bandwidth and achieve higher overall performance. Figure 2 shows a typical multi-application example with various high data rate applications running in the same PAN. The additional capacity provided by EDR supports the simultaneous operation of these consumer devices. Since EDR transmission provides a higher data rate, the radio device will operate for a shorter period of time, which will reduce power consumption and extend the battery life of wireless devices. 　 Figure 2: Simultaneous connection of multiple Bluetooth EDR devices operating in a personal area network

Packet Structure and Modulation Format ■1. Basic Rate and EDR Packet Formats A typical Bluetooth packet begins with an access code and a header. The access code is used to perform synchronization, DC offset compensation, and packet identification in the physical channel. It is also used for calling, querying, and stopping operations in the Bluetooth system. The header contains link control information, including the packet type. A total of 15 different packet types are available for three different logical transports. As mentioned earlier, these three logical transports, or link types, are ACL, SCO, and eSCO. The link type determines the format of the payload that follows the access code and header. The payload contains user and control information, where the user information can contain data or voice, or a combination of both. The payload may also contain control data for identifying the device and providing real-time clock information. In addition, the payload may also contain other information for error detection and recovery, such as cyclic redundancy check (CRC) and forward error correction (FEC) information. Figure 3 shows a general packet format, that is, the basic rate packet format. With the addition of EDR packets in v2.0+EDR, normal packets are now called basic rate packets. Basic rate packets are transmitted after the entire waveform has been modulated using Gaussian frequency shift keying (GFSK). The main feature of EDR packets is the change to differential phase shift keying (DPSK) modulation after the packet header, which requires additional timing and control information to maintain synchronization with the new modulation format. EDR packets use the same access code and header definition as basic rate packets, including the modulation format, but contain a short period of time after the header to allow the Bluetooth radio to prepare for the switch to DPSK modulation. This short period of time, the guard time, is specified to be between 4.75 μs and 5.25 μs. After the guard time there is a synchronization sequence consisting of a reference symbol and 10 DPSK symbols, which is used to synchronize the symbol timing and phase of the modulation type used in the EDR packet. The payload in the EDR packet can contain user and control information based on the type of packet transmitted. FIG. 4 shows the format of an EDR packet. 　 Figure 3: Bluetooth Basic Rate Packet Format 　 Figure 4: Bluetooth EDR packet format 　 ■2. Basic Rate and EDR Modulation Formats General or basic rate modulation refers to GFSK. Data is transmitted at a data rate of 1 Mb/s with one bit per symbol, so the symbol rate is 1 Ms/s. Data is modulated on the RF carrier using a shift or deviation in the minimum carrier frequency of 115 kHz. Binary 1 and 0 are represented by positive and negative frequency deviations, respectively. FSK modulated signals have a fixed envelope that can be used to improve the power efficiency of the transmit amplifier. The Gaussian pulse waveform provides Bluetooth spectral efficiency by maintaining a –20 dB bandwidth of 1 MHz. The EDR modulation format selects one of two DPSKs to process the payload portion of the packet. As shown in Figure 4, the EDR packet first uses GFSK modulation in the access code and header, but switches to DPSK modulation after the guard time. Switching to a DPSK format allows the data rate to be increased to 2 or 3 Mb/s because two or three bits are transmitted per symbol while maintaining the specified 1 Ms/s symbol rate. Figure 5a is a power vs. time measurement of an EDR packet that uses GFSK modulation in the access code and header portions and 8-DPSK modulation in the payload portion. Figure 5b is the same power vs. time measurement specifically focusing on the time to switch from GFSK to DPSK modulation. This plot shows the 5-μs guard time and the 11 sync bits at the beginning of the EDR payload. An interesting observation is that the amplitude appears fairly constant in the GFSK modulated portion of the packet, but has large amplitude fluctuations in the DPSK modulated waveform. Spectral efficiency is achieved using a root raised cosine pulse, resulting in a –20 dB bandwidth of 1.5 MHz, slightly larger than the bandwidth of the GFSK modulation format. 　 Figure 5a: Power vs. time measurement of an EDR packet using both GFSK and 8-DPSK modulation formats. This waveform was captured using the Agilent N4017A Graphical Measurement Application (GMA) and Option 205 for Bluetooth EDR testing. 　 The DPSK modulation format defined for 2 Mb/s transmission is π/4 rotation differentially coded quadrature phase shift keying (π/4-DQPSK). The advantage of differentially coded phase modulated signals is that the signal can be demodulated without having to estimate the carrier phase. Instead, the received signal at any given symbol time is compared to the phase of the previous symbol[4] and the amount of phase shift is used to estimate the received data. The π/4-DQPSK constellation can be viewed as two QPSK constellations superimposed on each other, offset by 45 degrees. The symbol phase at each symbol time is alternately selected from the two QPSK constellations, so that the relative phase difference between subsequent symbols is one of four angles between ±π/4 and ±3π/4. Converted to degrees, these phase angles represent ±45 degrees and ±135 degrees. The four possible data points in the constellation result in a transmission rate of two bits per symbol, or twice the data rate of the conventional GFSK modulation scheme. When the symbols switch from one constellation to another, the phase between the symbols must change, which makes clock recovery easier [5]. 　 Figure 5b: This power vs. time waveform shows the transition between GFSK and 8DPSK modulation modes 　 There are several advantages to using π/4-DQPSK modulation in mobile applications compared to other PSK modulations such as QPSK and offset QPSK (OQPSK). The π/4-DQPSK mode allows you to use a differential detector or frequency detector followed by an integrate-and-dump filter for demodulation. Both demodulators help reduce receiver complexity compared to other demodulators that require coherent detection. In addition, the transitions in the signal constellation of the π/4-DQPSK waveform do not pass through the origin compared to other QPSK waveforms, which improves spectral characteristics and reduces power consumption. Figure 6 is a π/4-DQPSK constellation of the EDR portion of a Bluetooth packet, showing eight constellation data points measured over many symbols. Note that only four constellation data points or transitions can be generated in any symbol time, and this figure is a combination of two different QPSK constellations that are offset by 45 degrees from each other. The second EDR modulation format defined for 3 Mb/s transmission is 8-DPSK. The key to the increased data rate of this modulation format is to add 4 more constellation data points to each symbol. All 8 constellation data points allow for a transmission rate of three bits per symbol, or three times the data rate of the GFSK modulation mode. This modulation shares many of the same advantages as π/4-DQPSK, including the non-coherent demodulation mode. Demodulation of 8-DPSK is done by examining the relative phase differences between successive symbols. These phase differences can result in phase angles of 0, ±π/4, ±π/2, ±π3/4, and π. However, the increased data rate comes at a price. Compared to π/4-DQPSK and GFSK signals, the 8-DPSK modulated signal has a smaller distance between the constellation data points and is more sensitive to noise. Furthermore, the transitions now pass through the origin, so the power amplifier must have better linearity. Finally, the requirement to achieve a zero phase transition between symbols eliminates the clock recovery advantages in a π/4-DQPSK demodulator. 　 Figure 6: Measured star diagram of EDR load using π/4-DQPSK modulation

New EDR Test Procedures and Test Cases With the introduction of EDR features in the Bluetooth core specification, some specific EDR measurements were added to the RF layer test procedures and specifications (TSS/TP) [2]. The test specifications developed by the Bluetooth SIG are intended to provide a set of conformance tests for interoperability between the air interface and different Bluetooth devices. The RF test cases from the TSS/TP can be used to perform preliminary tests of Bluetooth devices in non-return mode, which is very useful in the early stages of radio development. Transmitter EDR tests include relative transmit power, carrier frequency stability and modulation accuracy, and differential phase coding. Bluetooth receiver EDR tests include EDR sensitivity, EDR bit error rate (BER) floor performance, and EDR maximum input level. The TSS/TP specification uses the term test purpose (TP) which contains specific identifiers applicable to various tests. For example, TRM is the identifier for transmitter tests and RCV is the identifier for receiver tests. CA is a test subgroup that identifies the capabilities of the master device, C is used to describe the type of conformance test, and there is an integer that identifies the TP number. Each relevant section displays the TP ID, which can be used to reference other information in the TSS/TP document. The measurement examples are provided using the Agilent N4010A Wireless Connectivity Tester with Option 105 for Bluetooth EDR transmitter and receiver testing. Option 106 can be used instead for a more economical configuration of transmitter-only testing. The measurement results must be displayed using the PC-based N4017A Bluetooth Graphical Measurement Application (GMA) and Option 205 for EDR analysis. Figure 7 shows a typical measurement configuration for testing an EDR-capable Bluetooth radio. In this configuration, the RF signal is passed between the test instrument and the radio via a coaxial connection. Antennas can also be used at the test instrument and radio for over-the-air testing. The N4010A tester is controlled by the N4017A GMA, while the radio is controlled by a device driver resident on the PC. This configuration will be used for the measurement example of a Bluetooth EDR transmitter. 　 Figure 7: Measurement configuration for testing a Bluetooth EDR radio using the Agilent N4010A tester and N4017A graphical measurement application software

Bluetooth EDR Transmitter Test Case Bluetooth EDR transmitter testing must measure the modulation quality, carrier frequency stability, and power level of each component in the transmitted packet. Test instruments, such as the Agilent N4010A tester, must have the ability to demodulate the EDR waveform and measure the modulation accuracy of the DPSK signal. When developing a Bluetooth transmitter, TSS/TP can perform preliminary testing of the transmitter's performance in non-return mode. The Agilent N4010A tester with Option 105 can perform non-return testing of wireless transmitters. In this case, the Bluetooth transmitter modulates a pseudo-random bit sequence (PRBS) into the EDR packet, and the test instrument must demodulate the binary sequence and then measure the modulation accuracy and frequency stability of the transmitted EDR signal. Measurement examples are provided with the instrument configuration of Figure 7. ■EDR Relative Transmit PowerEDR relative transmit power verifies that the difference between the average transmit power for GFSK modulation and the average transmit power for DPSK modulation is within a specified range. The Bluetooth Core Specification specifies that the difference between the average power of the GFSK portion of the signal and the average power of the DPSK signal should be between -1 dB and +4 dB. The relative power is calculated by measuring the average power of at least 80% of the GFSK and DPSK portions of the packet and then taking the difference between the two. The test conditions require that the transmitter must be operated at maximum output power with frequency hopping and whitening turned off. These measurements are performed at low, mid, and high frequencies within the ISM band and repeated at the minimum transmitter output power. Figure 8 shows the relative transmit power measured at a mid-band frequency of 2441 MHz for a signal using π/4-DQPSK modulation and an RF carrier. As shown in the figure, the average power measurement results for the GFSK and π/4-DQPSK waveforms are –14.4 dBm and –16.22 dBm, respectively. Therefore, the relative transmit power can be calculated to be +1.82 dB, which falls well within the specified range of –1 dB to +4 dB. 　 Figure 8. Relative transmit power measurement of an EDR packet using π/4-DQPSK modulation 　 ■EDR Carrier Frequency Stability and Modulation Accuracy The EDR carrier frequency stability test verifies the frequency stability of the transmitter's RF center frequency carrier, while the EDR modulation accuracy test verifies the quality of the differential modulation and identifies the errors that can cause problems in a real differential receiver. Modulation accuracy is tested primarily through differential error vector magnitude (DEVM) measurements, which are very similar to the traditional error vector magnitude (EVM) measurements specified in other digital communication systems. The basic EVM measurement represents the error between the ideal signal and the actual received signal, while the Bluetooth core specification defines DEVM as the error between two received signals separated in time by one symbol. This error must be measured after all linear distortion in the received signal has been removed, including tracking the frequency drift of the carrier. DEVM measurements are made on the synchronization sequence and payload portion of the packet. The test conditions require that frequency hopping and whitening functions be turned off. Before calculating the DEVM value, the sampling sequence must be adjusted to compensate for carrier frequency drift and sampling timing phase errors in multiple blocks of 50 symbols each. A total of 200 non-overlapping blocks are required for each carrier frequency. A transmitter with no distortion other than the fixed frequency error will have a differential error sequence equal to 0. Modulation accuracy can be described by three different values, namely 99% DEVM, RMS DEVM, and Peak DEVM. 99% DEVM is defined as 99% of the measured symbols having a DEVM value less than 0.3 and 0.2 for π/4-DQPSK and 8-DPSK, respectively. RMS and Peak DEVM values are calculated using this same error sequence. RMS DEVM is calculated from 50 symbols in each block. Note that this calculation must include information from the previous symbol in the block to produce 50 differential error vectors. The worst RMS value found for the 200 measured blocks is the RMS DEVM. The maximum RMS DEVM for π/4-DQPSK and 8-DPSK modulation is specified as 0.2 and 0.13, respectively. Peak DEVM is the worst DEVM found from all symbols in the measurement block. The peak DEVM measurement limits for π/4-DQPSK and 8-DPSK formats are 0.35 and 0.25, respectively. Table 1 shows the specified limits for the three DEVM measurements for the two EDR modulations. The maximum values in the table are expressed as a percentage of the specified limits. Figure 9 shows the modulation accuracy of a Bluetooth packet using π/4-DQPSK modulation. The 99% DEVM, peak DEVM, and RMS DEVM values measured for this packet are 10.24%, 11.57%, and 5.5%, respectively. As shown in the figure, all DEVM values measured for this waveform meet the specifications required for the EDR packet type using π/4-DQPSK modulation. 　 Table 1: Maximum limits of DEVM 　 Figure 9. DEVM modulation accuracy measurement of a Bluetooth EDR packet. Also shown is the carrier frequency stability measurement for the entire packet. 　 The frequency stability is measured on the GFSK and DPSK parts of the waveform. The test conditions require that the frequency hopping and whitening functions are turned off, and the payload data uses the PRBS9 pseudo-random sequence. The measurement must first determine the starting center frequency error in the GFSK header using a bit sequence with low inter-symbol interference. Then the frequency deviation in the logic 1 and logic 0 bits is measured and expressed as Δω1 and Δω2 respectively. The start frequency error is calculated by finding the average frequency error of the logic 1 and logic 0 bits and expressed as ωi (ωi=[Δω1+Δω2]/2). The start frequency error is specified to be between ±75 kHz. The frequency error in the EDR part of the packet is corrected by the start frequency error ωi. The corrected waveform is divided into several blocks of 50 symbols in length, and the remaining frequency error in each block is expressed as ω0. The measurement must be performed on 200 non-overlapping blocks. The worst block frequency error ω0 is specified to be between ±10 kHz. Finally, the Bluetooth specification limits the maximum value of the combined frequency error ωi+ω0 to ±75 kHz. This value represents the maximum change in frequency error, including the start error in the access code and the frequency drift that may occur in the measurement block. Figure 10 shows the carrier frequency tolerance range of different parts of the EDR packet. In the access code (GPSK) part of the packet, the start frequency error limit of ±75 kHz is reached. The rest of the packet, including the header, synchronization, payload, and trailer symbols, has a limit of ±10 kHz after the start error and block frequency error correction. This figure also shows that the limit of the combined error, that is, the maximum change range, is ±75 kHz. The measurement example in Figure 9 also lists the frequency stability measurement results of the EDR waveform, where the start frequency stability is –5.997 kHz, the block frequency error is –0.857 kHz, and the combined frequency error is –6.854 kHz, all of which are within the specification. 　 Figure 10. Bluetooth EDR carrier frequency stability limit as a function of symbol position 　 ■EDR Differential Phase Encoding The differential phase encoding test verifies the operation of the differential PSK modulator. For the EDR payload, the modulator must correctly map the binary data stream to a set of specified phase angles in the complex plane. The test conditions require that the frequency hopping and whitening functions are turned off, and the output power level of the transmitter should be set to the maximum. The EDR payload is modulated with a PRBS9 sequence, and the packet error rate measurement must cover 100 packets. In this case, the test instrument must demodulate the payload data and verify that the packet contains the expected PRBS9 sequence. It is specified that 99% of the received packets must be free of bit errors, in other words, the packet error rate is less than 1%. In the measurement example, the Agilent N4010A tester is used to demodulate the PRBS9 payload data, and the N4017A software is used to display the measurement results. Figure 11 shows the packet error rate of an EDR packet using π/4-DQPSK modulation. In this example, no bit errors were detected, so the packet error rate is 0%. The N4017A data display also lists the total number of bits measured in this test and the associated bit error rate. Since packet errors can be caused by one or more bit errors in the packet, the bit error rate provides additional information about the total number of errors that occurred in the demodulated packet. Figure 11 also shows the results of the guard interval measurement. This measurement can be performed in all of the EDR transmitter tests discussed previously. The guard interval is generally expressed as the minimum, maximum, and average of all measured packets in the test. As shown in Figure 11, the average guard interval measured for 100 packets is 4.94 μs, while the minimum and maximum values are 4.9 μs and 5.0 μs, respectively. 　 Figure 11. Packet error rate and bit error rate measurements of EDR waveform

Bluetooth EDR Receiver Test Cases Bluetooth EDR receiver testing requires measuring BER performance using test signals that contain various frequency and timing defects. The test instrument typically uses an internal arbitrary waveform generator to generate signal defects in the EDR packet. These "dirty packets" must be provided to the Bluetooth EDR receiver for demodulation. When developing a Bluetooth receiver, TSS/TP can be used in non-return mode to perform preliminary tests of receiver performance. In this case, the test instrument modulates a PRBS sequence into the EDR packet and the receiver under test demodulates the binary sequence. The BER performance of the receiver can be calculated by comparing the received data with the original PRBS sequence transmitted by the test instrument. At this time, the software controlling the receiver under test is usually set to perform BER calculations. ■EDR Sensitivity (RCV/CA/07/C) Receiver sensitivity is measured using EDR packets that contain timing errors and frequency offsets in the transmitted carrier. The conditions for these defective packets are specified in the Bluetooth TSS/TP document [2]. The test instrument must send three groups of 20 packets, each with a different defect. The first group of packets contains no defects. The second group of packets contains a carrier frequency offset of +65 kHz and a symbol timing error of +20 ppm. The third group of packets contains a carrier frequency offset of –65 kHz and a symbol timing error of –20 ppm. The instrument repeats these groups of packets until a minimum of 16,000,000 data bits have been received. The received data is compared to the transmitted PRBS9 sequence and the BER is calculated. To simulate the worst-case carrier frequency stability in the transmitter, an additional signal is modulated at the beginning of the DPSK sync word in the EDR packet. The synchronous ±10 kHz sine waves used for this modulation are generated by the test instrument using an arbitrary waveform generator. Using a transmit signal with these frequency and timing imperfections, the receiver must achieve a BER performance of 10e-4. ■BER floor performance for EDR Receiver floor performance is the BER measured at a receive power level of –60 dBm. This test simulates the receiver performance when the transmitter and receiver are far apart or in certain non-line-of-sight transmission conditions. For example, using a Power Class 1 Bluetooth transmitter that transmits at the maximum specified power level of +20 dBm, a received signal of –60 dBm represents an ideal free space loss of 80 dB at a distance of 100 meters [6]. To perform a BER floor performance measurement, the test instrument transmits a π/4-DQPSK or 8-DPSK modulated signal using a PRBS9 load at an output power level of –60 dBm. The receiver demodulates the data sequence until 16,000,000 bits are received. The BER is calculated by comparing the received data with the transmitted PRBS9 sequence. Under these conditions, the BER performance measured at low, medium and high carrier frequencies is specified as 10e-5. ■BER floor performance of EDRThe maximum input level test shows the BER performance of the receiver at an input signal level of –20 dBm. This test shows the performance of the receiver under possible front-end compression when using high input power levels. In this measurement, the test instrument must transmit a π/4-DQPSK or 8-DPSK modulated signal using a PRBS9 load at an output power level of –20 dBm. The receiver must demodulate the data sequence until 16,000,000 bits are received. The BER is calculated by comparing the received data with the transmitted PRBS9 sequence. The BER performance measured at low, medium and high carrier frequencies is specified as 10e-3.

Bluetooth EDR Future Development Direction Higher data rates, lower power consumption, and the need for multimedia applications such as streaming video are expected to drive the trend of switching to EDR devices. The development of EDR will continue to emphasize the concept of personal area networks, allowing many devices to operate simultaneously in the same piconet. In addition, new portable devices are expected to combine several wireless interfaces, such as GPRS and WiFi plus Bluetooth EDR, to provide synchronized and integrated connections between multiple networks. The concept of Bluetooth personal area networks has also been extended to the field of automotive communications. Telematics integrates wireless communications, autonomous driving, remote diagnostics and GPS navigation into the car experience. According to research, 20% to 30% of mobile phone users will talk on the phone while driving, and the number of original Bluetooth hardware equipped in 2008 is expected to reach 22 million. Automakers no longer need to install mobile phones directly in the car. By using Bluetooth wireless connection, the car's voice system can connect to the user's selected mobile phone through a wireless link. In addition, portable navigation systems, MP3 and WAN devices can also be integrated in the automotive environment by taking advantage of the higher data rates provided by EDR devices. Bluetooth v2.0+EDR is just one step in the evolution of this special short-range technology. In May 2005, the Bluetooth SIG announced that Ultra-Wideband (UWB) technology will become an integral part of the Bluetooth specification. With the addition of UWB, Bluetooth will be able to meet the industry's future needs for high-quality streaming images and the transmission of large amounts of data between wireless devices. The Bluetooth SIG is currently working on the details of incorporating UWB into the next generation of systems and making it backward compatible with v2.0+EDR and older devices.