«Carles Gómez Josep Paradells José E. Caballero Edita: Fundación Vodafone España Autores: Carles Gómez Montenegro* Universitat Politècnica de ...»
The SHR is composed of a four-byte preamble, which allows the receiver to synchronize with the received message and a one-byte start-of-frame delimiter that indicates the beginning of the packet data.
The PHR contains the number of bytes in the PHY payload. The PHY payload, also called PHY Service Data Unit (PSDU), contains the MAC layer frame. The size of the MAC layer frame cannot be larger than 127 bytes. In consequence, the maximum size of an IEEE 802.15.4-2003 PPDU is 133 bytes. The size of the SHR varies in subsequent versions of the standard.
126.96.36.199. Modulation methods for the 868/915 MHz bands
Fig. 3.8 shows a block diagram of the operations performed for data transmission in the 868/915 MHz bands. First, the raw bits to be transmitted are differentially encoded in order to solve a problem exhibited by the BPSK modulation. If the communication channel introduces a 180º phase shift, the receiver detects the received symbols as valid ones. To overcome this ambiguity, differential encoding is used so that the transmitted symbols correspond to differences in the information bits.
Once the information bits are differentially encoded, each of them is mapped onto a 15-bit chip sequence, which constitutes the DSSS technique applied in this case. The resulting chip sequences are the input of the BPSK modulator. While BPSK carries only one bit (one chip in this case) per symbol, this modulation is the simplest and most robust of all PSK modulations.
The reason for this is that the signal must be strongly affected to lead the demodulator to an incorrect decision.
Fig. 3.8. Block diagram of the DSSS/BPSK spreading and modulation techniques used in the 868/915 MHz band 188.8.131.52. Modulation methods for the 2.4 GHz band Fig. 3.9 shows a block diagram that models the modulation mechanisms used in the 2.4 GHz band. In this case, every four raw bits are grouped together to form one symbol. Each symbol is then mapped onto a 32-chip sequence (there exist 16 such sequences, i.e. one for each symbol). Each chip sequence is then modulated using a variant of the QPSK modulation. This modulation carries two information bits (chips in this case) per symbol and increases the bandwidth efficiency of the BPSK modulation. To reduce the maximum phase shift of QPSK and facilitate implementation, O-QPSK is used. The resulting signal has a bandwidth of roughly 2 MHz.
184.108.40.206. Physical layer measurements The IEEE 802.15.4-2003 PHY is responsible for several measurements that may assist in a number of upper level tasks. Some of these measurements are Energy Detection (ED) and Link Quality Indication (LQI).
ED is a measurement performed by the receiver, which may be used by upper layer services to select the channel to be used (e.g. for avoiding the use of busy channels). ED is encoded as an 8-bit value.
The LQI represents the strength and/or quality of the signal at the reception of a data unit. The measurement may be implemented by using receiver ED, a signal-to-noise (SNR) estimation or a combination of both. LQI is also represented as an 8-bit value. This parameter can be used by upper layer services (e.g. as an input to routing metrics, in order to choose the highest quality paths in a multi-hop IEEE 802.15.4 network).
On the other hand, the 802.15.4-2003 PHY carries out the Clear Channel Assessment (CCA), which is a functionality used by the MAC sub-layer in order to implement the CSMA/CA (Carrier Sense Multiple Access with Collision Avoidance) algorithm (see Chapter 4). In particular, this mechanism requires verification of the absence of signal before a device starts to transmit data. Three methods can be used for the detection of a busy medium: a) the detection of any energy above an ED threshold, b) detection of a signal with the characteristics of IEEE 802.15.4 and c) detection of a signal with the characteristics of IEEE 802.15.4 and energy above the ED threshold.
3.5.2. IEEE 802.15.4-2006 PHYs
4-2006  introduces two new optional PHYs for the 868/915 MHz bands, in addition to those defined by IEEE 802.15.4i) a scheme based on O-QPSK combined with DSSS, and ii) another that employs Amplitude Shift Keying (ASK) and Parallel Sequence Spread Spectrum (PSSS). These optional PHYs increase the bit rate offered by the IEEE 802.15.4-2003 868/915 MHz bands (see Table 3.3).
Since in IEEE 802.15.4-2006 there exist various PHY options for the same bands, the PHYs supported and that currently used by a device are identified by channel pages, which are defined as shown in Table 3.4.
Channel pages specified in IEEE 802.15.4-2006 The standard does not specify the use of channels 11 to 26 in channel pages 1 and 2.
220.127.116.11. Modulation methods for the optional 868/915 MHz bands The ASK/PSSS scheme introduced in IEEE 802.15.4-2006 is illustrated in Fig. 3.10. The raw bits to be transmitted are first converted to bipolar data, where the value “0” is replaced by the value “-1”. Each bipolar bit is then multiplied by a 32-bit bipolar sequence. In the 868 MHz band, there are 20 branches where this operation is performed, while in the 915 MHz band the number of branches is 5. The sequences resulting from each branch are then summated, thereby leading to a single sequence. In consequence, a symbol is formed by 20 input bits in the 868 MHz band, while in the 915 MHz band a symbol is formed by 5 input bits. The resulting single sequence would be the input of the ASK modulator, and hence would determine the amplitude of the transmitted signal. However, pre-coding is used before the sequence enters the ASK modulator. The reason for this is that a constant value must be added or subtracted in order to ensure that the average values of that sequence is zero.
The DSSS/O-QPSK method introduced in IEEE 802.15.4-2006 is similar to the DSSS/BPSK specified in the 2003 version of the standard. In this case, every four raw bits are grouped together to form a symbol and each symbol is mapped to a 16-bit chip sequence.
3.5.3. IEEE 802.15.4a PHYs IEEE 802.15.
4a is an amendment to IEEE 802.15.4-2006 that was approved in 2007. This new specification defines two alternative PHYs. The first one, which is based in DS-UWB and operates in unlicensed UWB spectrum, enables precision ranging and can be useful for applications requiring location awareness (with accuracy of 1 meter or better). The second, which uses Chirp Spread Spectrum (CSS) and operates in the unlicensed 2.4 GHz band, offers extended range and enhanced mobility. Both PHYs are robust for multipath environments.
18.104.22.168. Channel pages The new PHYs specified in IEEE 802.15.4a make use of two new channel pages, which identify the channel numbers and the corresponding bands that can be used by a device. Table 3.5 shows the new channel pages included in IEEE 802.15.4a.
22.214.171.124. UWB PHY of IEEE 802.15.4a The frequency spectrum used by IEEE 802.15.4a UWB is divided into three band groups: the sub-GHz band (250 - 750 MHz), the Low Frequency
Band (LFB) (3.1 – 5 GHz) and the High Frequency Band (HFB) (6 – 10.6 GHz).
Table 3.6 shows the channels and their main features for each band group.
It is only compulsory to support channels 0, 3 and 9. As can be seen in Table 3.6 and Fig. 3.11, the bandwidth reserved for most channels is equal to 499.2 MHz. However, the bandwidth for channels 4, 7, 11 and 15 is greater than 1 GHz and includes the bandwidth specified for more than one of the 499.2 MHz-wide channels. The purpose is for the UWB interface to provide a ranging/location capability of 1 m when using a 499.2 MHz channel, and better than this when using channels 4, 7, 11 or 15.
Note that there exist frequency gaps between consecutive bands. Their purpose is to avoid interference with other systems working in the same environment.
Fig. 3.12 shows a block diagram for the UWB modulator. First, Reed-Solomon coding (which allows error correction) is applied to the payload bits of the PPDU, while a single error correction, double error detection (SECDED) coding is applied to the PPDU header. Further convolutional coding is then applied to the resulting bits. The combination of the Reed-Solomon and convolutional coding actually implements a Forward Error Correction (FEC) technique. Spreading is then applied according to a spreader sequence generated by a Linear Feedback Shift Register (LFSR). The preamble is then prefixed to the header and payload bits. The PHR and PSDU are modulated using Burst Position Modulation BPSK (BPM-BPSK).
Fig. 3.12. Block diagram of the modulator used in the UWB PHY
With BPM-BPSK, each symbol is divided into two equal periods called BPM intervals. Each one is in turn divided into two halves. The first half of a BPM interval can contain a burst of pulses (also called chips). The second half is left as a guard interval in which there is no signal in order to minimize the inter-symbol interference that can be caused by multipath. Each half is divided into an integer number of possible burst positions, but only one burst is transmitted in a symbol. An example of this scheme is shown in Fig. 3.13.
Each symbol carries two information bits. One is based on the BPM interval that contains the burst of pulses (also called chips). The second is based on the polarity of the burst. In effect, the BPM-BPSK modulation is a DS-UWB technique.
Fig. 3.13. Example of the location of a burst in a symbol in the BPM-UWB modulation Each UWB channel supports various bit rates. The standard specifies a mandatory data rates for the PSDU of 851 kbps, as well as optional data rates of 110 kbps, 6.81 Mbps and 27.24 Mbps. The different data rates are obtained by modifying the number of chips that can be contained in a burst, which modifies the symbol duration accordingly. The rate at which the PSDU is transmitted is indicated in the PHR. The PHR is transmitted at 851 kbps for all PSDU data rates, except for 110 kbps. In this case, the PHR is also transmitted at 110 kbps.
Chapter 3. Physical layer
DecaWave  has announced a UWB transceiver on a single integrated circuit (IC). This circuit will be IEEE 802.15.4a-compliant and will offer 110 kbps, 850 kbps, 6.8 Mbps and 27 Mbps transmission rates, at a distance between nodes from 0 to 30 m . In 2007 IMEC  presented the first UWB transmitter designed in accordance with IEEE 802.15.4a.
126.96.36.199. CSS PHY of IEEE 802.15.4a
The CSS PHY uses the 2.4 GHz band and has been created with the aim of providing good performance in terms of robustness to interference, mobility (up to 170 km/h), coverage and power consumption. Compared to IEEE 802.15.4, these improvements are possible thanks to the use of a larger bandwidth.
The centre frequency, Fc, of each channel8, expressed in Megahertz, can
be obtained as follows:
Fc (MHz) = 2412 + 5·(channel number -1), for channel numbers between 1 and 13 Fc (MHz) = 2484, for channel number 14 (5) Therefore, CSS PHY uses 5 MHz radio channels. The CSS PHY supports a mandatory data rate of 1 Mbps and an optional data rate of 250 kbps.
Fig. 3.14 illustrates the modulation procedure of the CSS PHY. First, every first bit and every second bit are separated into two different branches, where the same operations are applied. In each branch, three bits are grouped together and mapped onto an orthogonal 4- or 16-bit codeword. (The first case leads to a data rate at the modulator output of 1 Mbps, while the second leads to a data rate of 250 kbps). The resulting sequence of bits is interleaved and serialized. Then, the serialized bits from the two branches are the input to a Differential QPSK (DQPSK) modulator. Finally, the DQPSK The IEEE 802.15.4a standard is inconsistent with regard to the use of channel numbers.
The channel page definition (see Table 3.5) cites channels between 0 and 13, while the centre frequency of each channel assumes channel numbers from 1 to 14.
symbols are modulated onto a stream of chirps produced by a Chirp-Shift Keying (CSK) generator. A chirp is a signal in which the frequency increases or decreases with time.
Fig. 3.14. Block diagram of the modulator used in the CSS PHY As of writing, Nanotron Technologies  is the only manufacturer that implements this option of the standard. The company reports ranges up to 570 m and bit rates of 2 Mbps, which are higher than the 1 Mbps proposed by the standard . The Nanotron solution also offers ranging facilities.
3.5.4. IEEE 802.15.4c, IEEE 802.15.4d, IEEE 802.15.4f and IEEE 802.15.4g
At the moment of writing, other amendments to IEEE 802.15.4 focusing mainly on PHY issues are being developed. In particular, Task Group 4c and Task Group 4d of IEEE 802.15 were created with the goal of amending IEEE 802.15.4 for supporting operation in new Chinese (315, 432 and 783 MHz) and Japanese (953 MHz) bands, respectively. Task Group 4c has reached an agreement with the Chinese WPAN Standards body to define a Multiple PSK (MPSK) PHY and an O-QPSK PHY.
Furthermore, Task Group 4f of IEEE 802.15 is currently developing a PHY (and MAC) amendment to IEEE 802.15.4-2006 for active RFID applications. Finally, Task Group 4g is currently developing another PHY amendment for smart utility networks.
3.6. Physical layer of Bluetooth Low Energy Bluetooth Low Energy (BT-LE) operates in the 2.4 GHz ISM band.
The band is divided into 40 RF channels with 2 MHz channel spacing.
This scheme is different from that of traditional Bluetooth9, which defines 79 frequency channels, separated by 1 MHz, also in the 2.4 GHz band.