Section 1.2. What Makes UWB Unique?

1.2. What Makes UWB Unique?

1.2.1. Time Domain Design

UWB has a very unique set of design requirements, and attempting to apply the principles for traditional narrowband or even broadband communications to the design of I-UWB systems can be misleading. Analysis of I-UWB systems often means examining the impulse response of the system as opposed to the steady state response, particularly when examining the antenna response. Time domain effects can include frequency dependant pulse distortion imparted by RF components or the wireless channel, pulse dispersion produced by the antenna, or timing jitter generated by non-ideal oscillators. For traditional communication systems, these transient effects are only a small fraction of the symbol duration and may often be ignored. In I-UWB systems, these effects directly impact the performance of the overall communication system. For example, timing jitter will lead to imperfect correlation at the receiver or potential loss of data and system synchronization for modulation schemes where data is transmitted in the precise position of a pulse.

1.2.2. Impact of the Antenna

One of the challenges of the implementation of UWB systems is the development of a suitable antenna that would enhance the advantages promised by a pulsed communication system. I-UWB requires antennas that can cover multi-octave bandwidths in order to transmit pulses on the order of a nanosecond in duration with minimal distortion. Because data may be contained in the shape or precise timing of the pulse, a clean impulse response (that is, minimal pulse distortion) can be considered as a primary requirement for a good I-UWB antenna.

While it may be more intuitive for communication engineers to think of the performance of an antenna in terms of its frequency domain characteristics, the response of an antenna to a I-UWB pulse stream can best be described in terms of its temporal characteristics. An ideal UWB antenna needs to be relatively efficient across the entire frequency band with a Voltage Standing Wave Ratio (VSWR) of at most 2:1. To prevent distorting the pulse, an ideal UWB antenna should produce radiation fields with constant magnitude and a phase shift that varies linearly with frequency [5]. An antenna that meets these characteristics will radiate a signal which is only a time derivative of the input signal.

In reality, due to size and cost constraints, practical UWB antennas may not meet the previous requirements. It must also be noted that the antenna induced distortion can change with elevation and azimuth angle. Thus, we assume that such effects will ultimately be included in the assumed channel model. Chapter 3, "Channel Modeling," and Chapter 4, "Antennas," detail channel modeling and antenna effects, respectively.

1.2.3. Propagation and Channel Models

To perform systems-level engineering, UWB propagation characteristics must be considered. UWB differs from conventional communications in that the signal may be overlaid on top of interference. This interference must be considered in the link budget and, in fact, can often be the primary reason for performance limitations. Another issue is the introduction of large numbers of multipath signals that were not resolvable in narrowband communication systems. Measurements of typical UWB channels have revealed dense, multipath-rich environments, allowing for RAKE receivers that can harvest a tremendous amount of energy. Additionally, UWB propagation is highly dependent on the effect the antenna has on the shape and duration of the transmitted pulse.

UWB propagation measurements and modeling are the subjects of ongoing debate in the engineering community; as such, this book does not claim to resolve that debate. Rather, it discusses the basic concepts behind several UWB channel models and some of the differences between narrowband and UWB signal propagation.

1.2.4. Transmitter and Receiver Design

RF design for UWB systems is distinct from traditional narrowband or broadband systems in several ways. The extremely wide bandwidth of a UWB necessitates RF components that have flat frequency responses. Significant deviation, or ripple, in the frequency response of RF components as well as the nonlinearities present in all RF devices will introduce distortion to the UWB signal. UWB transmitted signals also have a very high peak-to-average power ratio (PAPR). As RF components are peak power limited, it becomes important to ensure that all RF devices have a power handling capacity at least as great as the peak power in the UWB signal.

Furthermore, the coexistence of UWB and existing services means that narrow-band interfering signals will be detected by the receiver. These narrowband signals can either corrupt the pulse or saturate the RF front-end, decreasing the receiver's dynamic range and effectively limiting the range of the UWB system. Introducing notch filters at the receiver is a potential solution; pulse-shaping techniques, such as those described in [22], provide an alternative method for mitigating narrowband interferers without distorting the UWB waveform.

Most UWB receiver techniques require highly accurate synchronization with the transmitter as well as stable oscillators to maintain synchronization. With certain I-UWB modulation schemes, data may be conveyed by the precise position or timing of the pulse, and a loss of precise synchronization could result in a loss of data.

1.2.5. Difficulties in Using DSP Technology

Designing an I-UWB transmitter to broadcast short pulses is much simpler than designing a receiver to demodulate those pulses. For instance, assuming a pulse width of 250 picoseconds and 2 samples/pulse requires a sampling rate of 8 Gigasamples per second. Assuming 6 bits per sample, the receiver must process a data stream of 48 Gbps; at 8 bits per sample, the data stream increases to 64 Gbps. At the time of this writing, only the most technologically advanced FPGAs and ASICs are capable of handling such a huge amount of data.

Another problem is the limitations inherent in practical Analog to Digital Converters (ADCs). Most mass-produced commercial grade ADCs have analog input bandwidths^[2] less than 1 GHz. Regardless of the sampling clock frequency, the ADC can only sample signals that fall within its input bandwidth. The highest performance commercially available ADCs can have input bandwidths, which extend into several GHz and have a maximum sampling clock frequency in the low GHz range. It is quite obvious, therefore, that in order to sample a UWB signal which lies in the 3.1-10.6 GHz range, the ADC must, at the very least, have an analog input bandwidth equal to or greater than the highest frequency component of the input signal (that is, an input bandwith of 10.6 GHz). The use of high-performance (and high-cost) FPGAs, DSPs, and ADCs are, however, an anathema to engineers who have heralded UWB as a low-cost, simple communication system.

^[2] Analog Input Bandwidth is defined as the frequency at which the sampled output of the ADC falls 3 dB below the full-scale input amplitude.

1.2.6. Networking Issues

A primary driving application of UWB is a high rate Wireless Personal Area Network (WPAN) confined to a small coverage area (less than 10 m radius). The network should be a self-organized, dynamic, ad hoc network, which means the network is formed without advanced planning and that users can join or leave at any time. Network security is also an important issue. Even though UWB signals may have a Low Probability of Intercept (LPI), it is still important to provide authentication, confidentiality, integrity, and availability. Variable modes of operation should allow for both long-range, low data rate communications and short-range, high-speed connections for multimedia or large data transfers.

UWB communications presents some unique challenges for a wireless network's Medium Access Control (MAC). As discussed in Chapter 9, "Networking," as the signal bandwidth becomes significantly greater than the data rate, a hybrid CDMA and Time Division Multiple Access (TDMA)-based MAC becomes a more optimal approach than a traditional TDMA MAC. This hybrid technique provides greater flexability and adaptabilityan important advantage for UWB networks that may need to meet a variety of Quality of Service (QoS) requirements. Furthermore, the unique nature of I-UWB communications means that several additional features should be built into the MAC layer. Ranging information will assist in the formation of piconets by excluding users that fall outside a predetermined radius of operation. The need for strict synchronization between transmitter and receiver and the ability to generate accurate channel estimates must also be addressed by the MAC. Implementing a decentralized MAC provides the ability to incorporate UWB into consumer electronics and mobile phones that can operate over ad-hoc networks. Finally, different modes of operation, such as high data rate, long-range, or distributed sensor networks, each have somewhat different design constraints, suggesting that multiple approaches to the MAC design may be necessary to develop an optimal MAC layer for a particular application.

1.2.7. Future Directions

At the present time, the FCC is content to allow UWB devices to develop within the limitations of their First Note and Order [14]. As the technology matures, it is possible that the FCC may relax both the transmitted power level and bandwidth restrictions for UWB operation. Such modifications will most likely be a result of detailed investigations that demonstrate the minimal impact that higher power UWB devices will have on the QoS of existing users. In particular, major concerns still exist about the potential interference of UWB emissions to GPS and air traffic control signals.

A potential future application of UWB communications is low power, low data rate distributed sensor communications, similar to the 802.15.4/ZigBee standard. Because the duty cycle of I-UWB pulses is inherently very small, an I-UWB-based extension of the 802.15 standard would help to conserve valuable battery life [15]. Also, the extremely low power spectral density and short time duration of the pulse makes the transmitted signal difficult to detect and intercept, which is a definite advantage for ensuring a secure network.

Another potential application for I-UWB signals is the field of medicine. Microwave and radar monitoring of physiologic functions is an idea that has been around in concept since the 1970s [7, 20], but its development was hampered by the cumbersome and expensive technology of the time. With sufficiently short pulse duration (on the order of 100 picoseconds), an I-UWB radar would be capable of monitoring the movements of internal organs such as the heart or lungs without the need for direct skin contact or constraining the patient in space. Additionally, research is underway that analyzes the backscattered signals from a UWB pulse to detect cancer [6, 21]. Although I-UWB imaging may not provide the resolution of CT (Computed Tomography) or MRI (Magnetic Resonance Imaging) scans, it has the potential to cost-effectively provide critical information and determine, based on those results, whether further diagnostics are required.