Global Sources
EE Times-Asia
Stay in touch with EE Times Asia
EE Times-Asia > Networks

DS-UWB vs. 802.11n: What's the best connectivity option?

Posted: 12 Sep 2005 ?? ?Print Version ?Bookmark and Share


In fact, there are two main reasons that signal bandwidth differences lead to higher transmit power for 802.11a/g/n systems relative to DS-UWB. The first is a consequence of the requirement to drive high data rates through modulation in a relatively narrow radio channel, and the second is the result of the basic physics of RF propagation in a multipath channel.

Modulation format describes the manner in which data are encoded into an RF signal for transmission through a wireless medium. For 802.11a/g systems, achieving a data rate of 54Mbps in an RF channel of 17MHz bandwidth requires the use of "high-order modulation" to achieve high spectral efficiency. In particular, 802.11a/g (and 11n) use 64-quadrature amplitude modulation (64-QAM) to map 6 data bits into each transmitted symbol. 802.11a/g technology combines this 64-QAM with orthogonal frequency division multiplexing (OFDM), but the underlying idea is the same.

When the overhead for forward-error-correction (FEC) and OFDM pilots and prefixes are factored in, 802.11a/g achieves about 3.3 bits/s for every Hertz of spectrum occupied. The cost of achieving this higher spectral efficiency using 64-QAM is that the receiver requires a higher signal to noise ratio (SNR) in order to demodulate the signal with the same level of error rate performance (relative to a baseline BPSK or QPSK system). The newer 802.11n technology will also use 64-QAM for its highest data rates, but also adds more sophisticated processing to achieve even higher spectral efficiencies using multi-antenna technology.

DS-UWB operates in a different way than 802.11a/g or 801.11n technology. Because of the wide bandwidths available, DS-UWB uses binary-phase-shift-keying (BPSK) to provide power efficient modulation. There is a significant difference in power efficiency between BPSK and 64-QAM. For instance, the Eb /N0 required at the receiver is 9.6 dB for BPSK at 10-5 bit-error-rate (BER) and almost 10 dB higher for 64-QAM at the same BER. It is important to note that these numbers are for uncoded systems operating in a pure additive white Gaussian noise (AWGN) channel, but the basic result is that high order modulation requires higher transmit power in order to provide equivalent BER at the receiver. In real operating systems, there are many other factors that impact receiver SNR requirements, including the use of sophisticated FEC. In addition, a key environmental factor that impacts actual operating system requirements is multipath propagation effects.

Impact of Multipath Fading on receiver SNR requirements

One of the key concerns for indoor wireless channels is multipath propagation (the dispersion in time of RF signal energy due to multiple propagation paths). OFDM technology has been traditionally used for wireless communications over indoor multipath channels, and it is used in 802.11a/g/n as well. One well-understood benefit of OFDM is that it can help prevent the effects of inter-symbol interference (ISI) for a digital communications system. This ISI effect occurs when the symbol length of the system is shorter than the multipath delay spread of the channel. In such cases, the individual symbols are "smeared" into each other, which typically requires an equalizer to compensate at the receiver.

OFDM can largely avoid this ISI effect by transforming its operating frequency band (approximately 17MHz) into a bank of narrower parallel channels (such as 48 data channels of 312.5kHz for 802.11a/g) and sending data symbols over these narrow in parallel. With 48 channels instead of one, the data symbols in each channel can then be 48 times longer (for the same data rate), and they are now much longer than the delay spread of the channel. As a result, ISI is prevented, and there is little penalty for the use of the OFDM approach relative to single-carrier narrowband approaches.

For UWB channels, however, the relationship between the signal bandwidth and the multipath channel is much different. In UWB channels, single-carrier and multi-carrier approaches (such as OFDM) result in very different multipath fading statistics for the received signal. Different multipath arrivals at the receive antenna result in a change in the perceived signal power at the receiver. If different components combine constructively, there is effectively higher signal power at the receiver, on the other hand, if the components combine destructively, there is lower signal power in that particular band at the receiver.

For some OFDM systems, this constructive or destructive combination at the receiver, or multipath fading, results in different signal power levels for each of the OFDM tones at the receiver. This variation across the parallel OFDM channels or tones occurs because each has a different center frequency and experiences a different pattern of constructive or destructive fading for the same multipath arrival times and amplitudes.

For DS-UWB, the effects of multipath fading are very different. Because of its wide signal bandwidth, the DS-UWB receiver is able to separately resolve multipath components to largely prevent the destructive combinations from occurring. The result is that the "deep fades" that occur in narrowband or OFDM systems do not occur for DS-UWB systems.

Figure 2 shows the difference between the typical fading distributions of narrowband or OFDM signals (4MHz per band/tone or less) and wideband signals such as UWB. In this example, the probability distribution plots show that the narrowband system experiences deep fades (of 15dB or more) in multipath and, as a result, some tones are almost completely "wiped-out" at the receiver. For the DS-UWB system, however, the received signal power varies little over different multipath channels, seldom fluctuating more than one or two dB.

Figure 2: Probability distribution plots showing the different fading statistics for Narrowband or OFDM signals and DS-UWB signals. These distributions are derived from indoor multipath channel models developed by IEEE 802.15 TG3a.

Multipath fading causes a variation in the received signal power over different channels which must be compensated for in the receiver. The use of strong FEC can help to average over the variations, but will still result in a higher SNR requirement in multipath channels than in AWGN (non-multipath) channels. For instance, in a punctured rate-3/4 convolutional code, a narrowband or OFDM system using low-order BPSK can require 6dB to 10dB higher SNR in multipath due to the fading effects. This effect can be even higher for 64-QAM. For UWB systems, the variation in received power also requires higher SNR, but the effect is limited to 1dB or so due to the smaller fading variance for UWB channels.

previous page

next page

Article Comments - DS-UWB vs. 802.11n: What's the best ...
*? You can enter [0] more charecters.
*Verify code:


Visit Asia Webinars to learn about the latest in technology and get practical design tips.

Back to Top