The challenge for design engineers then becomes modeling their system to account for these impairments. Fortunately, there are a number of models available today that designers can employ during the system design process.

In this paper we'll detail one of these models, a multipath model, that will enable designers to compare the performance of 802.11, 802.11b, and 802.11a system designs. The model presented here is statistically verified using a Simulink/Matlab platform. The effect of delay spread for these transceiver waveforms using the multipath channel model will be presented. Bit error rate (BER) curves for various delay spreads will also be highlighted.

A Little History
Before diving into the system-level modeling technique, let's provide a brief background on the WLAN sector. The 802.11 committee holds the responsibility of defining physical layer (PHY) and media access control (MAC) layer duties in a WLAN system. The first specification delivered by this group was the initial 802.11 spec, which supported data rates of 1 or 2 Mbit/s. The 1-Mbit systems employed a binary phase-shift keying modulation scheme while the 2-Mbit/s systems operated off a quadrature phase-shift keying (QPSK) modulation scheme.

Once the 802.11 spec was completed, designers quickly began working on the "b" version of the standard. Released in the latter half of the 90s, the "b" specification introduced a complimentary code keying (CCK) modulation scheme and moved WLAN performance into the 11-Mbit/s range.

The 802.11a specification is one of the latest incarnations out of the IEEE. This specification delivered some major changes to existing WLAN operation. First, it moved the frequency band to the 5 GHz range. Second, this spec delivered yet another new modulation scheme— orthogonal frequency division multiplexing (OFDM). Finally, the spec upped the data rates again, pushing them into the 54-Mbit/s range.

With that background in hand, let's move onto the multipath modeling example, starting with a look at the channel model.

Channel Modeling
Since WLAN systems operate in unlicensed bands, noise can be a big concern for system designers. In a noise-corrupted channel, like the 2.4 GHz band, the job of the demodulator is to retrieve the transmitted data from the received waveform as nearly error free as possible. Noise in a communication system corrupts the signal in an additive fashion so the noise is modeled using an AWGN channel. The AWGN channel is very important in defining the noise added to the transmitted signal, but is inadequate in characterizing signal transmissions over radio channels whose transmissions change with time.

Channels whose characteristics change with time are called multipath channels. Multipath conditions occur when a signal arrives at the receiver via different propagation paths with various delays. The signal fading created by multipath channels cause the transmitted signals to destructively add.

In a multipath environment, the delay of the reflected paths is known as the delay spread. Measured in nanoseconds, delay spread introduces a phenomenon known as intersymbol interference (ISI) to the receiver. ISI is introduced if the symbol period is shorter than the delay spread of the channel.

To accurately measure the delay and fading caused by multipath, we've chosen the Naftali model, which is a consistent channel model is required for comparison of different WLAN systems. Using this model, we can compose the channel impulse response of complex samples using random uniformly distributed phase and Rayleigh distributed magnitude (Figure 1).

The mathematical Naftali model for the channel is given below.

(1)

(2)

(3)

where is a zero mean Gaussian random variable with variance produced by generating an N(0,1) and multiplying it by , where is chosen so that the condition is satisfied to ensure the same average received power.

Simulation/Results
Now that we've laid the groundwork, let's look at the actual simulations. We'll start with simulations from an 802.11 system.

The 802.11 PHY layer defines the modulation and signaling characteristics for the 1- and 2-Mbit/s data rates. As stated above, a BPSK modulation scheme is used for 1-Mbit/s systems while a QPSK scheme is employed for 2-Mbit/s systems.

The transmitter and receiver of an 802.11 system were modeled using Simulink, which runs on a Matlab platform. Once the two rates were modeled, simulations were run to determine both systems performance in the presence of an AWGN channel and a multipath channel. BER curves were generated with delay spreads of 0, 25, 50, and 75 ns. The generated plots for the 1- and 2-Mbit/s rates are shown below in Figures 2 and 3, respectively.

The generated curves show theoretical BER vs. simulated BER. Also shown are simulated BER curves with various delay spreads. These curves show how received signal is degraded due to the effects of multipath.

Now we'll model the 802.11b PHY. The 802.11b PHY layer specification defines the modulation and signaling characteristics for 5.5- and 11-Mbit/s data rates. These rates are achieved using a CCK modulation format. CCK modulation was derived from M-ary orthogonal keying and uses the same in-phase (I) and quadrature (Q) channelization scheme as the 802.11 PHY layers. CCK codes were chosen as an appropriate high-rate waveform because they can reliably support high data rates and they perform well in a multipath environment.

Simulations were run on 802.11b both data rates and BER curves were generated with delay spreads of 0, 25, 50, and 75 ns. The generated plots for the 5.5- and 11-Mbit/s rates are shown in Figures 4 and 5, respectively.

The previous two generated curves also show theoretical BER vs. simulated BER. Also shown are simulated BER curves with various delay spreads. These curves show how received signal is degraded due to the effects of multipath. Even with increased data rate the effects of multipath are about the same as for the 1 and 2 Mbps data rates

The 802.11a PHY spec. defines the modulation and signaling characteristics for the data rates from 6 to 54 Mbit/s operating in the 5 GHz frequency bands. As stated above, 802.11a use OFDM as their modulation format. OFDM is a special case of multicarrier transmission, and is used to increase the robustness of a communication link and to combat frequency selective fading or narrowband interference.

Simulations were run on the 802.11a data rates of 12 and 24 Mbit/s and BER curves were generated with delay spreads of 0 and 75 ns. These generated plots for the 12- and 24-Mbit/s rates are shown below in Figures 6, and 7, respectively.

The curves for the OFDM PHY layer show the simulated BER for the 12 and 24 Mbps data rates. The figures also show these rates with a delay spread of 75 ns. The OFDM almost completely mitigates the effects of multipath.

Concluding Thoughts
As WLAN systems continue to evolve, interference will continue be a thorn in designers' sides. Using the techniques discussed above, however, designers can better model potential interference problems and, in turn, ease the overall system design process.

Author's Note: Thanks to Dr. Donald Malocha for his help on this paper.

Editor's Note: This paper is based on a presentation made at the 2002 Communications Design Conference.

References

1. R.E. Zeimer, W.H. Tranter, Principles of Communications, John Wiley & Sons, Inc. New York, 1995.
2. B. O'Hara, A. Petrick IEEE 802.11 Handbook, A Designer's Companion, IEEE Publication, 1999.
3. R. Van Nee, R. Prasad, OFDM for Wireless Multimedia Communications, Artech House Publishers, Boston, 2000.
4. A.A.M. Saleh, R.A. Valenzuela, "A Statistical Model for Indoor Multipath Propagation", IEEE Journal on Selected Areas in Communications, vol. SAC-5, no.2, pp. 128-137, 1987.
5. T. S. Rappaport, Wireless Communications Principles and Practice, IEEE Press, Prentice-Hall, New Jersey, 1996.

About the Authors
S. Maurice Nabritt is a graduate research assistant at the University of Central Florida. He earned a BSEE at Southern College of Technology in 1996 as well as an M.S. and Ph.D. in Electrical Engineering from the University of Central Florida in 1998 and 2002, respectively. Maurice can be reached at mnabritt@ara.com.