IEEE P 802 . 11 Wireless LANs Modulation and Coding in Wireless Local Area Networks

This report documents the results of a preliminary study on the application of various modulation and forward error control (FEC) techniques for application to wireless local area networks (WLAN) applications. The report considers the existing quadrature phase shift keying (QPSK) -Barker code based system. The notion of bandwidth and signal to noise ratios Es/No and Eb/No are reviewed. It is shown that the existing system obtains all of the processing from a loss in rate (a simple trade of energy for rate) which has no coding gain. Two proposed systems, one based on offset pulse position modulation (OPPM) and the other based on M-ary Orthogonal Keying (MOK) (a short block code known as a Hadamard code) are considered. It is shown that both systems exhibit a modest coding gain. However, it is argued that they pale in comparison to other modern FEC systems. January 1998 doc.: IEEE P802.11-98/24 Submission page 2 Chris Heegard, Alantro Communications 1. Existing Barker System 1.1 Bandwidth, Symbol Rate and Information Rate The information rate of a coded modulation signal is the product of two factors: Rinformation = Rsymbol ∗ Rmodulation The first term is the symbol rate, Rsymbol = 1 / Tsymbol , which varies inversely to the symbol period Tsymbol . The symbol rate is measured in symbols per second (IEEE 802.11(ds) uses a symbol rate of 11 Msps). The bandwidth B = (1+ α )∗ Rsymbol is proportional to the symbol rate. The excess bandwidth parameter α > 0 is related to the bandwidth efficiency of the modulated transmitted waveform. In the existing IEEE 802.11(ds) system, the excess bandwidth is almost 200% since an 11MHz symbol rate occupies a >30 MHz bandwidth. One method of improving the information rate of the system is to decrease the excess bandwidth parameter. With the existing QPSK modulation, the α parameter can be improved with pulse shaping. However, the economics of the existing transmitter power amplifier designs requires operation with significant non-linear (saturation) operation. This precludes the use of meaningful pulse shaping. An alternative approach is to use a constant phase modulation (CPM) such as minimum shift keying (MSK). This approach may improve the information rate though the increase of the symbol rate. However, (1) the practical benefit should be demonstrated and (2) such an approach would not be backward compatible with the existing IEEE 802.11(ds) standard. This report is mainly concerned with improvements in the information rate though the second term: Rmodulation the modulation rate (measured in bits per symbol). The study described assumes the use of QPSK modulation, however, the suggested improvements could also be achieved with other forms of modulation including MSK. January 1998 doc.: IEEE P802.11-98/24 Submission page 3 Chris Heegard, Alantro Communications 1.2 QPSK, BPSK and the Barker Sequence In existing 802.11 (DS) wireless local area networks (LAN) systems, quadrature phase shift keying (QPSK) modulation is combined with an n = 11 Barker code. As a basic form of modulation, the bit error rate (BER) of the system, as a function of the signal to noise ratio (SNR) Es / N o is presented in Figure 4. In this case, the SNR is a measure of the received signal power over the power of the noise in the bandwidth of the signal. It is assumed that the noise is white in this bandwidth (the noise power spectral density (PSD) is flat across the band). The noise is also assumed to be Gaussian (Normal) in distribution.