Presented at ISART, March 1-3 2005, Boulder, CO Propagation and Throughput Study for 802.16 Broadband Wireless Systems at 5.8 GHz Thomas Schwengler, Member IEEE Qwest Communications, 1860 Lincoln street 11 th floor, Denver CO 80295 USA. (phone: +1 720-947-1184; fax: +1 720-947-1194; email: thomas.schwengler@qwest.com). Niranjan Pendharkar University of Colorado at Boulder 530 UCB, CO 80309-0530 USA (phone: +1 720-352-2227; email: niranjan.pendharkar@colorado.edu) Abstract— This paper presents propagation studies and analyses of OFDM signals, following the 802.16-2004 standard, at 5.8GHz. Throughput measurements are conducted first in a controlled faded multipath environment in the lab, then in a suburban area. Results are analyzed and compared. Index Terms—SUI propagation models, fading channels, OFDM. I. INTRODUCTION T HIS paper presents the results of a study measuring data throughput of an OFDM radio system through various fade models. The radio system is 802.16-2004 compliant [1], using 256 FFT at 5.8 GHz. Only one sector is used, therefore no other cell interferences are considered. A single 20MHz channel is used for the sector and multiple access is obtained by time slot allocation to all units within the sector. The first part of the paper presents a study of this system through a controlled environment, where radio multipaths and their resulting fades are generated by a channel emulator. The fading models are based on Stanford University Interim (SUI) channel models. The second part of the paper presents the same radio system tested for throughput in a suburban area in Denver. In that case radio interferences are verified to remain consistent and fairly minimal in order to focus on channel variations similar to those considered in the controlled environment. II. LAB TESTING A. Test Setup The radio system under test comprises one base station sector (BS) and several subscriber units (SU’s). This study is interested in fixed broadband wireless communications in various propagation environments; consequently fixed models are considered, rather than the usual mobile propagation models. Tests were conducted to measure the throughput in different modulations. Devices were tested in a part-cabled environment and part-unbounded media as shown below. The cabled environment undergoes different fading channels programmed in a fading emulator. The air interface is a short direct line of sight with the BS and the SU’s at a distance of 10 feet. This is done in order to couple signals of four SU’s over the air onto one sector. The Fading emulator allows us to emulate two separate channels, each comprised of several multipaths, each of which is independently faded and delayed. Fade statistics for the direct path are either Rayleigh or Ricean, fade statistics for the delayed paths are all Rayleigh. Finally additive white Gaussian noise is added to the overall channel (C/I=30dB). As in many wireless LAN devices, our radio devices are TDD and have duplex ports: transmit and received signals go to one unique antenna. In our test, the fading emulator fades the transmit and receive paths independently, the two paths are therefore separated by circulators. Finally the fading emulator required some careful calibrating of power levels (especially due to the high peak to average ratio of OFDM signals). Radio transmit power levels were adjusted and additional attenuation (pad) was added where necessary. Figure 1 shows the detailed setup.