| Wireless communication systems in numerous applications are being asked to perform at levels of integrity more stringent that ever before. Similarly, signal data rates and bandwidths have also been steadily increasing. In light of this, it is more important than ever to have a good understanding of the wireless channel for signal design and fading mitigation purposes. To this end, we have been investigating channels that induce “severe” fading upon transmitted signals. Due to the prevalence of the Rayleigh fading model as a near-worst-case channel condition, we term this severe fading “worse than Rayleigh.”
One of the earliest reportings we have found for such severe fading is [1], where fading was observed in the measurement of ionospheric scintillation in the HF band. These authors presented empirical fits to fading data using the Nakagami-m distribution [2], with some m-factors less than unity. More recent examples of severe fading include measured results for the “RF backscatter” channel, encountered in two-way RF identification applications [3], and also for urban and suburban cellular settings [4]. For these applications, the lognormal or Weibull distributions provide good empirical fits. Our recent work measuring and modeling channels in the 5 GHz band at airport surface areas [5], and in vehicle-to-vehicle transmissions [6] also shows such behavior; we too employ the Weibull distribution for fitting measured results. Some analytical models that yield severe fading have also been proposed, e.g., [7].
In this paper we review these results, and provide additional results from our measurements and modeling. One such model employs a statistically non-stationary random process that “switches” between two distributions, akin to the multi-state models proposed for land mobile satellite channels, e.g., [8]. The other model we propose is a multiplicative model of two small scale fading processes. This model is analogous to the “amplify and relay” model proposed for relay applications, and for and “pin-hole” MIMO channels [9]. The full paper elaborates upon these models, provides some physical justification for the severe fading, and proposes analytical models for such channels that can be used by researchers investigating communication system designs in severe fading environments.
|
| David W. Matolak received the B.S. degree from The Pennsylvania State, the M.S. degree from The University of Massachusetts, and the Ph.D. degree from The University of Virginia, all in electrical engineering. He has over nine years of industrial experience, and more than six years experience in academia. His industrial experience includes research and development at AT&T Bell Laboratories, on analytical and empirical characterization of nonlinearities and their effect on QAM transmission; R&D for Lockheed Martin Tactical Communication Systems, where he was Lead System Engineer on the development of a wireless local loop synchronous CDMA communication system; research for the MITRE Corporation, on the analysis and modeling of various aeronautical and satellite communication systems; and R&D for Lockheed Martin Global Telecommunications, on mobile satellite communication system analysis and design. In 1999, he joined the School of Electrical Engineering and Computer Science at Ohio University, where he is an Associate Professor. His research interests are radio channel modeling and techniques for communication over non-stationary fading channels, multicarrier transmission, and CDMA. Prof. Matolak is a member of Eta Kappa Nu, Sigma Xi, and a senior member of IEEE. He has published over 35 refereed articles, dozens of other articles and industrial technical reports, and has 6 patents. |