Estimation of the time-varying paramets characterizing the wireless channels in cellular systems

点击翻译按钮获取中文摘要

My main research objective, is to propose an efficient iterative estimation module for the wireless channel coefficients and associated parameters (SNR, maximum Doppler shift, time and frequency offsets, angular spread, K-factor associated with the line-of-sight (LOS) , LOS Doppler shift,...) in future high data speed cellular networks. I will focus my efforts on wideband systems that use OFDM, even if many of the developed estimation algorithms are expected to be applicable, in addition, to systems for which a narrowband channel assumption is appropriate. While most studies focus on specific parameters while using simplifying assumptions for other parameters, our ultimate goals will be to develop a full estimation module capable of estimating the channel and its parameters under practical conditions. The estimators will therefore have to perform adequately with non-constant envelope modulation and with colored interference scenarios. The developed module will rely on the presence of pilot symbols, and various pilot allocation and signaling strategies will be evaluated, which will need to be tailored to the type of multi-access (OFDMA, TDMA and hybrid versions will be studied), smart antenna (beamforming, spatial multiplexing, etc) and link adaptation strategies. Both time division duplex (TDD) and frequency division duplex (FDD) will be considered, and possible uplink/downlink channel symmetry exploited. Recursive or turbo-like approaches will be studied to obtain a global estimation module of relatively low complexity. Furthermore, since we observe a proliferation of GPS-enabled cellular terminals, I will investigate the possibility of enhancing the channel parameters estimation process by exploiting the possible knowledge (with the appropriate signaling) of the position/speed/trajectory of the various transmitters/receivers. The algorithmic designs will first be based on the link-level performance, but will be fine-tuned based on a multi-user multi-cellular system-level performance evaluations conducted on a simulator that will need to be constructed with appropriate simplifying assumptions. Implementation aspects will also be considered.