Link Enhancement for Reconfigurable Software Defined Radio (SDR) Systems

Featuring Chitaranjan Pelur Sukumar

Ph.D. Candidate

Electrical Engineering and Computer Science

The Henry Samueli School of Engineering, UC Irvine



Free and open to the public



Abstract:

Orthogonal Frequency Division Multiplexing (OFDM) has become a modulation of choice for many state-of-the-art wireless standards due to its inherent ability to combat multipath. However, one area that has not received sufficient attention is the means by which power is divided between pilots and subcarriers. If too much power is invested in pilots, then there is insufficient energy in the data subcarriers for decoding. Conversely, if there is not enough power in the pilots, then channel estimation becomes unreliable and equalization becomes hard. In this work, three modes of multiple antenna transmission were modeled: a) Multiple Input Single Output (MISO) b) Single Input Multiple Output (SIMO) and finally, c) Multiple Input Multiple Output (MIMO) spatial multiplexing. For the cases where the channel rank is unity (i.e., for MISO and SIMO) systems, we demonstrate how to split the power between data and pilot subcarriers optimally using thesymbol MSE metric. However, for higher rank channels, because of the difficulty in solving the joint optimization, we demonstrate how to design the training symbols, provided the precoder has already been defined. Through simulations, it will be shown that the proposed technique performs better (in a symbol error rate sense) than the conventional technique. In the second part of the talk, we will discuss the improvements in capacity and bit error rate performance that can be achieved by beam tilting reconfigurable antennas. The architecture of such an antenna that utilizes Micro Electrical Mechanical System (MEMS) switches to reconfigure the antenna will be discussed. Training techniques used to train such an antenna will be presented with simulations results documenting the training overhead and performance improvements.



About the Speaker:

Chitaranjan Pelur Sukumar was born in Chennai, India in 1981. He received his Bachelor of Engineering degree from the University of Madras in the area of electrical and communication engineering in 2003, and a Master of Science in electrical engineering from North Carolina State University in 2005. Currently, he is a Ph.D. candidate at the University of California, Irvine. His research interests include wireless system design and signal processing. He was a recipient of the Center for Pervasive Communications and Computing (CPCC) Fellowship in 2007.