Optimization of limited feedback in two-way communications
Abstract
Wireless communications are becoming more and more popular nowadays. Cell phones, wifi and other wireless communication methods seem to be in every corner of people's daily life. The use of the wireless communications is becoming more diverse than before. Instead of simply talking on a phone, video chatting is more popular. Sharing a text message is not as well-liked as sharing a photo or a video. Cloud computing and remote storage are also developed to take over the place of local disk, etc. All of these needs require a highly increased data rate. However, the low utilization rate of transmitting packet gives a limit of increasing the data rate. To solve this problem, a two-way communication model is set up in the thesis to optimize the data rate with limited feedback. This model is under the condition that two nodes wishing to exchange information over a frequency division duplex, wireless additive white Gaussian noise with Rayleigh fading channel. The tradeoff between bits spent on learning channel, feedback and data must be carefully incorporated into the overall performance. In order to analyze the tradeoffs, a generalized framework is presented. Each packet contains three parts: a) training for estimating channel, b) feedback and c) data, where the feedback will contain ARQ (for requesting a re-transmission when a packet got lost) only or both ARQ and CSI (for transmitting channel state information). Under those conditions, an expression for the overall expected throughput in each direction is provided, which varies with the different allocating resources in a packet, channel conditions, SNR, transmitter protocols and receiver feedback protocols.
Degree
M.S.
Advisors
Smida, Purdue University.
Subject Area
Electrical engineering
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.