Please use this identifier to cite or link to this item:
http://hdl.handle.net/10266/2954
Title: | 8X8 MIMO System Using Precoded and Convolution Encoded Linear Dispersion Codes |
Authors: | Kaur, Satinder |
Supervisor: | Kansal, Ankush |
Keywords: | MIMO;LDC;CSI |
Issue Date: | 14-Aug-2014 |
Abstract: | As the desire for wireless transfer of high speed data in small, handheld units is increasing, consumer demands have exceeded the technological ability of existing wireless communication systems. In order to fulfill the desire for faster, more reliable communications, it is mandatory for the next generation wireless communication systems to achieve very high throughputs and improved reliability, even in the presence of harsh channel conditions. Unfortunately, consumer demands for higher data rates have rapidly surpassed the capabilities of existing wireless technologies. Present generation cellular phones cannot achieve higher data rates to satisfy most wireless consumers who are adapted to the extremely high data rates of wired networked systems. This has raised a need for new research into methods of supporting higher data rates for next generation wireless systems. In order to achieve outstanding reliability and exceptionally high capacity the most promising technique is to use Multiple Input, Multiple Output (MIMO) systems. MIMO systems utilize multiple antennas at the transmitter and receiver sides. LTE-advanced aims to use 8x8 MIMO and 128 QAM and promises to deliver 1Gbits/s at fixed speeds and 100Mbits/s to mobile users. Since the need of the present era is to improve data rates thereby reducing error rates. Major challenges for next generation wireless systems is the development of efficient coding and signal processing techniques so that spectral efficiency for large number antenna systems i.e. 8x8 systems is improved. Space-time codes create redundancy by spreading symbols in space and time, which improves bit error rates (BER) but rate is reduced for large number antennas. Spatial multiplexing techniques improve capacity but do not optimize BER performance. Linear Dispersion Codes (LDCs) are linear codes in structure that spread data symbols using dispersion matrices. High rates and good error performance is guaranteed for any arbitrary number of transmitter antennas. In conventional LDC techniques it is assumed that there is no knowledge of the channel at the transmitter. However, in order to convey channel state information (CSI) feedback from receiver to transmitter can be established. Such systems are called precoding systems. To achieve high data rates and low error rates for larger number of antennas, the focus of this work is to investigate the performance of 8X8 MIMO systems using Linear Dispersion Codes and Precoded Linear Dispersion Codes when CSI is used at the transmitter. The effect of channel known is only improving the performance at low SNRs. Precoding improves the symbol error rate performance of linear dispersion codes at lower SNR’s i.e. up to SNR 18dB. In precoded 8X8 LDC coded MIMO symbol error rate of 〖10〗^(-1.5) is obtained at SNR of 7dB whereas without precoding at 13 dB, hence there is 6dB system performance improvement. In order to combat the burst errors due to channel convolution encoding and bit interleaving is used along with Linear Dispersion Codes and performance is improved at higher SNR’s i.e. above 15 dB. Using convolution encoding symbol error performance of 〖10〗^(-4.0) is obtained at SNR of 20dB whereas without channel coding at same symbol error performance is obtained at 24 dB, hence 4dB performance improvement is obtained. |
Description: | Master of Engineering-ECE-Thesis |
URI: | http://hdl.handle.net/10266/2954 |
Appears in Collections: | Masters Theses@ECED |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.