QUANTIZED PRECODING AND POWER ALLOCATION IN FEEDBACK CHANNEL A DISSERTATION Submitted in partial fulfillment of the requirements for the award of the degree of MASTER OF TECHNOLOGY in ELECTRONICS & COMMUNICATION ENGINEERING (With Specialization in Communication Systems) By DEEPANKAR DEBNATH 4 of TECfi4fj <d ç:p4 1°ç - DEPARTMENT OF ELECTRONICS AND COMMUNICATION ENGINEERING INDIAN INSTITUTE OF TECHNOLOGY ROORKEE ROORKEE - 247 667 (INDIA) JUNE,

Abstract

There has been large increase in wireless data rates over last few years because of vast impru ements in the field of coding, scheduling, modulation, easy implementation of complex algorithms on IC chips. Behind all this improvements is the estimation of wireless communicatic channel environment as it determines the performance of wireless communication systems. In comparison to static and predictable behavior of wired channel, wireless channel is more dynamic and unpredictable which makes the task of design of the wireless communication system difficult. As we already know that feedback has huge role in the field of control systems, communication theory and source coding. In this report we concentrate on quantized feedback in MIMO systems in the presence of Rayleigh fading channel. The results obtained here can also be extended to multicarrier systems and multipath fading channels with slight modifications. An important characteristic of wireless channel is the phenomena of fading. Fading is characterized as a non-additive signal disturbance in wireless channel. Multiple-input multiple-output (MIMO) system have the ability to improve transmit rate or diversity gain in the presence of fading. The performance enhancement of multi-antenna systems depends on the availability of channel state information (CS!) at the transmitter and the receiver. If transmitter can achieve perfect CSI, the : system can obtain diversity as well as additional array gain by transmitting beamforming. Channel state information (CS!) can be obtained at the receiver side which is then transmitted back to the transmitter. Exploitation of such channel information allows increasing channel capacity, improving error performance while reducing hardware complexity