Please use this identifier to cite or link to this item: http://idr.iitp.ac.in/jspui/handle/123456789/285
Title: FPGA Implementation of Instantaneous Frequency Estimator Using DHT and CORDIC
Authors: Pal, S.
Keywords: Electrical Engineering
Issue Date: 2015
Abstract: In a communication system, Carrier Frequency Offset (CFO) refers to the difference in carrier frequency at transmitter and receiver. CFO is a major contributor to the inter-carrier interference in communication systems. The performance of Orthogonal frequency-division multiplexing (OFDM) signals is very sensitive to frequency offset errors as CFO leads to loss of orthogonality among the sub-carriers of OFDM signals. Thus, it is quite important to compute the instantaneous frequency of the received signal and compensate for the offset. In this work, a hardware design of an instantaneous frequency estimator using discrete Hilbert transform alongside vector-mode CORDIC is presented. Discrete Hilbert transform (DHT) is used to shift a sinusoidal signal by a phase of π=2. The DHT signal and the original signal are then given as input to the vectoring-mode CORDIC block so as to get the phase of the signal. The differentiation of the phase information is used to calculate the instantaneous frequency. MATLAB simulation result of the proposed algorithm which is quite able to estimate the instantaneous frequency is presented. The design is further implemented using System Generator and user defined HDL models. The HDL simulation shows that the results obtained by this fixed point method follows the trend of MATLAB simulation results. The final HDL code is implemented on commercially available FPGA device “XC5VLX110T-1“. To our knowledge, hardware implementation of the proposed methodology is efficient in terms of latency, maximum throughput and hardware usage compared to existing ones available in literature. The main constraint of this methodology is to compute the instantaneous frequency of the order of 2n where `n` is an integer.
URI: http://hdl.handle.net/123456789/285
Appears in Collections:03. EE



Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.