Reference no: EM133120470 , Length: word count:1000

ME606 Digital Signal Processing - Melbourne Institute of Technology

Assignment: Understanding the role of digital signal processing in state-of-the-art communications technologies and utilizing MATLAB or Python to simulate LTI systems for frequency/time domain analysis

Learning Outcome 1: Development and implementation of signal processing algorithms in Matlab or Python

Learning Outcome 2: In-depth design of digital filters

Learning Outcome 3: Understand the design of multirate signal processing and their applications

Learning Outcome 4: Implementation and applications of DFT/FFT

Assignment Description

This assignment focuses on the application of digital signal processing (DSP) tools and methods in solving practical problems in modern wireless communications. OFDM (orthogonal frequency division multiplexing) is one of the most important transmission methods in modern wireless communications. So, students are expected to possess an expertise in it. Programming in languages such as MATLAB and Python is another must have of communications engineers. The ability to understand and apply algorithms to the design and analysis of communications systems is another desirable skill, and so is R&D. The assignment motivates students to acquire these four skills in an integrated manner. This unit and its constituents build upon the knowledge acquired in ME502. This assignment is a foundation to Assignment 2, so a proper understanding of it is essential.

This assignment requires students to review the literature on channel estimation and equalisation of wireless channels using OFDM transmission. Channel estimation and equalisation is jointly one of the most important applications of digital signal processing (DSP) algorithms and tools in wireless communications. These are needed for the estimation and decoding of transmitted data at the receiver. While completing this task, you will demonstrate the application of DSP tools studied in this unit, such as convolution, DFT, time shifting (delay or advance), scaling, averaging and scaling.

Wireless channels are liable to multipath signal propagation which induce frequency-selective fading into the transmitted signal. Such channels may also experience time-selective fading if the Doppler spread/shift is high. Channels for high-speed trains, drones, high data rate applications (as in e.g. LTE- Advanced and 5G), underwater acoustic applications, and aeronautical communications are liable to high Doppler spreads owing to the high relative movements between the transmitting and the receiving antennas or the clutter around them. Channels liable to both frequency selectivity and time selectivity are referred to as doubly-selective channels. OFDM transmission has been adopted to counter frequency selectivity in wireless channels. However, OFDM is useful only if the orthogonality between its subcarriers is maintained. Wireless technologies using OFDM transmissions (e.g. LTE and 5G) presume that the wireless channel's variations within an OFDM symbol duration is negligible. However, this is not the case on doubly-selective channels due to the high Doppler spread, resulting in intercarrier interference (ICI) within an OFDM symbol. Algorithms estimating and equalising doubly- selective channels are much more involved than those for singly-selective channels.

Your main task is to study and simulate using MATLAB or Python an article on an algorithm to estimate doubly-selective channels for the purpose of equalisation to reduce bit error rate.

1. Download from the Moodle site for this assignment the article named "Article_for_assignment_01"

2. Study and critique the paper

3. Using either MATLAB or Python, simulate the algorithm proposed in the article to verify the results in the article. Your study and report should include the following:
a. An algorithm for estimating the length Lc of the channel's impulse response. You will need to search on Google scholar and/or library for useful material. Clearly indicate the value of Lc used in your simulation.
b. Select one basis expansion model (BEM) and clearly justify your choice. The common BEM include complex exponential BEM (CE-BEM), generalised CE-BEM (GCE-BEM), polynomial BEM (P-BEM) and discrete prolate spheroidal sequences BEM (DPSS-BEM).
c. Your estimated carrier frequency offset (CFO) and OFDM symbol timing offset (STO).
d. Assume maximum relative velocity between the transmitting and receiving antennas of 350 km/h and estimate the maximum Doppler spread needed to estimate the number, Q, of BEM basis functions.

4. Your submission should be a single pdf document and should include your MATLAB or Python script as the appendix.

In your simulation assume the following parameters:
1) N=601 (maximum number of subcarriers in each OFDM symbol). This is equivalent to a system using 10 MHz with 15-kHz subcarrier spacing.
2) 15-kHz subcarrier spacing
3) Fs=30.72 MHz sampling frequency
4) QPSK, 16-QAM or 64-QAM baseband modulation (those you studied in ME502)

Your submission should follow a structure like the article being simulated. Specifically, it should be structured into the sections:
a) Title
b) Abstract
c) Introduction
d) Analysis
e) Simulation results and discussions
f) References

Article - Joint Maximum Likelihood Timing, Frequency Offset, and Doubly Selective Channel Estimation for OFDM Systems by Hamed Abdzadeh-Ziabari, Wei-Ping Zhu and M.N.S.Swamy

Attachment:- Digital Signal Processing.rar