| Teaching Signal Processing and Communications with MATLAB, Simulink, and RTL-SDR RadioBob Stewart, MathWorks Professor of Signal Processing, University of Strathclyde, Scotland, UK Abstract: Low-cost hardware provides many opportunities for hands-on learning where students can reinforce theory with practical results and experimentation. For students of signal processing and communications, for less than $20 students they now can acquire the RTL-SDR USB receiver. (For the last few years, FPGA enabled software defined radio (SDR) hardware has been available of course, but it has always been too expensive for students to acquire.) The RTL-SDR device can now provide a hardware platform that will allow I/Q sampling in channels of a few MHz bandwidth over the radio frequency (RF) range from 50MHz to 1.7GHz and digitize directly onto a computing platform (Windows, Linux or Mac). And by connecting the RTL-SDR to a computer running MATLAB and Simulink alongside appropriate drivers and toolboxes we can give students an instant laboratory for exploring RF signal properties, mastering the fundamentals of DSP enabled radio design, and building real-time software-defined radio receivers directly on their own computers. Therefore direct from their desktop, students can perform off-the-air RF signal viewing in the time and frequency domains, design full DSP enabled SDR receivers for FM and AM signals, view mobile, ISM, satellite and frequency bands. As their mastery of DSP and digital communications progresses, then later they can implement complete first principle designs of QPSK type receivers with PHY layers, synchronization, tuning, control layers and so on; the full spectrum of digital communications! We have demonstrated the effectiveness of this hands-on approach in undergraduate laboratories and some recent tutorial workshops at Strathclyde. More recently we have integrated our teaching materials into a free 650 page downloadable book with 120 structured examples. This contains all of the material needed to learn the rudiments of MATLAB, Simulink, and DSP enabled radio design with the RTL-SDR. In this talk we will present findings from this experience, outline some of the examples and pedagogy of our DSP enabled SDR courses and our insights for future curriculum development. Robert Stewart is the MathWorks Professor of Signal Processing at the University of Strathclyde. Since August 2014 he has been Chair and Head of the Department of Electronic and Electrical Engineering, with more than 55 academic staff and almost 300 researchers. His interests in teaching, knowledge exchange and research over the last 20 years has focused around signal processing and communications. In recent years he has been working on digital communications and software defined radio, with specific interest in radio standards and most recently on wireless white space radio utilising TV spectrum frequencies. From 2006-2012 he was the Xilinx Professor of DSP and Digital Logic at Strathclyde, and from since 1997 he has been a visiting Professor at the University of California, Los Angeles (UCLA) Extension School. In 2004 he founded of the technology company Steepest Ascent Ltd, based in Glasgow with an office in Los Angeles. The company was acquired by MathWorks in 2013. |