Print Settings
 

EE 264: Digital Signal Processing

This is a course on digital signal processing techniques and their applications. Topics include: review of DSP fundamentals; discrete-time random signals; sampling and multi-rate systems; oversampling and quantization in A-to-D conversion; properties of LTI systems; quantization in fixed-point implementations of filters; digital filter design; discrete Fourier Transform and FFT; spectrum analysis using the DFT; and parametric signal modeling. The course will also discuss applications of DSP in areas such as speech and audio processing, autonomous vehicles, and software radio. An optional (1 extra credit hour) lab will provide a hands-on opportunity to explore the application of DSP theory to practical real-time applications. For more information, see the course web page at ee264.stanford.edu. Prerequisite: EE102A and EE102B or equivalent.
Terms: Win, Sum | Units: 3-4 | Grading: Letter or Credit/No Credit

EE 373A: Adaptive Signal Processing

Learning algorithms for adaptive digital filters. Self-optimization. Wiener filter theory. Quadratic performance functions, their eigenvectors and eigenvalues. Speed of convergence. Asymptotic performance versus convergence rate. Applications of adaptive filters to statistical prediction, process modeling, adaptive noise canceling, adaptive antenna arrays, adaptive inverse control, and equalization and echo canceling in modems. Artificial neural networks. Cognitive memory/human and machine. Natural and artificial synapses. Hebbian learning. The Hebbian-LMS algorithm. Theoretical and experimental research projects in adaptive filter theory, communications, audio systems, and neural networks. Biomedical research projects, supervised jointly by EE and Medical School faculty. Recommended: EE263, EE264, EE278.
Terms: Win | Units: 3 | Grading: Letter or Credit/No Credit

EE 392E: VLSI Signal Processing

DSP architecture design. Study of circuit and architecture techniques in energy-area-performance space, design methodology based on a data-flow graph model that leads to hardware implementation. We explore automated wordlength reduction, direct and recursive filters, time-frequency analysis and other examples. The project focuses on architecture exploration for selected DSP algorithms. Useful for algorithm designers who consider hardware constraints and for circuit designers who prototype DSP algo-rithms in hardware. Prerequisites: EE102B and EE108A; Recommended: EE264 and EE271.
Terms: not given this year | Units: 3 | Grading: Letter or Credit/No Credit
© Stanford University | Terms of Use | Copyright Complaints