EC ENGR 212A
Theory and Design of Digital Filters
Description: Lecture, three hours; discussion, one hour; outside study, eight hours. Requisite: course 113. Approximation of filter specifications. Use of design charts. Structures for recursive digital filters. FIR filter design techniques. Comparison of IIR and FIR structures. Implementation of digital filters. Limit cycles. Overflow oscillations. Discrete random signals. Wave digital filters. Letter grading.
Units: 4.0
Units: 4.0
1 of 1
Most Helpful Review
Fall 2024 - Never have I simultaneously vitriolically hated and holistically loved a class like this one. The homeworks are a**. The questions can be unclear at times, difficult spikes randomly, and there are a LOT of questions. Not much guidance is given on how to solve them, meaning you're at the mercy of the TA. Our TA didn't record lectures nor posted any solved questions. Additionally, the lectures sometimes don't contain the full process of solving questions. Your best bet is going to the professor. Otherwise, expect to spend LONG hours (~18-24 hours per homework that's due in like 10 days) The final project's initial stages are not well executed. You're given instructions on how to benchmark and adjust your filter, but the designing and implementation is totally up to you to find out. You're given some initial research papers, but while not explicitly stated in the specs, you're totally expected to branch out and find other papers. In other words, you're forced to play the research paper gacha (or lottery), an exercise in luck and frustration. (Use IEEExplore's search!) The lectures... are wonderful. The professor is clear and follows logical trains of thought. He also doesn't rush through material, so it's very easy to follow. Additionally, his lectures are posted online and are legible. They typically are enough to complete most of the problems such that you really don't need discussion. If you were to bind them together, they'd probably better than most textbooks on the matter. In all, take this class only if you need to. It'll be more difficult than the most difficult class you've taken (unless you've already had the joy of taking a class with Pamarti)
Fall 2024 - Never have I simultaneously vitriolically hated and holistically loved a class like this one. The homeworks are a**. The questions can be unclear at times, difficult spikes randomly, and there are a LOT of questions. Not much guidance is given on how to solve them, meaning you're at the mercy of the TA. Our TA didn't record lectures nor posted any solved questions. Additionally, the lectures sometimes don't contain the full process of solving questions. Your best bet is going to the professor. Otherwise, expect to spend LONG hours (~18-24 hours per homework that's due in like 10 days) The final project's initial stages are not well executed. You're given instructions on how to benchmark and adjust your filter, but the designing and implementation is totally up to you to find out. You're given some initial research papers, but while not explicitly stated in the specs, you're totally expected to branch out and find other papers. In other words, you're forced to play the research paper gacha (or lottery), an exercise in luck and frustration. (Use IEEExplore's search!) The lectures... are wonderful. The professor is clear and follows logical trains of thought. He also doesn't rush through material, so it's very easy to follow. Additionally, his lectures are posted online and are legible. They typically are enough to complete most of the problems such that you really don't need discussion. If you were to bind them together, they'd probably better than most textbooks on the matter. In all, take this class only if you need to. It'll be more difficult than the most difficult class you've taken (unless you've already had the joy of taking a class with Pamarti)
1 of 1
