# Tag Info

8

You've got a pretty good set of circumstances here; you should be able to meet your goal without too much trouble. I don't see anything in your description that would eliminate a whole class of modulation (e.g. phase-shift keying, frequency-shift keying, etc.). Some of the factors that would go into the choice of a suitable format would include: The ...

6

Since what interests you is the "embedded system" part, and since you have a low budget (this excludes anything that requires proprietary compilers), I'd recommend building yourself a board with an ARM MCU and a codec, like this one. There's less than $50 of parts - the processor, the codec and the bare minimum to get them to work. I'm recommending this ... 2 Matched filtering, or cross-correlation, is a pretty cheap and easy way to determine the similarity between signals. Otherwise you could try implementing the algorithm Shazam uses. 2 If I understand you correctly, then your issue is connected with transient that is present when switching on your device, because it is a RLC circuit. Most likely these spikes are not changing over time (because values of passive components are constant), if so, then you can: Create the template of such spike and use the normalised correlation for detection.... 2 The tf2sos function takes an input filter of order$N$, given by$H(z)=\frac{\sum_i^{N} b_iz^{-i}}{\sum_i^{N} a_iz^{-i}}$and returns coefficients for$N/2$second-order filters$H_k(z)=\frac{b_{k0}+b_{k1}z^{-1}+b_{k2}z^{-2}}{a_{k0}+a_{k1}z^{-1}+a_{k2}z^{-2}}$and a gain$g$such that $$H(z) = g \prod_kH_k(z)$$ So, regardless of what normalization you ... 1 Looks like you've done a lot of work on your projects. As @MarcusMüller said, by far the majority of people start with ReLU and go from there. It doesn't have the "vanishing gradient" problem that tanh has for example. All your questions are open ended but common for designing neural networks. There are so many "nobs to turn" to try and make your network be ... 1 I understand the sampling process is completely handled by the sensor itself, i.e. it delivers digital data. Then you just collect the desired number of samples from the sensor and calculate its FFT. The frequency range depends only on the sampling rate ($f_{\text{max}}=f_s/2)$, the frequency resolution depends only on the number of samples$N$, in your case ... 1 Basically you want to implement a filter with batch processing, right? Is this an FIR filter or an IIR filter? What is the sampling rate? Number of taps? Acceptable latency between output and input as batch processing adds some processing latency. 1 The function you are refering to only implements an order-1 filter. There is another function for order-2 filter. If you want to implement an order-3 filter, you will have to split it in an order-1 filter and an order-2 filter. There are many ways to do it, but cascading the filters, as we would do in the analog world, is a good solution, albeit not optimal ... 1 I'm not sure why you oppose statistical signal processing to digital signal processing, they are not mutually exclusive. If you perform statistical signal processing with non-continuous quantized samples, you perform digital signal processing. Second of all, if you are using a microcontroller, you are likely to have limited RAM resources. Can you clarify ... 1 There's an old trick you can use: $$1\cdot x=x$$ So you don't need to multiply with a factor that equals$1\$.

1

Generally, Arm Cortex-A9 type processor can be used for Audio signal processing blocks for effective results. For your case, basic arm processor type can support common floating point operations without any external floating point unit. But, in case of special floating point operations like saturation,rounding and truncation, you shall use some advanced arm ...

1

Some quick partial answers: If you simply want to learn embedded, get an Arduino. If you want to learn a DSP, get EZDSP If you want something that's 32-bits, 1 Ghz, get a Raspberry Pi or pcDuino. Off the top of my head I don't have anything that is a proper DSP that meets every one of your requirements. pcDuino (not sure about Raspberry Pi) has SIMD and ...

1

I do not know about the sinus values in embedded processors, since pre-computed values can be used. The principal problem with the interpolation with the cardinal series is that it is highly, one could even say extremely, acausal. To correctly implement it, you have to go back to times before the big bang and forward to the times after the dissolution of our ...

1

What your professor probably meant is that in most audio converters a small amount of aliasing is permitted (for example if you put 24.1khz into a 48khz converter you will see an output component at 23.9kHz that is not very heavily attenuated). The reason for this is that manufacturers like to use so- called half-band filters in the decimation chain because ...

Only top voted, non community-wiki answers of a minimum length are eligible