If I uses Low Pass Filter instead of Band Pass Filter,Is some information will be lost?
does this do any changes on the shape of IQ?
You would not lose any more information in a low pass filter than a band pass filter if you are using the same cutoff frequencies for the low cut. There are many ways you can lose information and it all depends on the type of filter you use, what parameters, and what your cut off frequencies are. All of this will ultimately have an effect on your IQ, the goal is to do so with the least amount of artifacts as possible while still achieving your intended purpose of filtering.
Related
I want to design a series of optimal filters for ultrasonic signals. The filter order is fixed at 16 and I want the filter to have a passband for a list of ranges, i.e. (16kHz-38kHz, 17kHz-39kHz etc.). I've been using Matlab's Filter Designer and trying to different design inputs in a trial and error manner. What method can I use to find the optimal filter design algorithm (Remez, least-windows or something else) rather than haphazardly throwing inputs to the wall and seeing what looks best (I'm looking for a filter that best attenuates in the stopbands and has a flat frequency response in the passband, given my filter order constraint).
16 taps is not much to work with for an FIR filter, so I'm a dubious whether satisfying results are possible with any design. That said, I suggest trying Remez, aka Parks–McClellan optimal FIR filter design. See Matlab's firpm function. FIR filters designed by the Remez algorithm are "optimal" in the sense of minimizing the maximum pointwise error. Anthor plus about Remez is it allows setting different importance on each band, which could be useful to balance the tradeoff between passband flatness vs. stopband attenuation.
i have quick simple question about neuron network. As we all know, it is better to make the network deeper instead of wider. So what will happen if i set each hidden layer to be just one neuron and expand my network really deep?
This question is raised because i had a lecture about CNN today. The reason why we use CNN is that we want to extract the features of images and decrease the dimensions of the input data. Since we keep making the input for each layer smaller and smaller, why not just use one neuron and make network deeper? Or something bad will happen?
thanks!
Obviously, the single-neuron example doesn't work -- otherwise, that's what we'd use.
The kernels of each layer in a CNN harness spatial relationships and evaluate those juxtapositions with non-linear functions, which are the main differentiation of a CNN over a simple linear-combination NN. Without those non-linear operations, the CNN layers are merely a programming convenience.
If you immediately collapse your input to a single value, you have a huge problem in trying to write the cascading non-linearities that comprise the output evaluation. Yes, it's theoretically possible to write a function with, say, 28x28x3 inputs and exactly the output you want -- and to train the multitiude of parameters in that function -- but it's very messy to code and nearly impossible to maintain.
For instance, imagine trying to code an entire income tax form in a single function, such that the input was the entire range of applicable monetary amounts, Boolean status info, scalar arguments (e.g. how many children are live at home), ... and have the output be the correct amount of your payment or refund. There are many Boolean equations to apply, step functions with changing tax rates, various categories of additional tax or relief, etc.
Now, parameterize all of the constant coefficients in that massive calculation. Get some 10^6 real-life observations, and train your model on only that input and labels (tax/refund amount). You don't get any intermediate results to check, only that final amount.
It's possible, but not easy to write, debug, or maintain.
I am currently using Mallet for training a sequential model using CRF. I have understood how to provide features (that solely depend on input sequence) to the mallet package. Based on my understanding, in mallet, we have to compute all the values of the feature functions (upfront). Now, I would like to use indicator functions that depend on the label of a token. The value of these functions depends on the output label sequence and during training, I can compute the values of these indicator functions as the output label sequence is known. But, when I am applying this trained CRF model on a new input (whose output label sequene is unknown), how should I calculate the values for such features.
It will be very helpful to me if anyone can provide me any tips/relevant documents.
As you've phrased it, the question doesn't make sense: if you don't know the hidden labels, you can't set anything based on those unknown labels. An example might help.
You may not need to explicitly record these relationships. At training time the algorithm sets the parameters of the CRF to represent the relationship between the observed features and the unobserved state. Different CRF architectures can allow you to add dependencies between multiple hidden states.
Goal
I'm trying to have a second signal (caused by force applied to a sensor's surface area), once it reaches a set minimum, to determine a servo motor's position. The motor's position is otherwise determined by a first signal (generated by EMGs that are measured with electrodes placed on a subject's body) that make give the servo preset positions when the signal crosses a certain threshold.
Schematic
I cannot seem to get the two signals to work together. Any corrections and recommendations on how to make the logic within Simulink function as intended would be immensely appreciated. I know little of the program and cannot find a good approach.
In your diagram you go from numeric values (Sensors) to logical ones (>=301) back to numeric (*90) and again to logical (AND). What you are doing may be possible but hides the intention. First of all I would recommend you to clearly distinguish between logical conditions and values, similar to the following model. To help you with that you can turn on data-type display from Format > Port/signal display > Signal data types followed by an Update (Ctrl-D).
Alternatively you could reduce the conditions as much as possible and operate only with values, such as
However I do not fully understand your requirements and I'm also not sure what the input of the servo is representing (Speed, position, ...)? Maybe you could explain the desired behaviour a bit more detailed? In particular how do you come up with the limits (301, 300, ...).
I've noticed in Apple's accelerometer sample code they use both regular and adaptive low pass filters, what is the difference?
They are both IIR 1st order low pass filters (simple, and laggy in responsiveness when compared with other DSP techniques). The adaptive filter switches to a higher frequency roll-off (and thus becomes even less smoothing but more responsive) for larger accelerations.
There are other higher quality (and more complicated) DSP filtering techniques for motion sensing often used by portable game developers.
In general, any adaptive filter will adjust itself based on the input signal. I'm not sure if you'll notice much difference in practice. Just try it out and see if one gives better response to what you're trying to do.
IIR Adaptive has a flatter response curve which means you have a higher fidelity output. Or, in other words, what you put in is what you get out.