why is the wiener-deconvoluted signal mirrored?

304
April 07, 2017, at 03:06 AM

I am trying to decovolute a signal with Wiener-deconvolution. Myfunction is below:

    def wiener_deconvolution(signal, kernel):
        kernel = np.hstack((kernel, np.zeros(len(signal) - len(kernel)))) 
        H = np.fft.fft(kernel)
        deconvolved = 
       np.real(np.fft.ifft(np.fft.fft(signal)*np.conj(H)/(H*np.conj(H))))
        return deconvolved

What I get is a signal that is mirrored to what it is supposed to be. Can anyone recognize an error in my code?

Rent Charter Buses Company
READ ALSO
Cookie Cutter Automation

Cookie Cutter Automation

For those of you who are familiar cookie cutter, please help me outI have currently created a cookie cutter its up and working

379
How can I encrypt the same message on different machines?

How can I encrypt the same message on different machines?

I'm working with some sensitive data and need to encrypt identification numbers (9 digits)

273
Asynchronous AMQP producer and consumer

Asynchronous AMQP producer and consumer

I have a bidirectional AMQP service that both consumes messages from a queue and publishes to a fanout exchangeClients, symmetrically, consume from the exchange and publish to the queue

471