# Positive slope of BER with a greater frequency deviation?

I am trying to implement a frequency-modulated communication channel on MatLab. However, there is something I don't understand about the sensitivity of the frequency modulator (or the frequency deviation as my signal has an amplitude equal to 1). My message has a 500Hz frequency and is sampled at 20kHz. I added white gaussian noise to my bandpass signal (carrier of 6kHz).

Why does the BER of the demodulated message have a positive slope after 250Hz ?

NB : it is written [dB] for the BER on the graph but it is not, I forgot to remove it

• what is the demodulator you use? – Marcus Müller Dec 19 '19 at 12:19
• I implemented a frequency discriminator composed of a pair of slope circuits followed by envelope detectors. – Théo Dec 19 '19 at 12:29
• ah, interesting! Care to share formula or code? – Marcus Müller Dec 19 '19 at 12:30
• (by the way, I'm assuming you're doing a discrete-frequency FSK so that it's easy to calculate BERs, is that right?) – Marcus Müller Dec 19 '19 at 12:44
• @MarcusMüller Here are the formulas I used link – Théo Dec 19 '19 at 12:45

I see from the link you provided that the slope is indeed constant over a limited usable range. So specifically in your case I assume as $$k_f$$ was increased, it would cause a proportionate amount of the signal to go beyond the discriminator range (where your response is 0).
• Thank you ! I divided the result by the gain $$\frac{1}{A_m4\pi k_f}$$ and it now works – Théo Dec 19 '19 at 13:09