BFSK Tx and Rx Synchronization Problem in USRPN210

[ITUbahadir]

   Hi,

   I try to do TX and RX implementation of BFSK in USRPN210 using Labview. When i sent desired information bits with some start/stop flag, I success to take bit sequences in receiver. However, I have 2 important problems.

   One of them is that, when I searched received bit sequences, two bits are lost. One of two bits is start of the sequnces, the other bit is end of the bit sequences. For example, when I transmitted 64 bits, I received 62 bits.

   The other problem is that approximately 20-30 bits are solved wrongly until USRP is ready. After these wrong bits are received, receiver begins to solve bits truely.

    For example I want to transmit these sequences,

    1111100000111110000011111(start bit) 0011110011000001(info bit) 1111100000111110000011111(stop bit)

    When I transmit these sequences continuous, I received the bits as,

    01010101110(This sequence is wrong solved approximately 20-30 bits)....1111000001111100000111110011110011000001111110000011111000001111(in this sequence start and end of the bits are lost)...   

   How can I solve this problem?

[fromm8]

Hello ITUbahadir

Could clarify the API you are using for this implementation of BFSK? What VIs exactly are you using? in order to check the configuration...

Regards

Frank R.

[ITUbahadir]

   Hello fromm8, 

   I forgot to share .VI . I added the important blocks as an attachment. The blocks which are not put, is not necessary for impact the system.

   I wrote the code in Matlab in order to seperate transmitted bits in receiver using Write the Measurements File. 

   I hope, you can help me.

[fromm8]

Hello ITUbahadir

I was comparing the code shown in the attachment and it looks like to examples provided by the driver. The only difference is that you are not connecting the channel list. Were you using that shipping example as reference?

Have you tried to increase or decrease the number of bits transmitted? What happens?

How are you running those codes? Are you doing something to synchronize them? You run one, then the other?

What happens if you use other kind of modulation? What about PSK? Same result?

Regards

Frank R.

[ITUbahadir]

   Hello fromm8,

   - Yes, I used some parts from sample as example, however, I added some special part such as eye diagram and bit detection codes from matlab. 

   - I increased the bit sequence number, but anything does not change. whatever I try bit sequence, two bits are lost.

   - Just I used AX.25 protochol for  synchronization. I wrote the code in matlab for AX.25 and I integrated the code to LAbview. Also, I can try the MIMO cable for bit flowing, but the distance is far away between usrp and I dont want to use this cable as practise.

   - I tried PSK instead of FSK, but result is same like FSK.  again, two bits are lost.

[fromm8]

Hello ITUbahadir

Are you able to reproduce the behavior in a very simple example? and share it in order to analyze it?

Regards

Frank R.

[ITUbahadir]

Hello fromm8,

It is my BFSK demodulator basically, please help me 😄

[ITUbahadir]

I found the example for PSK modulation and demodulation. If I apply the this example to FSK, maybe I can solve problem. However, I am waiting answer for my problem . 

Link:

https://decibel.ni.com/content/docs/DOC-18801

[fromm8]

Hello ITUbahadir

Sorry for the delayed response but it took me a while to get equipment and set up the system. I was able to test the example provided by the community in the link you attached. Is it working for you as well?
In order to test your demodulator, could you facilitate your modulator too? In case you still need to solve the issue with that code.

Regards

Frank R.

[ITUbahadir]

My modulator is added as attachment.