Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

BER measurement

Status
Not open for further replies.

Maitry

Junior Member level 1
Joined
Sep 17, 2020
Messages
16
Helped
0
Reputation
0
Reaction score
0
Trophy points
1
Activity points
127
Hello support team,

I have my vector signal generator that can generate PN sequence, provide PSK modulation and transmit via RF signal as well at the receiving end, I have demodulator. I want to verify the BER performance of my demodulator. But, i do not have BER Tester intsrument with me.

I have my transmit PN sequence file as well received data bits file. So, kindly guide on how I can measure BER without the use of BER tester?

Is there any possible way to compare both the files ( transmitted and received) and plot the EbNo vs BER plot via any online tool or simulator?

Awaited your guidance.
 

matlab does this but to compare theory from practise, you need to modulated quadrature signals with AM/PM p-p random noise to reduce the BER and accelerate errors to generate the slope and offset (if any) of the log curve. Then compare dB/decade slopes and asymptote offset and find root cause of errors.

With sync delays you can compare Tx to Rx with XOR and counter. Using precision p-p detector on injected noise signal with variable register delays and amplitude of PRSG analog signal to simulate Ricean echo BER errors and Raleigh noise do the same. Ricean errors are more dominant in fringe SNR testing in real situations.

In theory, there is no difference.
In practise, there is.
 

Hello,

Actually modulation on the data is being done with my vector signal generator( Transmitter) and On the receive end, I have my hardware in which there is a demodulator IP core. This core output the demodulated data that I can see in the oscilloscope.

That means , All the things that are the requires of BER testing includes transmitter( PN generator, modulation) , Real time noise addition, receiver( demodulator) etc. But the only thing I don't have is BER tester.

So, I have one reference information PN sequence which is modulated and transmitted by transmitter. I have received data from the receiver.

Now, I want to measure demodulator performance via BER testing without BER Tester simulator.

For that, I have attached the the MATLAB code I am referring. as well Transmitted PN9 sequence.

In this code, Below are the things that I do not understand.
1) When I import the attached file(1.txt), MATLAB do not take it as a binary file and provide input_file as inf. MATLAB only understand the text file in which data is written in vertical like the attached file(txdata1.txt). Is there any way to import the file like 1.txt into MATLAB.

2) Other thing is that, This code includes transmitter, AWGN noise as well demodulation(detection), but my only requirement is as per below.

- Import of the transmitted file( such as 1.txt)
- Import of the received file( taken from oscilloscope)
- Compare between the 2 files for the bit errors
- Plot the SNR vs BER to verify the received data BER

Kindly request you to help and guide for the above.

Awaited your guidance.
 

Attachments

  • 1.txt
    544.9 KB · Views: 238
  • txdata1.txt
    1.4 KB · Views: 242
  • MATLAB.txt
    514 bytes · Views: 236

Please apply to the Rx discriminator, not Tx. you can measure Window Margin =which is an alternative to BER or to compare with theoretical EbNo vs BER. But I found a it was easier to design and build a BER test set.

PRSG delayed and XOR with Rx PRSG data out to count errors. Then add jitter to clock or data.

The easiest method is 3 tap delay for data and PLL to centre then pseudo random select Early/Late/nominal to =+/- TBD% clock delay to accelerate BER by adding quantum random jitter
 

Although expensive for digital video, this describes what I did for old 10 Mbps HDD's and 1.544 Mbps telecom DS1 streams in measuring BER vs SNR https://noisecom.com/resource-libra...eid/84/noisecom-ufx7000a-series-demonstration

Matlab simulation https://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.415.4700&rep=rep1&type=pdf

What is your bit rate and medium?
--- Updated ---

You can do your research to find something that matches your yet to be defined design of a Eb/No test set https://ntrs.nasa.gov/api/citations/19870012669/downloads/19870012669.pdf
 
Last edited:

Hello,

My received bit rate is 4.8 Kbps . and the received data is coming from header pins and I can check the same on oscilloscope. It is not possible for us to purchase the BER tester immediately. So, In general, I have to do it via simulation only.

Can you plz suggest the modification in given MATLAB code?
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top