bit rate

[wicak]

so my question is simple, how to set the IQ rate, so that my code could send 2Mbits/s , while I want the bandwidth to be 5 Khz (means that I  have to set my IQ rate 5M/s)

Assumed that I used QPSK that sent 1 symbol consist of 2 bits, 

Roughly this is my calculation, I set the samples/symbol = 8. so that if I set IQ rate to be 8M/s, means that I have 1M symbols, if I set the modulation to be OQPSK, then I will have 2Mbits/s

Am i right? 

But, this way, I will have 8Mhz bandwidth?

So, I have to use IQ rate 5M/s , while set the samples/symbol = 5

So basically, I want to do this

How to set the labview so I could get the proper bit rate..

Please correct me if I'm wrong,

Regards.

[Kaitlin-N]

Hi wicak,

Your calculation is correct.

However, there are a couple of things to consider:

Your IQ Rate will approximately set your Maximum Bandwidth.

Your Symbol Rate (second equation) will set your approximate actual bandwidth.

So in the example you described above, your bandwidth would be approximately 1MHz = 8M * (1/8).

When you said that you want the bandwidth to be 5KHz, did you mean to say 5MHz?

If you want an approximate BW of 5M when using QPSK, your bit rate would need to be 10Mb/sec.

Hopefully these equations will help you find something that will work for you. If you are interested in the theory behind this, you may find this Wikipedia article useful:

Shannon–Hartley theorem:

http://en.wikipedia.org/wiki/Shannon%E2%80%93Hartley_theorem

Please let me know if you have any more questions about this and I will be happy to help.

Best regards,

[wicak]

Hi there, 

I greatly appreciate your help, thank you. I find your explanation help me a lot.

Actually, the important thing is the bit rate (or in my case, the chip rate, since I want to spread 1 symbol over 32 chips).

One more thing I want to ask, is about the OQPS with half sine pulse shapping, This is the constelation point of my code for OQPSK

I think it's weird that, I did not see any changes in constelation point. SInce I know that it is OQPS right? I mean, It has 4 symbol maps, but the difference is the phase of OQPSK will change 90degree. 

What do you thing? is there something wrong with my modulation?

This is the pulse shaping filter that I use.

I know I might send few data ( I will multiply the Y array several times later)

But, maybe you have some insight about my code,

Thank you

[HustLiliAn]

Hi, wicak . Whether are you implementing the 802.15.4 with USRP? 

[wicak]

Yes I do 🙂

[ErikL]

I find the re-sampler very helpful in the modulation toolkit. It will allow you to resample from an ideal symbol rate to a IQ rate that the USRP supports.

The GPS simulation example in the USRP example community uses the same method. The GPS signal is something like 2.048 MS/s. We can use the IQ resample or two resample it to a slightly larger IQ rate that you USRP supports which turns out to be a 2^n divisor of the 100 MS/s sample clock. Luckily you don't have to calulate supported IQ rates, you can just use the coerced IQ rate output of the USRP configure block.

[wicak]

but I found your example for GPS simulation used coerced IQ rate to the MT resample, (not a calculation one)

That means, you just used the IQ rate that the USRP support,