I want to design a diversity reciver (in a single input multiple output (SIMO)) where i have the following specifications.
1. i have 100MHz RF bandwidth
2. i use simple BPSK
3. No of transmitting antennas are only one.
4. I want to achive 1Gbps speed
I want to know is there any clear procedure to find the number of antennas required to achieve the desired data rate. Please let me know if any one has done such a design with an SDR board or any thing. Thank you.
Dear Bharath Reddy Gudibandi , because you have only one transmitter then increasing of quantity receivers antennas can increase SNR and anti-jammer protection. If you have M jammers (multi-path signals) then the minimal quantity of receivers antennas should be M+1. If you need to increase SNR N times, you need to use N receiver antennas. With a big SNR, you can have the opportunity to use QAM256 modulation instead of simple BPSK. In this regard, if you want to achieve 1Gbps speed then you need to use OFDM signals with a few thousands sub-carriers.
Well, Vadym gave you too detailed of an answer and stole my thunder, but the simple answer is, "One!" =) In a pure line of sight environment with no multipath or interference, you don't need more than one antenna to get what you're asking for. Adding more antennas just gives you increased SNR by either power or voltage gain (the latter if you create a phased array).
Once you add multipath to the environment now you get into a complicated scenario based on the particular environment, but the goal is to combat spatial fading by having another antenna (or more) to fill in holes in the spectrum at a given location of the first antenna. (i.e. antenna diversity) This gets into more complicated discussions including spatial correlation, where it's not just the number of antennas but their separation that matters.
Then adding interferer's to the puzzle, we're now talking about SINR, and while straight diversity schemes can help, it's the ability to do beamforming that gives a multiple antenna system its real power. Two antennas can give me 6 dB of gain, but more importantly they can give me an infinite null to point in the direction of an interferer. The SIR is impacted much more by the null depth than the beam peak.
Thank you Michael D. Foegelle for remembering me the interferer's role.
Dear Bharath Reddy Gudibandi ,
One uses the receive diversity to improve the signal to noise and interference ratio. So one sends say one signal and receives it by N antennas. Assuming the transmission coefficient of the nth antenna is hi then the received signal r can be expressed by
r= sum (hi S +ni) ,
where S is the sent signal amplitude ,
The system capacity C = BW Sum (log ( 1 + Ihi SI^2/ ni^2) from i=1 to N
The most simple idealized solution of this problem is to assume that the paths are all similar and the above equation reduces to
C= N BW log ( 1 + Ihi SI^2/ ni^2) This is assuming 10 separate receivers.
It the recieved signal power is combined directly after the antenna then
the recieved power will R sum of Ri^2 across the N channels, where Ri= hiS
Then the recieved combined power R= Nhi^2S^2 assuming identical paths.
and the capacity C
C= BW log2(1 + Nhi^2 S^2/ N) where N is the noise power of the combined receiver.
This rate is much less than the first rate using N separate receivers.
The rate may decrease under this rate by considering the spatial variations of hi.
Best wishes
- Badges
- Science topic