Somewhere on the internet, it's reported the maximum diversity gain limit theoretically can be achieved is 10dB. Some papers said it increases with more antennae and some said it will decrease.
Greetings, may the link for literatures validating the following point i.e. increase in DG beyond 10dB with increase in number of antenna radiators be shared here....
Hi,
According to chapter 10 of Wireless Communications by ANDREA GOLDSMITH:
Without any multiplexing gain (MG), the maximum diversity gain (DG) is equal to the multiplication of the number of TX and RX antennas. For example for a two-antenna system maximum DG is equal to 2*2=4(=6dB) without any MG. There exists a trade off between MG and DG (Fig. 10-8 of the reference) such that for the MG=r, DG=(M-r)(N-r) where M, N denote the number of TX and RX antennas, respectively. The value of r is always less than min{M,N}. So, if M and/or N are increased while r is kept constant, then DG would increase. In the case that r is Not kept constant, then increase or decreas of DG=(M-r)(N-r) depends on how r is changed along with M and N.
The MG itself depends on the rank of channel matrix, i.e., the number of the parallel independent virtual channels obtained through singular value decomposotion of channel matrix, as well as the gains of those parallel channels, and whether power is optimally allocated to the channels according to their gains. If M and N are increased, the rank of the new channel matrix and its SVD might result in more virtual channels with good conditions (high gains) and higher MG if power is optimally allocated over all of these new virtual channels. By controlling power allocation over less number of channels with good conditions and/or reducing the number of parallel data streams fed to the precoding process in TX (Fig. 10-3 of the reference), MG could be kept unchanged or even decreased after increasing M and N.
Hope this was helpfull.
- Badges
- Science topic
- Similar topics
- Filing