In this paper: https://www.researchgate.net/publication/325008791_Beamforming_Design_for_Max-Min_Fair_SWIPT_in_Green_Cloud-RAN_with_Wireless_Fronthaul

Here is its system model

enter image description here

In this model,it said the received signal in each DR is

$y_k=\mathbf h_k^H \mathbf w_k s_k^D+\sum\limits_{i=1,i \neq k}^{K}\mathbf h_k^H \mathbf w_i s_i^D+\sum\limits_{j=1}^{J}\mathbf h_k^H \mathbf v_j s_j^E+n_k$, Notice that due to the energy symbol $s^E_j$ is randomly generated, which carries no information but only satisfies the RF regulations;$s_k^D$ is data symbol for DR k .

Honestly,i don't understand about $s^E_j$,i mean i know the received signal of DR1,

$y_{DR1}=$(signal from RRH1) + (signal from RRH2 and RRH3,etc)+(noise),that is

$y_k=\mathbf h_k^H \mathbf w_k s_k^D+\sum\limits_{i=1,i \neq k}^{K}\mathbf h_k^H \mathbf w_i s_i^D+n_k$

but i don't know where does $\sum\limits_{j=1}^{J}\mathbf h_k^H \mathbf v_j s_j^E$ come from

I have a thinking before,the signal RRH transmit is information+RF signal,that is $s_k^D+s_j^E$,but in the other paper,they will only mention the $s_k^D$,like this paper: https://arxiv.org/pdf/1805.08898.pdf ,it just said $y_k=\mathbf h_k^H \sum\limits_{j=1}^K \mathbf f_j s_j+n_{a_k}$,so does anyone know about what the $s_j^E$ is?


Well, it seems the authors assume a setup where there are separate data receivers (DR) and energy receivers (ER) ("To achieve a balanced user experience for separately located datareceivers (DRs) and energy receivers (ERs) in the network, joint transmit beamforming vectors are optimized [...]"). In that setup, $s_k^E$ would be symbols that are used for sending energy to the ERs whereas $s_k^D$ are data symbols for the DRs.

Therefore, if you are a DR and want to receive data, you care about two types of interferences: data symbols for other DRs and energy symbols for the ERs. Note that you didn't copy equation (1) exactly: the second term should be $s_i^D$ (data symbol interference). Then you have the third term with interference of energy symbols for the ERs and the fourth term for the noise.

As an ER, you wouldn't care about interference, it's actually additional power, if you like.

Other papers use other models where one node may want to receive data and harvest energy at the same time. As they say: "all models are wrong, but some are useful..."

  • $\begingroup$ I think i know the difference,if the receiver consist of information decoder and energy harvester,we need a power splitter to separate the received signal,that is ,in this model,we will receive the signal (information+RF ),so we will have the power splitter coefficient in the SNR formula,and think the RF is not a noise.However,if the information decoder and energy harvester are not in the same device,i mean,they are truly separated, and we won't have the power splitter coefficient in the SNR formula,and think the RF is a noise now.Is my thinking right? $\endgroup$ Jul 1 '19 at 23:46
  • $\begingroup$ Well, yeah, there are no power splitters in this paper as far as I've seen. You're a DR that cares about receiving its own data. The data symbols of other users as well as the energy symbols of other users represent interference ("noise") to you. $\endgroup$
    – Florian
    Jul 2 '19 at 6:45

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.