Achieve the full rate–energy boundary for co-located MIMO SWIPT receivers

Establish transmitter and receiver architectures for MIMO broadcasting with co-located energy harvesting and information decoding (i.e., G = H) that achieve all boundary rate–energy pairs of the rate–energy region defined by R ≤ log|I + H S H^H| and Q ≤ tr(H S H^H), tr(S) ≤ P, S ≽ 0, overcoming the limitation that current energy harvesting circuits cannot directly decode information, so that simultaneous full RF power harvesting and capacity-achieving information decoding is feasible for the same transmit covariance.

Background

For co-located energy harvesting and information decoding receivers (same MIMO channel H), the paper derives an outer bound on the achievable rate–energy (R–E) region using a modified water-filling solution, but explains that practical EH circuits cannot decode RF-band information directly.

Consequently, except for trivial extremes (maximum rate with zero energy, and maximum energy with zero rate), the full boundary of the theoretical R–E region cannot be achieved by current designs. The authors explicitly pose achieving the remaining boundary points—i.e., simultaneous full energy harvesting with capacity-level decoding—as an open problem.