Capacity Optimization in Heterogeneous Home Area Networks With Application to Smart Grid

Abstract

This paper considers the problem of optimizing the downlink capacity in a heterogeneous home area network (HAN) with a beamforming technique at the smart meter for smart grid applications. We model the communication scenario between the smart meter and in-home appliances as a heterogeneous multiuser network. The optimal power-allocation algorithm is developed under the constraints that 1) each user satisfies individual signal-to-interference-plus-noise ratio (SINR) requirement for successful heterogeneous communication, 2) the sum of transmit power allocated to each user is equal to the permissible total transmit power at the smart meter, and 3) the allocated transmit power to each user is feasible. The optimization problem is mathematically shown to be convex, and the optimal power allocation is thus derived. By employing the theorem that the sum of independent exponentially distributed random variables follows Erlang distribution, the probability distribution function (pdf) of the smallest allocated transmit power is mathematically obtained from the properties of downlink indoor Saleh–Valenzuela (S–V) channels. It is analytically shown that the allocated transmit power has a lower limit, which is determined by the SINR threshold and the total number of active users in the HAN. Furthermore, numerical results verify the capacity performance improvements of the proposed optimal power-allocation scheme. It is also shown that beamforming technique contributes to the optimal power-allocation scheme.