Control-Oriented Power Allocation for Integrated Satellite-UAV Networks

TL;DR

This paper introduces a control-oriented power allocation strategy for a satellite-UAV network with integrated sensing, communication, computing, and control, optimizing control performance rather than traditional communication metrics.

Contribution

It formulates a convex optimization problem for power allocation that minimizes control cost, providing a closed-form solution and highlighting differences from classic methods.

Findings

Convexity of the power allocation problem is established.

Optimal power relates to the intrinsic entropy rate of SC3 loops.

Control-oriented power allocation outperforms capacity-oriented methods.

Abstract

This letter presents a sensing-communication-computing-control (SC3) integrated satellite unmanned aerial vehicle (UAV) network, where the UAV is equipped with on-board sensors, mobile edge computing (MEC) servers, base stations and satellite communication module. Like the nervous system, this integrated network is capable of organizing multiple field robots in remote areas, so as to perform mission-critical tasks which are dangerous for human. Aiming at activating this nervous system with multiple SC3 loops, we present a control-oriented optimization problem. Different from traditional studies which mainly focused on communication metrics, we address the power allocation issue to minimize the sum linear quadratic regulator (LQR) control cost of all SC3 loops. Specifically, we show the convexity of the formulated problem and reveal the relationship between optimal transmit power and intrinsic entropy rate of different SC3 loops. For the assure-to-be-stable case, we derive a closed-form solution for ease of practical applications. After demonstrating the superiority of the control-oriented power allocation, we further highlight its difference with classic capacity-oriented water-filling method.