Allocation of Heterogeneous Resources of an IoT Device to Flexible Services

TL;DR

This paper formulates and solves the complex problem of optimally allocating heterogeneous network resources of IoT devices to multiple services, providing algorithms and simulations for practical scenarios.

Contribution

It introduces a precise mathematical model for resource allocation in IoT devices with heterogeneous interfaces and develops approximation algorithms for large instances.

Findings

Optimal solutions for small instances using the formulation.

Two approximation algorithms for large instances.

Simulation insights into service splitting across interfaces.

Abstract

Internet of Things (IoT) devices can be equipped with multiple heterogeneous network interfaces. An overwhelmingly large amount of services may demand some or all of these interfaces' available resources. Herein, we present a precise mathematical formulation of assigning services to interfaces with heterogeneous resources in one or more rounds. For reasonable instance sizes, the presented formulation produces optimal solutions for this computationally hard problem. We prove the NP-Completeness of the problem and develop two algorithms to approximate the optimal solution for big instance sizes. The first algorithm allocates the most demanding service requirements first, considering the average cost of interfaces resources. The second one calculates the demanding resource shares and allocates the most demanding of them first by choosing randomly among equally demanding shares. Finally, we provide simulation results giving insight into services splitting over different interfaces for both cases.