Service Embedding in IoT Networks

TL;DR

This paper presents a MILP-based framework for embedding business process workflows into IoT networks, optimizing power consumption and latency in smart applications.

Contribution

It introduces a novel optimization framework that maps virtual process workflows onto physical IoT networks, minimizing power use and latency.

Findings

Reduces IoT network power consumption.

Decreases communication latency.

Enhances overall network performance.

Abstract

The Internet of Things is anticipated to participate in the execution of a variety of complex tasks in the near future. IoT objects capable of handling multiple sensing and actuating functions are the cornerstone of smart applications such as smart buildings, smart factories, home automation, and healthcare automation. These smart applications express their demands in terms of high-level requests. These requests are characterised by the different requirements of sensing actuating functions, processing and memory needs, activation zones, latency, etc. In service-oriented architecture-based IoT, application requests are translated into a business process BP workflow. In this study, we model such a BP as a virtual network containing a set of virtual nodes and links connected in a specific topology. These virtual nodes represent the requested processing and location where sensing or actuation are needed. The virtual links capture the requested communication requirements between nodes. In this paper, we introduce a framework, optimised using mixed integer linear programming MILP, that embeds the BPs from the virtual layer into a lower-level implementation at the IoT physical layer. The proposed framework results in a physical plan that optimally allocates the processing needed and provisions the sensing and actuation at the required locations requirements to an appropriate set of IoT nodes. The optimisation goal is to minimise the IoT layer total power consumption and optimise the traffic distribution in a manner that minimises the traffic latency of each IoT node. Our results show that the proposed framework enhances the network performance by reducing the power consumption and latency.