A Comprehensive Survey of Operating Systems for Smart Home Networks Based on IOT

A Comprehensive Survey of Operating Systems for Smart Home Networks Based on IOT

Suparna N, Manjaiah D H

Department of PG Studies and Research in CSc, Mangalore University Mangalore, Karnataka

ABSTRACT

The entire world is leaping with the biggest technological steps and moving towards automated lifestyles with either Artificial Intelligence networks or IoT networks. The researchers in the field of Computer Science are steering extensive studies to provide various services and Smart Home Automation is one such service. This study guides those enquirers who want to choose a safe and secure smart home system with suitable operating system (OS) that assists the development of reliable application software for home automation by providing a convenient and safe abstraction of IoT devices. The study has taken up comparison of existing surveys and available Operating Systems for Smart Home Networks.

KEYWORDS: Smart Homes, Cooja, Constrained devices, TinyOS, Contiki.

1. INTRODUCTION

   Researchers in the fields of science and technology always focus on making the fruitions of their projects revolutionize various fields such as homes, health, education, construction, automobiles, a nations infrastructure, and agriculture [1]. The field of IoT is such an affluence in technology, where anything and everything can be connected to the Internet and controlled remotely. The use of sensors to accumulate data without any human intervention has made IoT a ubiquitous field and has amplified the process of transforming the human lifestyle. Studies in this field fall into numerous categories. The The extensive survey of research papers in the field of IoT shows that most of the researchers have taken up common issues in the fields such as security, performance of IoT networks, performance analysis of various protocols, the IoT Eco system, the performance of Operating Systems etc. The OS plays a major role in IoT networks because most of the components and devices have constrained resources. IoT OS has varied hardware constraints such as low memory, less computational power, limited resources, and low battery life, and studies have shown that the IoT OS must be equipped to handle these constraints; however, the complexity of the OS must be kept low because the MCU will work at very low clock cycles. Although a number of Operating Systems for IoT are now available, many of them require optimization with reduced complexity. This study contributes to a distinct comparison perspective on IoT OSes.

2. THE IMPETUS FOR THE STUDY

   Home automation is no longer a dream or part of a science-fiction movie. Smart homes have come to existence, and the number of families adapting smart home technology is exponentially increasing worldwide. IoT Home Automation controls the electrical or electronic appliances of our homes using Internet-connected systems. The Operating Systems used in IoT networks are called Smart Home Operating Systems and are designed to coalesce all the connected devices across the

   home and control them from a single platform. A comprehensive smart home OS must be able to create an intelligent system by providing technical support to all types and categories of IoT devices. An example of choosing a smart home OS is the need of the hour, and thus we conducted a critical survey of Smart Home Operating Systems. We also studied existing survey papers and listed the features of these surveys.

3. REVIEWS OF LITERATURES

   1. IoT Ecosystem: A Survey on Devices, Gateways, Operating Systems, Middleware, and Communication [2]: In this article, Bansal et al. categorized IoT as high-end and low-end, and further formed sub-categories such as Linux-based OS and Non- Linux-based OS. In this study, the authors listed and briefly explained the design features for lightweight OS, such as Architecture, Scheduling, Memory management, interfaces and communication protocols, interfaces and communication protocols, simulation ability, Security, Development model, power management, and multimedia features.

   2. Internet of Things (IoT): operating system, application and protocol design, and validation techniques [3]: In this paper, Zikria et al. stated that the important features to be considered when selecting a Lightweight OS are Energy Efficiency, Memory Footprint, Support for Heterogeneous Hardware, Network Connectivity, Interoperability and Security features. 3.3 WSN OPERATING SYSTEMS FOR INTERNET OF THINGS(IOT):

      A SURVEY [4]: In this study, Yaqoob et al. listed design features such as Architecture, Programming Model, Scheduling, Memory Management, Resource Sharing and Real-time Application Support for some popular OS such as TinyOS, Contiki, MANTIS, Nano-RK, LiteOS, and RIOT. They also prepared a comparative analysis table of these features.

      1. Survey on Operating Systems for the Applications of the Internet of Things [5]: Kausar Parveen et al. studied TinyOS, Contiki, MANTIS, NanoRK, and RIOT, and summarized these operating systems for the IoT domain according to the properties of resource constraints, Architecture, Real-time requirements, Programming Model, Scheduling, Memory Management and Protection, Communication Protocol Support, Resource Sharing, Portability, Failure handling, safety, security, privacy, scalability, and upgrading for operating system software.

      2. IoT Operating Systems and Security Challenges [6]: This paper presents a brief study of IoT operating systems and the current security challenges in IoT using RPL and 6LoWPAN (IPv6 over low-power WPAN) protocols.

      3. Comparative Analysis Of Different Operating Systems Used For Low-End IoT Devices [7]: This study focused on lightweight operating systems designed for low-end IoT devices. This paper presents a comparative analysis of various operating systems and discusses the key strategies to consider in their design. These strategies include general models, scheduling approaches, hardware considerations, flexibility, and system capabilities.

      4. Survey of Operating Systems for the IoT Environment

         [8]: This paper explores diverse operating systems designed for resource-constrained IoT environments. It delves into supported platforms and available developer tools, and enables communication protocols, offering a comprehensive overview.

      5. A Comparative Study Between Operating Systems (OS) for the Internet of Things (IoT) [9]: Hicham et al. discussed the important features of OS for IoT, such as Architecture, Modularity, Communication Protocol Support. The authors listed the advantages and disadvantages of the well-known lightweight OS, namely TinyOS, Contiki, Nano-RK, LiteOS, FreeRTOS, and RIOT, and compared the features of these Operating Systems.

      6. An Overview of the Internet of Things Closed-Source Operating Systems [10]: In this paper, the authors present an overview of the common and existing closed-source Oss for IoT. Each OS is described in detail based on the set of design and development aspects that we established. These aspects include architecture and kernel, memory management, scheduling, power consumption, networking protocol support, security, programming models, and multimedia support.

4. PARAMETERS FOR SELECTING SUITABLE IOT

   OS An Operating System (OS) provides services to users to develop applicationsoftware with a convenient and safe abstraction of hardware resources. In Servers and personal computers, the OS allocates threads to processors, virtual addresses to locations in memory, and operates storage devices, peripherals, network devices, and media on behalf of the users application. Generally, IoT networks and devices make use of embedded

   Operating Systems, but many researchers have developed specific and specialized OSes for IoT. As IoT devices are built for specific usage, there is no one-size-fits-all approach to choosing the OS. If the OS chosen is just adequate for the time being, then it becomes tedious in future to change or alter the OS, if the user needs to add some more technologically advanced things or devices Therefore, the Operating System must provide all the necessary hardware, applications, and connectivity requirements of the product, now and in the future. When selecting the operating system for our Home Automation IoT project, it is vital to consider some key elements and common features of various currently available Operating Systems. Scalability, Portability, Memory Footprint, Modularity, Security, Compatibility, Simplicity, Flexibility, Reliability and Consistency are the key features that facilitate the IoT OS selection.

5. IOT ECOSYSTEM

   A set of interconnected devices such as processors, sensors, actuators, and communication hardware enabled by Internet connection constitutes an IoT ecosystem. The basic functioning of IoT system is to acquire, transmit and perform some tasks on the data they obtain from their environments.[11] The Internet Engineering Task Force (IETF) has classified constrained devices used in the IoT field into different [12] according to the required memory for storing code and data[2]. Because the Operating System inhabits a major portion of memory, researchers are thriving to develop tiny Operating Systems suitable for resource-constrained devices.

6. ARCHITECTURES OF OPERATING SYSTEMS

   In an effort to reduce the memory footprint, researchers are striving to lessen the complexity of the IoT OS, which has led to no consensus on the architecture for IoT, agreed universally. The structural design of IoT frameworks and protocols defines their architectures. This outline specifies the functions and principles of the physical components of IoT. Currently, researchers are using five well-known architectures of IoT OS: monolithic architecture, microkernel architecture, three- and five- layered architecture, service-oriented architecture, cloud, and fog-based architecture [2],[13]. Monolithic architecture is a combination of the necessary OS components and applications. The services are implemented separately, and each service has an interface for another service. The monolithic approach resulted in an underprivileged design choice for the OS. The micro- kernel architecture provides minimum functionality in the kernel. The application and the OS were built as a set of interacting modules. Therefore, the kernel size was reduced. Another type of OS architecture is virtual architecture, which works on the principle that a virtual machine is exported to user programs that resemble hardware[14]. The three-layered architecture consists of a perception, network, and application layers. As their names suggest, data are sensed and gathered at the perception layer, transported at the network layer, and processed, and the final product is provided at the application layer. In the five-layered architecture, along with the three basic layers of the three-layered

   architecture, two more layers were added to provide more abstraction to the IoT architecture. The five layers are perception, transport, processing, middleware, and application [2]. SOA, or service-oriented architecture, is a concept designed to build systems that provide services to applications. It is a design pattern and not restricted to any programming language A service is a well- defined self-contained function that represents a unit of functionality. A service can exchange information with another service. Here, the API does not change even if the inner technology and code are changed. This is not dependent on the state of the other services. It uses a loosely coupled message-based communication model to communicate with the applications and other services. In fog-based architecture, four layers are present between the physical and transport layers: monitoring, pre- processing, storage, and security. The monitoring layer observed and checked the data obtained from the sensors. The preprocessing layer performs operations on the sensed data. The storage layer gathers all the processed data. The security layer is responsible for the integrity and privacy of data.

7. OPEN-SOURCE VERSES CLOSED SOURCE OS IOT

operating systems employ microcontroller units (MCUs) to perform fundamental computing tasks on Internet- connected devices. These operating systems fall into two main categories: open-source software (OSS) and commercial or closed operating systems, the latter of which are also referred to as proprietary operating systems. OSS code is accessible to everyone, enabling users to inspect, modify, and enhance it to meet their specific needs. Many prefer OSS to proprietary systems because they tend to be more secure, stable, easily upgradable, and offer greater user control. Some popular OSS include TinyOS, RIOT, Contiki, Mantis OS, Nano RK, LiteOS, FreeRTOS, Apache Mynewt, Zephyr OS, ARM mbed, Yocto, and Raspbian[15]. Linux has released several lightweight operating systems that are specifically designed for IoT and WSN networks. These operating systems can be categorized based on the storage space required for their installation. Versions such as Xubuntu, Zorin OS Lite, Arch Linux, Bunsen Labs Linux Lithium, Bodhi Linux, and Linux Lite require only 1GB storage space. In contrast, Porteus and Puppy Linux occupy less than 500MB. Notably, SliTaz and Tiny Core Linux are the most intriguing options, as they require a maximum of only 100MB of storage space. An Operating System whose source code is not accessible by the public is called CSS or closed Software OS. Only the individual or institution that has created the OS can change the source codes of the OS, and it needs a valid license before installation into any computer. The major names in CSS are Android Things, Windows 10 IoT, WindRiver VxWorks, Micrium µC/OS, Micro Digital SMX RTOS, MicroEJ OS, Express Logic ThreadX, TI RTOS, Freescale MQX, Mentor Graphics Nucleus RTOS, Green Hills Integrity, and Particle[10]. The IoT OS must ensure security, connectivity, interoperability, networking, storage management, and remote-device management.

Therefore, the development of the IoT OS has become a competitive task for developers, and several projects have been set up to perform this tedious task. Some of the projects are developed by individual researchers such as Richard Barry, J P Norair, Dave Hudson and Adam Dunkels. Several studies and surveys have been conducted, and forums and mail lists have been formed to improve the functionalities of these projects. In this study, we examined 18 important Operating Systems for IoT projects. Table 1 summarizes some IoT OS projects and papers published on these projects.

Fig 1: Survey results for operating systems used for IoT devices

(Source: IoT Developer Survey 2016)

IoT operating systems are designed to connect devices seamlessly. With built-in support for various communication protocols, they effectively bridge the gap between different wireless technologies, allowing for effortless connectivity. Furthermore, these systems are incorporated with highly secure communication technologies like Bluetooth, WiFi, Ethernet, WiMAX, LoRa, Z-Wave and Zigbee etc. and also improve security by implementing encryption, authentication, and data integrity measures. Thus, IoT OS ensures that IoT devices communicate securely, protecting them from unauthorized access and tampering.