RT-OCF: A Lightweight Device-to-Device Framework for the Internet of Things

Abstract

Responding to rapid growth of Internet of Things (IoT) services and devices, many IoT platforms and frameworks are presented for successful IoT realization. Recently, Open Connectivity Foundation (OCF) which consists of a group of industry leaders emerges to create new standards for IoT platform and deliver an open source implementation and a certification program. IoTivity open-source software framework is one of the project sponsored by the OCF which has been developed to provide interoperability among heterogeneous IoT devices. It enables seamless Device-to-Device (D2D) connectivity and targets various application domains such as home and manufacturing automation, health care, and social networks. IoTivity uses, however, rather lots of memory in view of small devices which have limited hardware resources and runs Real-Time Operating System (RTOS). In this paper, we propose a light-weight IoT framework, RT-OCF that optimizes memory consumption and provides a memory tracer to prevent memory leaks. It has a layered architecture which consists of resource layer, messaging layer, and platform adaptation layer together with modules for security and utility. RT-OCF is able to run not only on Linux but also on TizenRT, an open source RTOS platform runnable on ARTIK053 board. The experiment performed on both Linux and TizenRT shows that more than 20% of peak memory is reduced when compared with IoTivity while preserving a packet latency for GET operations.