Software engineering for internet of underwater things to analyze oceanic data

Razzaq, A. and Ahmad, A. and Malik, A.W. and Fahmideh, M. and Ramadan, R.A. (2023) Software engineering for internet of underwater things to analyze oceanic data. Internet of Things (Netherlands), 24: 100893.

Text (SEforIOT)
SEforIOT.pdf - Accepted Version
Available under License Creative Commons Attribution.
Download (3MB)

Abstract

Internet of Things (IoTs) represents a networked collection of heterogeneous sensors – enabling seamless integration between systems, humans, devices, etc. – to support pervasive computing for smart systems. IoTs unify hardware (embedded sensors), software (algorithms to manipulate sensors), and wireless network (protocols that transmit sensor data) to develop and operationalize a wide range of smart systems and services. The Internet of Underwater Things (IoUTs for short) is a specific genre of IoTs in which data about ocean ecosystems is continuously ingested via underwater sensors. IoUTs referred to as context-sensing eyes and ears under the sea operationalize a diverse range of scenarios ranging from exploring marine life to analyzing water pollution and mining oceanic data. This paper proposes a layered architecture that (i) ingests oceanic data as a sensing layer, (ii) computes the correlation between the data as an analytics layer, and (iii) visualizes data for human decision support via the interface layer. We unify the concepts of software engineering (SE) and IoTs to exploit software architecture, underlying algorithms, and tool support to develop and operationalize IoUTs. A case study-based approach is used to demonstrate the sensors’ throughput, query response time, and algorithmic execution efficiency. We collected IoUT sensor data, involving 6 distinct sensors from two locations including the Arabian Sea, and the Red Sea for 60 days. Evaluation results indicate (i) sensors’ throughput (daily average: 10000–20000 KB data transmission), (ii) query response time (under 30 ms), (iii) and query execution performance (CPU utilization between 60%–80%). The solution exploits SE principles and practices for pattern-based architecting and validation of emerging and next-generation IoUTs in the context of smart oceans.