[Paper Review] Implementation of Sensor Network using Efficient CAN Interface

Yeshwant Deodhe, Swapnil Jain+1 more|arXiv (Cornell University)|Apr 26, 2014

Bluetooth and Wireless Communication Technologies4 references3 citations

TL;DR

This paper proposes an efficient sensor network architecture using Controller Area Network (CAN) for real-time monitoring of industrial parameters such as temperature, pressure, speed, and torque. By leveraging CAN's high reliability, low latency, and robust communication at speeds from 20 kbit/s to 1 Mbit/s, the framework enables distributed sensor nodes to exchange data reliably in industrial environments, demonstrating CAN’s suitability for scalable, cost-effective sensor networks with strong real-time performance.

ABSTRACT

Sensors monitored by centralized system, that may be used for controlling and monitoring industrial parameters (Temp, Pressure, Speed, Torque) by using CAN interface. In this paper we presents a comprehensive overview of controller area networks, their architecture, protocol, and standards. Also, this paper gives an overview of CAN applications, in both the industrial and nonindustrial fields. Due to CAN reliability, efficiency and robustness, we also propose the extension of CAN applications to sensor network. In this paper, a framework of sensor network for monitoring industrial parameters is explained where sensors are physically distributed and CAN is used to exchange system information. CAN (Controller Area Network) is a high integrity serial bus protocol that is designed to operate at high speeds ranging from 20kbit/s to 1Mbit/s which provide an efficient, reliable and very economical link

Motivation & Objective

• To design a scalable, cost-effective sensor network for industrial parameter monitoring using CAN as the communication backbone.
• To evaluate CAN's suitability for distributed sensor networks beyond traditional automotive applications.
• To demonstrate the feasibility of using CAN's high-integrity serial protocol for real-time, reliable data exchange in industrial sensor systems.
• To extend CAN applications into broader industrial and nonindustrial monitoring scenarios through a structured framework.

Proposed method

• The authors implement a centralized monitoring system where multiple sensors are physically distributed across an industrial environment.
• CAN protocol is used as the communication backbone, enabling data exchange between sensors and the central controller at speeds from 20 kbit/s to 1 Mbit/s.
• The system leverages CAN’s built-in error detection and fault confinement mechanisms to ensure high integrity and reliability.
• The framework is designed to support real-time monitoring of dynamic parameters such as temperature, pressure, speed, and torque.
• A layered architecture is proposed, integrating sensor nodes, CAN transceivers, and a central controller for data aggregation and control.
• The implementation focuses on minimizing hardware cost while maintaining robustness and real-time responsiveness.

Experimental results

Research questions

• RQ1 Can CAN be effectively used as a communication backbone for distributed industrial sensor networks?
• RQ2 How does CAN’s reliability and efficiency support real-time monitoring of critical industrial parameters?
• RQ3 What are the key architectural and protocol-level advantages of using CAN in sensor network deployments?
• RQ4 To what extent can CAN be extended beyond automotive applications into broader industrial monitoring systems?

Key findings

• The proposed CAN-based sensor network enables reliable, real-time communication among distributed sensors at data rates up to 1 Mbit/s.
• CAN’s inherent error detection and fault tolerance mechanisms significantly enhance system robustness in industrial environments.
• The framework supports efficient monitoring of multiple industrial parameters, including temperature, pressure, speed, and torque, with low latency.
• The use of CAN reduces system cost while maintaining high integrity and scalability for industrial applications.