Article Preview
Top1. Introduction
The Internet of Things (IoT) was designed to connect everything with everyone in an integrated global network. People, machines, natural resources, production lines, logistics networks and virtually every other aspect of social and economic life have the potential to be linked via sensors and software to the IoT platform. The Internet of things at its most generalized level has 3 main characteristics: i) Comprehensive sense: involving RFID, sensors, two-dimension code, etc. to collect information of objects anytime, anywhere; ii) Reliable transmission: accurately delivering object information in a real-time environment using a variety of telecommunications networks and the Internet; and, iii) Intelligent processing: using cloud computing to perform intelligent computations and fuzzy identification to process and analyze massive amounts of data and information (Yun & Yuxin, 2010). The purpose of our paper is to examine RFID technology and propose a peer-to-peer (P2P) network that will enhance the capabilities of traditional RFID technologies. We will also examine in depth our system in the context of supply chain management systems.
Radio frequency identification (RFID) begins with a small tag that contains an integrated circuit chip along with an antenna. The antenna then responds to radio waves transmitted from an RFID reader enabling it to send, process, and store information (Wu, 2006). RFID technology is important and essential because it is the future of IOT and has outstanding economic potential. “In today’s world, RFID technology is being used in many fields such as logistical tracking [in retail outlets such as Giant and Walmart], production monitoring and maintenance, product safety and quality control, access control, tracking & tracing of individuals, infant tracking in the hospitals, eHealth care, public services, and so on” (Siror et al., 2011).
RFID-based tracking systems specifically can be made more practical and efficient by integrating them with a Peer-to-Peer (P2P) network. P2P networks are a better choice because traditional RFID technology and their respective networks are facing several definitive problems. First, the effective reading distance of the reader is very small, and the reading speed of the reader is very slow, in addition to their networks suffering from single node failure, root ONS overload, etc. Second, different countries could have different RFID codes, so it is necessary to implement a peer-to-peer solution to avoid a “language barrier” that disrupts the RFID network. Finally, while we consider security, some enterprises and countries are not willing to disclose product information, creating a situation where a security vulnerability may be present unbeknownst to the rest of the traditional network until damage is incurred from an exploitation. A distributed network that isn’t based on a client/server model and works by combining the different nodes in a network (each with server and client capabilities) is a P2P network. In this type of network, all the computers communicate with each other directly and by exchanging the information in a flexible ad-hoc basis. In the proposed approach of implementing a RFID-based location tracking system, all the RFID readers, or nodes, are connected to other nodes to form a P2P network. The network can be formed using a ZigBee protocol which allows new nodes to join the network and exchange information in a simple manner. In short, a P2P-based system simplifies the process of network installation, technical maintenance of the system, and future additions or subtractions of nodes on the network.
There are various peer-to-peer techniques that could be used. One such technique is the Napster algorithm, which maintains the indices of peers and files that are shared by those peers using a central server. The server also performs resource lookup and provides coordination between peers. This centralized architecture used by Napster is simple and efficient in discovering any resource and looking up peers, but it doesn’t scale well, and the server is subject to a single point of failure. Gnutella (Androutsellis-Theotokis & Spinellis, 2004) on the other hand is a decentralized pure P2P network, which does not depend on any centralized server for lookup and discovery. The network architecture is unstructured, and these networks implement a flooding technique for peer discovery and resource lookup; however, this technique is expensive and time consuming.