RFID – The Driver of Big Data Generation

By: Denekew A. Jembere

Creative Commons License Licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.

Introduction

With the advent of cheap and easily deployable Radio Frequency Identification Devices (RFID) technology, automated data generation and collection, triggered by an event, has become possible. RFID is the core technology in the Internet of Things (IoT), which is modernizing our surroundings into “smart things” as smart building; smart parking; smart campus; smart cities; and more. RFID’s automatic identification capability is being leveraged in tracking things for purposes such as inventory management; and, for tracking, shopping contains to model customer behavior in groceries, fashion, and the apparel industry.   

For applications related to human beings, the use of RFID devices has raised data security concerns and, researchers and government agencies clearly described consequences that RFID presents in the privacy and data integrity. In this regard, this article will briefly summarize the application of the RFID in the United States Marine Corps (USMC) Passive RFID project; the exponential growth of the digital universe as a result of the introduction of RFIDs; and, the coping mechanisms proposed by researchers to analyze big data streams being generated by smart devices and RFIDs.

Overview and history of the project

According to USMC (2009), in response to a US secretary of defense policy memorandum, USMC envisioned to apply RFID technology to increase business process efficiency, improve data quality and increase asset visibility throughout the Maritime Prepositioning Maintenance Cycle (MPMC).  In relation to this, USMC staged a project having six (two supporting and four primary) sub-projects, each involving the assessment of the current business process; needs and site analysis; proof of principle deployment; revised business process; summary and conclusions with anticipated next steps.      

With four stockholders: the client and three software/hardware providers; four primary and two supporting sub-projects, USMC (2009) reported that the implementation of the project delivered several passive RFID tracking capabilities. Among the major accomplishments detailed in the USMC (2009) report include:

  • Ship Backload (BL) Asset Tracking – Passive RFID infrastructure, with no failure due to battery, enabled to track assets being loaded to Maritime Preposition Ships (MPS).
  • Consolidated Memorandum Report (CMR) Inventory – Passive RFID significantly reduced manhour for conducting a physical inventory.
  • Container Load / Mobile Load (CL/ML) Asset Tracking – Passive RFID reduced manhours required for the manual data recording.
  • Deployable Automated Cargo Measurement System (DACMS) Asset Tracking – Although this is not fully tested, the report authors claimed reliability of asset identification of Passive RFIDs would be similar to the Ship Backload.

In addition to the required system integration testing in each sub-project, according to USMC (2009), BICmd performed different RFID-specific testing in the project, intended for purposes such as: choosing the right passive RFID; make appropriate adjustments in the RFID placement on objects, etc. Among the major tests are:

  • Tag range performances test – On 30 commercially available Passive RFIDs.
  •  Tag placement testing – size and location of the passive RFID tags
  • Tag Readability testing – reading RFID tags on different types of surfaces.

Project evolution

According to USMC (2009) report, the evolution of the passive RFID project in the USMC, Blount Island Command (BICmd), is outlined as follows:

  • July 2004 – Policy Memorandum issued by the deputy under US secretary of state that required each service to develop an Implementation Plan for the use of Active and Passive RFID technologies in logistics business processes
  • January 2008 – A letter of intent (LOI) signed on for joint cooperation, between the Commanding Officer of the BICmd and the VP of Sales, Alien Technology Corp, to develop mutually beneficial passive RFID.
  • August 2009 – With four stockholders: the client, BICmd, and three software/hardware providers; four primary and two supporting sub-projects, the implementation of the project supported several capabilities.

Project status

Although the report fails to mention the completion and the evaluation dates of the project, according to Layher (n.d.), the state of the project after its evaluation for a year, is summarized as follows:

  • Passive RFID Application Results
    • Consistent reliability in tracking assets loaded aboard ship
    • Validated the reliability and range performance of Passive RFID
  • Real-time Precision Accounting & Reporting of MPS assets
    • 100% accountability of assets being loaded into containers/vehicles
    • Compared to 68% using Active RFID
    • Up to 75% reduction in man-hours required to conduct a physical garrison property inventory

The information generated to date

According to IDC (2014), the digital universe is growing 40% a year into the next decade, expanding to include both the increasing number of people and enterprises doing everything online and the smart devices connected to the Internet, Internet of Things (IoT).  Based on IDC’s prediction in 2014, the digital universe would exponentially grow from 4.4ZB in 2013 to 44ZB in 2020. Among the 40 different types of devices, contributors of this estimated exponential growth of the digital universe, as noted by IDC, are RFID tags and sensors; supercomputer and supercolliders; PCs and servers; cars and planes.

Although it is relatively very easy to deploy smart devices or their networks and collect generated data, the use of that data seems to be a challenge, due to the usefulness of the data.  In relation to this, IDC (2014) reported that only 22% of the 4.4ZB data in 2013 or 37% of the 44ZB data in 2020 would be useful if tagged and analyzed. The attempt to tag and analyze such huge data introduces a challenge for the current computing infrastructure, data processing techniques and algorithms and prescribes a change into our existing tools and techniques.

Decisions made as a result of the information

In response to the growth of data, as predicted by IDC (2014), and summarized above, researchers are proposing new techniques of data analysis. For instance, creating data mining models using most of the algorithms require loading the full dataset in memory.  However, in a big data scenario where the data continuously gets generated by smart devices, like RFIDs and sensors, re-loading the full dataset including the new data segment would be impossible due to the frequency of data generation. In relation to this, Fong et al (2016) proposed a new data stream mining approach, Stream-based Holistic Analytics and Rezoning in Parallel (SHARP).

The current state of the technology

According to Landt (2005), RFID is used in hundreds of thousands of applications such as smart theft prevention; smart tolls without stoping; smart traffic management; smart get to buildings; smart parking, etc. Landt predicted that the pace of developments in RFID would continue to accelerate where the future potential requiring advancements in other areas such as the development of applications software; privacy policies and considering other legal aspects. As predicted, RFID has become the core technology for the advent of the Internet of Things (IoT) which has created limitless possibilities for digitally interconnecting different aspects of human life, providing the ability to control and track these aspects from a remote location. However, as noted by IDC (2014), these limitless possibilities are causing the generation of the limitless size of data with an unstructured format that pauses challenges for computing and analysis.

In addition to its role as the core technology for IoT, RFID technology also supplements other technologies providing cost-effective tracking of items. According to Pearson (2017), the U.S. Marine logistic group is actively pursuing the use of RFID technology to track cargo in a cost-effective way, where the last known location of the cargo can be tracked using the RFID tag.  However, for near real-time tracking, the Marine logistic group is also using SHOUT Nano, a two-way satellite GPS that even provides text messaging in an emergency scenario and enable the Marine logistic group to track convoys. As per Pearson, the Marine logistic group did a comparative test for cost-effectiveness and real-time data availability for the application of the RFID and SHOUT Nanotechnologies, to ensure and the use of different assets to improve visibility with intermodal transportation.

References

IDC. (2014, April). Data Growth, Business Opportunities, and the IT Imperatives. Retrieved from EMC Digital: https://www.emc.com/leadership/digital-universe/2014iview/executive-summary.htm

Fong, S., Liu, K., Cho, K., Wong, R., Mohammed, S., & Fiaidhi, J. (2016). Improvised methods for tackling big data stream mining challenges: case study of human activity recognition. Journal of Supercomputing, 72(10), 3927–3959.

Landt, J. (2005). The history of RFID. IEEE Potentials, Potentials, IEEE, 24(4), 8-11

Layher, L. (n.d.). RFID for Tracking of Military Vehicles and Material. Retrieved from rfidjournal.net: https://rfidjournal.net/virtual_events/VE_management/09-2011/RFIDJ_YdMgmt11_USmarineCorps.pdf

Pearson, L. (2017, Feb). Marines and soldiers train with RFID. Retrieved from Marines: https://www.marines.mil/News/News-Display/Article/1083702/marines-and-soldiers-train-with-rfid/

USMC . (2009). Passive Radio Frequency Identification Project, United States Marine Corps Blount Island Command. Jacksonville, FL: ALIEN, STANLEY. Retrieved from https://www.omni-id.com/pdfs/US-Marines_Passive_RFID_Report_BAK.pdf