"No one is harder on a talented person than the person themselves" - Linda Wilkinson ; "Trust your guts and don't follow the herd" ; "Validate direction not destination" ;
Showing posts with label RFID. Show all posts
Showing posts with label RFID. Show all posts

June 23, 2019

Outside Retail, RFID in Smart Cities

Past few years I have looked at RFID in retail perspective - Inventory Management. We work on RFID solutions that enable real-time inventory management and improving stock efficiency by tracking
  • Restock
  • Misplaced Items
  • Replenishment Items 
RFID is a major player in Retail for Inventory Management. Millions of items are tagged and tracked using RFID based Inventory Solutions.
I had an opportunity to visit smart city solution provider. The solutions they covered were
  • Waste Management
  • Parking Management
  • Smart Lighting
  • Video Monitoring
  • Environment Management
  • Surveillance
  • Command Centre
Waste Management - The bins are equipped with volume sensors and RFID tags, They are designed with batteries lasting 10 years. They provide details when the Truck and the bin connects, the RFID event is alerted

Intrusion Detection - Sensors placed for Motion detection during the area to alert if Someone comes closer to the area

Environment Management - This is basically measuring pollutants in the air and reporting the trends/patterns

Traffic - This is more of heat maps to reflect the patterns / crowded areas

Command Centre - All the calls are routed/tracked here. Provides a lot of data on issues/queries/ complaints / accidents / reporting any incidents.

Parking - This is based on sensors placed above / below parking lots

State of Art - All of this implementation are mostly based on sensors / RFID. This visit provided a lot of insights into data collection, framework, real-time alerts of end-to-end infrastructure.

Where is Video Analytics here?
Before the visit I was under the assumption a lot of video analytics, people counting use cases would be there. Actually, most of the things are achieved with RFID, sensor-based events. Video Analytics is yet to make a mark in implementation. But down the line in the next few years, we would see a lot of video analytics on the data collected.
  • Crowd Detection
  • Use of Drones for Monitoring
  • Gender Detection from Surveillance Cameras
  • Vehicle Number Detection and Reporting
  • Face Detection and Indexing at Landmarks
  • Loitering
  • Audio-based analytics to detect gunshots/sounds/ accidents
A lot of use cases, RFID + Video Analytics will further strengthen Smart Cities Solutions portfolio.
Happy Learning!!!

January 18, 2019

RFID Cross Reads

RFID Cross Reads

Read1 - Link
  • Identify and control stray reads
  • Lay around tags inside an extended interrogation zone
  • Stray reads from handling units which do not go through the RFID gate
  • Stray reads from handling unit going through neighboring RFID gate
Passive RFID tags do not have any internal power available
Passive tags have typically very long shelf lives

Advances in eliminating cross Reads
  • Impinj provides lay around tag suppression as an proprietary extension to LLRP protocol
  • Thingmagic has patented methodology to determine tag distance based on phase angle change over frequency change
  • Received Signal Strength Indicator (RSSI)
  • Doppler shift based tag movement detection methods have been explored by Sensormatic. The company has patented Doppler shift based methodology to be used in fork lift truck application
  • A camera can also be used as a motion trigger for RFID reader
  • Triggered reading reduces the amount of stray reads, especially when RFID gates are adjacent
Data Sources - Monitored variables are:
  • RSSI
  • Mean RSSI
  • Standard deviation of RSSI
  • Phase angle rotation
  • Standard deviation of phase angle rotation
  • Doppler shift
  • Mean of Doppler shift absolute values
Variables measured for each RFID read event are
  • RSSI
  • Doppler shift
  • Phase angle rotation
  • Receiving antenna
  • Channel
Thoughts - Identifying Cross Reads, Eliminating Cross Reads
  • Data Sources from Multiple Reads
  • Identify those reads that are cross Reads
  • Label the cross reads
  • Analyze and identify if they follow any pattern
  • Feature Variables - Received Signal Strength Indicator (RSSI), Last Read Zones
Anti-Collision Algorithms: A Cross-Layer Approach
Read 2 - Link1
Read 3 - KDE, Link2, Link3

Happy Learning!!!