Radio Frequency Identification Systems New Ideas and Algorithms

Radio Frequency Identification Systems New Ideas and Algorithms Department of Computer Science University of Virginia Leonid Bolotnyy and Gabriel Robins School of Engineering & Applied Science lb 9 xk@cs. virginia. edu, robins@cs. virginia. edu Introduction to Radio Frequency Identification (RFID) Systems ® RFID Primer ® EPC System Architecture u Three types of RFID tags Local Server Reader ² Passive ² Active ² Semi-Active Tag ID ® Applications www. cs. virginia. edu/robins ® Reader-Tag Communication tag Tags Tag ID tag signal u Operational Frequencies antenna Tag ID ² 125 KHz - 5. 8 GHz ID Info u Operational Range Tag ID Inductive Coupling Server IP Object Server ² 5 mm - 15 m ® Major Research Issues u Standardization Bodies ² International Organization for Standardization ² EPCglobal, Inc Far-Field Propagation u Reducing the cost of tags u Providing security and privacy u Standardizing the technology ONS Server Infrastructure ONS Server Multi-Tag RFID Systems ® Attach more than one tag to an object ® Voltage on a tag u Redundant Tags u Dual-Tags ² Private memory only ² Shared and private memory u n-Tags ® Benefits of Multi-tags ® Optimal Tag Positioning ® Dual-Tags Coordinated Reply Request 1 B-field 3 2 Reader 4 Tag 1 Tag 2 Tag 1: Tag Inductive Coupling: Far-Field Propagation: Tag 1, Tag 2 Data 1, Power 1 Data 2, Power 2 Tag 1 if(Data 1 == Data 2) { Data 1 if(Power 1 >= Power 2) { } Data 1, Error else { ® Expected Largest Angle of Incidence Reader } Tag 2: Same procedure as Tag 1 (note: probability that the Power 1 == Power 2 is tiny) u Increased expected voltage on a tag u Increased expected communication range u Increased memory u Increased reliability u Increased durability ® Reliability and Dependability u Object’s detection is more likely u Failure of a redundant tag ² leaves the system functional ² is detectable in some systems ® Applications of Multi-Tags ® Effect on Singulation Algorithms ® Security Enhancement u Supply chain management ² to increase chances of object detection u Luggage tracking ² regulations require different algorithms u Preventing illegal deforestation ² tagging of trees to prevent illegal logging Randomized Tree Walking Algorithm Binary Variant Randomized STAC Slotted Aloha Redundant Tags No Effect Doubles Time** Causes DOS Doubles Time** Dual-Tags No Effect* No Effect* * If Dual-Tags communicate to form a single response ** Assuming an object is tagged with two tags u n-Tags send “chaff” hiding the real IDs u Recycled IDs are good “chaff” source u “Chaffing and winnowing” has a cost ² extra tag functionality ² overhead to create and filter “chaff” Randomized PRF Tree Walking Algorithm Goal: Efficiently solve reader-tag authentication problem in the presence of many tags ® Steps of the algorithm ® Properties Reader Tag Eavesdropper 1. Each tag generates a random number, and the reader performs a tree-walk on these numbers Backward Range Forward Range Secure Binary Tree-Walking i. Each tag generates a random number ii. Reader performs a tree-walk iii. Selected tag transmits real ID ² stealing a tag ² tracking and hotlisting ® Major questions: u. How to deal with collisions on the tags’ real-IDs? u. How to choose the optimal length for random numbers? u. How to select the threshold? ® Optimal Random Number Length u Allows tags addition and removal from the system u Provides security against active eavesdroppers u Offers security against active readers u Enables dynamic tradeoff between security, privacy, and singulation time u Effective against active attacks: ® Time and Space Complexity n is the total number of tags in the system 2. Once a tag is selected, the reader and the tag engage in a tree-waking private authentication protocol Use average n over many traverse runs : represents related work improvement : represents our improvement as shown : represents our improvement with some modifications ® Random Number Generation Hardware V Random Bits 3. The reader moves the tag to a different position in a tree. No Connect ® Threshold Selection u. Start the threshold at 2 u. Increase threshold by 1 if a collision occurs u. Decrease threshold by 1 if no collisions occur for entire traversal The voltage signal is amplified, disturbed, stretched, and sampled, resulting in random bits. ® Future Work u. Field testing of Multi-tags u. Identifying new applications of Multi-tags u. Improving hardware complexity of the algorithm u. Developing new efficient authentication algorithms