RFID technology (I)-introduction, market application and prospects, working principles

The course report for a course in the undergraduate course period ~~

The content includes introduction to RFID, market application and prospects, working principles, risks, security and privacy policies, and a total of 5 points.

Because the length is too long, we will divide it into two blog articles. The first part of this article is the introduction, market application and prospects, and working principles.

1. RFIDIntroduction

RFID (radio-frequency identification)WirelessRFID is a non-contact automatic identification technology..It uses RF signals to automatically identify the target object and obtain relevant data. It does not require human intervention and can work in various fields of society. It is small in size, low in cost, and easy to embed into objects, it can be read in a large amount without any need for access. A complete RFID system consists of three parts:

1. Tag(TAG): It is mainly composed of on-board coupling components (coils and microantennas) and chips. The coupling element is used to receive and send electromagnetic signals and obtain energy and time sequence from electromagnetic waves. The chip is used to store electronic code (and some also have logical computing functions ). A label is usually relatively small, and it is attached to an object to uniquely identify the target object. A tag is usually in sleep state and can be activated or disabled by a reader.
2. Reader(Reader): A device that performs read/write operations on RFID tags. The device can activate or disable tags. The device consists of two parts: RF module and digital signal processing unit. The device can be designed to be handheld or fixed. Readers are the most important infrastructure in RFID systems. They can send electromagnetic signals (providing energy and time series) to RFID tags, receive weak electromagnetic signals returned by tags, and convert them to digital signals, then, the returned information is obtained through processing and sorting, and then uploaded to the application software system.
3. Application SoftwareUnified(Application Software System): Using computer software and networks to process the obtained information requires the participation of background databases, which contain a large amount of tag-related information. The system provides the end user with a visual human-computer interaction interface to help the user obtain Tag Information and complete instructions on the reader. Since the system needs to be specially formulated according to different enterprises in different application fields, it is difficult to have universality.

1.1 working methods

The basic work of RFID is divided into the following four steps (table 1 ):

1. The reader sends a query signal to the tag through broadcast (the tag is activated ).
2. After obtaining the query signal, the tag sends its electronic code (ID) to the reader in the form of electromagnetic waves.
3. After obtaining the electronic code (ID), the reader queries the corresponding database through the network (the tag is disabled ).
4. The database sends the product information corresponding to the electronic code (ID) to the reader or a terminal device over the network.

Figure 1 basic working method of RFID

1.2 tag Classification

FollowRFIDDifferent Application FrequenciesWhich can be divided into low frequency (LF), high frequency (HF), ultra high frequency (UHF), and microwave (MW). The Representative frequencies are as follows: low frequency below 13.56 kHz, high frequency MHz, high frequency 860m ~ 960 MHz, microwave 2.4g ~ 5.8 GHz.

FollowRFIDDifferent Supply Modes of tag energyWhich can be divided into Passive Tag (passive), active tag (active), and semi-active tag (Semi-active ). Currently, about 80% of tags on the market are passive, and less than 20% of tags are active and Semi-active.

1. Passive TagNo built-in battery is passive. The operating power of tags is obtained from the electromagnetic signal sent by the reader. Generally, the tag information is transmitted to the reader through reflection modulation. It can only work within the reader's RF range, and the read/write distance is close (about 1mm ~ 1 m), long service life, smaller and lighter, low price.
2. Active tagsThe built-in battery is active. The working power supply of the label is completely supplied by the internal battery, and the energy of the battery is partially converted to the RF energy required for the tag to communicate with the reader. It can provide a further read/write distance (about 100 meters ~ 1500 m), large volume, high cost, need to replace the battery after energy consumption.
3. Semi-active tagsThe built-in battery and battery supplies power to the circuit that requires data and the work of the chip in the label. The label remains in sleep state until it enters the working state, which is equivalent to a passive label. The battery energy consumption inside the label is very small and can be maintained for several or even 10 years. When the tag enters the reader's RF area, it is stimulated by the RF signal sent by the reader and enters the working state, the energy required for information exchange between tags and readers is mainly based on the RF energy provided by the reader (reflection modulation mode). The function of the battery is to make up for the insufficient RF field strength at the tag location, the battery energy is not converted to RF energy. Its read/write distance, life, and cost are between the two.

1.3 coding system

The electronic coding structure of RFID tags mainly includesEPCCoding SystemAndUID CenterCoding System.

2. EPCThe full name is electronic product code. The Chinese name is product electronics.Code. EPC is the only electronic code assigned to an item. Its length is usually 64-bit or 96-bit, and can also be expanded to 256-bit. Different encoding formats are required for different applications. EPC code is a set of numbers consisting of four fields: version number, domain name management, object type, and serial number (usually in hexadecimal format), suchTable 1. The version number identifies the version number of the EPC, which allows later EPC to have different lengths or types for better compatibility. Domain Name management is the information of the manufacturer related to this EPC, for example, The Coca-Cola Company records the exact type of the product, for example, the 330ml canned sugar-free cola produced in the United States. The serial number uniquely identifies each item, it will precisely tell us which kind of cola is in this category. The European and American EPC coding system limits the application field to the logistics field and focuses on the successful large-scale application.
4. UID CenterThe architecture of ubiquitous identification is composed of four parts: ubiquitous identification code (ucode), Information System server, ubiquitous communication device, and ucode resolution server. Ucode is the unique identifier assigned to any physical object in the real world. It has ample capacity of 128 bits and can be further expanded to 128, 256, or even 384 bits in 512 bits. The biggest advantage of ucode is that it can accommodate the metadata design of the existing encoding system and be compatible with multiple encodings. Ucode tags come in various forms, including bar codes, RF tags, smart cards, and active chips. The ubiquitous recognition center classifies tags and sets up nine levels of different Authentication standards. The information system server stores and provides ucode-related information. The ucode parsing server determines the information related to ucode is stored on that information system server. The communication protocol of the ucode parsing server is ucoderp and ETP, and ETP is a password authentication communication protocol based on etron (PKI. The ubiquitous communication device consists of tags, readers, and wireless wide-area communication devices. It sends the read ucode to the ucode Resolution Server and obtains relevant information from the Information System server. The Japanese-dominated uid center is dedicated to the application of RFID technology in various fields of human production and life, and promotes the popularization of RFID technology through a wide range of application cases.

Table 1 96-bit EPC code example

EPC-96ISerial number
01	0000a89	00016f	000169dc0
Version Number Field	Domain name management	Object category	Serial number
8-digit	28-bit	24-bit	36-digit

 

2.RFIDMarket application and prospects

2.1 Origin and Development

The RFID technology originated from the Second World War. It was designed for the Royal Air Force to identify the enemy and me Planes in air combat. After the British invented the radar, they found a big problem in application: on the radar screen, they couldn't figure out which planes are enemies and who are themselves. As a result, they installed the identification system (identificationfriendorfoe, IFF) on the plane and installed the corresponding identification device (secondary radar) on the radar ). After a radar discovers a target, it will automatically send an inquiry signal. If it is a plane of its own, the enemy and me identification device will automatically answer the question, so that the radar can distinguish the enemy and the enemy from the plane. Each military aircraft now has an enemy-me identification system.

After decades of development, RFID technology has not only achieved great success in the military field, but also demonstrated its powerful application capabilities in commercial and other social fields. Today's human manufacturing technology makes RFID tags smaller and smaller, and the wide application further reduces its cost. A common tag may only need less than 5 cents, A reader only costs $100. At the same time, the cost reduction also accelerates the application of RFID in various fields.

2.2 advantages

RFID advocates believe that it inherits and develops the optical bar code function that consumers are familiar with and will replace the bar code in the future and become the main force of identity recognition technology. Compared with bar code, RFID has two unique advantages:

1. Uniqueness. A barcode only shows the type of an object. RFID can uniquely identify each object. We can use the electronic code in the RFID tag to search for the object in the database for details.
2. Automation. The bar code requires close-range optical recognition (that is, line-of-sight contact) and precise positioning. Most scanning tasks are performed manually, resulting in a slow speed. On the contrary, RFID does not require line-of-sight contact and precise positioning. It uses electromagnetic waves distributed in the air for scanning, and basically does not require human participation. Therefore, it can be automated and the scanning speed is very fast.

2.3 Application and prospects

RFID applications can be roughly divided into the following three categories:

1. Logistics and Supply Chain Management-- Labels that require fast, low-latency, and easy-to-read, require large-capacity reading capabilities, and require low security mechanisms.
2. Consumer Electronics-- High requirements on security mechanisms, requiring only lightweight reading capabilities.
3. Vertical applicationsProgram-- A custom application for a specific business.

Currently, RFID is widely used in society, including:

2. Large retailers (such as Walmart) use a large numberRFIDLabel to achieve fast and accurate logistics management and counter checkout business.The company can effectively track and maintain the warehouse receiving, warehouse picking, and stock of goods. Customers only need to push the trolley filled with goods to the cashier to complete the inspection of all goods, this greatly reduces the waiting time and the theft rate. In addition, this is also conducive to the after-sales service and consumption of goods information understanding.
4. Large consumer goods manufacturers (such as Procter & Gamble) will also use a large numberRFIDLabels for fast and accurate production and logistics management.The company can effectively track and maintain the production, warehouse receiving, warehouse picking, and stock of goods.
6. Digital LibraryUse RFID tags to manage and track books.
8. Boeing and AirbusIt is planned to use RFID tags for efficient identification and tracking of aircraft parts.
10. Contains RFID tagsContactless smart cardIt can be used in smart bus systems, non-contact access control systems, open attendance systems, conference sign-in and non-contact commercial applications.
12. Automatic charging systemUse RFID tags for fast billing. For example, the E-ZPass system allows drivers to charge their roads quickly through the toll station without delaying the driver's journey and reducing energy consumption, in addition, this avoids road congestion caused by FEE queuing.
14. Payment credentialRFID technology can also be introduced. In the United States, the speedpasstm gas station uses RFID-containingFuel dispenser. In addition,Credit CardRFID technology is also used, such as expresspaytm of American Express and paypasstm of MasterCard.
16. Automotive Electronic Security SystemAlready highly integrated with RFID technology. The smart keys in the anti-theft system contain RFID tags.
18. In the concept of IOT, consumers can useRFIDSmart appliances enable automatic notification and purchase of commodities, and provide rich life functions.The drug cabinet can track whether the drug is taken at the right time and whether the dose is correct. The wardrobe allows you to recommend appropriate clothes Based on the weather conditions and the style of your clothes. The refrigerator can send a warning to the user when the milk is about to expire or the amount is insufficient, or even automatically send a shopping list to the specified store to request door-to-door delivery. When working, the washing machine automatically checks the clothing label, understands its fabric composition and washing requirements, and makes the best washing options to avoid serious damage to it.
20. RFIDTechnology has broad application prospects in the identification and tracking of people, animals and livestock.The owner can implant RFID tags in the pet or collar to quickly retrieve their pets when they are lost. Similarly, large farms, especially those in stock, can track each livestock to prevent loss, loss of control, and disease control. In a crowded environment, we can use RFID to quickly determine whether or not a person is in the crowd or even find their location.
22. We can useRFIDTags track assets, especially the management of valuables.This helps to determine whether an item is correctly placed and to find an item. It also helps in the recovery and identification of theft and loss.
24. Consumers canUse built-inRFIDThe mobile phone of a reader or tag performs interactive applications.For example, a consumer can use a mobile phone to scan a movie poster containing tags and then learn the details of the movie, including directors, actors, venues, and fares. Consumers can even scan any product at any time to obtain detailed information about the product and the manufacturer, and provide a basis for purchase.
26. InMedical Management,Medical Accident PreventionThe RFID technology provides information and prompt risks conveniently. In the process of drug taking, injection, surgery, Ward monitoring, RFID can check the drug devices and check the matching between patients and drugs. At home, patients can also use the system to check whether the drugs they take are correct, which is very important for the elderly.
28. Product anti-counterfeiting SystemRFID tags can be used to prove the authenticity of the product and help crack down on counterfeit and shoddy products.
30. InID card(Passport, ID card, etc ),Banknotes,Air TicketUsing RFID on other items, but achieving efficient query, verification and tracking.
32. CountriesDefense Military SystemRFID technology is widely used, such as the enemy and me identification system and personnel and materials inspection system.

Some of the above applications have been implemented and popularized, some are still in the pilot phase, and some may be just concepts and ideas, but I believe they will become reality in the near future, in addition, more and more powerful applications are derived. It is predicted that the world's RFID market will exceed RMB 2016 in 270 billion, including tags, devices and related services. At the same time, the Internet of Things will gradually take shape, and people's lives will be full of networks and electronic information. Life will become very convenient and work will be more efficient.

3. RFIDWorking principle

Based on the structure and technology used, RFID readers can be divided into read/write devices, which are information control and processing centers of RFID systems. Tags are information carriers of RFID systems. A reader is usually composed of a coupling module, a transceiver module, a control module, and an interface unit. Generally, information exchange is performed between the reader and the tag through half-duplex communication. The reader coupling provides energy and time sequence for the Passive Tag. In practical applications, management functions such as collection, processing, and remote transmission of object recognition information can be further implemented through Ethernet or WLAN.

The basic working principle of RFID is not complex: For a non-source and Semi-active tag, when it enters the magnetic field range of the reader, it can receive the RF signal of a specific frequency from the reader, the information stored in the chip is sent out in the form of electromagnetic waves with the energy obtained by the induction current, and the active tag will actively send signals at a certain frequency. The reader receives and decodes the signals from tags, and then sends the signals to the application software system through middleware for data processing. In addition, some readers can read and write product information into tags and change the information stored in tags in the form of RF signals. From the communication between the reader and the tag and the energy sensing method, the sensor coupling and backscatter coupling can be roughly divided into two types: induction coupling and backscatter coupling, generally, the first method is used for low-frequency RFID, while the second method is mostly used for high-frequency RFID.

3.1 multi-tag anti-collisionAlgorithm

In real life, if multiple RFID tags receive query signals and send information at the same time, the signals received by the reader will interfere with each other, the reader can only process one electronic code at a time, so the tag reading conflict will inevitably occur. RFID technology must be able to quickly and efficiently identify multiple tags. Currently, there are two anti-conflict algorithms to solve the RFID tag read conflict problem:

1. Time Slot-basedAloha.It uses the random selection of the sending time. The reliability of the system identification is relatively poor, but it is easy to design and implement. Therefore, low-cost tags are generally designed using this algorithm. How to improve the recognition reliability of the algorithm system has become the focus of the current low-cost tag application system.
2. A tree structure-based anti-conflict algorithm.The binary tree search algorithm is used to ensure high system recognition reliability. However, the system must be associated with the electronic coding information of tags during implementation. The hardware design is complicated. How to simplify the design and reduce the cost has become the research direction of this algorithm.

3.2 conflict prevention Algorithm Based on Time Slot ALOHA

The Time Slot ALOHA algorithm (framed slotted ALOHA) is short for FSA. It is a random time division multiple access method for sending and receiving User Information Communication algorithms. It uses information frames to represent the channel, information frames are divided into many slots. Each tag selects a random time slot to send its own identifier information. During the entire Information Frame Time, each tag only responds once, as shown in table 2.

Figure 2 Information Frame time division multiplexing of the Time Slot ALOHA algorithm

In Figure 2, each circle represents the electronic encoding information sent by a tag. In this way, the tag reply encountered by the reader during the entire Information Frame process involves three situations: Success, idle, and conflict, they may indicate a response with a tag, no tag, or more than two tags in a time slot. In actual situations, due to the distance between tags and the reader, the information sent by the tag at the close distance may overwrite the information sent by the tag at a long distance, even if the time slot conflicts, the reader may also correctly identify the close-range tag information. Similarly, due to the impact of other environmental noise, even if there is only one tag response in one time slot, the reader may not be able to read successfully. Without considering the two unsatisfactory conditions (capture effect and environmental noise), if the number of time slots of the entire information frame is set to F, when reading n tags, the number of time slots for successful (A1), idle (A0), and conflicting (AK) in each information frame is:

 

3 charts (1)

Therefore, the read throughput of the RFID system (also known as recognition efficiency, that is, the proportion of tags that the reader can successfully identify within a message frame for a long time) can be expressed:

4-way chart (2)

Through the MATLAB simulation experiment, it can be found that when the number of tags is close to the Information Frame Length (that is, the number of tags is close to the number of time slots), the system throughput is relatively high, this is consistent with the result obtained through the Differential Calculation in formula 2. In RFID system applications, the number of RFID tags read by the reader is often unknown. Based on the analysis results of the above multi-tag RFID anti-conflict algorithm, to implement a system solution with the RFID anti-conflict algorithm function, the system must first predict the number of tags in the field. Generally, the following prediction methods can be used:

1. Lowbound ). If there is a conflict in reading, there are at least two tags, and the number of tags that conflict can be predicted to be at least 2 × AK.
2. Schout prediction. If the time slot selected by each tag in each Information Frame complies with the Poisson distribution of λ = 1, the average number of tags responding to each conflicting time slot in the information frame is about 2.39, in this way, we can predict that the number of unrecognized tags is 2.39 × AK.
3. Vogt prediction. By comparing the actual number of successful, idle, and conflicting time slots with the number of theoretically successful, idle, and conflicting time slots, the minimum error is obtained to predict the number of unknown tags, namely:

5 charts (3)

C1, C0, and CK are the actual values of successful, idle, and conflicting time slots. In the range of N tags, [C1 + 2 × CK ,..., find the smallest ε value in 2 × (C1 + 2 × CK)], and the corresponding N value is the number of predicted labels.

The simulation results of MATLAB show that the throughput of the system reading based on the number of tags prediction is significantly improved compared with the FSA (that is, the length of the information frame is fixed to 256. But in general, when there are a large number of tags on site (especially when the number of tags is greater than 500), the formula (2) is used) it is inappropriate to set the optimal length of Information frames based on the number of predicted tags. Therefore, some people proposed to adopt the group response method, that is, when the number of tags exceeds 354, the tags are grouped and the first response of the 1st groups is made, after recognizing 1st groups, you can recognize 2nd groups, and so on. Table 6 shows the relationship between the number of groups and the number of tags.

Chart 6 Relationship Between tags of grouping algorithms and Information Frame Length

The simulation results of MATLAB show that the system throughput of grouping algorithms can reach a high value when the number of tags is greater than 500, while other prediction methods are quickly reduced. Therefore, grouping algorithms can effectively improve the system's recognition efficiency in large-scale tag recognition.

3.3 tree structure-based conflict prevention Algorithm

Compared with the ALOHA algorithm, the binary tree algorithm has a good recognition rate, which is widely studied at home and abroad. Among them, the binary search algorithm BS (Binary Search) and the Dynamic Binary Search Algorithm DBS (Dynamic Binary Search) have a high system overhead. Based on these two algorithms, a binary search algorithm based on the return type is generated, which can effectively reduce the number of system queries. In addition, there are query Tree algorithms QT (query tree) and the improved conflict tracking tree (CTT) (collision tracking tree ). The latter uses the conflicting bits detected by the reader to update the query prefix, which can effectively avoid the disadvantages of increasing the query prefix by bit in the query tree algorithm. Compared with the query tree algorithm, the improved algorithm can effectively reduce the system running time and communication complexity, and reduce the amount of data transmitted by the system. After checking the information on the Internet, I learned the idea and performance of the query tree algorithm and the conflict tracking tree algorithm. The algorithm IDEA is very similar to the tree-walking algorithm provided by the teacher.

Chart 7 tree-walking Algorithm

 

Chart 8 tree-walking algorithm flow

3.3.1 query Tree Algorithm

Algorithm Protocol

The query tree algorithm is a non-memory algorithm that uses binary sequences. tags do not have to store previous queries, thus reducing tag costs. First, the reader sends a query command (a K-bit prefix) to the tag to check whether the tag contains the query prefix. If multiple tags respond, the reader can determine that at least two tags contain the prefix. Therefore, the reader adds a 0 or 1 prefix after the current query prefix to form two new prefixes and then continues to ask. When there is only one tag to respond to the query prefix, the tag will be identified. Repeat the preceding process until all tags are identified.

The specific algorithm protocol is described as follows:A Indicates a binary sequence set with a length not greater than K, and the reader State is represented by P (Q, m. The queue Q is the Character Sequence in A, and the Register M is a binary sequence, which stores successfully queried tag IDs. The reader's query prefix Q is the binary sequence in a. The tag response W indicates the binary sequence whose length does not exceed K.

Reader

Assume that the initial value of Q in the query queue is empty, and the initial value of register m is empty.

1. Set q = {Q1, q2, Q3 ,... , QN}, M = Q1
2. The reader broadcasts the query prefix Q1 to the tag.
3. Update the query queue q = {Q2, Q3 ,... , QN}
4. Receive tag response
5. If the received response is W, the reader checks for conflicts based on the tag response. If a conflict exists, update q = {Q2, Q3 ,... , QN, qn0, qn1}; otherwise, W is inserted to the end of register m, and a tag is found successfully. If no response is received, no action is taken.
6. Repeat the preceding process until Q is null.

Tag

Sequence w1w2w3... WK indicates the tag ID. If Q is null or Q = w1w2w3... WQ, that is, the tag ID matches the query prefix, And the tag will set the sequence WQ + 1wq + 2... WK is sent to the reader.

Algorithm performance

The query tree algorithm performance is mainly measured in two aspects: time complexity and communication complexity. The time complexity is expressed by the number of queries required by the reader to fully recognize n tags. If ts is used to represent the number of tags identified in the query Tree Protocol, when the number of tags is N or greater than 4, the statistical relationship is used:

2.881n-1 ≤ E [ts] ≤ 2.887n-1 (1)

In formula (1), E [ts] indicates the mathematical expectation of the number of identifiers. In the worst case, the number of queries required to identify n tags is:

N ≤ n × (L + 2 −log N) (2)

In formula (2), L indicates the length of the tag ID.

The Communication Complexity of a reader is the total number of BITs sent by the reader by recognizing n tags. Assume that the length of the tag ID is l, so the number of BITs sent by the reader each time cannot exceed l bit, while E [ts] ≤ 2.887n-1. Therefore, the reader's communication complexity has an approximate relationship:

LR ≤ 2.89ln (3)

3.3.2 conflict tracking Tree Algorithm

Core Idea of Algorithms

For the query tree algorithm, when the reader broadcasts a query prefix and multiple tags respond, the tags will pass their remaining IDs to the reader. Assume that the first n-1 bits in the ID sent by each tag are the same and the nth bits conflict occurs. Therefore, before detecting the nth bits in the QT algorithm, the reader must repeat at least N-1 digits. This increases the system's time complexity and Communication Complexity to some extent. Can we reduce the number of queries in this case to reduce the system overhead? Therefore, the improved QT algorithm is generated.

The improved algorithm is based on the query tree algorithm. The difference is that the algorithm uses conflict tracking. The core idea is that the reader sends a K-bit query prefix. If the first K-bit of the tag ID is the same as the one sent by the reader, the tag sends the id after K-bit. The ID of the tag sent by the reader. If no conflict exists, the reader continues to receive the ID. If the nth bit is detected to have a conflict, the reader sends an instruction to the tag, stop the tag from sending an ID number to the reader. Unlike the query tree algorithm that adds only one query prefix each time, the improved algorithm updates the query prefix only when the reader detects a conflict, after the updated query prefix is the current query prefix, add all the bits before the newly received conflicting bits, and add 0 or 1 at the end. We call this improved algorithm the CTT (collision tracking tree) of the conflict tracking tree ).

Algorithm flow

1. The reader broadcasts the K-bit query prefix to the tag.
2. When the first K of the tag ID is the same as the K-bit query prefix of the reader broadcast, the tag sends the id after the K-bit; otherwise, the tag does not respond.
3. The reader receives the id after the K-bit. If the nth-bit conflict is detected, an ACK command is sent to the tag to stop the tag from sending the ID number to the reader.
4. The reader processes tags according to the following States:
   1. ) When the nth bit conflicts, the reader saves two new query prefixes in lifo, all the bits before the original prefix plus the received conflicting bits (the k + 1 bits to the n-1 bits), and add 0 or 1 at the end.
   2. ) When the last vertex of the tag response conflicts, the reader considers that there are only two tags, because the IDs of any two tags are different.
   3. ) If no conflict occurs, the tag is identified.
   4. ) Repeat the above process until lifo is 0.

The process table 9 of the conflict tracking tree algorithm is shown in.

Chart 9 conflict tracking tree algorithm process

Algorithm performance

Compared with the query tree algorithm, the conflict tracking tree algorithm reduces the System Computing workload and communication traffic, thus reducing the system's time complexity and Communication Complexity to a certain extent.

To be continued ......