Design of a Quaternary Query Tree ALOHA Protocol Based on Optimal Tag Estimation Method

Transcription

1 information Article Design of a Quaternary Query Tree ALOHA Protocol Based on Optimal Tag Estimation Method Zhihui Fu, Fangming Deng * and Xiang Wu School of Electrical and Automation Engineering, East China Jiaotong University, Nanchang , China; fuzhihui@sohu.com (Z.F.); zgxiangyu@163.com (X.W.) * Correspondence: dengfangming@ecjtu.jx.cn; Tel.: Academic Editor: Willy Susilo Received: 30 October 2016; Accepted: 26 December 2016; Published: 30 December 2016 Abstract: Radio Frequency Identification (RFID) technology is one of most promising technologies in IoT (The Internet of Things) era. Many RFID systems have been used in supermarkets or warehouses. There are two challenges for RFID anti-collision algorithms. The first challenge is accurately estimating number of tags; or is improving efficiency of RFID systems. This paper proposes an optimal tag estimation method in which tags respond to reader in assigned time slots instead of responding randomly. In order to improve performance of RFID system, a 4-ary query tree Additive Link On-line HAwaii (ALOHA) protocol is presented that combines merits of query tree algorithm and frame slotted ALOHA, and avoids ir weaknesses. Simulation results show that proposed algorithm has a higher tag identification efficiency compared to or dynamic frame slotted ALOHA algorithms, and it can overcome tag starvation phenomenon, because it traces each tag until all of m are identified successfully. Keywords: RFID technology; anti-collision algorithms; 4-ary query tree 1. Introduction Many large-sized chain supermarkets, such as Wal-Mart and Metro AG invest in innovative technology to improve shopping experience for consumers. Supermarket smart payment systems (SSPS) based on Radio Frequency Identification (RFID) technology is an outstanding substitute for traditional payment patterns. With development of low-cost RFID tags [1 3], bar code will soon be eliminated [4]. RFID technology enables objects to be distinguished from a long distance [5], and furrmore, passive RFID tags which collect energy from RFID reader offer several advantages such as battery-less operation, wireless communication, high flexibility, low cost, and fast deployment [6]. Because re is only one communication channel between RFID reader and RFID tags, collision is a familiar problem in RFID systems, easily resulting in missing information and inefficient use of resources particularly hardware resources [7]. Anti-collision is essential for an RFID system. According to difference of channel classification method, tag anti-collision algorithms can be divided into four types: Frequency Division Multiple Access (FDMA), Code Division Multiple Access (CDMA), Space Division Multiple Access (SDMA), and Time Division Multiple Access (TDMA). Most RFID anti-collision algorithms in previous works have adopted TDMA. Among m, Additive Link On-line HAwaii (ALOHA)-based algorithms and tree-based algorithms are more popular. A slotted ALOHA algorithm can decrease probability of collision compared to a pure ALOHA algorithm [8]. In dynamic frame-slotted ALOHA (DFSA) [9 11], length of frame dynamically changes with number of unidentified tags. In frame slotted ALOHA protocols, when length of frame is equal to number of tags, maximum system efficiency is attained, which is about 36.8% [7,12]. Tags select time slots to send ir data packages Information 2017, 8, 5; doi: /info

2 Information 2017, 8, 5 2 of 10 randomly, so performances of existing frame slotted ALOHA protocols are not stable, and tag starvation phenomenon may occur. Recently, plenty of improved ALOHA-based anti-collision algorithms have been published. Reference [10] proposes a new Enhanced Dynamic Framed Slotted (EDFSA) ALOHA algorithm in which unread tags are divided into different groups, and only one group responds meanwhile. An anti-collision algorithm named Collision-Group-Based (CGA) was proposed in [13]. This algorithm has two reading cycles. In first cycle, reader sets frame size. In second cycle, reader decides group size. The tag randomly selects a group, n it selects a slot to send its ID to reader. EDFSA and CGA can improve efficiency to some extent, but y still cannot avoid tag starvation phenomenon and break through restriction of 36.8%. Reference [14] presents a splitting Binary Tree Slotted ALOHA (BTSA) algorithm, which can adjust frame length to a value close to number of tags, and as a result, its identification efficiency can achieve restriction of 42.5%. However, some tags may not be identified for a long time. The tree-based algorithms are deterministic algorithms which can accurately identify all tags. Among m, binary tree anti-collision algorithm (BT) [15] and query tree algorithm (QTA) [16,17] are more well-known. Tree-based algorithms can be expressed as a B-ary tree, where B should be in form of 2 n (n 1) [17]. The 4-ary tree algorithm can decrease number of collision cycles, and efficiency of 4-ary tree outperforms that of 2-ary tree algorithm. This paper presents a novel algorithm to break through restriction of existing methods, where tags respond to reader in assigned time slots instead of responding randomly. Considering advantages of ALOHA-based algorithms and tree-based algorithms, this paper proposes an improved 4-ary query tree ALOHA protocol that combines 4-ary query tree with dynamic frame slotted ALOHA. Simulation results indicate that proposed algorithm can improve efficiency of RFID systems to a great extent, and it can make sure reader identifies all tags accurately. The remainder of article is organized as follows. First, architecture of SSPS will be presented. Secondly, optimal tag estimation method based on assigned time slots will be detailed. Thirdly, a 4-ary query tree ALOHA protocol based on optimal tag estimation method will be described. Then, simulation results will be provided and compared with previous algorithms. Finally, main conclusions will be highlighted. 2. An Optimal Tag Estimation Method When we estimate number of unidentified tags, we should calculate optimal frame size that will maximize system efficiency. The frame size greatly influences system efficiency. If frame size is too large, plenty of idle time slots will occur; if frame size is too small, re will be many collided time slots. In existing frame slotted algorithms, tags select time slots of a frame to send ir data packages randomly, which will cause uncertainty in number of collided time slots and idle time slots. Schoute presented a method to estimate number of unrecognized tags in multi-access RFID system in which tags choose time slot i of a frame to transmit ir data packets that is Poisson distributed, estimation of number of tags (n ) which have not been recognized by reader after current frame is n = 2.39C, where C is number of slots in which collisions arise [9]. Reference [10] proposes a new tag estimation method. Thus, if number of tags is relatively small, it works well; however, if number becomes large, it begins to show poor performance. Now, we will introduce a tag estimation method based on more reasonable probability analysis Description of Tag Estimation Method Given number of unread tags is N, number of time slots in a frame ( length of a frame) is L and p bit serial binary digits (named as assignment bits) in tag ID are used by each tag to select assigned time slot in a frame (as shown in Figure 1).

3 Information 2017, 8, 5 3 of 10 The frame length is decided by number of assignment bits, relationship between p and L is: L = 2 p or p = log 2 L (1) When re are m tags responding in a time slot simultaneously, it means that ir p assignment bits of tag ID are same. The probability can be defined as: P(m, p) = ( ) 1 m (2) L If re are m of all tags responding in a certain timeslot, probability can be defined as: ( ) 1 m ( P(N, m, p) = CN m 1 1 ) N m (3) L L If re is only one tag responding in a certain slot of frame (m = 1), probability is: ( ) 1 1 ( P(N, 1, L) = C 1 N 1 1 ) N 1 (4) L L where CY X is a binomial expression. S is number of time slots in each one of which only one tag transmits its data packet successfully; E denotes number of time slots in each one of which no tag transmits its data packet; and C denotes number of time slots in each one of which more than one tag transmits ir data packet. ( ) 1 1 ( S = E(N, 1, p) = L C 1 N 1 1 ) N 1 ( = C 1 N L L 1 1 ) N 1 (5) L We define throughput T as follows: ( E = L P(N, 0, p) = L 1 1 ) N (6) L C = N S E (7) T = S L = N L ( 1 1 L ) N 1 (8) We can obtain number of assignment bits (p) that gives maximum throughput by differentiating Equation (6). dt dl = N ( L ) N 1 + N ( (N 1) 1 1 ) N 2 1 L L L L 2 = 0 (9) Log 2 N may not be an integer, hence in order to getting more a reasonable frame length, we command that: { } log p = 2 N ; when log 2 N log 2 N < 0.5 (10) log 2 N ; when log 2 N log 2 N 0.5 As shown in Figure 1, when re are collided tags in a frame, all collided tags will respond to reader in next frame. If start bit of assignment bits is fixed, tags colliding once may collide again, which will result in tag starvation phenomenon. So, start bit of p 1 is start bit of tag ID, and start bit of p n+1 is next bit of end bit of p n.

4 Log2N may not be an integer, hence in order to getting more a reasonable frame length, we command that: log N ; when log N log N < p = log N ;when log N log N (10) As shown in Figure 1, when re are collided tags in a frame, all collided tags will respond to reader in next frame. If start bit of assignment bits is fixed, tags colliding once may Information 2017, 8, 5 4 of 10 collide again, which will result in tag starvation phenomenon. So, start bit of p1 is start bit of tag ID, and start bit of pn+1 is next bit of end bit of pn Example Analysis Figure 1. Schematic diagram of assignment bits in tag ID. Figure 1. Schematic diagram of assignment bits in tag ID. Information 2017, 8, 5 4 of Example Analysis We suppose that re are 10 tags A J. A: , B: , C: , D: , E: , F: We , suppose that G: re , are tags H: A J , A: , I: , B: , J: C: , We can get D: , E: , F: , G: , H: , I: , tag identification process as shown in Figure 2. J: We can get tag identification process as shown in Figure 2. Figure 2. Tag identification process. 3. A 4-Ary Query Tree ALOHA Protocol Based on Optimal Tag Estimation 3.1. QTA and 4-Ary QTA Figure 2. Tag identification process. 3. A 4-Ary Query Tree ALOHA Protocol Based on Optimal Tag Estimation 3.1. QTA and 4-Ary QTA Query Tree Algorithms (QTA) have advantage of easy execution due to ir simple structure and operating mode. In QTA protocol, reader asks tags wher ir IDs contain same prefix as query strings q. If prefix of tag ID is same as query strings q, tag will respond to reader, but when more than one tag answers, a collision is detected [13]. The reader n attaches bit 0 or 1 to generate longer prefixes in a queue, and reader repeats process until all tags are uniquely identified. In 4-ary QTA, if a collision occurs in a tree node, two bits 00, 01, 10, and 11 are used by reader to generate longer prefixes in queue, and it can decrease collided cycles compared to 2-ary QTA. Table 1 shows identification process. We suppose re are four tags in reader capture range; ir tag IDs are 0001, 0011, 1000, and 1101, respectively. In round 1, collision occurs because more than one tag responds, and in rounds 2, 5, and 7, re are no tags responding to reader. In rounds 3, 4, 6, and 8, only one tag can be successfully identified. Query Tree Algorithms (QTA) have advantage of easy execution due to ir simple structure and operating mode. In QTA protocol, reader asks tags wher ir IDs contain same prefix as query strings q. If prefix of tag ID is same as query strings q, tag will respond to reader, but when more than one tag answers, a collision is detected [13]. The reader n attaches bit 0 or 1 to generate longer prefixes in a queue, and reader repeats process until all tags are uniquely identified. In 4-ary QTA, if a collision occurs in a tree node, two bits 00, 01, 10, and 11 are used by reader to generate longer prefixes in queue, and it can decrease collided cycles compared to 2-ary QTA. Table 1 shows identification Table 1. The operating process. process We of 4-ary suppose query tree re algorithm are(qta). four tags in reader capture range; ir tag IDs are 0001, 0011, 1000, and 1101, respectively. In round 1, collision occurs because more than one tag responds, and in rounds 2, 5, and 7, re are no tags responding to reader. In rounds 3, 4, 6, and 8, only one tag can be successfully identified. Round Query Response Tag1 Tag2 Tag3 Tag4 (R to T) (T to R) (0001) (0011) (1000) (1101) Queue 1 Collision ,01,10, Collision ,10,11,0000,0001,0010, Empty ,11,0000,0001,0010, Table TheSuccess operating - process- of 4-ary 1000 query tree - algorithm 11,0000,0001,0010,0011 (QTA) Success ,0001,0010,0011 Round Query (R6 to T) 0000 Response (TEmpty to R) Tag1 -(0001) - Tag2 (0011) - Tag3 -(1000) Tag4 0001,0010,0011 (1101) Queue Collision Success , ,01,10, Collision Empty ,10,11,0000,0001,0010, Empty Success Empty 10,11,0000,0001,0010, Success ,0000,0001,0010, Success ,0001,0010, Empty ,0010, Success , Empty Success Empty

5 Information 2017, 8, 5 5 of Description of Proposed Algorithm QTA can accurately identify all tags, but in each tree node, only one tag can be identified. ALOHA-based algorithms are more efficient than tree-based algorithms, but tag starvation phenomena may occur. In this part, we present a 4-ary query tree ALOHA protocol (4QTAP) which combines 4-ary QTA with dynamic frame slotted ALOHA. Firstly, proposed algorithm is based on a 4-ary tree structure. Secondly, all tree nodes adopt dynamic frame slotted ALOHA, and tag estimation method presented in Section 3 is adopted in each node. Thirdly, if tag IDs are known, identification process can be predicted accurately, all tags can be identified efficiently. In 4QTAP, 4-ary tree is composed of initial node and leaf nodes, and y are frames consisting of some time slots. Each leaf node in query tree corresponds to each prefix in queue, and status of each node can be classified as follows. Empty node (E n ): There is no tag responding to reader s query, resulting in a waste of time slot. Success node (S n ): All response tags are successfully identified by reader, and re are no collided tags in S n. The success node does not have branch nodes in query tree, and corresponding prefix will not generate new prefixes in queue. Collision node (C n ): The collision occurs as multiple tags respond to reader s query simultaneously. The collided tags in C n will be recognized in its branch nodes, and reader will attach two binary bits 00, 01, 10, 11 to corresponding prefix to generate longer prefixes in queue. Before initial node works, query string is null, so procedure of initial node is identical to example presented in Section 3; each tag needs p initial serial binary digits to assign a response time slot, and number of time slots in initial node is L initial = 2 p initial. If initial node is S n, it means that all tags have been successfully identified by reader, and process of identification is over. If initial node is C n, two bits 00, 01, 10, and 11 are added to queue as longer query bits. We suppose number of collided tags is C initial, number of unread tags in each branch node of initial node is C initial/4 on average. The number of unread tags in each branch node of initial node is expressed as notation R xx, where xx is two serial binary digits on behalf of four branch nodes. Each collided tag in initial node needs p xx serial binary digits of tag ID to assign a response time slot, and p xx assignment bits are adjacent to p initial assignment bits in tag ID. The reader sends query bits and assignment bits p xx to all unread tags; if prefix of a tag matches query bits, it selects a branch node and a response time slot, n transmits its ID to reader. We suppose that a leaf node is determined by query bits q 1 q 2... q x q x+1 q x+2 q x+3, number of unread tags in this node is R q1 q 2...q x q x+1 q x+2 q x+3, it is equal to (1/4) C q1 q 2...q x q x+1 based on probability (where C q1 q 2...q x q x+1 is number of collided tags in tree node q 1 q 2... q x q x+1 ). Each tag needs p q1 q 2...q x q x+1 q x+2 q x+3 assignment bits to select a response time slot. The reader sends query bits q 1 q 2... q x q x+1 q x+2 q x+3 and assignment bits p q1 q 2...q x q x+1 q x+2 q x+3 to all unread tags; eligible tags will respond to reader. The assignment bits p are set from beginning of tag ID like Figure 1. Some collided tags in a tree node may have same prefixes, which generates a tag starvation phenomenon. In order to avoid this phenomenon, assignment bits should be set from end of tag ID (as shown in Figure 3), and y should keep away from prefixes of tags. The tag ID is composed of a series of binary bits r 1 r 2 r 3 r 4... r b, where b is number of bits in tag ID. When query bits from reader is q 1 q 2... q x, prefix of a tag is r 1 r 2... r x, where 1 x b.

6 Information 2017, 8, 5 6 of 10 Information 2017, 8, 5 6 of 10 Information 2017, 8, 5 6 of 10 Figure Figure The The prefix prefix and and assignment bits in tag ID ID during implementationof of 4-ary 4-ary query query tree tree ALOHA ALOHA protocol protocol (4QTAP). Figure 3. The prefix and assignment bits in tag ID during implementation of 4-ary query tree Flowchart ALOHA of protocol of 4QTAP (4QTAP) The Flowchart The reader s of 4QTAP query and tag s response of 4QTAPis isan an iterative processes as as shown shown in in Figures Figures 4 and 4 and In In order to to implement 4QTAP, necessary notationsare are listedas as follows. The reader s query and tag s response of 4QTAP is an iterative processes as shown in Figures Qb: 4 The and shorthand 5. In order notation to implement of query 4QTAP, bits. It is necessary a serial binary notations bits are qq q, where qx is 0 or 1, 1 2listed x as follows. Qb: The shorthand notation of query bits. It is a serial binary bits q 1 q 2... q x, where q x is 0 or 1, and 1 x b, is number of bits in tag ID. Every Qb determines a leaf node in 4-ary and Qb: 1 The x shorthand b, b is notation numberof ofquery bits inbits. It tag is a ID. serial Every binary Qb determines bits qq qa leaf, where node qx in is 0 or 4-ary 1, 1 2 x tree, and Qb has two statuses in initial node of 4-ary tree, Qb is null; in leaf nodes of 4-ary tree, tree, and and 1 Qb x has b, b is two statuses in number of bits initial node tag of ID. 4-ary Every tree, Qb Qb determines is null; ina leaf nodesin of 4-ary 4-ary tree, Qb is valid. Qbtree, is valid. and Qb has two statuses in initial node of 4-ary tree, Qb is null; in leaf nodes of 4-ary tree, p: The assignment bits of a tag; a tag can select a time slot according to p. In initial node, it is p: Qb Theis assignment valid. bits of a tag; a tag can select a time slot according to p. In initial node, it is expressed as pinitial; In leaf nodes of 4-ary tree, it is expressed as pqb. The reader can calculate expressed p: The assignment p based on Equation p initial ; In bits (11). leaf of a nodes tag; a of tag can 4-ary select tree, a time it isslot expressed according as pto Qb p.. In The reader initial can node, calculate it is p based expressed Queue: onthe Equation as pinitial; In storage (11). leaf nodes of 4-ary tree, it is expressed as pqb. The reader can calculate space in reader, where all Qbs are stored. If a Qb is used, it will be Queue: p based deleted The on from storage Equation Queue. space (11). If ina node reader, is Cn, four where longer all query Qbs bits are Qb00, stored. Qb01, If aqb10, isand used, Qb11 it will be deleted Queue: be added from The to storage Queue. space Queue, If and ain node reader, reader is C n, will four where broadcast longer all query Qbs longer bits are stored. Qb00, Qbs Qb01, If a Qb to Qb10, is used, unread andit tags Qb11 will be in will be deleted subsequent added from to process. Queue, Queue. and If a node reader is Cn, will four broadcast longer query bits longer Qb00, Qbs Qb01, to Qb10, unread and Qb11 tags will in be added to Queue, and reader will broadcast longer Qbs to unread tags in subsequent Lookup table: process. In process of identification, a lookup table is used to store all assignment subsequent process. Lookup bits pqb. table: In process of identification, lookup table is used to store all assignment Lookup table: In process of identification, a lookup table is used to store all assignment bits p Qb. bits pqb. Initially Qb=null,Pinitial Initially Reader broadcasts query Qb=null,Pinitial package{qb,pinitial} Reader broadcasts query Tags select a time slot between 1~2 Pinitial package{qb,pinitial} According to assignment bits Pinitial Tags select a time slot between 1~2 Pinitial According to assignment Cn? bits Pinitial YES Cn? Add xx(00,01,10 and 11) to Qb respectively, Reader calculates P YES Qb and records m. Add xx(00,01,10 and 11) to Qb respectively, Reader broadcasts calculates query P Qb and package{qb, records m. P Qb } to all unread tags. Delete Qb from queue. Reader broadcasts query package{qb, P Qb } to all unread tags. Delete Qb YES from queue. Cn? YES Cn? Is queue empty? YES Is queue empty? END YES Figure 4. The reader s END flowchart of 4QTAP. The status of each tag can Figure Figure be divided The The into reader s four flowchart flowchart types: of of 4QTAP. 4QTAP. The status of each tag can be divided into four types:

7 Waiting status: Tags wait to receive reader s query package {Qb, pqb}. Activated status: If prefix of a tag matches query bits (Qb), tag is activated, and it select an assigned time slot to send its ID to reader. Success status: There is only one tag responding in a time slot, tag is identified successfully. Information 2017, Collision 8, 5 status: More than one tag responds in a time slot, a collision occurs, and all collided 7 of 10 tags in this time slot are not identified by reader. Waiting status Receive query package Information 2017, 8, 5 {Qb,p} from reader 7 of 10 Waiting status: Tags wait to receive Tag compares reader s its query prefix package {Qb, pqb}. to Qb Activated status: If prefix of a tag matches query bits (Qb), tag is activated, and it select an assigned time slot to send its ID to reader. Matched? Success status: There is only one tag responding in a time slot, tag is identified successfully. Collision status: More than one tag responds in YESa time slot, a collision occurs, and all collided tags in this time slot are not identified by Activated reader. status Tag selects Waiting a time status slot according to assignment bits p, and sends Receive its ID to query reader package {Qb,p} from reader YES Collision status? Tag compares its prefix to Qb Success status Matched? ENDYES Activated 5. status Figure 5. A tag s flowchartof of 4QTAP. 4QTAP. Tag selects a time slot according 3.4. Example Analysis to assignment bits p, and sends The status of each tag can be divided intoits four ID to types: reader We suppose that re are 10 tags A J. A: , B: , C: , Waiting D: , status: Tags E: , wait to receive F: YES , reader s G: query , package H: {Qb, , p Qb }. I: , Collision status? Activated J: status: We can If get prefix tag ofidentification a tag matches process query as shown bits in (Qb), Figure 6. tag is activated, and it select an assigned time slot to send its ID to reader. Success status Success status: There is only one tag responding in a time slot, tag is identified successfully. Collision status: More than one tag responds inend a time slot, a collision occurs, and all collided tags in this time slot are not identified by reader. Figure 5. A tag s flowchart of 4QTAP Example Analysis 3.4. Example Analysis We suppose that re are 10 tags A J. A: , B: , C: , D: , We suppose that re are 10 tags A J. A: , B: , C: , E: , F: , G: , H: , I: , J: We can get D: , E: , F: , G: , H: , I: , tagj: identification We process can get as shown tag Figure identification 6. inidentification Figureprocess 6. process as shown of 4QTAP. in Figure Simulation Results and Discussion We evaluate performances of 4QTAP and or former ALOHA-based algorithms by matlab simulation. The total time to identify all tags is equal to total number of time slots multiplied by slot time, which is an important evaluation factor for RFID anti-collision algorithms. Since slot time is constant, we only take total number of time slots into consideration. The smaller total number of time slots, better performance of algorithm. The system efficiency (throughput) is defined as ratio of slots filled with one tag to number of all time slots, which is anor Figure 6. Identification process of 4QTAP. Figure 6. Identification process of 4QTAP. 4. Simulation Results and Discussion 4. Simulation Results and Discussion We evaluate performances of 4QTAP and or former ALOHA-based algorithms by matlab We simulation. evaluatethe total performances time to identify of 4QTAP all tags and is equal or to former total ALOHA-based number of time slots algorithms multiplied by matlab by simulation. slot The time, total which time is an toimportant identify evaluation all tags is factor equalfor to RFID total anti-collision number algorithms. of time slots Since multiplied slot by slot time time, is constant, which iswe anonly important take evaluation total number factor of time forslots RFIDinto anti-collision consideration. algorithms. The smaller Since total slot time is number constant, of time weslots, only take better total performance number of of time algorithm. slots intothe consideration. system efficiency The (throughput) smaller total is defined as ratio of slots filled with one tag to number of all time slots, which is anor

8 Information 2017, 8, 5 8 of 10 number of time slots, better performance of algorithm. The system efficiency (throughput) is defined as ratio of slots filled with one tag to number of all time slots, which is anor evaluation factor that should be taken into consideration in our simulations. When durations of collision slot, idle slot, and successful slot are same (t 0 = t 1 = t k ), system efficiency is same as channel Information usage 2017, 8, efficiency 5 presented in paper [18], so system efficiency in our paper is8 aof special 10 case of channel usage efficiency. We evaluation compare factor 4QTAP that should with be taken Schoute into DFSA consideration algorithm in our [9], simulations. EDFSA [10], When Query durations Tree Split of (QTS) ALOHA collision protocol slot, idle [19], slot, and and Q successful algorithm slot [20]. are The same (t0 basic = t1 ideas = tk), of se system algorithms efficiency is are described same as channel usage efficiency presented in paper [18], so system efficiency in our paper is a as follows: special case of channel usage efficiency. We compare 4QTAP with Schoute DFSA algorithm [9], EDFSA [10], Query Tree Split (QTS) (1) EDFSA: The algorithm estimates number of unidentified tags first, n compares with ALOHA protocol [19], and Q algorithm [20]. The basic ideas of se algorithms are described as given maximum frame size. If number of tags is much larger than one that gives follows: optimal system efficiency, it divides unread tags into some groups and allows only one group (1) to respond. EDFSA: The algorithm estimates number of unidentified tags first, n compares with given maximum frame size. If number of tags is much larger than one that gives (2) QTS ALOHA: The length of a frame is chosen in set (8, 16, 32, 64, 128, 256). If size of optimal system efficiency, it divides unread tags into some groups and allows only one next identification frame is selected, total number of time slots will equal to all frames group to respond. (2) multiplied QTS ALOHA: by ir The size. length of a frame is chosen in set (8, 16, 32, 64, 128, 256). If size of (3) Schoute next DFSA: identification The DFSA frame algorithm is selected, was total presented number by of Schoute time slots inwill 1983, equal andto all frames number of collided multiplied tags isby equal ir to size times number of collided time slots. (4) (3) Q algorithm: Schoute DFSA: In this The algorithm, DFSA algorithm value was of presented Q is updated by Schoute slot by in 1983, slot according and number to of status of collided preceding tags is received equal to 2.39 slot, times which can number determine of collided time frame slots. size that can maximize tag (4) Q algorithm: In this algorithm, value of Q is updated slot by slot according to status of identification efficiency. preceding received slot, which can determine frame size that can maximize tag identification efficiency. Figure 7a plots number of total time slots, success time slots, idle time slots, and collision time slotsfigure in 4QTAP 7a plots when number number of total of tags time increases slots, success fromtime 5 toslots, 100. idle Figure time 7bslots, furr and describes collision relationship time slots between in 4QTAP when number number of time of tags slots increases and from number 5 to 100. of Figure tags 7b infurr range describes 100 to Firstly, relationship number between of total number time slots of time isslots equal and to number sumof oftags success in range time 100 slots, to idlefirstly, time slots, number of total time slots is equal to sum of success time slots, idle time slots, and collision and collision time slots. Secondly, as number of tags increases, total number of time slots may time slots. Secondly, as number of tags increases, total number of time slots may be invariant, be invariant, because when re is a small difference in number of tags, corresponding frames because when re is a small difference in number of tags, corresponding frames may have may have same number of time slots. same number of time slots. Figure Figure 7. The 7. The number number of total of total timeslots, timeslots, success timeslots, idle idletimeslots, and and collision collision timeslots timeslots in in 4QTAP vs. number of tags. (a) The number of tags increases from 5 to 100; (b) The number of tags 4QTAP vs. number of tags. (a) The number of tags increases from 5 to 100; (b) The number of tags increases from 100 to increases from 100 to Figure 8 shows efficiency of tag identification for different algorithms. When number of Figure tags increases 8 shows from 5 efficiency to 100, efficiency of tag identification of Schoute DFSA for different and Q algorithms. rapidly When become stable, number of tags increases but efficiency from 5 of to EDFSA 100, and efficiency QTS ALOHA of Schoute slowly DFSA increase. andwhen Q algorithms number rapidly of tags become is beyond stable, 100, we can observe that efficiency of each of 4QTAP, Schoute DFSA, QTS ALOHA, and Q algorithm becomes stable. For a large number of tags, efficiency of 4QTAP varies around 40%, and that of or four algorithms is only about 30%. In general, 4QTAP has better performance

9 Information 2017, 8, 5 9 of 10 but efficiency of EDFSA and QTS ALOHA slowly increase. When number of tags is beyond 100, we can observe that efficiency of each of 4QTAP, Schoute DFSA, QTS ALOHA, and Q algorithm becomes stable. For a large number of tags, efficiency of 4QTAP varies around 40%, and that of Information or2017, four8, algorithms 5 is only about 30%. In general, 4QTAP has better performance than or 9 of 10 ALOHA protocols. When all tag IDs are determinate, procedure of 4-ary query tree and tag than or ALOHA protocols. When all tag IDs are determinate, procedure of 4-ary query tree estimation method presented in this paper can be accurately predicted, so all tags will be traced in and tag estimation method presented in this paper can be accurately predicted, so all tags will be 4QTAP until each of m is identified. traced in 4QTAP until each of m is identified. Figure 8. Throughput of different algorithms vs. number of tags. (a) The number of tags increases from 5 to 100; (b) The number of tags increases from 100 to Conclusions This paper presents a novel RFID anti-collision algorithm. When an RFID system identifies multiple tags, tag collision will happen. A 4-ary query tree ALOHA protocol with optimal tag estimation method is is proposed, demonstrated, and discussed. and discussed. In order In torder breakto through break through constraint constraint of traditional of frame traditional slottedframe ALOHA slotted protocols, ALOHA we design protocols, an optimal we design tag estimation an optimal method. tag estimation The frame method. length changes The frame withlength number changes of tags, with and number each tagof selects tags, and assigned each tag time selects slot an to respond assigned totime slot reader to respond accordingto to reader assignment according bits p. to Simulation assignment results bits indicate p. Simulation that tag results estimation indicate method that is tag moreestimation efficient. ALOHA-based method is more algorithms efficient. havealoha-based tag starvation algorithms phenomenon. have We combined tag starvation a 4-ary query phenomenon. tree algorithm We combined with a dynamical 4-ary query frame tree algorithm slotted ALOHA. with Collision dynamical tagsframe in a frame slotted (node) ALOHA. will Collision be identified tags in in a frame branch(node) nodes. will Thebe simulation identified results in show branch that nodes. whole The simulation performances results of 4QTAP show that is better than whole performancesof of4qtap currentis ALOHA-based better than algorithms. performances of current ALOHA-based algorithms. Acknowledgments: This work was supported by National Natural Science Foundation of China ( ), Acknowledgment: Natural ScienceThis Foundation work was of Jiangxi supported Province by (20161BAB212051) National Natural Science and Foundation Key Research of and China Development ( ), Program Natural of Jiangxi Science Province Foundation (20161BBE50076). of Jiangxi Province (20161BAB212051) and Key Research and Development Author Program Contributions: of Jiangxi Province Zhihui (20161BBE50076) Fu and Xiang Wu undertake most of this work. Fangming Deng provides instructions and helps during design. All authors provide helps in revisions of this manuscript. All authors Author Contributions: Zhihui Fu and Xiang Wu undertake most of this work. Fangming Deng provides have read and approved final manuscript. instructions and helps during design. All authors provide helps in revisions of this manuscript. All authors Conflictshave of Interest: read and The approved authors declare final no manuscript. conflict of interest. Conflicts of Interest: The authors declare no conflict of interest. References 1. Dagan, H.; Shapira, A.; Teman, A.; Mordakhay, A.; Jameson, S.; Pikhay, E.; Dayan, V.; Roizin, Y.; Socher, E.; Fish, A. A Low-Power Low-Cost 24 GHz RFID Tag With a C-Flash Based Embedded Memory. IEEE J. Solid-State Circuits 2014, 49,

10 Information 2017, 8, 5 10 of 10 References 1. Dagan, H.; Shapira, A.; Teman, A.; Mordakhay, A.; Jameson, S.; Pikhay, E.; Dayan, V.; Roizin, Y.; Socher, E.; Fish, A. A Low-Power Low-Cost 24 GHz RFID Tag With a C-Flash Based Embedded Memory. IEEE J. Solid-State Circuits 2014, 49, [CrossRef] 2. Jang, S.; Kim, S.; Tentzeris, M.M. Low-cost flexible RFID tag for on-metal applications. In Proceedings of IEEE International Symposium on Antennas and Propagation & USNC/URSI National Radio Science Meeting, Memphis, TN, USA, 6 11 July 2014; pp Wang, J.; Li, H.; Yu, F. Design of Secure and Low-cost RFID Tag Baseband. In Proceedings of International Conference on Wireless Communications, Networking and Mobile Computing, Shanghai, China, September 2007; pp Preradovic, S.; Karmakar, N.C. Chipless RFID: Bar code of future. IEEE Microw. Mag. 2010, 11, [CrossRef] 5. Want, R. An Introduction to RFID Technology. IEEE Pervasive Comput. 2006, 5, [CrossRef] 6. Deng, F.; He, Y.; Li, B.; Zuo, L.; Wu, X.; Fu, Z. A CMOS pressure sensor tag chip for passive wireless applications. Sensors 2015, 15, [CrossRef] [PubMed] 7. Liu, L.; Lai, S. ALOHA-Based Anti-Collision Algorithms Used in RFID System. In Proceedings of International Conference on Wireless Communications, Networking and Mobile Computing, WiCOM 2006, Wuhan, China, September 2006; pp Cheng, T.; Jin, L. Analysis and Simulation of RFID Anti-collision Algorithms. In Proceedings of 9th International Conference on Advanced Communication Technology, Gangwon-Do, Korea, February 2007; pp Schoute, F.C. Dynamic frame length ALOHA. IEEE Trans. Commun. 1983, 31, [CrossRef] 10. Lee, S.R.; Joo, S.D.; Lee, C.W. An enhanced dynamic framed slotted ALOHA algorithm for RFID tag identification. In Proceedings of International Conference on Mobile and Ubiquitous Systems: Networking and Services, MOBIQUITOUS 2005, San Diego, CA, USA, July 2005; pp Chen, W.T. An Accurate Tag Estimate Method for Improving Performance of an RFID Anticollision Algorithm Based on Dynamic Frame Length ALOHA. IEEE Trans. Autom. Sci. Eng. 2009, 6, [CrossRef] 12. He, Y.; Wang, X. An ALOHA-based improved anti-collision algorithm for RFID systems. IEEE Wirel. Commun. 2013, 20, Lin, C.F.; Lin, Y.S. Efficient Estimation and Collision-Group-Based Anticollision Algorithms for Dynamic Frame-Slotted ALOHA in RFID Networks. IEEE Trans. Autom. Sci. Eng. 2010, 7, [CrossRef] 14. Wu, H.; Zeng, Y.; Feng, J.; Gu, Y. Binary Tree Slotted ALOHA for Passive RFID Tag Anticollision. IEEE Trans. Parallel Distrib. Syst. 2013, 24, [CrossRef] 15. Myung, J.; Lee, W.; Srivastava, J. Adaptive Binary Splitting for Efficient RFID Tag Anti-Collision. IEEE Commun. Lett. 2006, 10, [CrossRef] 16. Yang, C.N.; Hu, L.J.; Lai, J.B. Query Tree Algorithm for RFID Tag with Binary-Coded Decimal EPC. IEEE Commun. Lett. 2012, 16, [CrossRef] 17. Kim, Y.; Kim, S.; Lee, S.; Ahn, K. Improved 4-ary Query Tree Algorithm for Anti-Collision in RFID System. In Proceedings of 2009 International Conference on Advanced Information Networking and Applications, Bradford, UK, May 2009; pp Shakiba, M.; Singh, M.J.; Sundararajan, E.; Zavvari, A.; Islam, M.T. Extending birthday paradox ory to estimate number of tags in RFID systems. PLoS ONE 2014, 9, e [CrossRef] [PubMed] 19. Yan, X.; Yin, Z.; Xiong, Y. QTS ALOHA: A Hybrid Collision Resolution Protocol for Dense RFID Networks. In Proceedings of 2008 IEEE International Conference on E-Business Engineering, Xi an, China, October 2008; pp EPCglobal Standard Specification. EPC Radio-Frequency Identification Protocols Class-1 Generation-2 UHF RFID Protocol for Communications at 860 MHz 960 MHz Ver ; EPCglobal Inc.: Lawrenceville, NJ, USA, 2005; pp by authors; licensee MDPI, Basel, Switzerland. This article is an open access article distributed under terms and conditions of Creative Commons Attribution (CC-BY) license (