16. multiple rfid tags access algorithm

Multiple RFID Tags Access AlgorithmWeilian Su, Senior Member, IEEE, Nikolaos V. Alchazidis, and Tri T. Ha, Fellow, IEEE

Abstract—One of the main problems that affect the data integrity of passive RFID systems is the collision between the tags. A popular

anticollision algorithm which dominates the standards in HF and UHF passive RFID systems is Framed Slotted Aloha (FSA) and some

variations of FSA. Throughput and average time delay of the RFID system which determines the performance/efficiency of the system

are reduced rapidly when the number of tags inside the interrogation zone is increased. Using larger frame sizes is not always the

solution. This paper discusses and compares the existing protocols, and proposes a variation of FSA, called Progressing Scanning

(PS) algorithm. The PS algorithm divides the tags in the interrogation zone into smaller groups and gives the reader the ability to

communicate with each of them. For performance analysis, the PS algorithm was evaluated with the parameters of a typical passive

RFID system at 2.5 GHz. The results showed that the PS algorithm can improve the efficiency of the RFID system and provide a

reliable solution for cases with a high density of tags in the area (over 800 tags).

Index Terms—Passive RFID systems, tags, framed slotted aloha, collisions, data integrity, progressing scanning algorithm.

Ç

1 INTRODUCTION

CURRENTLY, a revolution is occurring in Radio FrequencyIdentification (RFID) technology, and many companies

create new implementations of RFID systems and newproducts related to this technology daily. The main advan-tage of RFID technology is the automated identification anddata capture that promises wholesale changes across a broadspectrum of business activities and aims to reduce the cost ofthe already used systems such as bar codes. For this reason,although RFID technology was discovered many years ago, ithas advanced and evolved only during the last decade sincecost has been the main limitation in all implementations.

The main advantages of RFID systems compared to bar

codes are the following:

. In RFID applications intended to replace bar codes,contact with the item to be identified is notnecessary, and even the line-of-sight (LOS) is oftennot necessary. Thus, it is no longer necessary to openshipping boxes and scan their contents.

. RFID systems work over long distances.

. RFID provides full automation of the supply chainand can reduce the cost of the vendor using it.

. It can be implemented in different environmentalconditions, such as in rain or with dust and dirt andstill operate extremely well.

. Also, while data stored in bar codes are fixed andcannot be changed, in most RFID systems, this ispossible by changing the data inside their electronicmemory.

. RFID systems are capable of multiple simultaneousscans of items which reduce the time needed tocollect the data.

. RFID systems can be used to track people andanimals in real time, while this cannot be done withbar codes.

. A bar code is the same for all similar items, whilewith RFID technology, the same items can havedifferent data, such as a different expiration date.

A disadvantage of RFID technology is that the manu-facturing cost of the main components is still not cheaperthan simple bar codes. Therefore, bar codes will coexistwith RFID systems in some applications. Due to the relativehigh data rates and the long tracking distance, RFIDsystems are being examined to ascertain whether theycould be employed for tracking people and supplies inmilitary operations. The most important issues to be solvedare the integrity of data collection and the security of datatransferred in the RFID system. In general, RFID technologyhas vulnerabilities in securing the data between the maincomponents of the RFID system.

This paper proposes to investigate RFID technology andevaluate the performance/effectiveness of the RFID systemsin collecting data. It intends to discuss and evaluate theperformance of the different protocols used today forcommunication between the main components of the RFIDsystem: the tags and the reader. It focuses on RFID systemswhich work in the microwave frequency band of 2.45 GHz,without the use of a battery supply for the tags. The goal ofthis research is to discover ways to increase the performanceof data collection for such systems under the constraints oftime delay, throughput, and finally, the working distance.

The paper is organized into the following sections:Section 2 presents the existing anticollision protocols usedin RFID systems, and it evaluates the performance of eachone and compares them in terms of throughput and timedelay. Section 3 presents the proposed hybrid anticollisionprotocol, and Section 4 discusses the performance of theproposed protocol. Finally, we conclude in Section 5.

2 DATA INTEGRITY AND ANTICOLLISIONS

PROTOCOLS

This paper focuses on passive RFID systems using back-scattering modulation in microwave (2.4 GHz) and

174 IEEE TRANSACTIONS ON MOBILE COMPUTING, VOL. 9, NO. 2, FEBRUARY 2010

. The authors are with the Department of Electrical and ComputerEngineering, Naval Postgraduate School, 833 Dyer Road, Rm. 452,Spanagel Bldg. 232, Monterey, CA 93943-5121.E-mail: {weilian, nalchazi, ha}@nps.edu.

Manuscript received 24 Mar. 2008; revised 28 Oct. 2008; accepted 6 May2009; published online 29 May 2009.For information on obtaining reprints of this article, please send e-mail to:[email protected], and reference IEEECS Log Number TMC-2008-03-0108.Digital Object Identifier no. 10.1109/TMC.2009.106.

1536-1233/10/$26.00 � 2010 IEEE Published by the IEEE CS, CASS, ComSoc, IES, & SPS

ultrahigh-frequency (UHF) bands, where the transmissionrate is very high. Therefore, the following discussion refersto those RFID systems.

The main disadvantage in RFID communications is themultiaccess that occurs in uplink, where multiple tagsrespond to the reader’s signal command. Multiple re-sponses at the same time on the Radio Frequency (RF)communication channel means that the reader cannotidentify the data transmitted from the tag. This event iscalled tag collision and is responsible for the low tagidentification efficiency [1] in RFID systems.

Data integrity in RFID systems also depends on thefollowing parameters:

. Power received from the tag. This power should beefficient to energize the circuit of the tag and alsotransmit the data to the channel.

. Signal-to-Noise Ratio (SNR) in the reader. Thereceived signal in the reader should be high enoughso the reader will be able to verify the data sequence.Some kinds of Error Detection (usually Cyclic Re-dundancy Check (CRC)) need to be used in the reader.

In the following discussion, it is assumed that the aboveparameters are satisfied in the development of an efficientanticollision algorithm.

Space Division Multiple Access (SDMA) and FrequencyDivision Multiple Access (FDMA) are not generally used inRFID systems because of the high cost to implement them.Both of the above anticollision techniques are limited to “afew specialized applications” and therefore, are not suitablefor passive RFID systems, which need to have a lowimplementation cost and complexity.

Time Division Multiple Access (TDMA) is an excellentchoice for RFID systems with a small number of tags. If thenumber of tags is large and not known during the processof identification, the Packet Radio (PR) technique is by farthe best anticollision technique used in RFID systems. Theadvantage of PR technique is that it allows the reader toidentify a large number of tags with a small amount ofoverhead. Also, it is not a requirement for the reader toknow the number of tags in the RFID system.

Two main procedures of PR are used: asynchronous andsynchronous procedures [1]. In the asynchronous proce-dures, the reader does not control the tags inside theinterrogation zone. Asynchronous procedures are alsocalled transponder drive; the most important of this kindof PR is the PURE ALOHA procedure which is used inpassive RFID systems for its simplicity. On the other hand,in the synchronous procedures, which are reader driven,the reader identifies all the tags inside the interrogator areaby a unique serial number assigned to each tag. Reader-driven procedures are divided into polling and binarysearch procedures. Most of the standards for the UHF RFIDsystems propose Aloha-based anticollision algorithms thatare probabilistic and binary tree search anticollision algo-rithms that are deterministic [3].

In passive RFID systems, tags are powerless and statelessdevices that cannot sense the channel. They do not knowabout the existence of other tags in the neighboring area andcannot detect when a collision occurs. Thus, the reader isresponsible for implementing an anticollision algorithm

and controlling collisions. This algorithm must be simplewithout numerous computations in the transponders. Inorder to increase the low efficiency of the PURE ALOHAalgorithm, a variation of it is used which is called SlottedAloha. In a Slotted Aloha algorithm, the reader controls andsynchronizes the tags before they respond.

This section focuses on the description of those algo-rithms and an evaluation of their performance in terms ofthroughput efficiency and time delay, which is the timeneeded by the reader to read all the tags in the area.

2.1 ALOHA-Based Algorithms

2.1.1 Basic ALOHA Procedures

The Aloha procedure or algorithm is a probabilisticprocedure, which can be used in the multiaccess uplinkcommunication from the tag to the reader to avoid collision.It is very simple to implement, and thus currently, it is verycommon in passive RFID systems with read-only tags.

In RFID systems using Aloha anticollision algorithms,the time each tag uses for the transmission of data is a smallfraction of the repetition time and long pauses betweentransmissions from the same tag occurs. Thus, the commu-nication between the reader and the tags is not continuous.In addition, each tag occupies, in general, a different periodof time to transmit the data [1], which depends on theamount of data to be transmitted.

As a result of the simplicity of the Aloha algorithm, thereis a high possibility of collision between tags because tagscan transmit their data to the reader randomly at any time.This possibility increases while the offered load G isincreased. The average utilization or throughput of thechannel is given by (1).

S ¼ G � eð�2GÞ: ð1Þ

The Aloha algorithm has a maximum of 18.4 percentutilization at G ¼ 0:5, and for this reason, the Alohaalgorithm has been modified to improve efficiency up to36.8 percent. This modification is called Slotted Aloha. Theaverage utilization in Slotted Aloha is given by (2) and has amaximum value of G ¼ 1.

S ¼ G � eð�GÞ: ð2Þ

In Slotted Aloha, the time of the channel is divided intouniform slots with size equal to the transmission time.Currently, tags transmit the data packets only at thebeginning of each slot [1], [2]. Consequently, synchronizationis necessary in the Slotted Aloha algorithm. The necessarysynchronization is provided by the reader, and therefore,Slotted Aloha is a reader-driven TDMA procedure [1].

According to (2), it can be shown that as the offered load(G) increases, throughput (S) increases until the maximumvalue of 36.8 percent and then falls rapidly for values of Ggreater than 1. This is a main disadvantage of the SlottedAloha algorithm because the system is unstable, and withlow efficiency.

A variation of Slotted Aloha is used in RFID systems andhas been proposed by the International Organization forStandardization (ISO) and the Electronic Product Code(EPC). This variation of Slotted Aloha in RFID systems iscalled Framed Slotted Aloha (FSA). In the following section,the FSA and dynamic FSA are reviewed and examined.

SU ET AL.: MULTIPLE RFID TAGS ACCESS ALGORITHM 175

2.1.2 Framed Slotted Aloha (FSA) Algorithm

The FSA algorithm is a Slotted Aloha in which the availabletimeslots where the tags can respond to the readercommands are organized into time frames. Each frame isdivided into a number of slots (usually powers of 2) andeach timeslot is long enough for the tags to transmit theirdata. Those time frames have duration equal to the timebetween two REQUEST commands of the reader [1], [3].Thus, the efficiency of the FSA remains the same as inSlotted Aloha.

A modification of FSA is the Dynamic Framed SlottedAloha (DFSA). The DFSA algorithm dynamically changesthe frame size to increase tag identification, and thus,increases the efficiency in collecting the data from the tags.

2.1.3 Dynamic Framed Slotted Aloha (DFSA)

The DFSA algorithm was first introduced by Schoute [5] formultiuser channel environments and proven to increase theupper bound of the FSA algorithm to 42.6 percent; it alsoincreases the stability of the multiuser channel. In [6], [7],DFSA was introduced for passive RFID systems where thenumber of tags is unknown.

In specific applications where the number of tags isknown, constant, and not too big, FSA can be used;otherwise, DFSA is the solution. In DFSA, the reader hasthe flexibility to vary the frame size. Hence, it varies thenumber of available slots for the tags. If the reader does notdetect tags, which means that collisions occur, it increasesthe frame size until an efficient number of tags can bedetected. As long as tags are detected, it decreases the framesize and so on [3].

The reader tries to identify all the tags in the interrogationzone in multiple read cycles. The amount of time in one readcycle is equal to the time elapsed between two REQUESTcommands sent by the reader. The subject of this research isto investigate how the reader can detect and read themaximum number of tags with a minimum number of readcycles and a maximum probability of detection.

Much research has been done about the criterion whichmust be used in order to change the frame size. In passiveRFID systems, the reader waits for the tags to respond andchanges the frame size after each read cycle according to thenumber of tags in the interrogation zone. Thus, the criterionin DFSA is that the reader needs to estimate the number oftags in the previous read cycle and then adjusts the framesize accordingly. In the following section, the systemefficiency of DFSA and FSA is studied.

2.2 System Efficiency in RFID Systems with DFSAand FSA Algorithms

In order to measure the efficiency of DFSA in a passiveRFID system, the following assumptions must be made todecrease the complexity of the problem in the next sections:

. Tags that have been read once from the reader in aprevious read cycle, after activation, will not sendtheir data again (Identification Number) if theyreenter the reader’s field. This is like flipping an“inventoried flag” from A! B or B! A during aninventory round according to EPCGlobal Gen 2 UHFRFID [4], so the inventoried RFID does not reply

again. If this assumption is not valid, as it is for tagsthat need to be read more than once, the efficiency ofthe DFSA algorithm is less than the followingcalculations. It will be assumed that tags after readingwill not respond to future reader requests [3].

. In the estimation of the number of tags in thereader’s field, the Capture Effect, is assumed to benegligible. This effect helps tags near the reader totransmit their data although collision had occurredin the timeslot they used. This happens because theirsignal is stronger than the farthest tags due tochannel attenuation. Capture effect increases thethroughput of the DFSA algorithm and thus in-creases the overall efficiency of the RFID system [1].

. The communication channel, both uplink and down-link, is assumed to be noise free. Increasing noisedecreases the ability of the reader to read the datafrom the tags, and thus decreases the performance ofthe system.

In DFSA, the reader estimates the total number of tags inthe interrogation zone by using the received informationfrom the slots in each frame as a feedback control. Thus, acontroller is necessary. This is not a problem because allmodern readers have an onboard controller.

2.2.1 Estimation of Frame Size and Number of Tags in

RFID Area Using the DFSA Algorithm

In the description of the DFSA algorithm that follows, theapproach and methodology of [3] is used. First, theperformance of FSA is discussed and then it is applied toDFSA since DFSA just enhances FSA by dynamicallychanging the frame size. Let N be the random variablerepresenting the number of slots that have been used by thereader at the previous read cycle. Also, let R and C be therandom variables representing the number of slots of theframe, which are selected by only one tag and by more thanone tag, respectively. Finally, n denotes the number of tagsin the interrogation zone, which is also a random variable.

Slots that have been selected by only one tag are the slotsthat the reader can read the data from the tag and identify it.In slots that have been selected by more than one tag,collision has occurred and the reader cannot identify the tags.The probability p of a tag selecting a specific slot is given by

p ¼ 1

N; ð3Þ

and the probability for a tag to transmit its data (Pread) inthis slot is given by

Pread ¼1

N

� �� 1� 1

N

� �n�1

; ð4Þ

because the selection of the slot from the tags is random,and every tag selects a slot independent of the rest of then� 1 tags with the same probability.

When the reader uses a frame size of N slots, theprobability that k out of n tags to select a specific slot andnot the others is binomially distributed [5]. The number ofthe tags k that occupy a specific frame slot is known as theoccupancy number of the slots [3].

The time delay T which is necessary for each tag totransmit its Identification (ID) successfully is T ¼ N � i,


where i is the number of read cycles or number ofretransmissions. Let Pempty be the probability that a slot isempty, and Pc is the probability that a collision has occurredin a slot. Then, Pempty ¼ ð1� 1

NÞn and Pc ¼ 1� Pempty � Pread.

The optimal frame size Noptimal for the next read cycle canbe calculated in two ways [3], as given below:

1. Maximizing the throughput S of the RFID system,which is defined as

S ¼ Probability of Reading a TagPtotal

¼ PreadPc þ Pread þ Pempty

¼ Preadð1� Pempty � PreadÞ þ Pread þ Pempty

¼ Pread

therefore S ¼ 1

N

� �� 1� 1

N

� �n�1

:

ð5Þ

2. Minimizing the time delay T which is given in [3] as

T ¼ N

ð1� 1NÞ

n�1: ð6Þ

Fig. 1 shows the behavior of the RFID system if the FSAalgorithm and (5) are used. It is obvious that when the numberof tags is greater than the selected frame size, the system’sefficiency decreases and the RFID system is unstable.

Both of the above methods provide the same Noptimal

solution [3], which is n. Therefore, p ¼ 1=n. Thus, in DFSA,if the reader allocates a frame size equal to the number oftags in the interrogation area, the efficiency of the RFIDsystem increases. The only problem now is that the readerneeds to estimate this value n before it transmits the nextREQUEST command.

Fig. 2 illustrates the stability in throughput efficiencyprovided that the DFSA algorithm is used; so, Noptimal ¼ n isselected at the reader at every duty cycle. The figure showsthat throughput is always equal to the maximum theoreticalvalue of 36.8 percent of the Slotted Aloha protocolregardless of n. Moreover, there is a spike at the beginning

of both subplots because when there is only one tag in thearea, the probability for a tag to collide in practice is equalto zero and the reader always read this tag and so S ¼ 1.After some tags are added, throughput falls very quickly toits final value.

Fig. 3 illustrates the time delay (T ) in both FSA fordifferent frame sizes and DFSA for n ¼ 0� 512 tags. InDFSA, it is assumed that the reader’s initial frame size isequal to the optimal, which in practice is not always true.Fig. 3 shows that the DFSA algorithm increases theperformance of the RFID system in terms of the time delay.T is greater if the FSA algorithm is used. The performanceof the FSA algorithm decreases as the difference between nand N increases. The FSA with N ¼ 512 and DFSA seem tohave the same behavior; this is only true when the numberof tags is near 512 (actually for n � 350 tags). For a smallnumber of tags (n � 220 tags), FSA with N ¼ 512 gives theworst performance.

2.2.2 Developments in DFSA Algorithm

In [8], Lee et al. proposed an alternative probabilistic Aloha-type anticollision protocol for RFID systems. It is calledEnhanced Dynamic Framed Slotted ALOHA (EDFSA), which issimilar to the DFSA algorithm with one difference. After theestimation of the number of tags:


Fig. 1. Throughput S of FSA for different frame sizes.Fig. 2. Throughput S of DFSA with optimal frame size selection.

Fig. 3. Performance of RFID system in terms of time delay.

. if the number of tags n is greater or equal to 177, itdivides the tags into smaller groups (called “Modulooperation”); the number of smaller groups isdnumber of unread tagsN e, and thus, a smaller frame size isused (always N ¼ 256 slots) or,

. if the number of tags is smaller than 177, then theDFSA algorithm is used.

Although the author’s simulations in [8] showed im-provement in system efficiency versus frame size, thismethod needs extended calculations in the reader andincreases the tag’s complexity and power consumption,something which is critical in passive RFID systems,because tags are powerless. For this reason, the next sectionproposes a new type of DFSA anticollision protocol basedon simplicity and the constraint of minimizing powerconsumption in the transponder (tag).

Moreover, in both DFSA and EDFSA algorithms, thereader can only calculate the number of tags that respondswithout collision which is represented by the randomvariable R. Also, the reader can calculate the number of slotswith collision (given by random variable C) but how itestimates the number of tags that collides is unknown. It isknown that for every slot that collision occurs at least two tagshave transmitted their data to it. Thus, the number of slotsNthat the reader must use in the next frame is given by [7]

N ¼ Rþ 2C: ð7Þ

3 A PROGRESSING SCANNING TECHNIQUE

This section presents an alternative hybrid anticollisionprotocol based on the FSA algorithm for microwave RFIDcommunication systems at 2.45 GHz. The parameters of thecommunication system will be according to those estab-lished in [9] by the ISO and the International Electrotechni-cal Commission (IEC) for MODE 1 systems, which are thepassive backscatter RFID systems of interest.

The proposed protocol takes into account the physicallink and Media Access Control (MAC) parameters. Themost important of these parameters are the following [9]:

. The maximum transmitted power (Pr;max) measuredat the reader’s antenna is 4 Watts (36 dBm) EIRP.

. The modulation, which is used in the forward link(reader to tag), is Amplitude Shift Keying (ASK).ASK is the simplest waveform and can be detectedeasily by the tag with a simple circuit.

. The modulation, which is used in the reverse link(tag to reader), is Backscatter Modulation (BM).

. The data coding is Manchester for the forward linkand FM0 (Bi-Phase Space) for the reverse link. FM0is actually a differential Manchester technique.

. Both the tag and the reader have error detectioncapability by using a CRC with 16 bits.

. The data bit rate should be between 30 and 40 Kbpsin both directions.

. Tags use the reader’s signal for synchronization.

. The maximum occupied channel Bandwidth (BW) is500 KHz.

. The memory size of the tags varies from 8 bytes to64 bytes.

Moreover, it will be assumed that the chips on the tags areusing the current semiconductor technology, which de-creases the power consumption in the tag, and a minimumpower received (Ptag;min) equals to 50 �watts is enough [1]for the tag to transmit the data stored in its memory.

In the following sections, a detail understanding of themaximum distance of the RFID system, the effects ofpassive backscattering modulation on the frame size, andthe benefits of capture effect on FSA type algorithms arepresented. These in-depth understanding allows us todevelop the proposed PS algorithm. For example, themaximum distance constraints how much power can betransmitted, and the frame size affects the tag collisionprobability, which dictates the number of tags within a scan.

3.1 Maximum Distance of the RFID System

First, it is important to evaluate the maximum distance (rmax)of a typical passive RFID system in the microwave band. Thedistance rmax refers to the maximum distance a tag can beplaced to receive the necessary power from the reader. Apart of this power supplies the inner circuit of the tag toperform all the necessary operations to wake up the tag, andanother part is used to transmit the data back to the reader.The maximum distance is smaller than the actual readingdistance due to the attenuation in the reverse link.

The Friis free space equation capturing the relationshipbetween the transmitted power from the reader (Pr) and thepower received by the antenna of the tag (Ptag) is as follows:

Ptag ¼ PrGrGtag�

4�r

� �2

; ð8Þ

where Gr is the transmitter antenna gain, Gtag is the receiver

antenna gain, � is the wavelength in meters, and r is the

distance between the transmitter and antenna; (8) assumes

that there is no loss from the transmitter and receiver

equipments. By using the Friis relationship for the free

space path loss, the one way (reader to tag) path loss aF isPrPtag

. Substituting Ptag with Ptag;min, (9) gives the maximum

allowable path loss that the RFID system can experience and

is still capable of transmitting the stored data to the reader.

aF ¼Pr

Ptag;min: ð9Þ

From (9), the maximum distance for a tag with a dipoleantenna and a reader with maximum EIRP of 4 Watts at theoutput of the reader’s antenna is calculated and plotted inFig. 4. Fig. 4 illustrates the reading distance versus thetransmitted power from a reader. The maximum distancermax is 3.5 m, and this result agrees with [9], which specifies4 m for a typical passive RFID system with data rates up to30 Kbps. Also, Finkenzeller [1] specifies 3 m as the lowerbound for backscattering RFID systems (UHF and Micro-wave bands).

The results in Fig. 4 are for LOS communication betweenreader and tags, which is almost a prerequisite formicrowave RFID systems. If the RFID system is inside abuilding, for a line-of-sight communication, the path lossexponent is less than 2. Usually, a value of 1.6-1.8 is used forthis case [10], and therefore, the maximum reading distance


can be higher than 3.5 m. To determine the maximumdistance in an indoor environment, it is first necessary tocompute the path loss aF at a reference distance ro from thereader. The selection of the reference distance ro is notarbitrary for the RFID system and is determined in thefollowing sections.

3.1.1 Evaluation of Reference Distance (ro)

The reference distance ro could be any distance in the far-field region of the reader’s antenna, which is much smallerthan the maximum distance of this specific system [10]. It isknown from antenna theory and design that the far-fielddistance or Rayleigh distance of the antenna is given by

rff ¼2D2

�; ð10Þ

where D is the maximum size of the antenna in meters and� is the wavelength of the operating frequency. Anydistance r that meets the following requirements lies inthe far-field region of the antenna.

. r > rff ,

. r� D, and finally

. r� �.

Table 1 gives the Rayleigh distance for different types ofreader antennas at 2.45 GHz, where � is 0.1224 m. For anantenna in the microwave band, the far-field region isdetermined by the wavelength. The selection of ro ¼ 0:5 m(more than four times �) satisfies the previous require-ments and is eligible to be selected as a reference distancefor the evaluation of the reference path loss from the Friisequation. The reference path loss at reference distance ro,aF ðroÞ, is 32 dB.

3.1.2 Maximum Distance for RFID Systems in Different

Environments

From the theory of large-scale path loss at distance r, aF isgiven in [10] by

aF ðdBÞ ¼ aF ðroÞ þ 10 � n � log10ðr=roÞ: ð11Þ

Using (9) and (11) and solving for distance r, the followingrelationship is found:

r ¼ ro � 10

�ð10�log PrPtag;min

Þ�aF ðroÞ

10�n

�; ð12Þ

where n is the path loss exponent.Fig. 5 shows the reading distance of two cases: 1) an

indoor environment with path loss exponent n ¼ 1:6, and

2) an urban environment with n ¼ 3. For applications, such

as scanning products in a store, n ¼ 1:6 is usually the case;

thus, a reading distance of up to 5.5 m can be achieved.

3.2 Selection of Backscattering Modulation

In passive RFID systems at the microwave band, the reader

transmits the ASK modulated carrier to the shared wireless

channel. This carrier provides the tag with enough power to

energize it and is also used by the tag as a carrier for

transmitting its ID in the reverse link by using the

backscatter modulation. The tag reflects the reader’s signal

by changing the impedance of its antenna according to the

bits that are transmitting, or in other words, it changes the

gain of the antenna [11], [12]. As a result, small fluctuations

occur in the amplitude of the carrier’s signal.When the signal returns, the reader needs to “peak-detect”

the modulation of the tag in the carrier and then decode it. A

high value in the envelope of the carrier is represented by a

binary one “1” and a low value of a binary zero “0.” If this is

the only change that occurs in the reader’s signal, this type of

backscatter modulation is called direct modulation and it is

simply an ASK modulation. In addition, the tag can also

change the phase or the frequency of the carrier signal, and

thus, create a PSK or FSK modulated signal.


Fig. 4. Reading distance of a typical passive RFID system.

TABLE 1Rayleigh Distance for Different Antenna Types at 2.45 GHz

Fig. 5. Maximum reading distance for different values of path lossexponent.

Fig. 6 illustrates a direct modulated backscattering

carrier signal from the tag to the reader. A carrier signal

is an ASK modulated sine wave with amplitude 100 V.

The tag creates a drop of 100 mV in the amplitude of the

carrier for each transmitted binary zero “0” [13]. The

reader peak-detects this signal, decodes it, and thus

identifies the tag.For the case of passive RFID tags in the microwave band,

this kind of ASK-backscattering modulation is selected

because of the simplicity either in detection from the reader

and in computations inside the tags, which, as a result,

reduces the cost of the tag. In addition, direct modulation

(ASK) provides higher data rates (up to 40 Kbps) than PSK

and FSK backscatter modulation [11], [13]. General passive

tags must have as simple functions as possible to reduce the

power consumption, and thus, to increase the maximum

working distance of the RFID system [14].Also, the messages transmitted from the tags must be as

short as possible for two reasons:

. Shorter messages mean less power consumption inthe tag.

. Shorter messages have lower probability of error intransmitting the tag’s ID as discussed in thefollowing section.

3.2.1 Probability of Error in Transmitting Tag’s ID with

ASK Backscattering

When ASK is the only modulation (backscattering) used in

the reverse link, then the Bit Error Probability (BER) Pb is

given by (for coherent detection):

Pb ¼ QffiffiffiffiffiffiEbNo

r� �; ð13Þ

where EbNo

is the average signal-to-noise ratio per bit andQðxÞis the Q-function. In the above expression, an Additive WhiteGaussian Noise (AWGN) channel with no fading is assumed.

The probability for a tag to transmit its ID withouthaving any bit in error is given by

PF ¼ ð1� PbÞL: ð14Þ

In the above equation, L is the length in bits of the framethat the tag is using to transmit its ID and is assumed to bethe same for all tags. Finally, the probability for a frame tobe in error is given as

Pe;F ¼ 1� PF ¼ 1� ð1� PbÞL: ð15Þ

Fig. 7 illustrates the Pb for different values of EbNo

. Notethat Pb becomes negligible for Eb

Nogreater than 7 dB (less than

1 percent error).Equation (15) shows that shorter tag messages (smaller

L) result in lower probability of frame error Pe;F , or in otherwords, a lower frame error rate. Fig. 8 plots (15) for threeframe lengths:

. L ¼ 64 bits, which is the minimum frame lengthfrom the ISO standards, and also is the length of theUnique Identification (UID) of the tag,

. L ¼ 144 bits, which is the recommended framelength from ISO standards, and,

. L ¼ 512 bits, which refers to the maximum framelength used in typical passive RFID systems.

It shows that for smaller tag messages, the frame error ratedecreases, and moreover, for a 7 dB signal-to-noise ratio, theprobability a frame is received in error is now significant(8� 10�1 for L ¼ 144 bits).

Since only a 16 bit CRC is used for error detection andforward error correction coding is not applicable, it is veryimportant to use short messages and for the signal receivedin the reader to be as strong as possible to reduceretransmissions and overhead in the reverse link.


Fig. 6. Backscatter amplitude modulation signal.

Fig. 7. BER with ASK-backscatter modulation.

Fig. 8. Probability of tag ID error for binary ASK backscattering (directModulation).

3.2.2 Selection of Tags Frame Length

The frame that the tag transmits to the reader (response)consists of the following fields [9]:

. Quiet. The tag does not transmit for a specific periodof time determined from the protocol.

. Return Preamble. This consists of 16 bits in a specificsequence, which enables the reader to lock the datafrom the tag and start decoding the message.

. The data field with at least 64 bits for the UID plusthe rest data bits to transmit other kinds ofinformation stored in the tag’s memory.

. A 16 bit CRC for error detection.

Since it is important as previously mentioned to keep thetag’s message as short as possible, but on the other hand, aserror detection capability and the return preamble arenecessary, a frame length of 96 bits is selected. However,the memory size of the tags can be larger (144 bits is therecommended standard), and thus it could be compatibleeven with the EPC (96 data bits). Fig. 9 shows the frame errorrate for L ¼ 96 bits. For a signal-to-noise ratio less than 11dB, the frame error rate is still significant (7� 10�1 for 7 dB).A signal-to-noise ratio greater than 11 dB is necessary toachieve low frame error probability (lower than 10�2).

3.3 Increasing the Efficiency of FSA with CaptureEffect

The tag’s responses in the reverse link can be identifiedfrom the reader even if they collide (occupy the same slot).This can happen if the strength of one signal is higher thanthe rest of the signals in the same slot. This is known as thecapture effect [1], [10]. The capture effect is used very oftenin most common cellular systems. For this paper, it ispossible to take advantage of the capture effect and increasethe throughput S of the FSA algorithm by choosing theappropriate threshold in the reader, which acts as a filter forthe weak signals. As a result, the reader identifies the tageven if collision has occurred in the specific slot.

The throughput S in the Slotted Aloha algorithm withthe capture effect is given [15] by

S ¼ G � e� TG1þTð Þ; ð16Þ

where T is the selected threshold at the reader, which iscalled capture ratio and corresponds to how many timesgreater the received power in reader should be from aspecific tag as compared to the summation of the receivedpowers from the remaining tags that occupy the same slotin order to be identified by the reader. The throughputreaches its maximum value of 1þT

eT when G is 1þ 1T . The

value T in (16) is in decimal form instead of dB.Fig. 10 shows that a higher throughput can be achieved

in the RFID system by “filtering” the weak signals with thethreshold in the same slot and keeping only the survivorsignal. It can be seen from this figure, that for a thresholdvalue of 3 dB, maximum throughput increases to 0.55, andfor 6 dB, it increases to 0.46.

So as the threshold T decreases, the performance of theRFID system increases, and as this takes place for highervalues of G, so does the stability of the FSA algorithm. Onecould say that using a low threshold in the reader wouldeliminate the collision problem, but this cannot happen inpractice, because T is determined by the reader’s sensitivity.

The typical reader’s sensitivity requires 6 dB differencebetween the signal from the tag of interest and the channelnoise (white noise plus interference from other tags) inorder to identify the tag [1]. Therefore, only an increase upto 46 percent in performance in terms of throughput canbe achieved.

Table 2 shows the maximum theoretical throughput ofthe FSA algorithm for different threshold values. It isobvious that theoretically, even for very high thresholdvalues, such as 20 dB (a cheap reader with very lowsensitivity), the performance is still better than the max-imum of the FSA algorithm without using filtering in thereceiver of the reader.

3.4 Progressing Scanning (PS) Algorithm

The proposed PS algorithm considers two constraints of theRFID system:

. the power consumption in tags must be minimum inorder for the RFID system to achieve the maximumdistance, and

. the RFID system must be simple, especially the tags.


Fig. 9. Probability of tag error for 96 bits frame length.Fig. 10. Throughput of FSA with the use of capture effect.

In the PS algorithm, the reader takes advantage of the rangedifference between the tags and the reader’s antenna. Tagsthat are near the reader receive more power from the readerthan those which are further away.

In the PS algorithm, the reader starts transmitting from aminimum EIR power level Pr;min until the maximum Pr;maxthat is permitted by the regulations. Tags that are furtherfrom the reader do not receive enough power and thuscannot transmit their IDs. In each retransmission, the readerincreases the transmitted power by an increment k and thetags that are further in distance reply. This continues untilthe transmitted power reaches Pr;max. Then, a new cyclebegins and the procedure is repeated from the beginning.

The PS algorithm divides the number of tags n in thearea into smaller groups as in the EDFSA algorithmintroduced in [8], and therefore, this method has thebenefits of EDFSA since the reader does not use largeframe sizes that reduce the efficiency of Aloha-basedalgorithms [8]. Also, it does not have the complexity ofEDFSA to set up the groups.

A detail description of the PS algorithm is as follows:

. At first, the reader transmits with Pr ¼ Pr;min. Thetags at distance rt � rmax;pi, where rmax;pi is themaximum distance from Fig. 4 that corresponds tothis transmitted power, become energized, and replyusing the FSA protocol.

. Next, the reader increases the power level by k, andthe aforementioned procedure repeats, but nowwith transmitting power Pr ¼ Pr;min þ k. All newtags that entered the interrogator zone of the readerreply. Tags from the previous scanning do not replyto the reader’s command. This can be accomplishedif the reader transmits a command in the header thatinforms the tags, which have already transmittedonce, not to reply until the next cycle. Of course, thetags need to have been programmed to do so. Thisprogramming can be done from the manufacture byusing a flash memory in tags for quick loading tocompare its state, or by using tags with Read/Writememory. For example, this is like flipping an“inventoried flag” from A! B or B! A during ainventory round according to EPCGlobal Gen 2UHF RFID [4], so the inventoried RFID does notreply again.

. This aforementioned procedure continues with Pr ¼Pr;min þ i � k (i ¼ 1; 2; 3; . . . ).

. Finally, in the last scanning, the reader transmitswith Pr ¼ Pr;max.

. This is the end of the first cycle of the PS algorithm,which consists of dPr;max�Pr;mink e transmissions. Afterthis point, a new cycle with multiple scans beginsand the whole procedure repeats until there are nomore tags in the interrogation zone.

. After each scan, the frame size changes according tothe frame size estimation algorithm if dynamicframe size is used. At the beginning of each cycle,the first scan should use the minimum frame sizeselected from a set of available frame sizes; this is toavoid using a large frame size estimated at previousscan, which is the last scan of the previous cycleusing the maximum transmit power thus covering alarge area.

The PS algorithm is an alternative and simplex method

to divide the tags in the interrogation zone into smaller

groups like in EDFSA, without any involvement from the

tags. As a result, PS algorithm decreases the complexity of

the tags. Thus, PS has all the benefits of EDFSA in the

performance of the Framed Slotted Aloha.Fig. 11 shows that the PS algorithm successfully divides

the tags in the interrogation zone into groups with a fewer

number of tags. In the simulation, 1,000 tags were randomly

generated and uniformly placed around the reader with

distances of 0-3.5 m. In addition, two different minimum

transmitted power values Pr;min were used. The reason that

more tags are in the first group for both cases is the result of

the reverse proportional relationship between Pr and r. For

example, for Pr;min ¼ 1 Watt, this corresponds to r ¼ 1:7 m,

which is almost half the maximum distance. By increasing

the transmitting power by a factor of 0.2 or 0.5 Watts, the

increase in parameter r is almost insignificant; thus, fewer

tags are included in the groups following the first one.As Fig. 11 illustrates, the number of tags in each group

decreases as the minimum transmitted power from the

reader and the increment k decreases. However, both smaller

Pr;min and k must, as a result, increase the times the reader

needs to scan to identify all the tags in the interrogator zone,

which thus, negatively affects performance in the PS

algorithm as will later be demonstrated. On the contrary, if


TABLE 2Maximum Theoretical Throughput for Different Thresholds

Fig. 11. Number of tags on each cycle (uniform distribution around thereader).

the number of tags is too high, small values ofPr;min and k areneeded to decrease the number of tags in each group.

If instead of a constant value for the increment k, avariable one kv is used, this will decrease the number oftimes the reader needs to transmit in one cycle. The onlyrequirement is that the increment should increase when thetransmitted power is increased, or in other words, smallervalues of kv should be used with the first scans where moretags are involved in the identification process and highervalues when the transmitted power reaches Pr;max.

Fig. 12 compares the performance of the PS algorithm thatuses constant power step increase (k ¼ 0:2) with a variableone (k ¼ f0; 0:3; 0:9; 1:5; 2:5; 3:1; 3:6g). The tags are uniformlydistributed around the reader. The solid line corresponds tothe constant step increase and divides the tags into19 groups, while the dashed line represents the variable step.

As seen in Fig. 12, an increasing variable step decreasesthe number of scans of the PS algorithm, which has asimilar effect if a higher constant k value was used.Generally, a variable step can be avoided because itincreases the complexity of the system. However, whenthe distribution of tags is not uniform, it might be useful.For example, if the tag’s distance from the reader follow aGaussian distribution, smaller increments can be used nearthe mean distance and larger ones for tags that are manystandard deviations greater.

4 PERFORMANCE EVALUATION

Performance comparison between FSA and progressingscanning algorithm is discussed in Section 4.1. Afterward, adetail analysis of the PS algorithm and comparison of the PSand DFSA algorithms with dynamic frame size estimationare given in Sections 4.2 and 4.3, respectively.

4.1 Comparison between FSA and ProgressingScanning Algorithm

In order to compare the performance of the PS algorithmwith FSA, it is necessary to simulate in discrete time usingMatlab as the frame transmitted from the reader and alsothe random selection of a slot from the tags. A fixed framesize is used both for the FSA and PS algorithms. Thedistance of the tags from the reader is assumed to follow auniform distribution with minimum at 0 m and maximum

at 3.54 m, which is the maximum distance for an outdoorline-of-sight RFID system at a frequency of 2.45 GHz.

The number of tags n in the interrogation zone is 1,000and the simulation is ran 1,000 times to increase theaccuracy of the results. The scope of the simulation is toevaluate the performance in terms of the time delay neededfor the identification of all the tags. The time delay T forFSA in terms of the number of slots is given by (6). Thus, thedelay in units of seconds is given by the following equation:

TFSA ¼N

1� 1N

� �n�1½slots� � L½bits=slot�

R½bits=sec�

� �ð17Þ

¼ N

1� 1N

� �n�1� L

R

� �½sec�; ð18Þ

where

. L is the length of the tag response in bits (L ¼ 96 bitsis selected).

. R is the bit rate from the ISO standards (a minimum30 Kbps is selected).

To calculate the delay from (18) for the PS algorithm, it isimportant to understand that the PS algorithm is just a FSAprocedure with the only difference being that the number oftags in the interrogation zone is divided into multiplegroups. The delay for each group which contains ni tags isthen given as

TPSi ¼N

ð1� 1NÞ

ni�1� L

R

� �½sec�; ð19Þ

where i indicates the group number, and ni is the number oftags in this group that the PS algorithm has created.

Thus, the total delay in seconds of the PS algorithm is thesummation of the individual delay for each group and isgiven by

TPS ¼Ximaxi¼0

N

ð1� 1NÞ

ni�1� L

R

� � !½sec�: ð20Þ

This average total delay does not include the time inwhich the reader needs to temporally inactivate the tagsafter identification. Such omission is also done for FSA.Since the time to inactivate the tags for both FSA and PSalgorithms is the same, comparing the performance of PS toFSA using (18) and (20) is fair.

Table 3 summarizes the simulation results of the PS andFSA algorithms for Pr;min of 0.2 Watt and step size k of0.2 Watt; thus, imax is 18 for this case. The third column of thistable shows how many times (power levels) the PS algorithmneeds to transmit to complete one cycle, while the fourthcolumn shows how many tags were identified in the firstcycle. Finally, the last column gives the average time delaycalculated from (18) and (20) for each case, but except for FSAwithN ¼ 64 andN ¼ 128 slots, which is from the simulation.

The results indicate that when a small frame size is used(N ¼ 64 or N ¼ 128 slots), the performance of the PSalgorithm is much better than that of FSA, which is unableto identify such a large number of tags due to collision.Moreover, the time delay introduced by the PS algorithm is


Fig. 12. Performance comparison of PS algorithm using constant powerstep with variable one.

reasonable, which is between 7 and 10 seconds for1,000 tags. When a larger frame size is used (N ¼ 256slots), the PS algorithm is able to read more tags in the firstcycle than can FSA with a greater frame size, but moretransmissions occur than in FSA. In addition, the averagetime delay is much lower in the PS algorithm than in FSA(56 percent lower), and thus, the PS algorithm rapidlyincreases the performance of Slotted Aloha in terms ofidentification time. It is important to notice from Table 3that FSA is ineffective in identifying the entire number oftags in the area when the frame size is small. On the otherhand, the PS method with the same frame size requires avery low average identification time.

In Fig. 13, the average time delay T is plotted for both theFSA and PS algorithms. It is evaluated for tags ranging from100 to 4,000. For a better presentation of the results, a linearscale was used for the horizontal axis (T ), while alogarithmic scale was used for the vertical axis (number oftags). According to Fig. 13, the PS algorithm performs muchbetter than the FSA algorithm when the number of tags inthe areas is too high; for example, for 4,000 tags, the PSalgorithm requires less than 30 seconds while the FSAalgorithm requires more than 120 seconds. However, if thenumber of tags is less than 790, the FSA algorithm has loweraverage delay; thus, the FSA performs better.

Fig. 14 more distinctly shows the variation of the averagedelay with the number of tags for the PS algorithm. Thisfigure shows that T increases very slowly while the number

of tags increases. This was expected because the PSalgorithm uses 19 power levels and thus divides the tagsinto 19 groups. Therefore, an increase of 10-20 tags does notaffect the performance since the tags are mapped into19 groups and not just one group (as in FSA).

According to Fig. 14, the PS algorithm has high initialdelay of around 15.8 sec for scanning 100 tags. This occursbecause the PS algorithm transmits 19 times instead of only1 time, and also after the first cycle, the reader needs totransmit again in all those different power levels corre-sponding to different distances or tag groups, even if someof those groups do not contain any more tags. To minimizethe effect of the initial delay, a small frame size should beused. This is shown in Table 3, where the delay is smallerfor small frame size when reading 1,000 tags.

Having said that, Fig. 15 illustrates the performance oftwo different frame sizes, 64 and 256 slots, according toTable 3. The results show that the frame size of 64 slots has alower delay than 256 slots for number of tags smaller than2,760. However, 2,760 tags is a big number. Normally, inmost applications, the number of tags does not reach thisnumber. For this reason, in most of the simulations, theauthors use up to 1,000 tags. As a result, a smaller framesize is the best choice when the PS algorithm is used.

In addition, a smaller frame size in the PS algorithm isnot only superior than using a larger frame size in the PSalgorithm, but it is also the best choice as compared to FSAwith a larger frame size. However, because of the initial


TABLE 3FSA Algorithm versus PS Algorithm with 1,000 Tags

Fig. 13. Comparison of performance (delay) between FSA and PS. Fig. 14. Delay incurred when using PS algorithm.

delay that the PS algorithm introduces (due to multiple

scans), it has a lower performance than the FSA for a small

number of tags in the interrogation zone (less than 480 in

this case as shown in Fig. 15).Furthermore, the PS algorithm performs better when a

larger step size is used, i.e., 0.4 watts as shown in Fig. 16. In

this case, the tags are divided into fewer groups than before.

Both frame sizes of the PS algorithm have a smaller initial

delay and exceed the performance of FSA sooner, e.g., at

338 tags for frame size of 64 slots.

4.2 Detail Analysis of PS

As observed from the plots described in the previous section,

the effectiveness of the PS algorithm against the FSA is not

always the same but depends on the following variables:

. the selected frame size N of the reader,

. the increment (step) k in watts used by the reader,which determines the number of scans in eachtransmitted cycle,

. the minimum transmitted power Pread;min, whichdetermines the number of scans in each transmittedcycle as well as the effectiveness of the algorithm inthe first scan, and

. the number of tags n in the interrogation zone.

To evaluate the performance of the PS algorithm in termsof the average time delay and to observe how thisperformance is affected by the number of tags in theinterrogation zone, the step size, and the frame size, the PSalgorithm is simulated and three-dimensional mesh plotsare created.

In the simulation, the minimum transmitted power isassumed constant and equals to 0.4 Watts. Also, either theframe size or the step size is kept constant, depending onwhich plot is referenced. Fig. 17 shows the performance ofthe proposed algorithm for different frame sizes of 32, 64,128, and 256 slots. A constant step size of 0.2 Watts is used,and the number of the tags is increased from 100 to 2,000.The horizontal y-axis represents the number of tags in theinterrogation zone, while the x-axis represents the selectedframe size. Finally, perpendicular to the xy plane, the z-axisgives the average time delay for the PS algorithm.

As Fig. 17 shows, the slope of the mesh plot with respectto the y-axis is smaller for a smaller frame size and increaseswhile the frame size increases. That means the initial delayis higher when a large frame size is used by the reader, andthe tags in the area are just a few. This was expected sincethe reader needs more time to transmit the larger frame andit also waits longer for the response from the tags. Moreover,the average time delay is increased at a higher rate for thesmaller frame size than the larger while the number of tagsis increased. This was expected as well, due to the increasingnumber of collisions which occurs when a small frame sizeis used. It is easy to observe that for 2,000 tags, the frame sizeof 32 slots has a delay of over 40 seconds, which is twice thatof 256 slots frame size. However, with 100 tags, this framesize is the most efficient. The above results show that theperformance of the PS algorithm is closely related to theexisting number of tags as in the FSA algorithm. Therefore,it is very important for the reader to estimate the number oftags before the next cycle, as in DFSA.


Fig. 15. Comparison of delay for PS and FSA.

Fig. 16. Comparison of delay for PS and FSA with larger step size.

Fig. 17. Performance of the PS algorithm with frame size(N ¼ 32; 64; 128, and 256 slots) and constant step (k ¼ 0:2 Watts).

Fig. 18 shows the performance of the proposed algorithmwhen the power transmission step size is not constant. Thestep sizes are 0.2, 0.4, 0.9, and 1.8 Watts, and the frame sizeis constant and equals to 128 slots. In this three-dimensionalmesh plot, the y-axis represents the number of the tags, andthe x-axis the different step sizes used. The vertical z-axisonce again gives the average time delay of the algorithm.

The first subplot has a maximum number of tags in theinterrogation area equal to 1,000, while in the secondsubplot, the maximum number of tags is 2,000. This wasdone for better presentation of the results when the numberof tags is increased rapidly.

The second subplot of Fig. 18 easily shows that when thenumber of tags is too high, the average time delay isincreased rapidly when a large value for the step size isused; thus, the performance of the PS algorithm decreases.This inverse proportional relationship between the delayand the step size is logical, because when a large value forthe step size is used, the PS algorithm is almost the same asthe FSA algorithm; thus, in order to identify a large numberof tags, more time is needed due to collisions, or largerframe size is used as in DFSA [3], [8].

In the first subplot, it is more obvious that the step sizeaffects the performance of the PS algorithm. If the tags areonly a few, a higher step size is better, and actually as shownin Figs. 15 and 16, the FSA is even better. However, when thenumber of tags is increased, smaller step sizes result in lessdelay. In this subplot, the step size of 0.4 Watts generallyseems to result in better performance for the PS algorithm.

The performance results indicate that in order for theproposed algorithm to be effective, it is important to selectthe right frame size as in DFSA as well as the right step size.To do so, it is essential to know the number of tags insidethe interrogation zone as this mainly affects performanceand is closely related to the other two variables.

Since the PS algorithm is more effective than the simpleFSA when the number of tags in the interrogation zone ofthe reader is high, only in that case it is necessary toimplement it. Thus, the selection of the step size is simple. Itmust be small enough (0.2-0.4 Watts) to divide the tags intomany small groups in order for the reader to be able to

identify them using the Aloha algorithm. The exact value ofthe step size depends on the number of tags and also theselected frame size.

For the case of the right frame size, it is essential for thereader to have a good estimation of the number of tags. Thisis the main problem from which both DFSA [3] and EDFSA[8] also suffer. The simplest estimation can be provided bythe lower bound of (7), which is N ¼ Rþ 2C.

After the first cycle of transmissions, the reader calculatesthe frame size N by counting the number of identified tags,given by R, and the number of collisions, given by C. Thisbound should work better for the PS algorithm since thenumber of collisions in each scanning is less than in FSA orDFSA using the same frame size, because the number of tagsis not as great as in those algorithms.

4.3 Comparison of PS and DFSA with DynamicFrame Size Estimation

Since both PS and DFSA can enhance the performance ofreading the tags by adaptively changing the frame size, theyare simulated and compared. Both of them use (7) toestimate the next frame size after each scan. The choices offrame size, N , are 64, 128, 256, and 512. If (7) gives a valuebetween two of the choices, the larger frame size is selected.The maximum frame size is 512 and minimum is 64.

As shown in Fig. 19, DFSA takes more than 300 secondswhile PS takes less than 50 seconds to scan 3,600 tags. Theperformance of PS improves as the step size k reduces.Having a smaller step size allows the reader to scan a smallerregion multiple times, thus, reducing the number of colli-sions. If the step size k is reduced lower than 0.01 Watts, thetime needed to read 3,600 tags will be smaller than 50 seconds.Furthermore, PS outperforms DFSA when the number of tagsis greater than 600.

In addition, Fig. 20 illustrates the energy consumed by thereader in order to read the tags. DFSA consumes 2,000 Jouleswhile PS with k set to 0.01 Watts consumes around100 Joules; that is around 20 times saving. It is interestingto note that the smaller the k value the less energy consumed.This corresponds to less time spent reading the tags.


Fig. 18. Performance of the PS algorithm with step size (k ¼ 0:2; 0:4; 0:9,and 1.8 Watts) and constant frame size of 128 slots.

Fig. 19. Total time to scan tags: PS versus DFSA. (Choices of N are 64,128, 256, and 512.)

If the number of nodes is much greater than 3,600, themaximum frame size should be greater than 512. If not,collisions will occur to a point where all the tags cannot beread. For example, as shown in Fig. 19, DFSA with a framesize of 512 seems to approach a scanning limit of 4,000 tags.

5 CONCLUSIONS

This paper proposes a variation of the FSA called theProgressing Scanning algorithm. The Progressing Scanningalgorithm improves the performance of the FSA when thenumber of tags in the area is too high by dividing the tagsinto groups and dealing with each group individually. Theparameters that control the performance of the PS algorithmare the minimum transmitted power level from the reader,the frame size, and finally the step size of increasing thepower in each cycle. Different values cause the PS algorithmto perform differently. Generally, the PS algorithm is betterthan FSA when the number of tags in the area is over 1,000.Also, it is better than DFSA if the number of tags is over 600.

Furthermore, higher initial frame sizes correspond to ahigher initial average delay, but can also handle more tagsdue to collisions. The most important conclusion for theperformance of the proposed algorithm is that it can providea high degree of data integrity in the RFID system, even withthe use of small frame sizes, while FSA and DFSA cannot.

REFERENCES

[1] K. Finkenzeller, RFID Handbook, Fundamentals and Applications inContactless Smart Cards and Identification, second ed. Wiley, 2004.

[2] S. Lahiri, RFID Sourcebook. IBM Press, 2005.[3] J. Cha and J. Kim, “Novel Anti-Collision Algorithms for Fast

Object Identification in RFID System,” IEEE Proc. 2005 11th Int’lConf. Parallel and Distributed Systems (ICPADS), vol. 2, pp. 63-67,July 2005.

[4] http://www.epcglobalinc.org, 2009.[5] F.C. Schoute, “Dynamic Frame Length ALOHA,” IEEE Trans.

Comm., vol. 31, no. 4, pp. 565-568, Apr. 1983.[6] H. Vogt, “Efficient Object Identification with Passive RFID Tags,”

Proc. Int’l Conf. Pervasive Computing, pp. 98-113, Apr. 2002.[7] H. Vogt, “Multiple Object Identification with Passive RFID Tags,”

Proc. IEEE Int’l Conf. Systems, Man and Cybernetics (SMC ’02),vol. 3, pp. 4-9, Oct. 2002.

[8] S. Lee, S. Joo, and C. Lee, “An Enhanced Dynamic Framed SlottedALOHA Algorithm for RFID Tag Identification,” Proc. Second Ann.Int’l Conf. Mobile and Ubiquitous Systems: Networking and Services(MobiQuitous ’05), pp. 166-172, July 2005.

[9] ISO/IEC, “18000 Part 4: Parameters for Air Interface Commu-nications at 2.45 GHz,” ISO, 2004.

[10] T.S. Rappaport, Wireless Communications—Principles and Practices,second ed. Prentice Hall, 2002.

[11] R. Bridgelall, “Bluetooth/802.11 Protocol Adaptation for RFIDTags,” Symbol Technologies, RFDESIGN, July 2002.

[12] G.D. Vita and G. Iannaccone, “Design Criteria for the RF Section ofUHF and Microwave Passive RFID Transponders,” IEEE Trans.Microwave Theory and Techniques, vol. 53, no. 9, pp. 2978-2990, Sept.2005.

[13] P. Sorrells, Passive RFID Basics, AN680, DS00680B, MicrochipTechnology, Inc., pp. 1-5, 1998.

[14] F. Zhou, C. Chen, D. Jim, C. Huang, and H. Min, “Evaluation andOptimizing Power Consumption of Anticollision Protocols forApplications in RFID Systems,” Proc. Int’l Symp. Low PowerElectronics and Design (ISLPED ’04), pp. 357-362, Aug. 2004.

[15] F. Borgonovo and M. Zorzi, “Slotted ALOHA and CDPA: AComparison of Channel Access,” Wireless Networks, vol. 3, pp. 43-51, 1997.

Weilian Su received the BS degree in electrical,computer, and systems engineering (ECSE)from Rensselaer Polytechnic Institute in 1997with Summa Cum Laude, and the ECSEdepartment’s Lockheed Martin Capstone DesignAward. He also received the MSECE and PhDdegrees in electrical and computer engineeringfrom Georgia Institute of Technology in 2001and 2004. He specializes in sensor and ATMnetworks under the guidance of Dr. Ian F.

Akyildiz in the Broadband and Wireless Networking Laboratory atGeorgia Institute of Technology. In 2003, he received the 2003 BestTutorial Paper Award from the IEEE Communications Society.Currently, he is an assistant professor at the Naval PostgraduateSchool. His current research interests are sensor networks, ad hocnetworks, quality of service in Internet, distributed networks, satellitenetworks, and cyber warfare. He is a senior member of the IEEE.

Nikolaos V. Alchazidis graduated from theNaval Postgraduate School in 2006 with themaster of science degree in electrical engineer-ing and the master of science degree insystems engineering. His research interest isin wireless networks.

Tri T. Ha received the BSEE and MSEEdegrees from Ohio University and the PhDdegree from the University of Maryland. Hejoined the Electrical and Computer EngineeringDepartment at the US Naval PostgraduateSchool in 1987. He is a fellow of the IEEE andcurrently holds a joint appointment in theDepartment of Electrical and Computer Engi-neering and the Department of Systems En-gineering. His current research interests include

spatial signal processing, interference cancellation in 2G, 3G, and 4G,robust detection algorithms for CDMA signals, Doppler correctionalgorithms, equalization for OFDM signals, and signal analysis in thepresence of interference.

. For more information on this or any other computing topic,please visit our Digital Library at www.computer.org/publications/dlib.


Fig. 20. Energy consumed to scan tags: PS versus DFSA. (Choices of Nare 64, 128, 256, and 512.)

Date post:	03-Jun-2015
Category:	Business
Upload:	ieeexploreprojects
View:	460 times
Download:	0 times

16. multiple rfid tags access algorithm

Business