Evaluation of the Effect of Gen2 Parameters on the UHF RFID Tag Read Rate

Transcription

1 International Journal of Latest Trends in Computing (E-ISSN: ) 160 Evaluation of the Effect of Gen2 Parameters on the UHF RFID Tag Read Rate Jussi Nummela, Petri Oksa, Leena Ukkonen and Lauri Sydänheimo Tampere University of Technology Department of Electronics, Rauma Research Unit Kalliokatu 2, FI Rauma FINLAND Abstract: The UHF RFID technology is nowadays widely spreading in different industries, e.g. in the supply chain and warehouse management of different goods among others. This paper presents passive UHF RFID tag read rate measurements using the EPCglobal Class 1 Gen 2 protocol. Tag encoding schemes such as FM0 and Miller, and Gen2 parameters such as Tari and Backscatter Link Frequency are varied to find out their effect on the tag s reading rate. Relating to these parameters alterable in the tag reader, this paper provides certain achievements which can be exploited in industrial RFID configurations and setups. Keywords: radio frequency identification, RFID, passive tag, read rate, tag encoding, Gen2 parameters 1. Introduction The UHF RFID technology has become more and more common in different industries during the last years. For example supply chain management of paper reels is one of the latest applications, as well as small scale positioning, in which promising results have been achieved. [1, 2] EPCglobal was formed in 2003 to carry on the non-profit standard framework supporting use of RFID in the supply chain and in other applications. As part of this mission, it sought to create a single global standard for the UHF RFID reader-tag air interface. This second-generation (Gen2) standard became publicly in 2005, and today numerous vendors are producing reader and tag models supporting EPCglobal Class 1 Gen2 -standard. [3, 4] Any passive RFID protocol must provide certain basic functions: Data Modulation: a set of waveforms that is understood as symbols by tags and readers. Packet Structure: preambles, training symbols, and timing conventions that enable tags and readers to synchronize and to recognize commands and data. Command Set: commands and responses that make it possible to read (and write) information stored in tags. This paper deals with Data Modulation issues of Gen2- protocol, and thus it is described in more detail in chapter II. Chapter III presents the measurements, where different Gen2- protocol parameters were varied and read rates observed. Chapter IV gives the measurement results and discussion, and finally Chapter V conclusion and the future work. 2. Data modulation in UHF Class 1 Gen2 - communication Modulation is the change made in a signal in order to send information. Each RFID standard employs one modulation scheme for Forward Link (reader-to-tag) and another for Reverse Link (tag-to-reader), because of different roles of reader and tag in backscatter communication. [5, 6] 1.1. Forward Link In Gen2-protocol reader sends signals by changing its output power level between two states, on state and off state. This is known as amplitude-shift keying (ASK). Gen2 uses two data symbols, Data 0 and Data 1, when former specification Class 0 had additional Null symbol. Both data symbols begin with an on state, followed by an off state. The length of the on state varies, but the length of the off state is always fixed, and called Pulse Width (PW). Symbols Data 1 and Data 0 are distinguished by varying the length of on state, in other words, the interval between off state pulses. Thus, this technique is sometimes referred to as pulse-interval encoding (PIE). [4, 5, 6] The total time of Data 0 is called Tari. Data 1 symbol is allowed to between 1,5*Tari or 2*Tari, and the length of the Tari may vary between 6,25 µs and 25 µs. Since the data symbols of Gen2 are not of the same length, there is no fixed Gen2 Forward Link data rate. Instead, there is an effective data rate assuming equiprobable data, which is, depending on Tari, between 27 and 128 kbps. Fig. 1 clarifies the explanation. [3, 4, 5]

2 International Journal of Latest Trends in Computing (E-ISSN: ) 161 Fig. 3. Miller-modulated subcarrier example with M=2 [3] Fig. 1. Gen2 Forward Link symbols [3] As seen, Gen2 offers a wide selection of allowable Forward Link waveforms. This permits tradeoffs between read rate, read range and transmit bandwidth, and allows convenient adaption to different operating environments Reverse Link As passive RFID tag does not have its own radio, it sends information back to reader by changing the impedance state of its antenna. The reader senses this as changes in the backscattered power from the tag. The tag encodes its data in the timing of the transitions between backscattering states. In Gen2-communication, there is a fundamental clock, known as the Backscatter Link Frequency (BLF), which specifies the pulse width of the shortest Reverse Link feature. Then there are four different ways to encode data symbols: FM0 baseband and three different Miller modulations of BLF subcarrier. The simplest encoding, FM0, has a state transition at the beginning of each data symbol. Data 0 symbol has an additional mid-symbol transition. Based on this, a long string of FM0 Data 0 symbols produce a square wave at BLF, and a long string of Data 1 symbols generates a square wave at BLF/2. For FM0 the data rate is same as BLF, and they share the same allowable range, from 40 to 640 kbps. FM0 compliant tags must support the whole range, but readers need not, and in many cases will not, implement all data rates. Figure 2 shows default Gen2 FM0 tag signaling. [5, 6] Fig. 4. Example of encoding the same data stream using Miller M=2,M=-4 and M=8 [3] MMS is supposed to provide a constant performance when a large number of readers are operating in the same facility at the same time. This is due to its narrower Reverse Link spectrum and a tendency to put it into the frequency region between readers Forward Link channels. [3, 5] 3. Measurements This paper presents the measurements where UHF RFID tag read rates were studied when varying certain Gen2 parameters as well as reader transmission power. In addition some interference was added with extra reader reading the same tags. Totally ten individual measurements setups (i.e. circumstances) were conducted. Within each measurement setup the differences between read rates were studied by modifying the Gen2 parameters similarly. The measurements were conducted with 1, 2 or 4 similar passive UHF tags, and a reader with one antenna (monostatic mode). The taken parameter changes are presented in table I. Each group represents one individual permanent set of parameter values used in this study. Fig. 2. Default Gen2 FM0 tag signaling [3] Miller modulated subcarrier (MMS) is more complicated encoding. MMS provides more state transitions per bit and is therefore easier to decode in the presence of interference. However it is slower with the same tag BLF than FM0. MMS exists in three different schemes: Miller-2, Miller-4, and Miller-8. Then number defines how many BLF periods define one data symbol. For example, using the slowest BLF of 40 khz, the data rate for Miller-8 is the BLF/8 = 5 kbps. Figures 3 and 4 clarify the MMS structure. [5, 6] PROTOCOL SETTINGS Table 1. The tested protocol parameters A B C D Tari (µs) 14,29 14, PIE 2,0:1 2,0:1 2,0:1 2,0:1 Forward Link PR-ASK PR-ASK PR-ASK PR-ASK Pulse Width 0,5 0,5 0,5 0,5 BLF (khz) Reverse Modulation FM0 Miller M=4 Miller M=4 Miller M=8 Each of ten measurement setups consisted of following adjustments: at first Gen2 parameters were set as presented in A column. The read rate was recorded based on the average values of ten individual 60 second measuring periods. Next Link Frequency was decreased to 320 khz and Reverse

3 International Journal of Latest Trends in Computing (E-ISSN: ) 162 Modulation changed from FM0 to Miller M=4 ( B), and read rate again recorded similarly. For third setup the length of Tari was increased to 20 µs ( C), and finally the Miller M=4 turned to Miller M=8 ( D). As mentioned, the above described measurement arrangement was carried out with totally ten different ways with varying the number of tags, the TX-power of the reader, or adding the extra reader to create interference, as presented in more detail in table II. In measurement setups 7-10 an additional reader was added to create interference. In setups 7-8 this extra reader was using 28 dbm TX power and 4 alternating channels without LBT (listen before talk) feature. The actual reader whose read rate was monitored used Reader Selects Frequency option, which made it possible to choose between the same four channels, and LBT. In setups 9-10 the actual reader was forced to act only in one single channel (866,3 MHz) with LBT, and the extra reader was transmitting with 30 dbm first in four channels (#9) and then in the same single channel (#10) again without LBT. The used RFID equipment was: Tags: o UPM ShortDipole EPC Class 1 Gen 2 tags with 240 bit EPC memory RFID reader: o Impinj Speedway R1000 UHF reader o Firmware version: Octane o Operating region: ETSI EN LBT Reader Antenna: Huber+Suhner SPA 860/65/9/0/V Measurement software: Impinj MultiReader Extra reader creating interference: o ThingMagic Astra EU reader, with integrated 6 dbi CP-antenna Table 2. Different measure arrangements Measurement setups Common parametres: - Tag distance: 1,0 m; Height: 1,2 m - Duration: 10 x 60s - RX sensitivity: -70 dbm - Search Mode: 0 - Dual target - Session 1 - Reader Selects Frequency (not in #9 and #10) - LLRP Protocol Varying parametres: - Setup #1: TX power 18 dbm, 1 tag - Setup #2: TX power 28 dbm, 1 tag - Setup #3: TX power 18 dbm, 2 tags - Setup #4: TX power 28 dbm, 2 tags - Setup #5: TX power 18 dbm, 4 tags - Setup #6: TX power 28 dbm, 4 tags - Setup #7: TX power 18 dbm, 4 tags, Extra reader - Setup #8: TX power 28 dbm, 4 tags, Extra reader - Setup #9: TX power 16 dbm, 4 tags, 1 channel, Extra reader - Setup #10: TX power 16 dbm, 4 tags, 1 channel, Extra reader in the same 1 channel The measurements were conducted in a laboratory, where the wall behind the tags was covered with absorbing material to cut down unintended interference. Otherwise the environment was close to real-life, not anechoic chamber. In these circumstances, the threshold power level was approximately 15,0-15,5 dbm. Figure 5 shows the measurement setup. Fig. 5. The measurement setup. Tags are attached on the styrofoam pillar and readers and antenna on the rack on right. Impinj reader at the bottom, antenna in the middle, and ThingMagic reader uppermost. 4. Results and Discussion As the purpose of this paper is to study the effect of adjusting the Gen2-parameters, the achieved read rates are compared to each other s within each measurement setup. This way the differences between each parameter values can be seen, as well as if the measuring circumstances cause any variation for their mutual interrelationships. The following graphs in figure 6 presents actual amount of successfully read tags in each measurement setup (#1 #10), and with each Gen2 parameter setup ( A D). Graphs #1 #6 prove clearly the theory that FM0 coding scheme ( A) achieves higher read rate than MMS coding schemes ( B D). In this study the increase in number of tags increased greatly the total read rate per second as well, especially with FM0 coding. The variation in Tari ( B vs C) had only a minor effect in tested situations. In graphs #7 #10, presenting setups where interference was created with extra reader, the benefits of MMS coding schemes can clearly be seen. The read rates stayed close or even increased when moved from FM0 ( A) to MMS ( B C).

4 International Journal of Latest Trends in Computing (E-ISSN: ) 163 Fig. 6. The actual amount of successfully read tags in each measurement setup (#1 #10) and with each Gen2 parameter setup ( A D) The figure 7 shows graphs about relative read rates of different tested Gen2 parameter setups ( A D), in each of ten tested environments (#1 #10). A, where FM0 coding scheme was used, is individually set as a reference value within each of 10 measurement setups. Percentage values between different setups (#1 #10) are not comparable in this figure. Graphs in figure 7 show the purpose and advantage of the MMS coding scheme. In setups #1 #6, where no interference was present, the read rates with MMS coding schemes were between 27% and 59% of the FM0 read rate. In setups #7 #10, where additional reader was present, the MMS coded read rate achieved higher percentages, and finally when channel switching was disabled (#9 and #10), over three times higher values than with FM0 coded reading. Fig. 7. The relative read rate of different Gen2 parameter setups

5 International Journal of Latest Trends in Computing (E-ISSN: ) Conclusions and Future Work Based on the results presented in this paper can be concluded that changes in UHF Gen2 parameters have effects on read rate. As long as the operating environment does not have any interference present, FM0 coding gives highest read rates, despite of number of tags (1, 2 or 4) or transmission power level. When one additional reader was brought nearby to read same tags, FM0 still achieved highest read rates, when reader was able to select its channel automatically. When it was forced to act in a one channel, and even further, when additional reader was acting on that same channel, the MMS coding scheme achieved notably higher read rates than FM0. The tested 40% increase in Tari value did not have a relevant effect on read rate, in these circumstances. If more interference and larger amount tags had been tested, the advantage of MMS coding might have been even more significant. This would be the interesting research topic in the future. Also the effects of varying Tari and PW values could be studied in more detail. References [1] Nummela, J., Lehto, A., Ukkonen, L. and Sydänheimo, L. Automatic reel identification by RFID technology in paper reel supply chains: examples from paper mill and sea port environments, International Journal of RF Technologies: Research and Applications,Vol 1, No 3, pp , September [2] Nummela, J., Nikkari, M., Soini, M., Ukkonen, L., Sydänheimo, L. (2009) A Novel Method for Indoor Positioning with Passive UHF RFID International Journal of Radio Frequency Identification Technology and Applications (IJRFITA). Inderscience. (ACCEPTED) [3] D. M. Dobkin, The RF in RFID: passive UHF RFID in practice. Burlington, MA USA: Elsevier Inc., [4] EPCglobal, Specification for RFID air interface: EPC radio-frequency identity protocols Class-1 Generation-2 UHF RFID protocol for communications at 860 MHz-960 MHz, version 1.2.0, 108 Pages, October [5] D. M. Dobkin, D. J. Kurtz, Overview of EPCglobal Class 1 Generation 2 and comparison with 1 st Generation EPCglobal standards, 10 Pages, March [6] P. V. Nikitin, K. V. S. Rao, Effect of Gen2 Protocol Parameters on RFID tag Performance, 2009 IEEE International Conference on RFID, April 2009.