Advances in Internet of Things, 2012, 2, 13-22 13

http://dx.doi.org/10.4236/ait.2012.22003 Published Online April 2012 (http://www.SciRP.org/journal/ait)

Design and Implementation of a Low-Power Active RFID

for Container Tracking at 2.4 GHz Frequency
Dawood Moeinfar, Hossein Shamsi, Fatemeh Nafar
Electrical Faculty, K. N. Toosi University of Technology, Tehran, Iran
Email: [email protected], [email protected], [email protected]

Received November 22, 2011; revised December 28, 2011; accepted January 10, 2012

ABSTRACT
In this paper a new active RFID system at 2.45 GHz based on the low-power system-on-chip CC2530 RF transceiver is
designed and implemented. Only by using of an integrated multi-channel fast chip, both the MCU and RF operations
are done which makes the RFID more reliable and reduces the complexity of the hardware and cost, vividly. This RFID
system utilizes the Zig-Bee IEEE 802.15.4 standard in the ISM band. A lot amount of energy is restored by setting Tags
in the sleep mode in the most of times. The maximum transmission range of 80 m at the output power of 4.5 dBm is
obtained. The main application of this system is for the container identification with precise operation and high accu-
racy. An active Tag with unique ID is mounted on each vehicle. By enabling the AUTOCRC error detection possibility,
minor errors are detected in the received frames. Receiver sensitivity of –97 dBm and current consumption of 1 µA in
the sleep mode and 29.6 mA in the active mode are reported.

Keywords: Radio Frequency Identification (RFID); CC2530; RSSI; Active Tags; Collision; TDMA; IEEE 802.15.4;
Sleep Mode

1. Introduction ognized by it, and afterward they enter to the relationship

with each other according to the user request. Unlike
Radio Frequency Identification technology had been wide-
some proposed RFID schemes, which have some Read-
spread in recent years, and encompasses an extensive
ers and Tags, our design is a centralized RFID with only
range of wireless applications such as distribution, trans-
a central Reader and numerous Tags. According to Fig-
portation, tracing, patient monitoring, military applica- ure 1, user interface (PC) is the main master of operation
tions and etc. RFID is an automatic real-time non-contact in RFID system. So, sub-master (Reader) commands
identification technology that deals with Identification Tags only by the request of the main master (PC). Tags
Numbers (ID) through scattered electromagnetic waves only respond to the commands, whenever they have been
in the air. It’s more reliable, efficient, secured, inexpen- investigated. Interrogated data from Tags are exhibited to
sive and accurate than other similar automatic identifiers the user with the serial interface.
such as sensor networks, imagery systems and so on. One important classification in RFID is based on the
These reasons are why the RFID technology is more method of power supply provision in Tags. Passive Tags
popular, recently [1]. don’t have any internal power source. They extract their
An electronic label dedicates to each identifiable ob- energy from the Reader scattered wave and only act
ject, which contains main information or specification within the magnetic domain of the Reader. This type of
about that object. This label is named Tag and restores
the data for a long time. Each Tag has a same structure
and function. It consists of a microchip and antenna. An-
other key part of an RFID system is Reader (data collec-
tor), which collects the data stored in the Tags. A Reader
includes the microprocessor, transceiver, antenna and a
peripheral interface for user commands and monitoring.
This two structured electronic devices which are at the
heart of every RFID system, communicate with each
other in the wireless manner [2]. Whenever a Tag is lo-
cated at the magnetic zone of the Reader, it will be rec- Figure 1. Centralized RFID system.

Copyright © 2012 SciRes. AIT

14 D. MOEINFAR ET AL.

RFID is used for short-range applications such as smart integrated chip [6]. So, it handles all the baseband and
cards. In addition to passive RFID Tags, active RFID RF processing functions together. In addition to mini-
Tags are designed having their own power supply, which mize the hardware size, it prevents from high frequency
is the basis of our work. They even act out the range of noise generated from connecting RF transceiver and
magnetic field of the Reader. RFIDs are designed at the MCU chips. This integrated chip benefits from the low
standard frequency ranges: low-frequency (LF), high- power consumption, high communication range, high
frequency (HF), Ultra High Frequency (UHF) and Mi- capacity memories, multiple channels, and high sensitiv-
crowave. In the comparison of different types of RFIDs, ity receiver [7]. Some critical issues in the active RFID
lower frequencies don’t penetrate or transmit around should be considered when developing the hardware [8].
metals, handle only small amounts of data and slow the The first key point of the development is the lifetime of
data transfer. The main advantages of microwave fre- the active RFID Tag. Because the active RFID Tag is
quencies are smaller size, smaller antenna and higher energized by the internal battery, the lifetime of the Tag
data range than other frequencies [3]. is mainly depends on the lifetime of the battery. At first,
By emerging advanced technology and improving the we choose the low-power components and implement a
economy, the automation of harbors is necessary. Using mechanism in which the processor can completely turn
RFID, a large amount of time and cost will be restored off the radio or simply put it in the sleep mode to save
and the safer transportation method is provided. Because the power. The second one is the ability of the anti-colli-
containers occupy a large area at the ports, the RFID in- sion, which is important especially in the vehicle trans-
terrogate zone should be large enough to cover a wide portation environment. To get to the high identification
range. So active Tags at the microwave frequency should rate for multiple Tags, the anti-collision is carefully ap-
be used for these applications [2,4]. plied. In Table 1, the key parameters of the system
Collision phenomenon is a common problem that we achieved by using of CC2530 are listed [7].
have encountered at the first step of designing a multi-
Tag RFID system. Whenever several Tags are presented 2.1. Reader Structure
in the interrogate zone of Reader and respond to the In Figure 2, the hardware of the RFID Reader is shown.
Reader command simultaneously, their data collide in the The Reader acts as a sub-master in this system so it exe-
air. So a lot amount of energy, bandwidth and time will cutes any command received from the PC. A personal
be wasted. Different anti-collision procedures have been computer is connected to the Reader through the RS-232
designed for avoiding from the collision. Because TDMA serial interface [9]. The block diagram of the Reader is
protocols are the largest applicable group of anti-colli- shown in Figure 3. The MCU unit directly controls the
sion, hence we focus on TDMA in this paper. [3]. RF transceiver to communicate with Tags and processes
Power consumption is always a remarkable criterion in all Tags response data [5]. For precise operation of the
the design of Tags, especially in the applications where radio part, an external oscillator with the frequency ac-
power supply is limited. One drawback of the active curacy of lower than ±40 ppm is required. We used
RFID systems is the precise synchronization requirement HXO-36B crystal oscillator with the frequency stability
between the Reader and Tags. In this work, we put Tags of ±25 ppm.
in the sleep mode most times, which all circuitry of Tags
are turned off during the sleep mode. A Tag is only acti- 2.2. Tag Structure
vated by the Reader command in the small fraction of
The amount of power and area consumption is a remark-
one reading cycle. So, in the major portion of time, Tag
able challenge in the design of portable Tags. Decreasing
is inactive.
the Tag’s power consumption is an important factor in
This paper is organized as below: at first the hardware
active RFIDs [10]. Choosing a low-power chip in hard-
is described in detail. Second, software algorithm and
anti-collision protocol are described, completely. After
that, experimental results are reported. Finally, we con-
clude our work.

2. Hardware Description
Unlike other previous similar works, which use some
modules to meet the requisite of a practical RFID system,
our hardware uses only a low-power high-accuracy trans-
ceiver for both Reader and Tags [4,5]. It has all capabili-
ties of a microprocessor and RF transceiver module in an Figure 2. The hardware of the reader.

Copyright © 2012 SciRes. AIT

 D. MOEINFAR ET AL. 15

Table 1. Key parameters of the system [7].

Parameters Value (Range) Unit

TX Output Power Range –22 - +4.5 dBm
Receiver Sensitivity –97 dBm
Integrated RF Chip Dimensions 6×6 mm2
Current Consumption in Sleep Mode 1 µA
Maximum Current Consumption In Active Mode 29 mA
Supply Voltage 2 - 3.6 V
Programmable Flash Memory 256 kB
RX/TX Turnaround 192 µs
Radio Baud Rate 250 Kbps
Channel Spacing 5 MHz
RSSI Range 100 dB
SRAM Memory 8 kB
Radio Ram Memory 384 kB
Communication Range 80 meter
Modulation Scheme - OQPSK
Channel Numbers 16 numbers
Tag Size 22 × 30 mm2

Figure 3. The block diagram of the reader hardware.

ware design is an aspect, and the software algorithm of when the received frames are from the Reader. After
Tags is also another key factor to decrease the power responding to the Reader command, Tag goes to the
consumption. There are three kinds of states in this algo- sleep mode. We compact the dimensions of Tags down
rithm as follows: sleep condition, survey condition, and to near 660 mm2. The block diagram of Tags is shown in
communication condition. The hardware of Tags is same Figure 4.
with that shown in Figure 2, but it does not need any
serial interface [2]. In general, in most of times, Tags are 3. Software Development
in the sleep condition. At this time all internal circuits of
3.1. Anti-Collision Protocol
the Tag are inactive except the low-frequency clock. The
Tag is woken up and interrupted by received data packets. Anti-collision protocol is critical in the performance of
Then Tag enters to the communication condition only RFID systems. Without using an anti-collision protocol,

Copyright © 2012 SciRes. AIT

16 D. MOEINFAR ET AL.

each cycle. The timing of SA-TDMA anti-collision pro-

tocol is shown in Figure 5.

3.2. Software Algorithm

As shown in Figure 6, we have considered three basic
commands for our software. In ID allocate command, we
dedicate an ID to each Tag. We search the Tag ID by the
polling command and clear the Tag ID by the clearing
Figure 4. The block diagram of tags. command. This algorithm is implemented at high-level
software which provides a user-interface tool on a PC.
the responses from the Tags would collide together and The ID allocation is performed as follows. A unique ad-
thereby prolong the identification process. So, collision dress and ID is dedicated to each Tag. The Reader re-
causes the bandwidth and energy wastage. There are four ceives the ID and address from the PC serial interface
main types of anti-collision schemes: space division mul- and then sends them to a Tag. This Tag saves ID as an
tiple access (SDMA), frequency division multiple access identification number in the permanent memory. After
(FDMA), time division multiple access (TDMA), and
receiving ID from Reader, it sends an acknowledge mes-
code division multiple access (CDMA). TDMA protocols
sage to the Reader. This message confirms the successful
are the largest group of anti-collision algorithms. These
ID allocation of the Tag. Then we allocate the ID to an-
algorithms could be classified as the Reader-driven or
other Tag. This process is repeated for all Tags by the
Tag-driven procedure. Our scheme is based on the Reader-
user until every Tag will have a unique identification
Talk-First (RTF) protocol, in which Tags remain silent
until commanded or searched by the Reader. Tag driven number.
procedures are slower than RTF. In the Pure-ALOHA The whole flowchart of the RFID Reader is shown in
(PA) procedure, Tags respond to the Reader at random Figure 7. The Reader waits for the user commands from
instants. So the possibility of the collision increases. In the serial port. In this situation, the RF transceiver of the
Slotted-ALOHA (SA) based RFID systems, Tags trans- Reader remains in the idle state. By reception of any
mit their ID in synchronous time slots. Our approach is command from PC, the Reader interprets the command
based on the SA protocol. By this method, the round time and executes it. By using the ID-Allocation command,
and probability of the collision decrease, greatly [3]. each Tag will have a unique identification number, and
Here, Tags respond to the Reader commands only at cer- subsequently the user can search Tags by their unique
tain moments. If a Tag is present in the duration of this IDs. When we want to remove an ID, we use the clearing
slide, it will respond to the Reader and then go to the Tag command. By performing this command, the given
sleep mode at the end of this slide until the next cycle. In Tag loses its ID and it will no longer has passing permis-
our algorithm, a command frame includes 20 bytes. So, sion.
at the data rate of 250 kbps, it takes 0.5 ms to be trans- All Tags have monotonous routine in the soft protocol.
mitted, completely. The slide duration in our protocol is They should react to the user commands that come from
about 5 ms. One Tag is active only for a short time in the Reader. In fact, Tags should be synchronous with the

Figure 5. The timing of the anti-collision protocol.

Copyright © 2012 SciRes. AIT

 D. MOEINFAR ET AL. 17

There are 4 types of MAC frames in this standard. In our

work, we use the data frame for communication between
the Reader and Tags. In Figure 9, the schematics of
MAC and physical layers of 802.15.4 are shown. SHR
(Synchronization Header) is added at the first of any
frame for synchronization between the Reader and Tag.
The SHR consists of the 4-byte preamble sequence and
1-byte SFD. These two fields are set automatically by the
hardware. In the PHR (Physical Header), the length of
MPDU (MAC Protocol Data Unit) is written. The SHR
and PHR are placed at the physical layer of 802.15.4
standard. The Radio has a FIFO memory with 128 bytes
capacity. FCF (Frame Control Field), data sequence, ad-
dress information, frame payload, and FCS (Frame Check
Sequence) create the MAC layer. The Radio supports the
automatic calculation of 2-byte FCS. By using of the
source and destination addresses, each device will have a
unique address, which helps us provide a precise timing
in the polling process. Moreover, IEEE 802.15.4 is
equipped with frame filtering and AUTOCRC protection.
It rejects non-compliant frames, so decreases the packet
Figure 6. The flowchart of the user-interface software. error rate [7].

Reader. In practice, Tags are always located in inactive 4. Experimental Results

mode. In this mode, only internal low-power 32.768 kHz In Figure 10, all the 16 channels are shown in the spec-
oscillator is active and all other components are dormant. trum analyzer with a high precision. By using of a pow-
In this mode, by using of the sleep timer, external events erful RF transmitter, all out of band frequencies are fil-
such as radio signals can be detected. Whenever SFD tered and the main frequency is amplified. The receiver
(Starting Frame Detector) from the received packet is sensitivity is a critical parameter for correct operation of
detected, an interrupt will be occurred. We implement RF transceivers. In CC2530, the receiver sensitivity is
this interrupt source for synchronizing the Tags with the –97 dBm [7]. So, if the received signal strength (RSSI)
Reader and for reducing the power consumption in Tags. becomes lower than this level, the receiver will not detect
After SFD detection, Tags Wake up and go to the survey the signal properly. In Figure 11, the RSSI versus dis-
condition. Now, Tags receive whole of the frame. If the tance between the Reader and Tags is depicted in a prac-
source address within the frame does not match with the tical test. For lower transmitter output power, the RSSI
Reader address, Tags do not pay attention to this frame value is decreased. This reduction in RSSI is often more
any further and go to the sleep mode immediately. But for long distances rather than short distances. As shown
when the source address is identical with the address of in Figure 11, the RSSI level is higher than the receiver
the Reader, Tags look for the type of the frame command sensitivity threshold for distances smaller than 80 meter.
within the frame. As shown in Figure 8, Tag goes to the The PER (Packet Error Rate) is another significant pa-
communication condition for these three commands: ID rameter in RFIDs. It is expressed in the form of percent-
allocation, polling, and clearing the Tag ID. When poll- age. As shown in Figure 12, the PER is increased with
ing and clearing commands are coming from PC, at first, increasing the distance, significantly. The experiments
Tag checks the Destination address with the address of were performed for the two different TX output powers
itself. Tag only acts when these two addresses are iden- to ensure the proper operation of this system. As shown
tical. After executing any of these commands, Tag goes in Figure 13, round time for collecting 170 Tags data is
to the sleep mode for other receptions. So, by this way, near 1 second. In Table 2, important parameters of this
Tags always listen to the channel and after processing system compared with similar works are shown.
any event they go to the low power consumption mode,
immediately. 5. Conclusion
In this work, a low-power high-range active RFID is im-
3.3. Zig-Bee IEEE 802.15.4 Frame Format
plemented by using the chip CC2530. Without using any
IEEE 802.15.4 is an efficient standard for MAC protocol. additional peripheral module, the identification process is

Copyright © 2012 SciRes. AIT

18 D. MOEINFAR ET AL.

Figure 7. The reader flowchart.

Table 2. System characteristics of similar RFIDs.

Modulation Carrie Frequency Data Rang Max Output Receiver Max Current in
Reference Anti-Collision Protocol
Typpe (MHZ) Rate (meter) Power (dBm) Sensitivity (dBm) TX Mode (mA)
[2] QPSK 2450 150 100 0 –95 Binary Search Algorithm 11.3 mA at 3.6 V
[4] OQPSK 2450 250 100 1 –95 Binary Search Algorithm 15.1 mA at 3.6 V
[11] ASK/FSK/MSK 800 - 928 500 30 +10 –110 Hybrid TDMA MAC 40 mA at 3 V
[12] FSK 433 153 10 +5 –85 Dynamic framed Slotted-ALOHA 40 mA at 3 V
[13] QPSK/MSK 2450 150 30 0 –90 Pure-ALOHA TDMA 13 mA at 3.6 V
This Work OQPSK 2405 - 2485 250 80 +4.5 –97 Slotted-Aloha TDMA 29.6 mA at 3.3 V

Copyright © 2012 SciRes. AIT

 D. MOEINFAR ET AL. 19

Figure 8. The flowchart of tags operation.

Copyright © 2012 SciRes. AIT

20 D. MOEINFAR ET AL.

Bytes: 4 1 1 2 1 2 2 2 n=2 2
Preamble Sequence SFD Frame Length FCF Data Sequence Number PAN ID Dest Address Source Address Payload FCS

Figure 9. The physical and MAC layers of IEEE 802.15.4 standard frames [7].

Figure 10. Frequency spectrum of the transmitting signal on spectrum analyzer.

-50
+4.5 dBm
-4 dBm
-55
Received Signal Strength Indicator(dBm)

-60

-65 RSSI @ +4.5 dBm

-70

-75

RSSI @ -4 dBm
-80

-85

-90

-95

Receiver Sensitivity
-100

-105
5 10 15 20 25 30 35 40 45 50
Distance(m)

Figure 11. Received signal strength (RSSI) versus distance.

Copyright © 2012 SciRes. AIT

 D. MOEINFAR ET AL. 21

1.8

1.6
Packet Error Rate(%) 1.4

1.2

0.8

0.6

0.4

0.2 +4.5 dBm

-4 dBm
0
0 5 10 15 20 25 30 35 40 45 50
Distance(m)

Figure 12. Packet error rate versus distance.

1.4

1.2

1
Round time(second)

0.8

0.6

0.4

0.2

0
0 20 40 60 80 100 120 140 160 180 200
Tag numbers

Figure 13. Round time versus tag numbers.

executed with high accuracy and minimum PER. By RFID-Based Material Tracking Information System,” 2007
programming Tags in the sleep mode in most of times, a IEEE International Conference on Automation and Lo-
gistics, Jinan, 18-21 August 2007, pp. 2922-2926.
lot amount of power is restored. This system is tested and
its operation is verified at outdoor environment until 80 [2] Y. Sun, Y. Zhang and P. Peng, “Design and Realization
meter distances, successfully. With a range extender de- of 2.45 GHz Active RFID System,” 2nd International
Conference on Intelligent Computation Technology and
vice, such as CC2591, the communication range up to Automation, Changsha, 11 October 2009, pp. 582-585.
300 m can be achieved. In our work, the collection time doi:10.1109/ICICTA.2009.147
for 170 Tags is near 1 second. By using of a low-power
[3] D. K. Klair, K.-W. Chin and R. Raad, “A Survey and
multi-purpose Chip, small dimensions for Tags are Tutorial of RFID Anti-Collision Protocols,” Communica-
achieved. In this architecture, compacting Tags down to tions Surveys & Tutorials, Vol. 12, No. 3, 2010, pp. 400-
near 660 mm2 is possible, which is required in many 421. doi:10.1109/SURV.2010.031810.00037
practical environments. This system will be used for ve- [4] X. Huang, G. Zhang, Q. Sun and D. Qing, “Container
hicle transportation and especially container tracking at Transportation System Using 2.45 GHz Active RFID
commercial ports. Technology,” International Conference on Remote Sens-
ing, Environment and Transportation Engineering, Nan-
jing, 24 June 2011, pp. 3030-3033.
REFERENCES doi:10.1109/RSETE.2011.5964953
[1] Z. Min, L. Wenfeng, W. Zhongyun, L. Bin and R. Xia, “A [5] W. J. Yoon, S. H. Chung, S. J. Lee and Y. S. Moon, “De-

Copyright © 2012 SciRes. AIT

22 D. MOEINFAR ET AL.

sign and Implementation of an Active RFID System for [10] X.-Z. Zheng, X.-H. Cao and W.-L. Zheng, “Design of
Fast Tag Collection,” 7th International Conference on Com- Electronic Tag in Active RFID Systems,” Port Operation,
puter and Information Technology, Fukushima, 16 Octo- Vol. 1, No. 2, 2008, pp. 850-853.
ber 2007, pp. 961-966. [11] Z. H. Xie and S. L. Lai, “Design and Implementation of
[6] H. Cho and Y. Baek, “Design and Implementation of an an Active RFID MAC Protocol,” International Confer-
Active RFID System Platform,” International Symposium ence on Wireless Communications, Networking and Mobile
on Applications and the Internet Workshops, Busan, 23 Computing, Shanghai, 21 September 2007, pp. 2113-2116.
January 2006, pp. 1-4. [12] H. Cho, H. Choi, W. Lee, Y. Jung and Y. Baek, “LITeTag:
[7] Texas Instruments, CC2530 User Guide and Datasheet, Design and Implementation of an RFID System for IT-
April 2009. http://www.ti.com Based Port Logistics,” Journal of Communications, Vol.
[8] Y. Hao and Z. Hong, “Development of a Low-Cost Ac- 1, No. 4, 2006, pp. 48-57.
tive RFID Platform,” IEEE 2007 International Sympo- [13] U. Bilstrup and P.-A. Wiberg, “An Architecture Com-
sium on Microwave, Antenna, Propagation, and EMC parison between a Wireless Sensor Network and an Ac-
Technologies for Wireless Communications, Hangzhou, tive RFID System,” 29th Annual IEEE International
16 August 2007, pp. 152-155. Conference on Local Computer Networks, Tampa, 16
[9] Maxim, MAX3232 Datasheet, 2007. November 2004, pp. 583-584.
http://www.maxim.com doi:10.1109/LCN.2004.17

Copyright © 2012 SciRes. AIT