1. Introduction
As climacteric fruits ripen, the starch in their pulp turns into
sweet-tasting sugar. It is necessary to harvest them at an appropriate
time, i.e., harvesting them too early and they will no longer ripen
properly, harvesting them too late then they must be consumed within a
few days before they are spoiled. Timing the harvest for sufficient
shipping time for export is absolutely crucial. Durio zibethinus Murray
(Durian) is a major export fruit that brings into Thailand large amount
of income every year. The conventional method for monitoring the
maturity stage of Durian is by counting the number of days after
full-bloom. However, the results are often inaccurate due to variations
in the environment, e.g., temperature and humidity [1]. Therefore, a
number of research projects have been conducted to develop sensors for
monitoring the maturity stage of Durian [2-4]. Yet, these are generally
only used to monitor post-harvest products.
This work presents a wireless sensor network for monitoring the
maturity stage of Durian in the pre-harvest monitoring scheme.
2. Dual-Polarization Coupled Patch Sensor
It has been shown in [5] that starch and reducing sugar in Durian
pulp vary, respectively, proportionally and inversely proportionally to
the days after full-bloom. The peel is relatively unchanged. These
changes gradually decrease the dielectric constant of the pulp in the
days following full-bloom [6,7].
Due to severe environment during Durian-growing season, which is
highly humid with heavy rains, a dual-polarization coupled patch was
selected to ensure that measurements were robust to environmental
changes. For the sensor, using coupled patch antenna, we simulated a
Durian fruit as a pulp covered by a prickle peel with CST microwave
studio software [8]. The Durian was modeled as an ellipsoid, see Figure
1(a), with the major and minor axis of 30 cm and 20 cm, respectively.
This is the standard size for export Durian. The thickness of the
prickle peel was 1 cm. Figure 1(b) shows the diagram of the
dual-polarization coupled patch sensor which was fabricated on an FR-4
substrate 1.6 mm high. The dielectric constant of the substrate was 4.4.
The right patch was used for transmitting a microwave frequency of 2.45
GHz whereas the left patch was used for receiving the coupled signal.
The probe along the same line with the transmitting probe was used for
receiving the parallel polarization signal whereas the one below it was
used for receiving the perpendicular polarization signal.
[FIGURE 1 OMITTED]
The dielectric constant of the peel was kept constant at 15 whereas
the dielectric constant of the pulp was varied from 62 to 22. The
variations of the magnitude of mutual coupling for parallel polarization
are shown in [9]. They are plotted together with those for perpendicular
polarization as shown in Figure 2. It can be seen that the mutual
coupling decreased as the dielectric constant decreased from 62 to 37,
but then increased slightly afterwards. Thus, the variation in mutual
coupling can be used as an indicator for monitoring the maturity stage
of Durian. It should be noted that the perpendicular polarization showed
higher mutual coupling due to the effect of near-field region of the
antenna.
3. Wireless Sensor Network
The architecture of the wireless sensor network is shown in Figure
3. It shows sensor nodes attached to Durian fruits sensing the magnitude
of the mutual coupling of the fruits and sending signal to a tree node
installed on each tree. The signals were then sent from each tree via
other tree nodes to a master node using a routing protocol similar to
the directed diffusion method [10]. For a large tree, in order to
overcome the path loss on the tree, it was possible to install several
tree nodes on the tree. Each tree node oversaw a number of sensor nodes
and communicated via other tree nodes to the master node. In order to
keep the cost low, readily available wireless communication modules from
local market were used for both sensing and communications. The
operating frequency was 2.45 GHz. An experiment was set up to measure
the path loss on a tree and a link budget was calculated for the system
using a TRW2.45 communication module operating at 2.45 GHz. It had an
output power of +4 dBm and a sensitivity of -80 dBm. From the
measurement, the path losses at various positions on the tree from the
stem 1.5 m above ground were around 65 dB. It was evident that a
sufficient margin of 27 dB was obtained.
[FIGURE 2 OMITTED]
[FIGURE 3 OMITTED]
4. Experimental Results
A dual-polarization coupled patch sensor was designed as detailed
above. It was equipped with two RF switches as shown in Figure 4. The
first RF switch (SW01) was connected with the transmitting patch and a
monopole antenna for communications. The TRW2.45 was used for
transmitting 2.45 GHz to the sensor and the monopole whereas the second
RF switch (SW02) was connected to a MAXIM MAX4004 power detector and the
parallel and perpendicular polarization ports of the receiving patch.
After converted to digital signal by a 12 bit ADC, the signal was input
into an MCS-51 microcontroller which was used for controlling RF
switches. Communications on a tree was tested on a 20-meter high Durian
tree at Nakornsrithamarat, Thailand, in July 2009.
Eight sensors were utilized for measuring the mutual coupling of
eight fruits of the same size. Measurements were conducted during 10 am
to 12 pm every day since the humidity was low and the rain was not as
heavy as that in the other periods of the day. The averaged results of
coupling voltage of parallel and perpendicular polarization from the ADC
throughout the data collecting period are plotted in Figure 5. It is
apparent that on the 94th day after full-bloom, the coupled voltage
increased from 1.9 V and 2.2 V for the parallel polarization and
perpendicular polarization, respectively. It reached the maximum on the
97th day where the parallel polarization and perpendicular polarization
were 2.15 V and 2.4 V, respectively. The coupled signals then decreased
gradually until the minimum was reached on the 100th day. Then, it
happened to increase again. Please note that, if only the parallel or
perpendicular polarization was used, there might be an error since the
coupled voltage for the parallel polarization was increasing while that
for the perpendicular polarization was fluctuating, probably due to
environmental changes. During the 94th-97th days, the variations in the
parallel polarization were quite large compared to those in the
perpendicular polarization. On the other hand, during the 100th-104th
days, the parallel polarization was increasing whereas the perpendicular
polarization was decreasing and then increasing slightly. When the
results of the parallel and perpendicular polarizations were averaged,
the measurement reliability increased. The averaged results of both
polarizations provided more reliable results than that from each single
polarization measurement. Together, the maximum and minimum coupled
signals on the 97th and 100th days after full-bloom were a good
indicator of the maturity stage of Durian.
[FIGURE 4 OMITTED]
[FIGURE 5 OMITTED]
During this data collecting period, three Durian fruits of the same
age were sampled every three days to investigate their physical
properties, i.e., starch concentration, reducing sugar concentration,
color, etc. This information was used to evaluate the maturity stage of
Durian by comparing it with the conventional method in [2]. The
horizontal axis of Figure 5 shows the maturity stage corresponding to
the days after full-bloom whereas the vertical axis shows the coupling
voltage from the ADC and % dry-weight of Durian pulp. Figure 5 shows
that the % dry-weight of 29% and 32% corresponds to the days when the
maximum and minimum coupling took place. It also shows that the maturity
stage of 60%-70% was indicated by the % dry-weight of 29%-32%. Hence,
the transition of the coupling voltage from the ADC from the maximum to
the minimum represents the maturity stage of 60%-70%.
It is recommended that sensors be installed on the Durian fruits
that are representatives of Durians of the same season around the 94th
day after full-bloom, which corresponds to 50% maturity stage. Coupling
voltage should be measured every day. The maximum voltage indicates the
maturity stage of 60%. The results of this experiment show that the
maturity stage increases from 60% to 70% within three days. It should be
pointed out that a harvesting date can also be predicted by calculating
the slope of the measured mutual coupling. And when the slope changes
from negative to positive, it is time that Durians should be harvested.
According to the conventional method of counting the days after
full-bloom, Durian is at the maturity stage of 75% around 120 days after
full-bloom [1]. However, Figure 5 shows that, actually, on the 101st
day, the maturity stage was 75% already. Therefore, under some
conditions, it is quite possible that fruit growers may harvest Durian
at the wrong time. The proposed sensor may be a good alternative for
pre-harvesting quality control of this fruit.
5. Conclusions
A dual polarization coupled patch antenna for sensing maturity
stage of Durian was proposed. The averaged mutual coupling signal can
indicate the appropriate time for harvesting Durians, which is
absolutely essential for controlling its quality for export. This work
demonstrates an application for Durian quality control. Future work will
apply to other climacteric fruits.
doi: 10.4236/wsn.2011.39034
6. Acknowledgments
This work was supported by the National Research Council of
Thailand (NRCT) and the Thailand Research Fund (Grant No.RTA5180002).
The authors appreciate P. Sooksumrarn, T. Limpiti, P. Yoiyod, P. Leekul
and T.Tantisoparak for implementing the system and data collection.
7. References
[1] P. Lertrut, B. Rattanachinakorn, S. Vijitranont, S. Suthiarom,
S. Pavenakarn, H. Hirunpradit, S. Juntarapunnik, and S. Sluckpetch and
Durian, Ministry of Agriculture and Cooperative, No.13/2547, 2003.
[2] K. Kalayanamitra, "Evaluation and Classification of Durian
Fruit Maturity and its Relationship with Chemical Constituents,"
Thai Journal of Agricultural Science, Vol. 38, No. 1-2, 2005, pp. 45-54.
[3] S. Ketsa and T. Dangkanit, "Firmness and Activities of
Polygalacturonase, Pertinesterase, [beta]-Galactosidase and Cellulose in
Ripening Durian Harvested at Different Stages of Maturity,"
Scientia Horticulture, Vol. 80, 1999, pp. 181-188.
[4] T. Rutpralom, K. Chamnongthai and P. Kumhom,
"Non-destructive Durian Maturity Determination by Using Microwave
Free Space Measurement," IEEE International Symposium on Circuits
and Systems, Island of Kos, 21-24 May 2006, pp. 441-444.
[5] S. Suttapa, J. Varith, M. Krairiksh, C. Noochuay and J.
Phimpimol, "Microwave Sensor Response in Relation to Durian
Maturity," Proceedings of the 5th CIGR Section VI International
Symposium on Food Processing and Monitoring Technology in Bioprocesses
and Food Quality Management, Potsdam, 31 August-2 September 2009.
[6] T. Motwani, K. Seetharaman and R. C. Anantheswaran,
"Dielectric Properties of Starch Slurries as Influenced by Starch
Concentration and Gelatinization," Carbohydrate Polymers, Vol. 67,
No. 1, 2007, pp. 73-79. doi:10.1016/j.carbpol.2006.04.017
[7] X. Liao, G. S. V. Raghavan, J. Dai and V. A. Yaylayan,
"Dielectric Properties of [alpha]-D-Glucose Aqueous Solutions at
2450 MHz," Food Research International, Vol. 36, No. 5, 2003, pp.
485-490. doi:10.1016/S0963-9969(02)00196-5
[8] CST Microwave Studio User's Manual, 2006.
[9] M. Krairiksh, J. Varith, A. Kanjanavapastit, C.
Phongcharoenpanich, A. Thanachayanont, P. Sirisuk and M.
Chongcheawchamnan, "Microwave Sensor for Durian Inspection,"
Proceedings of 2009 IEEE International Conference on Antennas,
Propagation and Systems, Johor, 3-5 December 2009, pp. 221-1-221-4.
[10] C. Intanagonwiwat, R. Govindan and D. Estrin, "Directed
Diffusion: A Scalable and Robust Communication Paradigm for Sensor
Networks," Proceedings of the 6th Annual International Conference
on Mobile Computing and Networking, Boston, 6-11 August 2000, pp. 56-67.
Monai Krairiksh (1), Jatuphong Varith (2), Apichan Kanjanavapastit
(3)
(1) Faculty of Engineering, King Mongkufs Institute of Technology
Ladkrabang, Bangkok, Thailand
(2) Department of Agricultural and Food Engineering, Faculty of
Engineering and Agro-Industry, Maejo University, Chiangmai, Thailand
(3) Department of Telecommunication Engineering, Faculty of
Engineering, Mahanakorn University of Technology, Bangkok, Thailand
E-mail: kkmonai@kmitl.ac.th, jatupong@mju.ac.th, apichan@mut.ac.th
Received August 2, 2011; revised August 26, 2011; accepted
September 5, 2011