Vol 14 No 3 - November 2014

ISSN: 1694-2507 (Print)
ISSN: 1694-2108 (Online)
International Journal of Computer Science
and Business Informatics
(IJCSBI.ORG)
VOL 14, NO 3
NOVEMBER 2014

Table of Contents VOL 14, NO 3 NOVEMBER 2014
A Review of Efficient Information Delivery and Clustering for Drip Irrigation Management using WSN.1
S. R. Boselin Prabhu, Dr. S. Sophia and A. Inigo Mathew
Dual Hybrid Algorithm for Job Shop Scheduling Problem ........................................................................ 14
Do Tuan Hanh, Vu Dinh Hoa and Nguyen Huu Mui
CThe Comparative Analysis of Power Optimization in Clustered Sleep Transistors................................ 25
M. Divya Sree, Y. Kranthi Kiran and Vijaya Vardhan Kancharla
Cyber Crimes Incidents in Financial Institutions of Tanzania ................................................................... 37
Edison Wazoel Lubua (PhD)
IJCSBI.ORG

International Journal of Computer Science and Business Informatics
IJCSBI.ORG
ISSN: 1694-2108 | Vol. 14, No. 3. October/November 2014 1
A Review of Efficient Information
Delivery and Clustering for Drip
Irrigation Management using WSN
S. R. Boselin Prabhu
Assistant Professor, Department of Electronics and Communication Engineering
SVS College of Engineering, Coimbatore, India.
Dr. S. Sophia
Professor, Department of Electronics and Communication Engineering
Sri Krishna College of Engineering and Technology, Coimbatore, India.
A. Inigo Mathew
UG Student, Department of Electronics and Communication Engineering
SVS College of Engineering, Coimbatore, India.
Abstract
In this paper a concise outline for improving throughput and average end to
end delay of information gathered from the agriculture field for precision
agriculture, using a distributed clustering mechanism has been outlined.
This algorithm offers a throughput of 180 bits/seconds. Besides delivery of
water level information packets/signals to base station, it also computes a
threshold as well as calculates the values based on transmission range. This
overall computational mechanism helps us to build a robust mechanism for
delivery of information to the base station, thus reducing the packet loss. A
wireless sensor network is a system consisting of sensor nodes, which
incorporates a radio frequency (RF) transceiver, sensor, microcontroller and
a power source. Recent advances in wireless sensor networking technology
have led to the expansion of low cost, low power, multifunctional sensor
nodes. Sensor nodes facilitate environment sensing together with data
processing, are able to network with other sensor systems and exchange data
with external users. Sensor networks are used for a variety of applications
including wireless data acquisition, environmental monitoring, irrigation
management, safety management and in many other areas. In this paper, a
review of incorporating a distributed clustering algorithm for an agricultural
application has been elaborated.
Keywords: Wireless sensor network, sensor node, distributed clustering,
water level monitoring, drip irrigation, precision agriculture.

IJCSBI.ORG
1. INTRODUCTION
A normal wireless sensor network (WSN) protocol consists of layers like
application layer, transport layer, network layer, data link layer, physical
layer, power management plane, mobility management plane and the task
management plane. Currently two standard technologies available for
wireless sensor networks are Zigbee, and Bluetooth both operates in
Industrial Scientific and Medical (ISM) band of 2.4 GHz, which provides
license free operation for scientific research and study purpose. In general,
increase in frequency increases bandwidth which allows high speed data
transmission. In-order to decrease the power requirement distance between
sensor nodes has to be reduced as possible [13-16]. Multi-hop
communication over the ISM will be promising technique in WSN, since it
consumes less power than traditional single-hop communication. A sensor is
intelligent to convert physical or chemical readings gathered from the
environment into signals that can be calculated by a system. A multi-sensor
node is intelligent to sense several magnitude values in the same device. In a
multi-sensor, the input variables may be temperature, fire, motion detection
sensors, infrared radiation, humidity and smoke. A wireless sensor network
could be a functional architecture for the deployment of the sensors used for
fire detection and verification.
The most imperative factors for the quality and yield of plant growth are
temperature, humidity, light and the level of nutrition content of the soil in-
addition with the carbon dioxide in the surrounding atmosphere. Constant
monitoring of these ecological variables gives information to the cultivator
to better understand, how each aspect affects growth and how to achieve
maximal crop productiveness. The best possible greenhouse climate
modification can facilitate us to advance productivity and to get remarkable
energy saving, predominantly during the winter in northern countries. In the
past age band, greenhouses it was enough to have one cabled dimension
point in the middle to offer the information to the greenhouse automation
system. The arrangement itself was typically simple without opportunities to
supervise locally heating, light, ventilation or some other actions which
were affecting the greenhouse interior climate. The archetypal size of the
greenhouse itself is much larger than it was before, and the greenhouse
facilities afford several options to make local adjustments to light,
ventilation and other greenhouse support systems. However, added
measurement data is also needed to put up this kind of automation system to
labor properly. Increased number of measurement points should not
dramatically augment the automation system cost. It should also be probable
to easily alter the location of the measurement points according to the

IJCSBI.ORG
particular needs, which depend on the definite plant, on the possible changes
in the external weather or greenhouse arrangement and on the plant
placement in the greenhouse. Wireless sensor network can form a helpful
part of the automation system architecture in contemporary greenhouses
constructively. Wireless communication is used to transmit the
measurements and to establish communicate between the centralized control
and the actuators located to the different parts of the greenhouse. In highly
developed WSN systems, some parts of the control system has to be
implemented on the field in a distributed so that local control loops can be
created. WSN is fast, cheap and easy compared to cabled network systems.
Moreover, it is easy to relocate the measurement points when needed by
immediately moving sensor nodes from one location to another within a
communication range of the coordinator gadget. If the greenhouse
vegetation is high and dense, the small and light weight nodes can be
hanged up with the branches. WSN is easy to maintain, relatively
inexpensive and trouble-free. The only other expense occur only when the
sensor nodes run out of batteries (figure 1) and the batteries have to be
charged or replaced. Lifespan of the battery can be increased to several
years if a proficient power saving algorithm is applied. In this work, the
very first steps towards the wireless greenhouse automation system by
building a wireless measuring arrangement for that purpose is taken and by
testing its feasibility and reliability with a straightforward setup [2, 3].
Figure 1: Various blocks of a wireless sensor node
These greenhouse parameters are generally essential in the case of
precision agriculture [4, 5]. In this paper, a mechanism for cluster formation

IJCSBI.ORG
for drip irrigation system for precision agriculture has been elaborated.
Keeping aside these greenhouse parameters which are essential for
cultivation, methodology for cluster formation in drip irrigation system shall
be discussed. A distributed clustering mechanism has been employed for
cluster formation [7, 8]. Also, few parameters that is necessary for wireless
sensor node deployment, for agricultural application has been discussed in
the subsequent sections.
2. PLANTATION MANAGEMENT USING WSN
For developing an efficient system for agricultural system management, the
foremost inputs to the system will be the availability of accurate data’s
(figure 2) like soil properties, agronomic, physicochemical parameters,
atmospheric data, etc., Data collection can be made flexible on a day-to-day
basis or even hourly basis based on the need. Normal laboratory analysis of
above mentioned parameters and manual decision-making take a long time
even with the most sophisticated analytical techniques. Most of the samples
have to be brought from the field to laboratories to analyze most of the time.
By the time the results are available and decisions are taken, the conditions
of the farm may change which results in inappropriate decision. Quick and
quality decision-making at the farm level will enhance agricultural
productivity and quality manifold which further needs accurate and real time
properties of the field.
Figure 2: WSN in Precision agriculture

IJCSBI.ORG
Computer-aided decision-making process can handle and analyze several
input parameters at the same time which involves large databases.
Monitoring of physical and environmental parameters including soil
moisture, soil temperature, nutrition content of the soil, leaf temperature,
relative humidity, air temperature, rainfall, vapor pressure and availability of
sun light is done through a wireless sensor network. WSN is comprised of
distributed sensors to monitor physical, chemical and environmental
conditions. WSN integrates sensors, wireless and processing tools whoch
are capable of sensing different physical parameters without any loss in
sensed data accuracy. The parameters are processed and wirelessly
transmitted to a centralized data storage system through a gateway from
where they may be remotely accessed and analyzed by the user [17]. The
system architecture of a WSN-based system consists of different sensors
interfaced to electronic hardware with a suit of data processing tools. The
electronic hardware is also equipped with wireless communication modules
which allows sensed data to be modulated and transmitted in accordance to
a selected protocol. These hardware nodes are called motes in WSN. Each
of these motes are interfaced with some set of wireless sensors depending
on the applications interest. The sensors may be programmed to sense in a
continuous or discrete manner. The needed parameters for precision
agriculture have been enumerated in the next section.
3. WSN IN PRECISION AGRICULTURE
WSN technology can broadly be applied into three areas of agriculture: a)
Fertilizer control, b) Irrigation management and c) Pest management. The
sensors are interfaced with the mote to sense parameters like temperature,
relative humidity, solar radiation, rainfall, wind speed and direction, soil
moisture and temperature, leaf wetness and soil pH. The following
parameters should essentially be accounted before deploying sensor nodes
in a wireless sensor field.
A. The Greenhouse Environment
A contemporary greenhouse can consist of copious parts which contain their
own confined climate variable settings. As a result, a quantity of
measurement points is also needed. This group of environment is
demanding both for the sensor node electronics and for the short-range
IEEE 802.15.4 wireless network, in which communication choice is greatly
longer in open environments [9, 10].

IJCSBI.ORG
B. Sensors
Speedy response time, squat power consumption and tolerance beside
moisture climate, relative humidity and temperature sensor forms an idyllic
preference and explanation for the greenhouse environment.
Communication among sensor nodes can be carried out by IIC interface.
Luminosity can be measured by light sensor, which converts light intensity
to equivalent voltage. Unstable output signal is handled by low-pass filter to
acquire exact luminosity values. CO2 measuring takes longer time than
other measurements and CO2 sensor voltage supply have to be within little
volts. The carbon dioxide assessment can be read from the ensuing output
voltage. Operational amplifier raises the voltage level of weak signal from
the sensor.
C. Greenhouses
A greenhouse is a pattern covering the ground frequently used for growth
and progress of plants that will revisit the owner’s risk, time and capital.
This exhibit is mounted with the purpose of caring crop and allowing a
better environment to its advancement. This defend is enough to guarantee a
higher quality in production in some cases. However, when the chief idea is
to achieve a superior control on the horticulture development, it is necessary
to examine and control the variables that influence the progress of a culture.
The chief role of a greenhouse is to offer a more compassionate
environment than outside. Unlike what happens in customary agriculture,
where crop conditions and yield depend on natural resources such as
climate, soil and others, a greenhouse ought to promise production
independent of climatic factors. It is noteworthy to view that even though a
greenhouse protects crop from peripheral factors such as winds, water
excess and warmth it may root plentiful problems such as fungus and
extreme humidity [1].
D. Temperature
Temperature is one of the main key factors to be monitored since it is
unswervingly related to the development and progress of the plants. For all
plant varieties, there is a temperature variety considered as a best range and
to most plants this range is comparatively varying between 10ºC and 30ºC.
Among these parameters of temperature: intense temperatures, maximum
temperature, minimum temperature, day and night temperatures, difference
between day and night temperatures are to be cautiously considered.
4TH WORLD CONGRESS ON COMICULTUR

IJCSBI.ORG
E. Water and Humidity
An additional significant factor in greenhouses is water. The absorption of
water by plants is associated with the radiation. The deficient in or low level
of water affects growth and photosynthesis of these plants. Besides air, the
ground humidity also regulates the development of plants. The air humidity
is interconnected with the transpiration, while the ground humidity is linked
to water absorption and the photosynthesis. An atmosphere with tremendous
humidity decreases plants transpiration, thereby reducing growth and may
endorse the proliferation of fungus. On the other hand, crouch humidity
level environments might cause dehydration [6].
F. Radiation
Radiation is an elementary element in greenhouse production and sunlight is
the key starting place of radiation. It is an imperative component for
photosynthesis and carbon fixing. Momentous radiation features are
intensity and duration. The radiation intensity is linked to plant development
and the duration is explicitly associated with its metabolism.
G. CO2 Concentration
CO2 is an indispensable nutrient for the plant development, allowing the
adaptation of carbon. The carbon retaining process occurs through the
photosynthesis when plants take away CO2 from the atmosphere. During
photosynthesis, the plant use carbon and radiation to produce carbohydrate,
whose purpose is to permit the plant development. Therefore, an enriched
air environment should add to plant growth, but it is also vital to note that an
intense carbon level may turn the environment poisonous.
4. THE DRIP IRRIGATION AUTOMATION SYSTEM
Conventional irrigation methods like overhead sprinklers and flood-type
feeding systems usually lead to wetness in and around the lower portion of
the leaves and stems of the plants. The entire soil surface becomes saturated
and often stays wet even after the irrigation is completed. Such condition
induces infections in leaf mold with fungus growth. Flood-type methods
consume a large volume of water, but the area between crops remains dry
and receives moisture only by the incidental rainfall. The drip irrigation
technique slowly applies some small amount of water to the plant's root
zone. Water is supplied frequently, or on daily basis to maintain favorable
soil moisture condition and prevents moisture-stress in the plant with proper

IJCSBI.ORG
use of water resources. WSN based drip irrigation system is a real time
feedback control system which continuously monitors and controls all the
activities of the drip irrigation system.
A typical drip irrigation system consist of delivery systems, filters, pressure
regulators, valve or gauges, chemical injectors, measuring
sensors/instruments and controllers. WSN framework installed in the field
could gather various physical and chemical parameters related to irrigation
and plant health. These includes various temperature parameters like
ambient temperature, ambient humidity, soil temperature, drip water
temperature, soil moisture, soil pH, water pressure, flow rate, amount of
water, energy calculation, chemical concentration and water level. The data
is sent to the central server wirelessly through the motes and gateways [11,
12]. Based on the data received, the central server generates necessary
control signals, which are routed to the respective controllers through
control buses which enables implementation of closed-loop automation for
the drip irrigation system. The main function of the system is to enable
switching on and off of the motors and gauges remotely. The system also
ensures that all the devices are checked for fault and only then the motor is
started.
5. ALGORITHM FOR DRIP IRRIGATION SYSTEM
In this paper, a mesh topology in which sensor nodes are placed in the farm
area have been properly reviewed and adopted [17]. Sensors in this topology
are mobile where as the base station is stationary and it collects the data
from sensor nodes and process them. Efficient clustering of sensors in the
wireless sensor field is considered as the basic operation in this research
work. This work elaborates how to forward the sensed data to the base
station effectively. For this purpose the farm area consisting of deployed
WSN nodes has been formed initially. Now set the position of sensor and
sink nodes in the farm and the monitoring station location. Set the
transmission range for each node. Now for each node, calculate distance
from: node to node, node to sink and node to forwarding node. Also
calculate: Angle a and predict min angle for next route based on fuzzy time
series, if the current angle a is available as predicted, continue with path
[find possible node (x, y)] or else hold packet for limited time. If
connections (i, j) = 1 i.e., there is a link based on transmission range, send
the packet information i.e., water level information and some other essential
details about field of interest. The packet reaches to the sink node and gets
stored there. Else connections (i, j) = infinity, end the structure. Therefore

IJCSBI.ORG
values of sensor nodes are stored in sink nodes. These sink node sends the
stored values to monitoring station. On the basis of water level information,
the switch is made on/off.
6. RESULTS AND DISCUSSIONS
Here we obtain information of water level by using wireless sensor nodes.
On the x-axis, we plot the number of readings of one sensor whereas on the
y-axis we plot the magnitude of water level (figure 3). Here we have taken
the ten water level readings of one sensor and the corresponding magnitude
of water level readings. The next step is to deliver the obtained water level
information to the forwarding node and then to the sink node (base station).
This process is known as Packet Delivery. In the percentage form it is
known as Packet Delivery Ratio (PDR) as expressed by equation 1.
Figure 3: Water Level Information
The greater value of packet delivery ratio means the better performance of
the protocol. The PDR is inversely proportional to Packet Lost Ratio (PLR)
as expressed in equation 2. Packet lost ratio is the total number of packets
dropped during the transmission.
100
Number of packets received
Number of packets sent
PDR


   1

IJCSBI.ORG
 
100
Number of packets lost
Number of packets lost Number of packets received successfully
PLR

 
 
… (2)
7. SIMULATION RESULTS
By using fuzzy time series algorithm, we get the initial throughput of 46
bits/sec, which is represented by blue line (figure 4). It is known as Fuzzy
THR. When Nuppy algorithm is employed, we get the throughput between
48 bits/sec. This is referred to as Nuppy THR, represented by red lines. In
this comparison, with 200 sensor nodes employed for throughput
calculation, we can clearly see that, the throughput of fuzzy time series
increases with increase in the number of wireless sensor nodes.
Figure 4: Throughput versus number of nodes
Figure 5, shows the Nuppy delay being given by blue line, whereas
fuzzy delay is represented by red line. In case of nuppy delay, the average
end to end delay is found to be between 6.5 msec to 3 msec. In Fuzzy
algorithm, the average end to end delay is found to occur between 3.8 msec
to 1.8 msec. Hence by the employment of fuzzy time series algorithm, better
average end to end delay could be attained which could be understood from
the results. Hence, in improved Fuzzy based algorithm, reliable water level
information could be attained in comparison with the Nuppy algorithm.

IJCSBI.ORG
8. CONCLUSION AND FUTURE WORK
Conventional Flood-type methods consume large volume of water, but the
area between crop rows remains dry and receives moisture only from the
incidental rainfall, whereas the drip irrigation technique slowly applies a
small amount of water to the plant's root zone. Therefore by using the fuzzy
based algorithm in wireless sensor drip irrigation technique, the wastage of
water can be controlled. Also by using wireless sensor network, laborers are
not essentially needed. In this paper, water level information is achieved by
the usage of 200 wireless sensor nodes. The comparison between the two
clustering mechanisms, clearly gives an idea to employ fuzzy time series
algorithm, when throughput is mandatory for a drip irrigation system. When
the number of sensor nodes is increased, there is a large amount of power
consumption by sensors to deliver the water/packet information to the
monitoring station. Thus it is mandatory to minimize the power
consumption by using some power control techniques. In future, an energy
efficient protocol will be employed in the same mechanism, to maximize the
network lifetime.
9. ACKNOWLEDGMENTS
Our sincere gratitude to the management of SVS Educational Institutions
and my Research Supervisor Dr. S. Sophia who served as a guiding light to
come out with this amazing research work.

IJCSBI.ORG
REFERENCES
[1] Iqbal Singh, Meenakshi Bansal, (2011), "Monitoring water level in agriculture using
sensor networks", International Journal of Soft Computing and Engineering, Volume 1,
Issue 5.
[2] Qingshan S, Ying L, Gareth D, Brown D, (2004), "Wireless intelligent sensor networks
for refrigerated vehicle", IEEE 6th Symposium on Emerging Technologies Mobile and
Wireless Communication, Shanghai, China.
[3] Mahir Dursun, Semih Ozden, (2011), "A wireless application of drip irrigation
automation supported by soil moisture sensors", Scientific Research and Essays, Volume 6.
[4] Baggio A, (2005), "Wireless sensor networks in precision agriculture", Proceedings of
the Workshop on Real-World Wireless Sensor Networks, pp. 50-51.
[5] Kim Y, Evans R.G, Iversen W.M, Pierce F.J, (2006), "Instrumentation and control for
wireless sensor network for automated irrigation", ASAE Annual International Meeting.
[6] Steven R, Evett R, Troy Peters, Terry A.H, (2006), "Controlling water use efficiency
with irrigation automation: cases from drip and center pivot irrigation of corn and soybean",
Southern Conservation Systems Conference, Amarillo TX.
[7] SRB Prabhu, S.Sophia, (2011), “A survey of adaptive distributed clustering algorithms
for wireless sensor networks”, International Journal of Computer Science and Engineering
Survey, Volume 2, Number 4, Pages 165-176.
[8] SRB Prabhu, S.Sophia, (2013), “Real-world applications of distributed clustering
mechanism in dense wireless sensor networks”, International Journal of Computing,
Communications and Networking, Volume 2, Number 4.
[9] Evans R.G, (2000), “Controls for Precision Irrigation with Self-Propelled Systems",
Proceedings of the 4th Decennial National Irrigation Symposium.
[10] Callaway E.H, (2004), "Wireless sensor networks: architectures and protocols”,
Auerbach Publications: New York, NY, USA.
[11] SRB Prabhu, S.Sophia, (2013), “Mobility assisted dynamic routing for mobile wireless
sensor networks”, International Journal of Advanced Information Technology, Volume 3,
Number 3.

IJCSBI.ORG
[12] SRB Prabhu, S.Sophia, (2012), “A research on decentralized clustering algorithms for
dense wireless sensor networks”, International Journal of Computer Applications, Volume
57.
[13] Evett S.R, Howell T.A, Schneider A.D, Upchurch D.R, Wanjura D.F, (1996),
"Canopy temperature based automatic irrigation control", Proceedings of the International
Conference on Evapotranspiration and Irrigation Scheduling, San Antonio, TX..
[14] Rajaravivarma V, Yang Y, Yang T, (2003), “An overview of wireless sensor network
and applications”, Proceedings of 35th
South Eastern Symposium on System Theory.
[15] Wanjura D.F, Upchurch D.R, Mahan J.R, (1992), "Automated irrigation based on
threshold canopy temperature", Transactions of the ASAE, Volume 35.
[16] Ning Wang, Naiqian Zhang, Maohua Wang, (2006), “Wireless sensors in agriculture
and food industry - recent development and future perspective”, Computers and Electronics
in Agriculture, Volume 5, pp. 1-14.
[17] Marron P.J, Minder D, Consortium E.W, (2006), “Embedded research roadmap”,
Information Society Technologies, Germany.
This paper may be cited as:
Prabhu, R. B., Sophia, S. and Mathew, A. I., 2014. A Review of Efficient
Information Delivery and Clustering for Drip Irrigation Management using
WSN. International Journal of Computer Science and Business Informatics,
Vol. 14, No. 3, pp. 1-13.

IJCSBI.ORG
Dual Hybrid Algorithm for Job Shop
Scheduling Problem
Do Tuan Hanh
Hanoi National University of Education
136 Xuân Thuy Cau Giay Hanoi Vietnam
Vu Dinh Hoa
Nguyen Huu Mui
Abstract
This paper presents a dual hybrid method to solve job shop scheduling problem
(JSP). We use a genetic algorithm (GA) combined with the dual hybrid approach to
obtain a new genetic algorithm (GTD-GA). The algorithm proposed new hybrid
approach called dual hybrid, and new data organization with the purpose of
maximizing the GT algorithm. They're building program for the experimental
schedule finding maximum job shop problem. In order to prove the effectiveness of
the algorithm, we ran on the standard given by Muth and Thompson and compare
with the results of Yamada used GT-GA for JSP. Moreover, we also compare them
with the local search genetic algorithm of M. Kebabla, L. H. Mouss for the job shop
problem.
Keywords: Jobshop Scheduling, Schedule, Genetic Algorithm.
1. INTRODUCTION
Job shop scheduling problem (JSP) is one of the well-known classic problems. There
are many approaches to solve it and the first work on this field is perhaps the paper
of Fiedman and Akers [3].

IJCSBI.ORG
JSP general statements for a set n jobs {Jj}1 ≤ j ≤ n and a set of m machines {Mi}1 ≤ i ≤
m must satisfy the following conditions:
1. At the same time each machine can process only a single job.
2. The set of n jobs {Jj}1 ≤ j ≤ n has processed on all m machines {Mi}1 ≤ i ≤ m.
3. Each job must be processed on each machine in a given order of
operations. The sequence of operations performed each task in turn on the
machine called sequential technology.
4. The handling of a job Jj at the machine Mi is called the operation Oij.
5. Each operation Oij executed must be uninterrupted on the given machine
Mi.
6. The beginning processing time and the finishing time of operation Oij are
denoted respectively by T_Sij and by T_Fij. The processing time operation
Oij is denoted by Time_jobij.
7. The time to complete the processing of all jobs is called makespan and is
denoted by Time_Fmax.
The requirement of JSP problem is to determine a schedule (order processing jobs on
each machine) with the possible smallest Time_Fmax.
Example: A JSP for 4 machines and 4 jobs including sequential technology of
machines for each job with the processing time of each job on each machine (in
parentheses) is given in Table 1.
Table 1. JSP 4 jobs, 4 machines
Jobs Machine (processor time)
1 1(1) 2(4) 3(3) 4(15)
2 2(3) 3(5) 4(12) 1(7)
3 4(8) 1(9) 2(16) 3(10)
4 2(11) 3(6) 1(2) 4(13)

IJCSBI.ORG
Job shop scheduling problem is a NP-hard problem and is one of the most difficult
combinatorial optimization problems. JSP problem can be solved with a polynomial
algorithm only in some special cases. For n = 2, Akers [2] proved that the problem
of finding JSP can be solve with a polynomial time algorithm using shortest path
algorithm. Using a method of Kravchenko and Sotskov [11], Brucker [5] give a
polynomial time algorithm for the JSP with m = 2 machines and n = k (k is a
constant) jobs. Except for some special cases mentioned above, the remaining cases
of JSP are NP-hard. This is the result of a research of Shaklevich Sotskov [15]. Table
2 summarizes the complexity of JSP.
Table 2. The computational complexity of JSP
Number of m machines
2 3 Constant Depending
on the
The
number n
jobs
2 P P P P
3 P P P NP-hard
Constant P P NP-hard NP-hard
Depending
on the
P NP-hard NP-hard NP-hard
JSP was published first time in 1955. A statement for this problem can be found in a
paper of Akers and Friedman. Followed them are the studies of Bowman (1959) and
Wagner (1959) in which there was only one job that can be handled on maximal 3
computers. The year 1963 marked an important milestone in JSP when Fisher and
Thompson [12, 7] studied a JSP with 10 jobs and 10 machines. This test was
considered to be one of the most intractable of JSP because the optimal solution was
found ever after more than 20 years later. They proposed GT algorithm as active
schedules. The result of H. Fisher and GLThompson [7] can be found in the book
"Industrial Scheduling", edited by JF-Muth and GL- Thompson [12] in 1963.
The method of solving the problem as 'branching method approach' introduced by
GH Brooks and CR White [4] who pioneered solving the job shop. Finally, in 1985,
E. Carlier and Pinson [6] succeeded in solving optimization problems Mt10 using
algorithms branch access. Finding optimal solution for JSP with unsightly test is a
major challenge because it is a NP-hard problem. Thus, approximate methods have
been developed. One of the approximate methods considered to be most effective is
genetic algorithm.

IJCSBI.ORG
John Holland [9] is considered as the founder of genetic algorithm. GA (Genetic
Algorithms) is characterized by a search strategy based on population and by the
genetic operators: selection, mutation and crossover. Nakano and Yamada [13]
developed the first classical hybrids method and obtained quite good results. Yamada
[16] developed GT-GA algorithm with many significant achievements. In 2012, M.
Kebabla, LH Mouss [10] offered “A local search genetic algorithm for the job shop
scheduling problem”. Recently, Nguyen Huu Mui and Vu Dinh Hoa [14] proposed a
new hybrid method coded calendars by natural numbers and a new crossover
operator combined 3 individual parent to an individual children.
After a period of researching methods to solve the problem, we find a new hybrid
method with new mutations which may be more efficiently. Our innovations are:
1. Reorganize database to reduce leverage calculations suitable grafting and
new mutations.
2. Use the GT algorithm taking good jealous of mutant to a better generation,
with the aim to get earlier convergence than the GT-GA.
3. Apply new crossover method: from 2 individual parents P1, P2 hybridized
according to GT algorithm to F1 and F2 offspring, then proceed hybridized form F1
and F2 to FF1 and FF2. With this new crossover approach, this starting personal
space can be expanded to diverse selection closer to a optimal experience.
2. CONTENT
In what follows, we present a method of organizing new data processing allows new
mutations and new crossover.
2.1. New data encryption
Suppose there are n jobs to be processed on m machines. T matrix is called matrix
sequentially technology. We will encode matrix C where Cij is 1 if the job j is done
on machine i at some time, and Cij is 0 if the job j is not done on machine i. L_T is
the matrix where L_Tij is the next machine which performs the job i, according the
sequential technology, after this job was performed on the j machine. F_T is the
matrix where F_Tij is the machine which performed the job i just before it has to be
performed on the j machine according the sequentially technology.

IJCSBI.ORG
For example is the problem of 4 jobs, 4 machines given in Table 1. T matrix and
transformation matrix C, L_T, F_T are given in Figure 1, respectively:
Figure 1. The matrix T, C, L_T, F_T
2.2. Mutation operator
Operator mutations on individual parent P were carried out by following these steps:
1. Choose a random matrix H where Hij is 0 or 1, apply the GT algorithm to
generate mutant P_C using the matrix H.
2. Create G, the set of the first stage in the sequence of all these technology
jobs, using the matrix C (only if Cij is 1) so that if Oij  G then T_Sij=0; and VT [i]
= 1.
3. Construct the set V_G G satisfying that if Oij V_G, then T_Sij is the
possible minimum value.
4. Construct the set E_G V_G satisfying if Oij E_G, then HiVT[i] =1 and j
= PiVT[i].
5. If E_G ≠ Ø, then choose randomly Oij  E_G. If E_G = Ø, choose Oij
V_G randomly.
6. P_CiVT[i] = j; VT[i] = VT[i]+1; Remove Oij from the set G; if k ≠ 0 (with k
= L_Tij), then add Oik in G.
7. Update to T_Sij if Oij G using the following formula:
r = P_CiVT[i]-1
k = F_Tij
T_Sij = Max( T_Sir + Time_jobir , T_Skj + Time_jobkj)
8. Go to step 3 until all stages are scheduled in mutant P_C.

IJCSBI.ORG
If mutant P_C is less adaptive than individuals P, then we repeat these steps to create
mutant P or stop once reaching the maximum number of mutations (through variable
input level).
For example, individual parent P and H matrix generated mutant P_C in Figure 2:
Figure 2. The P parent and mutant P_C matrix H
Time_Fmax Individual's parent P is: 66
Time_Fmax P_C of mutants is: 48
2.3. GTD Crossover
Operator 2 hybrid created on individual P1 and P2 parents was conducted according
to the following steps:
1. Choose a random matrix H which Hij the value 0 or 1. Using genetic
algorithms get better GT parents P1 and P2 through the matrix H to generate F1 and
F2 offspring. From the F1 and F2 be the application of the GT get good genes of
their parents through the matrix H to generate FF1 and FF2 offspring.
2. Create G is the set of the first stage in the sequence of all these
technologies jobs through the matrix C (Cij = 1 will be included in G), if Oij  G is
T_Sij=0; and VT [i] = 1.
3. Construct set V_G G satisfying if Oij V_G, then T_Sij is the minimum
value that can be obtained.
4. Construct set E_G_F V_G satisfying if Oij E_G, then HiVT[I ] = 1 and j
= P1iVT[i] .
5. If E_G_F ≠ Ø, then choose randomly Oij E_G, then trip to step 8.
6. If E_G_F = Ø, Set E_G_M V_G satisfied if Oij E_G, then HiVT[i ] = 0
and j = P2iVT[i] .
7. If E_G_M ≠ Ø, then choose randomly Oij E_G_M. Conversely, choose
Oij V_G randomly.
8. F1iVT[i] = j; VT[i] = VT[i]+1; Oij removed from the set G; if k ≠ 0 (with k =

IJCSBI.ORG
L_Tij) then add Oik in G.
9. Update to T_Sij if Oij G using the following formula:
r = P_CiVT[i]-1
k = F_Tij
T_Sij = Max( T_Sir+ Time_jobir , T_Skj+ Time_jobkj)
10. Go to step 3 until all phases are scheduled in individual F1 crossovers.
This methodology improves our differences. We produce F2 simultaneously in step 4
with additional conditions: j1=P2iVT[i], with additional conditions in Step 6: j1=P1iVT[i]
and in step 8 with F2iVT[i] =j1. Using F1 and F2 as the parents of FF1 and FF2 to
produce these steps as well as adopted hybrid matrix H. From two individual P1 and
P2 parents we are offering first creation of four new individual at F1, F2 , FF1, FF2
with the same operators number as the creation of our first offspring using GT_GA
algorithm. For example, we give two individual crossover parents P1 and P2 through
the matrix H as shown in Figure 3:
Figure 3. Parents P1, P2 and matrix H
Figure 4. The generated in children F1, F2, FF1, FF2

IJCSBI.ORG
Table 4. Time_Fmax of individual values P1, P2, F1, F2, FF1, FF2
Individual P1 P2 F1 F2 FF1 FF2
Time_Fmax 66 52 48 66 66 50
2.4. Operator selection
Operators for selectively choosing individual n_gen generation t + 1 are
described as follows:
1. Construct intermediate solution set P '(t) applying mutation and
hybridization:
+ Apply mutation operator for P (t) is P1 (t).
+ Apply crossover operator for P (t) is P2 (t).
+ P'(t) = P(t)  P1 (t)  P2(t).
2. Choose one individual with the best adapted from the first generation to
generation t, called Pmin, in the following way:
Choose individual which may be the most adaptable in the set P '(t), called P'min , if t
= 1 then Pmin= P'min. If t > 1 and if P'min have better adaptability than Pmin , then
replace Pmin with P'min..
3. Get n_gen - 1 random individuals remaining in P '(t) according to the
principles of the wheel.
2.5. Genetic Algorithms for JSP
Void GA_JSP ()
{
t  0 ; // t is the number of generations of evolution
Initialize P (t); // GT algorithm
while ( t < n_g ) do //n_g is generations
{
Construct intermediate solution set P '(t) by applying mutation and
hybridization:
+ Apply mutation operator for P (t) to obtain P1 (t) in Section 2.2
+ Apply crossover operator for P (t) to obtain P2 (t) in Section 2.3
+ P'(t) = P(t)  P1 (t)  P2(t)
Selective P (t) from P '(t) in Section 2.4
t = t+1 ;
}
}

IJCSBI.ORG
The correctness of the algorithm is confirmed by the following characteristics:
1. When initializing n_gen individual's initial population we use GT
algorithm to generate a schedule, so each individual children is born with an active
schedule [13] [16].
2. During mutation, mutant individuals are modified by taking a number of
individual genes fathers through the matrix H applying GT algorithm, so it remains
an active schedule.
3. Allow crossovers from two parent individuals P1 and P2 take individual
genes through the parent matrix H applying GT algorithm to generate children F1
and F2 schedule, so each individual children is born by this calendar will be an
active one. From F1 and F2 and through the matrix H obtained simultaneously
applied genetic algorithms GT with calendar children FF1 and FF2 is a schedule
which actively.
4. The algorithm will stop when it runs through all generations.
5. Since the algorithm maintains the best solution in the population, before or
after the selection, so the algorithm is converged to the global optimum which is not
reduced by the properties of the next generation. These results are discussed with
schema theorem and are proved in [8].
3. RESULT AND CONCLUSIONS
Based on the method proposed in this paper, we have designed a program in C++
language to run tests on a PC with a processor Core 2 Dual-speed 2 4GHz. The
performance of the algorithm was analyzed on a set of benchmarks on the job-shop
scheduling problem instances from literature. The size of the benchmark instances
varied from 10 to 20 jobs and from 5 to 20 machines. We consider (FFT6, FT10,
FT20) proposal by Fisher and Thompson [12]; three problems (ABZ5, ABZ6,
ABZ7) generated by Adams, Balas & Zawack [1]. Table 5 presents the result
compared our proposed GTD-GA algorithm with the GT-GA by Yamada [ 8].
Table 5. Comparison of GTD-GA and GTD-GA
Instance(size)
Best generated
solution with GT-
GA
Best generated
solution with
GTD-GA
Best known
solution
FT06 (6  6) 55 55 55
FT10 (10  10) 930 930 930
FT20 (20  5) 1185 1182 1165

IJCSBI.ORG
FT06 proposal for optimal results also FT10 for optimal results are obtained only
after 80 trials. GT-GA algorithm by Yamada [13] [16] proposed FT06 gives optimal
results after 2 trials, FT10 gives optimal results after 100 trials. Table 6 compares the
results of GTD-GA algorithm with the results of GLS algorithm proposed by M.
Kebabla, LH Mouss [10].
Table 6. Comparison of GT-GA and GLS
Instance(size)
Best
generated
solution with
GT-GA
Best generated
solution with
GA in 5
attempts
Best generated
solution with
GLS in 5
attempts
Best
known
solution
ABZ5 (10  10) 1234 1238 1234 1234
ABZ6 (10  10) 943 944 943 943
ABZ7 (15 20) 665 680 667 656
In the tests, ABZ7 GTD-GA algorithm gives better results than GLS algorithm [10].
The comparing test results may indicate that the method we propose has the
following advantages:
1. The organization using the 2 matrices L_T and F_T reduces the numbers
of search operations.
2. When we use a mutation associated with GT algorithms, it searches nearby
so quickly and find a better value mutation father and minimize the number of
calculations.
3. Our first children are 4 instead of getting only one child by the same
parents and with the same number of operators.
Our designed program is tested on a standard data of Muth and Thompson and gives
better results than the latest algorithm GA. However, for problems with large-size
test ABZ7 (15  20), the program has to find the optimal schedule. In the future we
will continue to develop out new methods with higher efficiency.
REFERENCES
[1] Adams, J., Balas, E. and Zawack, D., 1988. The Shifting Bottleneck Procedure for Job Shop
Scheduling, Management Science, Vol. 34, No. 3, pp. 391–401.

IJCSBI.ORG
[2] Akers, S.B., 1956. A graphical approach to production scheduling problems, Operations
Research, Vol. 4, No. 2, pp. 244-245.
[3] Akers, S.B. and Friedman, J., 1955. A non-numerical approach to production scheduling
problems, Operations Research, Vol. 3, No. 4, pp. 429-442.
[4] Brooks, G. H. and White, C. R., 1969. An algorithm for finding optimal or near optimal solutions
to the production scheduling problem, The Journal of Industrial Engineering, Vol. 16, No. 1, pp. 34-
40.
[5] Brucker, P., 1994. A polynomial time algorithm for the two machines Job Shop scheduling
problem with a fixed number of jobs, OR Spektrum, Vol. 16, No. 1, pp. 5-7.
[6] Carlier, J. and Pinson, E., 1989. An algorithm for solving the job-shop problem, Management
Science, Vol. 35, No. 2, pp. 164-176.
[7] Fischer, C. and Thompson, G. L., 1963. Probabilistic learning combinations of local job-shop
scheduling rules, In Industrial Scheduling, Prentice Hall, Englewood Cliffs, New Jersey, pp. 225-251.
[8] Günter Rudolph, 1994. Convergence Analysis of Canonical Genetic Algorithms, IEEE
Transactions on Neural Networks, special issue on evolutionary computation, Vol. 5, No. 1, pp. 96-
101.
[9] Holland, J. H., 1975. Adaptation in Natural and Artificial Systems, University of Michigan Press.
[10] Kebabla, M. and Mouss, L. H., 2012. A local search genetic algorithm for the job shop
scheduling problem, Revue des Sciences et de la Technologie – RST, Vol. 3, No. 1, pp. 61-68.
[11] Kravchenko, S. A. and Sotskov, Y. N., 1996. Optimal makespan schedule for three jobs on two
machines, ZOR - Mathematical Methods of Operations Research, Vol. 43, pp. 233-238.
[12] Muth, J. F. and Thompson, G. L., 1963. Solving Production Scheduling Problems, In Industrial
Scheduling, Prentice Hall, Englewood Cliffs, New Jersey, Ch. 3.
[13] Nakano, R. and Yamada, T., 1991. Conventional genetic algorithm for job shop problems, In
Proceedings of International Conference on Genetic Algorithms (ICGA ’91), pp. 474-479.
[14] Nguyễn Hữu Mùi and Vũ Đình Hoà, 2011. Một thuật toán di truyền lai mới cho bài toán lập
lịch công việc, Kỷ yếu hội nghị khoa học công nghệ quốc gia lần thứ V, pp. 239-249.
[15] Sotskov, Y.N. and Shaklevich, N. V., 1995. NP-hardness of shop-scheduling problems with three
jobs, Discrete Applied Mathematics, Vol. 59, No. 3, pp. 237-266.
[16] Yamada, T., 2003. Studies on Metaheuristics for Jobshop and Flowshop Scheduling Problems,
Japan: Kyoto University, pp. 42-70.
This paper may be cited:
Hanh, D. T., Hoa, V. D. and Mui, N. H. 2014. Dual Hybrid Algorithm for Job Shop
Scheduling Problem. International Journal of Computer Science and Business
Informatics, Vol. 14, No. 3, pp. 14-24.

IJCSBI.ORG
The Comparative Analysis of
Power Optimization in Clustered
Sleep Transistors
M. Divya Sree
ECE Department
Vignans Nirula Institute of Technology and Science for Women
Guntur
Y. Kranthi Kiran
ECE Department
Vignan’s Foundation for Science Technology & Research University
Vadlamudi
Vijaya Vardhan Kancharla
ECE Department
Vignan's Lara Institute of Technology & Science
Vadlamudi
Abstract
This paper concentrates on the various power reduction techniques for clustered sleep
transistors and in particular leakage and dynamic power of the gated circuits published
recently based on different logic styles. All the advantages and applications of the circuits
have discussed with the relevant proofs of results and analytical models. All the circuits are
designed and tested using spice simulation models and the screenshots of the different sleep
transistor circuits based on Multi-threshold CMOS approach have depicted and its
advantages over other methodologies have tested and its spice simulation results has
presented with the required analytical model and the final layout of clustered sleep
transistor has laid down.
Keywords: Dynamic power, Leakage power, MTCMOS, Low power, subthreshold
values.
1. INTRODUCTION
From the last three decades CMOS devices have been scaled for achieving
higher compact density, maximum performance and less power
consumption. About every two years 30% of transistor delay decreasing
with the advancements of the technology and thereby by increasing the
microprocessors performance by twice that of earlier. And at the same time
the supply voltage for the circuits have been maintain at the same level for
meeting the lower power consumption. In order to achieve this higher
performance factors the threshold voltage of the transistors should be

IJCSBI.ORG
reduced to in proportionate to the other model parameters. The
advancements of the portable electronic devices and high compaction of
VLSI circuits, power dissipation give the major challenge to the designers.
And also at the same time due to the advancements of the technological
practice the increase of the leakage current is increasing due to the scale
down of the supply voltage and also the threshold voltage.
In the deep submicron (DSM) regions, the reduction of dynamic power can
be done with the reduction of supply voltages which in turn plays the
important role for the reliability issues by using the advancements of the
technology. The designer has to concentrate more about circuit performance
as the reduction of supply voltages leads altering circuit’s performance. And
the mathematical notation for the subthreshold current can be denoted as
follows.
Tthgs V/)VV(
leakage eII

 0
(1)
The subthreshold leakage current Ileak in sleep mode will be determined by
the sleep transistor and is expressed as follows.
)e(eVe)L/W(CI T
ds
T
tHgs
V
V
V
VV
T
.
sleepoxnleak

 1281 

(2)
Where μn is the N-mobility, Cox is the oxide capacitance, VtH is the high-
threshold voltage (= 500 mV), VT is the thermal voltage = 26 mV, and n is
the subthreshold swing parameter. The reduction of leakage power has
achieved in the standby mode with the proposed methods [1]-[10]. By
varying the substrate bias voltage the threshold voltage can be controlled
which can be observed in Variable Threshold CMOS (VTCMOS) design.
The main advantage in this technique is low threshold voltage for all the
transistors used for design. And it has the two different controlling aspects
which are firstly, the delay variations can be minimized which cause
compensations for fluctuations and secondly, reduction of the leakage
current in the standby mode. These are the two things which need to be
concentrating while implementing this design. Even though these two are
crucial and plays the major advantages for design aspect but it has two
disadvantages which cause the circuit reliability in dilemma. those are can
be observed from its mathematical expression which is as leakage current is
proportional to square root of the substrate voltage the resultant value would
be about large change of effective values and final drawback is it require a
triple-well structure and charge pump circuit for producing substrate voltage
in the VTCMOS. These are the phenomenal causes for the VTCMOS circuit
design.

IJCSBI.ORG
And in contrast to the VTCMOS [4] technique, Multi voltage CMOS
(MVCMOS) has LVT transistors where PMOS gates larger sleep mode
voltages when compare to the NMOS gates which is different in the
VTCMOS design. To reduce the drawbacks caused by these techniques
virtual ground method has taken and by applying the clustering of the LVT
transistors for reduction of the dynamic power and leakage power of the
circuit which in turn reduce the leakage power of the sleep transistors has
achieved. And the logic block diagram of the nMOS sleep transistor can be
shown with the virtual ground and threshold voltage ranges in the Figure1.
Figure 1. Logic block with an nMOS sleep transistor [6].
2. Leakage Reduction Techniques
Figure 2. Charging of the zero output nodes with a pMOS pull-up.
The total power consumption can be computed in a CMOS circuit as the
sum of dynamic power and the static power in the active mode operation.
And due to the standby leakage current the power will be dissipated in the

IJCSBI.ORG
standby mode. Due to the switching power during charging and discharging
transitions of load capacitance and from the non zero rise and fall of input
waveforms which leads to short circuit power consumption. These both the
switching power and short circuit power combination is nothing but the
dynamic power and where as the leakage current through each transistor is
termed as static power of the CMOS circuit.
The dynamic power can be denoted in the mathematical notation as follows
2
ddD fCVP  (3)
ddLEAKLEAK VIP  (4)
3. Power Gating and Multi-Threshold CMOS
By turning off the supply voltage the leakage power can be reduced in the
Standby mode which is the frequently used technique for reduction of the
leakage power in the VLSI circuits. As shown in Figure 1 formation of
virtual ground and virtual power can be formed with the single PMOS and
NMOS transistors by connecting in series to the transistors of each logic
block which can be done with the single transistor while doing the design.
For this technique NMOS transistors are preferred because of their lower on
resistance.
Figure 3. Power gating modeling
From the figure 3 it can be visualized as it consists two portions upper one is
linear RL network and the lower one is linear RC network. By using
constant voltage sources RL network can be excited where as in the RC
network the excitation is with the time-varying current sources and it is
because of the currents drawn from the logic circuits. With the triangle
notation the current profiles can be modeled.
The entire system can be visualized as linear RLC network with the
excitation from the voltage and current sources. This can be done with

IJCSBI.ORG
replacement of vias where the sleep transistors are existed. In the ungated
chips for power network analysis this technique is used widely. This can be
overcome with the modified nodal analysis (MNA) formulation method.
And it has the problem with non linearity of the network which exists
because of the sleep transistors. With the inclusion of sleep transistors the
circuits will be no longer linear. In the spice simulations by performing
power gating noise analysis the above mentioned problem can be removed.
While doing so the sleep transistors are turned ON and OFF to observe the
voltage delivery behaviour.
The simplified power gating circuit can be modeled as shown in figure 4. It
contains pull up and pull down gates with the virtual VDD and virtual
Ground. the sleep transistors are connected into the necessary nodes with
the specified ports for the power rails and inputs and outputs.
Figure 4: Power gating circuit.
The functionality of the circuit is same in the active state as the sleep
transistor is in ON state. The overall phenomenon exists in the Standby
mode. In this state the sleep transistor disconnects the gate from the ground
as it turned into OFF state. Here care should be taken for achieving lower
leakage current which can be obtained with the consideration of larger
threshold voltage for the sleep transistors. Otherwise the sleep transistors
will give high leakage current and causes the power gating less effective if
the sleep transistor has the lesser threshold voltage. And also the additional
power savings can be achieved with the lowering the width of the sleep
transistor in compares to the total width of the pull-down network. For
achieving power gating VTCMOS or Multi Threshold CMOS (MTCMOS)
is used. The best practice for achieving higher power saving is using lower

IJCSBI.ORG
threshold voltage transistors for the logic implementation and using higher
threshold voltage transistors for the sleep transistors.
The expected amount of discharge current value can be obtained with the
technique as shown in figure 4. The switching activity of the gate can be
calculated as the ration of the multiplication product of the probability of the
gate output at zero by the probability of the gate output at one. the switching
activity of the gate would be accountable unless it leads to the negative
impact of the design exists and causes increased sizes for the sleep
transistors making the increased die size also the serious cause from
increase of leakage and dynamic power and finally due to this clustered
formation of gates produce worst case current discharges. By taking CLA
adder typical parameters with the application of all the inputs concurrently
to it can be reduced.
Figure 5: Using one sleep transistor for several gates.
In contrast to the usage of single sleep transistor for overall gate the sharing
of sleep transistor for the circuit can be done by applying the above
technique as shown in the figure 6.
Figure 6: Sleep transistor sharing.

IJCSBI.ORG
4. Leakage Control In Active Mode
Figure 7. Proposed design methodology.
By using the techniques which were discussed earlier the domination of the
leakage power compare to the overall power consumption of the circuit in
Standby mode can be reduced. if that is not the case and the leakage power
is minimum when compare to the overall power consumption the total
power consumption is needs to taken into account where the combination of
switching power dissipation and active mode leakage has to consider and
thereby optimization of circuit is to be considered. And it can be achieved
by considering the following techniques.
 Multiple Threshold Cells
 Long Channel Devices
 Minimum Leakage Vector Method
 Stack Effect-based Method
 Sizing with Simultaneous Threshold and Supply Voltage
Assignment
The proposed clustering algorithm [8] used the approach of constrained
minimization in which for the given timing constraints single cost function
has taken. The iterative row selection scheme has used in the clustering
algorithm. In the iterative row selection process all rows are selected in the
initial stage and are marked as gated rows. in the next step the second row is
taken and rows are progressively eliminated based on the given timing
constraints until it reach the maximal set of rows where the leakage cost
function can be minimized. The output of a gate is represented in the form
of vectors for the discharging currents for the convenience of vector
comparisons. And the same can be evaluated from the flow chart of the
figure 8.

IJCSBI.ORG
Figure 8: Clustering algorithm.
The proposed scheme [9] can be easily implemented to cope with the single
set of sleep transistors. the load dependent delay di can be denoted
mathematically as follows.


)VV(
VC
d
tLdd
ddLi
 (5)
And from the above equation the mathematical terms can be defined as the
load capacitance for the CL and low voltage threshold is about 350 mV for
the VtL and supply voltage Vdd is 1.8 V and α as the velocity saturation
index and which is about 1.3 for the 180nm CMOS technology.
The propagation delay of a gate in presence of the sleep transistor can be
expressed as

)VVV(
VKC
d
tLxdd
ddLi
sleep


2 (6)
And from the above equation the internal terminology is about Vx is for the
virtual rails, K is the proportionality constant. And from the above equation
the internal terminology is about Vx is for the virtual rails, K is the
proportionality constant. The logic gates operates at the virtual supply rails
about the magnitude of Vx from the supply rails on either side but not
operates at the true supply rails (Vdd) in the active mode. From this can
conclude that the effective supply voltage is Vdd-2Vx for the gate.
If the ON current of the sleep transistor in the active mode is IsleepON the
sleep transistor will be in the linear region of operation. And by using the
basic device equations the drain to source current of the sleep transistor in

IJCSBI.ORG
the linear region for the sleep transistor can be denoted in the mathematical
notation as
)
V
V)VV(()L/W(CI x
xtHddsleepoxnsleepON
2
2
 
(7)
xtHddsleepoxnsleep V)VV()L/W(CI ON
 
(8)
And also the sleep transistor leakage current in the sleep mode can be
identified with the mathematical expressions as follows
)e(eVe)L/W(CI T
ds
T
tHgs
V
V
nV
VV
T
.
sleepoxnleak

 1281

(9)
From the above equation the terms can be represented as µn is the N-
mobility, VT id the thermal voltage=26mV, VtH is the high threshold
voltage and n is the subthreshold swing parameter. And from the above the
relations between di sleep and Vx can be determined. By substituting the
notations of di sleep, (W/L)sleep from the equations (4)-(7) the relationship
between the leakage and delay can be denoted as follows
.
)VKC(d)VV(
d
)VV(C
I
)e(eVeCI /
ddL
/
sleeptLdd
/
sleep
tHddoxn
sleepV
V
nV
VV
T
.
oxnleak
ONT
ds
T
tHgs



 11
1
281
1





(10)
For examine the power reduction techniques six standard benchmarks are
taken into account as test benches. Those are 27 bit channel interrupt
controller (CIC), 32 single error correcting circuit, a four bit ALU/Function
Generator, 6 bit array multiplier design, 32 bit priority checker and finally 4
bit carry look ahead adder. These benchmarks provide the different fan outs
for the different gates for offering variety of different circuits. CLA adder
consists of 28 gates and all are designed and tested using 180 nm
Technology.
Figure 8. Variable length n-way partitioning algorithm.

IJCSBI.ORG
5. Simulation Results
Figure 9. Sleep transistor model 1
From the figures 1, 2 and 9 it can be concluded that the virtual ground line
actually help reduce the ground bounce which were adjoined by the wire
and junction capacitance acts as reservoir for current or local charge sink.
To reduce the offset effects of the sleep transistors if taken lower sized sleep
transistor by taking the extreme high value of capacitance. Where the RC
network will do the low pass filtering action and here it should meet the
value as large for the RC time constant in order to make the virtual ground
to the fraction of peak dc value.
Two constraints are needed to be considered for the ground bounce by
simultaneously reducing the dynamic and leakage power with first
constraint of maintaining the speed degradation to 5% and second constraint
as ground bounce never exceed to 50 mv. These two constraints are
sufficient to make the circuit for achieving speed and noise margins.
Figure 10. Output wave forms of sleep transistor
For scaling down the devices there are two mechanisms are existed viz. sub
threshold leakage and gate leakage. Unfortunately with the increase of the
technology scaling the sub threshold leakage and gate leakage would

IJCSBI.ORG
increase rapidly. Decaps made up of gate oxide MOS transistors with the
long wide channels are used for getting the large capacitance.
Figure 11. Sleep transistor model 2
Due to the continuous switching of the core logic and I/O buffers power grid
noise will exist and this power gating noise is because of the switching
action on and off of the sleep transistors.
From the above analysis the discharge current can be expected. Based on the
multiplication value of the probability of the gate will be at zero to the
probability of the gate will be at one gives the switching activity of the gate.
6. CONCLUSIONS
The clustered sleep transistors leakage power reduction techniques has
analyzed and discussed various advantages and improvements need to take
for the advancements of the sleep transistors for achieving the higher power
saving and for using maximum die area with the consideration of the
various algorithms proposed by the different designers which are
considering as the standard designs. And all are designed and tested with
180 nm technology by using the spice simulation. And the research has to
be done on the power scaling of the sleep transistors in the standby mode for
the better performance.
7. ACKNOWLEDGMENTS
The authors would like to thank anonymous reviewers for their valuable
inputs.
REFERENCES
[1] Sathanur, Ashoka, et al. Row-based power-gating: a novel sleep transistor insertion
methodology for leakage power optimization in nanometer cmos circuits. IEEE
Transactions on Very Large Scale Integration (VLSI) Systems, Vol. 19, No. 3, 2011,
pp. 469-482.

IJCSBI.ORG
[2] Homayoun, Houman, et al. MZZ-HVS: multiple sleep modes zig-zag horizontal and
vertical sleep transistor sharing to reduce leakage power in on-chip SRAM peripheral
circuits. IEEE Transactions on Very Large Scale Integration (VLSI) Systems, Vol. 19,
No. 12, 2011, pp. 2303-2316.
[3] V. Khandelwal and A. Srivastava. Leakage control through fine-grained placement
and sizing of sleep transistors. In Proc. IEEE/ACM Int. Conf. CAD (ICCAD), San
Jose, CA, Nov. 2004, pp. 533–536.
[4] J. Kao, A. Chandrakasan, and S. Narendra. MTCMOS hierarchical sizing based on
mutual exclusive discharge patterns. In Proc. ACM/ IEEE Des. Autom. Conf. (DAC),
San Francisco, CA, Jun. 1998, pp. 495–500.
[5] S. Mutoh, S. Shigematsu, Y. Matsuya, H. Fukuda, and T. Kaneko. 1-V power supply
high-speed digital circuit technology with multithreshold voltage CMOS. J. Solid-State
Circuits, vol. 30, no. 8, pp. 847–854, Aug. 1995.
[6] K. Roy, S. Mukhopadhyay, and H. Mahmoodi-Meimand. Leakage current mechanisms
and leakage reduction techniques in deep-submicrometer CMOScircuits. Proc. IEEE,
vol. 91, no. 2, pp. 305–327, Feb. 2003.
[7] M. Keating, D. Flynn, R. Aitken, A. Gibbons, and K. Shi. Low Power Methodology
Manual: For System-on-Chip Design. New York: Springer, 2007.
[8] H. Jiang, M. Marek-Sadowska, and S. R. Nassif. Benefits and costs of power-gating
technique. In Proc. IEEE Int. Conf. Comput. Des.(ICCD), San Jose, CA, Oct. 2005, pp.
559–566.
[9] M. Anis, S. Areibi, and M. Elmasry. Design and optimization of multithreshold CMOS
(MTCMOS) circuits. IEEE Trans. Comput.-Aided Des. Integr. Circuits Syst., Vol. 22,
No. 10, pp. 1324–1342, Oct. 2003.
[10]J. Kao, A. Chandrakasan, and D. Antoniadis. Transistor sizing issues and tool for
multi-threshold CMOS technology. In Proc. ACM/IEEE Des. Autom. Conf. (DAC),
Anaheim, CA, Jun. 1997, pp. 409–414.
[11]F. Fallah and M. Pedram. Standby and active leakage current control and minimization
in CMOS VLSI circuits. IEICE Trans. Electron., Vol. E88-C, No. 4, pp. 509–519, Apr.
2005.
Sree, D., M., Kranthi, K. Y. and Vardhan, V. K., 2014. The Comparative
Analysis of Power Optimization in Clustered Sleep Transistors.
International Journal of Computer Science and Business Informatics, Vol.
14, No. 3, pp. 25-36.

IJCSBI.ORG
Cyber Crimes Incidents in Financial
Institutions of Tanzania
Edison Wazoel Lubua (PhD)
Mzumbe University, Dar Es Salaam Campus College
Abstract
This paper investigates the trend of cyber-crimes in Tanzania. The purpose is to evaluate
challenges associated with cyber-crimes in mobile money services. The study acknowledges
the provision of mobile money services by both telecommunication companies and local
banks, the fact which poses a threat in the old fashion of addressing crimes. Data were
collected from the Foreigscic Section of the Tanzania Police Force and users of the mobile-
money services. The analysis used both descriptive techniques and the Pearson Correlation
model to establish different facts. The conclusion is based on observed evidence, and it is
placed in the last section of the paper.
Keywords: Cyber-crimes, cyber security, cyber laws, mobile money
Background of the Study
The use of the internet and other Information and Communication Tools (ICTs)
transforms the way our societies perform their day to day activities. As the result of
the importance of such uses in ICTs, the number of users is increasing in a daily
manner. In 2011, about 7 billion people were connected to the internet (through
computers and mobile phones) across the globe (KPM international, 2011). In
Tanzania, about 7,500,000 users were reported in 2012, this is about 17% of the
whole population (IPP Media, 2014). The number of people subscribed to the use of
internet facilities, is the reflection of many activities completed through this
platform. Therefore, the internet has transformed the way people perform different
activities in modern communities.
Although a large group of people use the internet for different activities, a significant
per cent of users do not use the facility for shopping purposes (Zickuhr, 2012)). In
the United Kingdom, about 1/3 of internet subscribers are not shopping online due to
fear of online security. Cyber-crimes reduce the confidence of consumers about the
level of online security facilitated by service providers (Digital Policy Alliance,
2013).
A study about cyber security showed that the fear of cyber crimes cost the world
community a large amount of money in establishing security protocols. In 2008 the

IJCSBI.ORG
worldwide cost of cybercrime was approximately USD 8 billion (Intersecurity
Magazine, 2013). Some of the recently characterized cyber crimes include corporate
cyber espionage, stealing of intellectual properties, financial scams, computer
hacking, downloading pornographic material from the internet, virus attacks, e-mail
stalking and creating websites that promote racial hatred (Singer & Friedman, 2014).
Many countries face difficulties in addressing issues arising from cyber crimes,
because they lack a concrete definition of computer crimes and how such crimes
differ from traditional crimes (Aslan, 2006; Mayunga, 2013). Aslam (2006) defined
computer crimes as a violation of criminal laws that involves the knowledge of
computer technology for its penetration, investigation, or prosecution.
Unfortunately, Africa has become the target of cyber crimes than the rest of the
world where more internet users are victimized by these crimes than before. It is
estimated that 80 percent of the computers in Africa are already infected with viruses
and other malicious software (Kumar, 2010). The noted level of infection increases
the level of vulnerability to cyber crimes in the region (Pack, 2013).
In the Tanzanian context, a significant percent of the population is connected to the
internet (Lubua E. , 2014; IPP Media, 2014). As the result the government
established a unit within the Police Force to address challenges of cyber crimes.
However, the impact of cyber crimes is still threatening the security of internet users.
In 2012, about 620 cases were reported to the cybercrime unit (Mayunga, 2013). The
most reported crime was online stealing of money. Other reported crimes include
obscene communications, computer forgery and life threatening messages.
It is evident that the increase of cyber crimes affects transactions which are
conducted online in the Tanzanian community. Nevertheless a number of controls
are introduced to address the challenge. Such controls include the use of
authentication methods, the use of surveillance cameras and awareness campaigns
about online safety. Some of the stakeholders are even proposing laws that allow
online patrol by the Police Force. This paper discusses factors influencing the safety
of mobile-money banking in the Tanzania context.
2. Statement of the Problem
The emerging ICT technology comes with a number of benefits to the community. In
financial institutions (banks), clients are able to access different services without
visiting bank premises. Nevertheless the use of ICTs for banking purposes comes
with the risk of cyber-attacks. The failure to control the online activities provides
people the room to conduct old crimes in a new way (Mayunga, 2013). Reports show
that Tanzania lost approximately 892.18 billion through online crimes in 2012
(Mwananchi, 2012). Similarly, a number of studies conclude that the lack of cyber

IJCSBI.ORG
crime laws creates a vacuum in the control of these crimes (IPP Media, 2014; Lubua
E. , 2014; Pladna, 2008). Other contributing factors include the low technological
literacy of users and technical security loopholes. In environments where the internet
is lowly controlled, criminals conduct crimes anonymously (Paganinip, 2012). In
Tanzania, the Chief of the Forensic Bureau suggests online patrol by Police officers
as effective and efficient in addressing these crimes (Majaliwa, 2011); however, the
international community is against this practice.
The government of Tanzania is currently implementing reforms aimed at addressing
cyber crime incidents. Such reforms include the strengthening of the
telecommunications regulatory body (through equipping it with modern
technologies), raising the awareness of the law enforcing body and that of online-
services users on cyber crimes. Despite these efforts, incidents of cyber crimes are
still increasing. In this study, we determine:-
i.) Whether the rate of response from the mobile-money officers on queries
from clients contributes to the level of security to clients.
ii.) Whether the nature of control of the Tanzanian mobile money platforms
adequately address the challenge of cyber crimes.
3. Significance of the Study
This paper establishes the following:-
i.) It improves the knowledge of stakeholders about the adequacy of
methods for addressing the challenge of cyber crime in Tanzania and
other developing countries. The more important part is where it identifies
employees of the mobile money companies and their clients as the most
important part of stakeholders in addressing the challenge.
ii.) It shows the loopholes brought by the lack of the legislation to administer
cyber issues in Tanzania.
4. Adopted Research Methods
The study used the Criminal Investigation Department of the Tanzania Police Force
as the key source for secondary data about the trend of cyber crimes in Tanzania.
Data were extracted from the cyber crime unit of the Police Force. This is the Police
Force section where cyber crimes are reported. Moreover, the study interviewed
employees and clients of the local mobile-money companies to understand their
perception about the security level of the mobile money services.
Generally, the population of the study included employees of the Tanzania Police
Force in the Department of Criminal Investigation (cyber crime unit). A sample
included 50 respondents where 20 were Police Officers and 30 respondents were
taken from employees of commercial banks in the online banking unit. Because

IJCSBI.ORG
data were in two main groups, the study used convenience sampling to exploit data
from respondents. The analytical model adopted is the Pearson Correlation Model.
The study ensured that data were collected from original sources and were clearly
audited for reliability reasons.
5. Cybercrimes in Tanzanian Financial Institutions
The use of ICT technologies in Tanzania is growing at a high rate. The rate of
growth reported between 2000 and 2010 is about 450% (Lubua & Maharaj, 2012). It
is further reported that about 45% of the Tanzanian population owns mobile phones
(Genuchten, Haring, Kassel, & Yakubi, 2012). The increase in the use of the internet
and mobile technologies has impacted the methods to which financial services are
offered to clients. The majority of local banks offers their services through both
traditional and online media. These banks have also incorporated the use of mobile
phones in effecting financial transactions. Additionally, telecommunication
companies do also offer financial services, which do not necessarily engage banks.
The use of mobile money services is in the upward trajectory in Tanzania. About
45% of the Tanzanian adults are reported to be using mobile money (Mayunga,
2013). A total of TZS 1.7 trillion was transacted through mobile money in 2012; this
shows a significant shift of financial transaction from traditional options to the use of
mobile money and internet (Ndulu, 2012). Unfortunately the increase of mobile
money uses in financial transactions comes with new challenges. The most noted
challenge is the impact of cyber crimes.
In 2012, about 627 cyber crime cases were reported. Figure 1 below shows the trend
of cyber crimes as reported to the Foreigscic section of the Tanzanian Police Force.
There is a steady increase of cyber crimes in Tanzania from 2009 to 2012. A simple
explanation to the observed trend is that the increase of mobile uses for banking
purposes do also increase the rate of crimes associated with mobile money. This is
because more users of the mobile phones were subscribed in every year.

IJCSBI.ORG
Figure 1: Cybercrimes trend
Source: (Tanzania Police Force, 2012)
The following are suggested by the literature to contribute to the increase of mobile
money crimes: the lack of cybercrime policy, inadequate system security and low
awareness of mobile money users (Mayunga, 2013). The results of the interview
with the Cybercrime unit of the Tanzanian Police Force proposed that online patrol
could address the challenge. Online patrol is the idea borrowed from the traditional
way that the Police Force uses to combat crimes. However, this suggestion receives
criticism from internet users across the world, since it allows the government to
scrutinize internet information from users. Besides, the suggestion remains to be a
theory rather than solution due to a number of unaddressed questions. One of the
questions is the efficiency of online patrol since the traditional police patrol does not
adequately address the challenges of crimes occurring in traditional societies.
While the security of online transactions remains as the primary concern of mobile-
money providers and other stakeholders, the most suitable solution is that which does
not intrude the privacy of users. Any proposed solution must balance the relationship
between mobile money users, service providers and policy makers in ensuring online
safety without intruding one's privacy.
5.1 Cyber Law
Tanzania adopted her first National ICT Policy in 2003. It aimed at increasing the
pace for providing different services to the society through the use of ICT tools
(Tanzania National ICT Policy, 2003). Since then, more people are using ICT tools
in their routines; nevertheless, this increase results to the rise of a new form of
crimes. The study by Mayunga (2013) emphasizes that the lack of cyber policy in

IJCSBI.ORG
Tanzania contributes significantly to the increase of cyber crimes. In this study, we
acknowledge improvements made by the legal system of Tanzania in addressing
online cases where online evidences are now accepted. Initially, the Court System of
Tanzania did not accept online evidences (Msuya, 2014).
It is the expectation of stakeholders that the instalment of the cyber law in Tanzania,
would address a number of issues concerning the safety of the mobile-money users.
First, the law would address the issue of privacy. The Tanzanian society is found
under the socialistic and the self-reliance ideology where the concept of
individualism had little importance (Nyerere, 1967). In addition, the majority of
users of the mobile money formerly used traditional methods for making different
transactions where privacy was not a serious concern. Currently, the lack of defined
levels of privacy to users (of mobile-money), exposes them to threats of online theft.
This is because online methods for accessing financial services require the storage of
several individual information to the database of the company. The information may
be used against the owner.
In certain incidents, the study witnessed the mobile money Kiosk operators who
required their clients to submit the login credentials to receive assistance in accessing
their money. Mobile money users happen to submit such information out of
ignorance. It is possible that some of the Kiosk operators use such data against their
clients, this results to the rise of online crimes. The interview found several clients of
the mobile money services who admitted to have submitted such credentials without
knowing the possible consequences. Moreover, about 30% of the surveyed sample
admitted that someone who pretended to be an officer of the mobile-money operator
once demanded the submission of login credential through email or SMS for some
reasons including updating user’s information. The combination of the socialistic life
and the illiteracy level of members of the community threaten the safety of users
who transact through mobile money.
Also, there is nothing that legally identifies who has to access the stored information
of clients, what type of information can be accessed and when should that access be
granted. The lack of guidance grants mobile money operators the unmonitored
privilege to the access of clients’ information. It is this loophole that the unfaithful
employees use to access clients’ information and money. The results from the survey
support this observation since about 44% of the mobile money clients reported an
incident where the loss of their money was contributed by the mobile money
operator.
Based on observations above, the establishment of a comprehensive legal system that
address issues associated with privacy would decrease challenges associated with

IJCSBI.ORG
cyber crimes.
Secondly, the cyber law would decrease the potential for online abuse. In the absence
of cyber laws, the unfaithful government leaders and business corporations use the
captured data of clients according to their wish. In Tanzania, mobile subscribers and
mobile money users are registered. The registration system stores the information
about users and transactions that they make. Economically, knowing expenditures of
a person lead to an easy prediction of the income, hence an easy access to the
information poses security threats to consumers of mobile money services. A number
of incidents were reported in the previous year about mobile money kiosk operators
and their clients who were ambushed by robbers. It is most likely that the thieves
were well informed of the cash flow of the mobile money operators and that of
clients. In such cases, the cyber law can prevent the abuse of clients’ information by
any individuals or organization.
Moreover, individual unsolicited information could be accessed through the mobile
money clients’ phone. This is because the majority of the passwords used by the
mobile-money users are easy to guess. This degree of complexity is closely
associated with both illiteracy and negligence levels. The study observed about 90%
of respondents to admit the use of passwords with the combination of either numbers
or alphabets only. Some of the mobile-money companies limit the password to four
digits. This is a very weak protection and could be the contributing factor to the
increasing number of cyber crimes in Tanzania. The cyber law could dictate the
complexity of the password used.
Another area which shows a potential for attack is the lack of a regular change of the
password by users of the mobile money. Above 55.4% of respondents have never
changed their password since registration. Further analysis reveals that 54% of
clients who never changed their password have the experience of using mobile
money for over 18 months. Again, some are not taking precautions out of ignorance
while some are doing out of negligence. Regardless of the factor which may have
contributed to a failure to take precautions (for self-protection against information
abuse), users are vulnerable and must be protected by the law. The cyber law could
impose the maximum time for the password to be used.
In an interview with advocates for criminal cases, respondents had a common
acknowledgement of changes attained in the court room on cyber crimes. Initially,
the court could not execute cases related to cyber crimes because of a number of
reasons: One was the absence of the law to guide the prosecution of online crimes,
and the other was the fact that online evidence were not accepted by the court.
Nevertheless, noticeable changes have taken place to the extent that the Court of

IJCSBI.ORG
Tanzania allows a court debate to be conducted through video-conferencing. This is
a good progress. Regardless of the absence of the cyber law, the court of Tanzania
judge cases about cyber crimes based on international laws and experience from
international bodies associated with the nation such as the commonwealth. While
this approach provides a temporal measure, a comprehensive and permanent
approach is needed.
5.2 Does the nature of control of the Tanzanian mobile money platforms
adequately address the challenge of cybercrimes?
The question of whether the nature of controls offered by the Tanzanian mobile
money platforms adequately address the challenges associated with cyber crimes
need urgent attention. This is because crimes associated with mobile money are
increasing annually (figure 1). A proper level of security control to mobile money
users allows narrowing the focus of scholars and other stakeholders to technical
variables influencing cyber crimes. This discussion regards the mobile money
platform to consist two sides: System Administrator and Users.
The nation lacks the cyber law to provide guidance to mobile money operators on
how, who and when should the system administrator access the account of the client.
It is obvious that these operators make such judgements based on personal intuition
or internal organization laws. This situation poses a threat because the majority of
the mobile money operators (employees) manage to transfer client’s money
(electronically) with a simple authorization written by a client on a paper. The
information proposes that the same transaction could be effected without
authorization from the owner. In this case the safety of the account of the client relies
much on the integrity of an employee. In an interview with anonymous employees of
the mobile money operating companies, they acknowledged that several incidents
were reported where employees with low integrity stole the money from clients.
Another factor which poses security threats to the account of the client is the ability
of mobile money administrators to change the password of the client without his
intervention. It is unfortunate that in most operators, whenever the client forget the
password the assigned employee simply creates another password and send it
through clients mobile phone. In the case where the mobile money operator is a
bank, the same employee is also able to change the mobile phone number where
password credentials are directed. This is attested by my own case where I requested
a change to my mobile-money login credentials from a bank through the unregistered
email while in a foreign country. Isn’t this making the mobile money users
vulnerable to criminals (hackers)?

IJCSBI.ORG
On the other hand, the controls administered to users of the mobile money are weak.
Unlike where the online banking is used, the mobile money system does only present
the user with a single login window. Optionally, the user could use the login
credentials provided earlier to access the second login window through the use of
temporal credentials sent through other methods such as emails or phone numbers.
While the password for the first window is known to the user, the second must be
temporal and be sent automatically to the user (through another phone number or
user’s trusted email). The other number or email should receive such information
because it was registered earlier.
The endless duration for password validity is another area that weakens protection in
the mobile money system. The study found that all mobile money subscribers were
not demanded by the system to change their password periodically. In most cases,
the password was valid for the undefined duration of time. There were cases where
the mobile money users ignorantly submitted their login credentials to the mobile
money Kiosk operator for help. In such cases, the unlimited life span of the password
could pose a threat to the user in case the Kiosk operator decided to temper with the
account.
Moreover, the study observed that (in most cases) the mobile money users were
supposed to use a four digit standard login credentials (password). The password
does not necessitate users to mix characters to improve its strength. The survey
found several cases where the year of birth of the user was used as a password. The
lack of methods to administer the use of strong passwords by the mobile money
users create a room for hackers to break through the mobile money accounts, hence
increase cyber crime incidents.
6. Cyber Crimes and Mobile Money Officers’ Response to Clients’
Queries
The study, thought of the possibility that the efficiency of the response of the mobile
money staff to queries from clients relates to the rate of cyber crimes in the country.
Initially, the survey found that about 33.9% of respondents were confident that they
were secure from the unsolicited use of their information submitted for mobile
money use. The low percent of respondents who are confident contributed by
individual experience in the miss-use of such information or the current trend of
online theft (Nyenyelwa, 2013).
It was further observed that about 43% of respondents perceive employees’ of the
mobile money companies to operate with low efficiency. The reported low efficiency
is due to delayed response to requests submitted to the organization for support.
With these statistics, many respondents are not comfortable with the efficiency of

IJCSBI.ORG
employees who attend them. The study by Snow (2011) provides the evidence that
low employees’ competency in using ICT equipments is among the reasons for low
efficiency in attending queries related to cyber crimes in mobile money. To address
the issue of knowledge in technical projects, the organization must conduct technical
training which is closely monitored to ensure that the acquired knowledge is
practiced (Lubua E. , 2014). It is equally important to acknowledge that the
unmonitored activities in the organization may result to such delays. Table 1 shows
the association between the perceived level of security and the help desk efficiency
toward responding to users.
Table 1: Correlations- Level of Security*Helpdesk Efficiency
Level of security from Unsolicited Use of
Client's Information
Help Desk
Efficience
Level of security from
Unsolicited Use of
Client's Information
Pearson
Correlation
1 .327**
Sig. (2-tailed) .008
N 65 65
Help Desk Efficience Pearson
Correlation
.327**
1
Sig. (2-tailed) .008
N 65 65
Additional information to the analysis (Table 1) showed a significant correlation
between the efficiency by employees in responding to reported queries and the
perceived level of security. The r-value is 0.327 and p < 0.05. With these results, the
increase of efficiency exerts about 33% of influence to the decrease of cyber crime
and vice versa. The increase of the efficiency of employees responding to clients’
queries is a proper strategy for addressing the challenge of cyber crimes. A quick
attendance of queries would prevent crimes which were about to occur. The mobile
money companies should assume the role of the natural Police Force, in preventing
online crimes to occur while responding to the needs of rescue from users.
Data collected through experimentation showed that the mobile money companies
take at least 10 minutes to start responding to a new call from the client. Interviewee
complemented these observations by suggesting the presence of some cases where
the call made never received a response from the help desk. Some of the mobile
money companies have initiated a priority service system, where the client of the
mobile money has to pay to receive the priority in services from employees. This is
not the right way of doing things, instead clients are to be treated equally and
efficiently.

IJCSBI.ORG
7. Conclusion
The study intended to show how the perceived rate of occurrence of cyber crimes
relate to the following factors: the factors are the efficiency of employees in
responding to clients’ queries, the nature of control by the mobile money platforms
and the absence of the cyber law. The study used survey and experimental methods
to obtain relevant data. Based on findings presented above, the study concludes that
the lack of cyber laws in Tanzania results to the violation of clients’ right of
confidentiality through allowing employees’ and other cyber stakeholders to use
clients’ information without limit. It increases insecurity to clients because of the
lack of protection against online uses of personal information by government entities
and business corporations.
Moreover, the study concludes that online platforms of mobile money companies do
not address the challenge of cyber crimes adequately because employees are able to
interfere with important mobile money information. In online banking, the ability of
employees to change the phone number used for online banking by the client, need
re-assessment. This is the number where the client receives an authenticating
password from the bank. The ability to change such important information must be
granted to clients only. Besides, in telecom companies, there must be a separation
between the ability to make a new SIM card and mobile money password setting.
This will address the challenge of miss-using the information from clients by
unfaithful employees. The password complexity must equally be addressed.
It is also necessary for the mobile money employees to increase their efforts in
attending queries from clients efficiently. The low rate of response from employees
stirs the rate of cybercrimes. Where necessary, employees are to be trained to meet
technological needs while ensuring that only competent employees are hired.
References
Aslan, Y. (2006). Global Nature of Computer Crimes and the Convention on Cyber Security. Ankara
Law Review, Vol. 3 No. 2, 129-142.
Digital Policy Alliance. (2013). CYBER SECURITY AND E-CRIME WORKING GROUP. Retrieved
May 6, 2014, from http://dpalliance.org.uk/cyber-security-wg/
Genuchten, R. v., Haring, W., Kassel, D. v., & Yakubi, K. (2012). Mobile phone use in Tanzania.
Amsterdam: vrije universiteit amsterdam.
Intersecurity Magazine. (2013). Global Cybercrimes Costs. Retrieved 5 6, 2014, from
http://www.infosecurity-magazine.com/view/33569/global-cybercrime-espionage-costs-
100500-billion-per-year/
IPP Media. (2014, February 2). Number of Internet users still low in Tanzania, says global report.
Retrieved June 6, 2014, from http://www.ippmedia.com/frontend/?l=64361

IJCSBI.ORG
Kumar, N. (2010). It is estimated that 80 percent of the computers in Africa are already infected with
viruses and other malicious . Retrieved May 6, 2014, from
http://www.psfk.com/2010/04/africa-could-become-the-cybercrime-capital-of-the-
world.html#!JoaOm
Lubua, E. (2014). Adoption of E-transparency in the Tanzanian Public Sector. Durban: University of
KwaZulu Natal.
Lubua, E., & Maharaj, M. (2012). ICT Policy and E-transparency in Tanzania. IST-Africa. Dar Es
Salaam: IIMC International Information Management Corporation.
Mayunga, J. (2013). Cybercrimes Investigation in Tanzania. Morogoro: Mzumbe University.
Msuya, N. (2014, June 4). Online Evidence are Accepted by the Tanzanian Court. (E. Lubua,
Interviewer)
Ndulu, B. (2012). Mobile money transactions top TZS1.7tn. Retrieved May 08, 2014, from
http://www.telegeography.com/products/commsupdate/articles/2012/12/13/mobile-money-
transactions-top-tzs1-7tn-bank-of-tanzania-reports/
Nyenyelwa, F. (2013, December 15). Security Confidence. (E. Lubua, Interviewer)
Nyerere, J. (1967). Education for Self-Reliance. Dar Es Salaam: Government Press.
Pladna, B. (2008). The Lack of Attention in the Prevention of and How to Improve It. Greenville:
University of East Carolina.
Singer, P., & Friedman, A. (2014). Cyber Crimes and Cyber War. New Yorl: Oxford Press.
Snow, G. (2011). Cyber Security Threats. Retrieved May 2, 2014, from
http://www.fbi.gov/news/testimony/cybersecurity-responding-to-the-threat-of-cyber-crime-
and-terrorism
Tanzania National ICT Policy. (2003). National ICT Policy. Dar Es Salaam: The Government of
Tanzania.
Tanzania Police Force. (2012). Annual Crime Report. Dar Es Salaam: Police Force.
Zickuhr, K. (2012). Digital differences. Retrieved May 6, 2014, from
http://www.pewinternet.org/2012/04/13/digital-differences/
Lubua, E. W., 2014. Cyber Crimes Incidents in Financial Institutions of Tanzania.
International Journal of Computer Science and Business Informatics, Vol. 14, No. 3,
pp. 37-48.

Vol 14 No 3 - November 2014

Recommended

Recommended

More Related Content

What's hot

What's hot (20)

Similar to Vol 14 No 3 - November 2014

Similar to Vol 14 No 3 - November 2014 (20)

More from ijcsbi

More from ijcsbi (20)

Recently uploaded

Recently uploaded (20)

Vol 14 No 3 - November 2014