Skip to main content
main-content
Top

About this book

This book constitutes the proceedings of the First International Conference on Information and Communication Technology for Development for Africa, ICT4DA 2017, held in Bahir Dar, Ethiopia, in September 2017.

The 31 revised full papers presented were carefully reviewed and selected from 72 submissions. The papers address the impact of ICT in fostering economic development in Africa. In detail they cover the following topics: e-services, natural language processing, intelligent systems, mobile and wireless communication, privacy and security.

Table of Contents

Frontmatter

ICT4DA Main Track

Frontmatter

Is Addis Ababa Wi-Fi Ready?

As we are heading towards future ubiquitous networks, heterogeneity is one key aspect we need to deal with. Interworking between Cellular and WLAN holds a major part in these future networks. Among other potential benefits it gives the opportunity to offload traffic from the former to the latter. To successfully accomplish that, we need to thoroughly study the availability, capacity and performance of both networks. To quantify the possibility of mobile traffic offloading, this work-in-progress presents the availability, capacity and performance investigation of Wi-Fi Access Points in the city of Addis Ababa. Analysis of the scanned data, collected by travelling through the highly populated business areas of the city, reveals the potential of existing Wi-Fi coverage and capability for many application domains.

Asrat Mulatu Beyene, Jordi Casademont Serra, Yalemzewd Negash Shiferaw

A Finite-State Morphological Analyzer for Wolaytta

This paper presents the development of a free/open-source finite-state morphological transducer for Wolaytta, an Omotic language of Ethiopia, using the Helsinki Finite-State Transducer toolkit (HFST). Developing a full-fledged morphological analysis tool for an under-resourced language like Wolaytta is an important step towards developing further NLP (Natural Language Processing) applications. Morphological analyzers for highly inflectional languages are most efficiently developed using finite-state transducers. To develop the transducer, a lexicon of root words was obtained semi-automatically. The morphotactics of the language were implemented by hand in the lexc formalism, and morphophonological rules were implemented in the twol formalism. Evaluation of the transducer shows as it has decent coverage (over 80%) of forms in a large corpus and exhibits high precision (94.85%) and recall (94.11%) over a manually verified test set. To the best of our knowledge, this work is the first systematic and exhaustive implementation of the morphology of Wolaytta in a morphological transducer.

Tewodros A. Gebreselassie, Jonathan N. Washington, Michael Gasser, Baye Yimam

Malaria Detection and Classification Using Machine Learning Algorithms

Malaria is one of the most infectious diseases, specifically in tropical areas where it affects millions of lives each year. Manual laboratory diagnosis of Malaria needs careful examination to distinguish infected and healthy Red Blood Cells (RBCs). However, it is time consuming, needs experience, and may face inaccurate lab results due to human errors. As a result, doctors and specialists are likely to provide improper prescriptions. With the current technological advancement, the whole diagnosis process can be automated. Hence, automating the process needs analysis of the infected blood smear images so as to provide reliable, objective result, rapid, accurate, low cost and easily interpretable outcome. In this paper comparison of conventional image segmentation techniques for extracting Malaria infected RBC are presented. In addition, Scale Invariant Feature Transform (SIFT) for extraction of features and Support Vector Machine (SVM) for classification are also discussed. SVM is used to classify the features which are extracted using SIFT. The overall performance measures of the experimentation are, accuracy (78.89%), sensitivity (80%) and specificity (76.67%). As the dataset used for training and testing is increased, the performance measures can also be increased. This technique facilitates and translates microscopy diagnosis of Malaria to a computer platform so that reliability of the treatment and lack of medical expertise can be solved wherever the technique is employed.

Yaecob Girmay Gezahegn, Yirga Hagos G. Medhin, Eneyew Adugna Etsub, Gereziher Niguse G. Tekele

Intelligent Transport System in Ethiopia: Status and the Way Forward

Vehicular transportation systems are used extensively to transport people and goods which is detrimental for faster, reliable and cost effective socioeconomic activity. However, there are major challenges associated with accelerated utilization of such systems. These include threat to safety of life and property; pollutions; congestion triggered reduction of road network utilization; reduced cost effectiveness of vehicles; and increased waiting and travelling times of passengers. This paper briefly surveys the above problems in international and national context. It then assesses deficiency of conventional methods of mitigating the problems. Next it proposes introduction of intelligent transport system (ITS) in Ethiopia as a better and cross cutting solution to the above problems. The paper analyses and presents verifications of the hypothesis that if ITS is introduced, the nation would achieve better safety to life and property; less pollution; more efficient mobility traffic control and management; and better utilization of road networks and vehicles.

Tezazu Bireda

Survey on Indoor Positioning Techniques and Systems

Navigating different devices and human beings in indoor scene has become very crucial for number of tasks specially in automated system. The efficiency of outdoor positioning has become excellent due to the development of GPS. However lots of mass market applications require very excellent positioning capabilities in almost every environments. As a result, indoor positioning has attracted the researchers attention and has been a focus of research during the past decades. This paper presents an overview of the four typical indoor localization schemes namely triangulation, trilateration, proximity and scene analysis are analyze and discussed. Moreover it gives a detailed survey of different positioning systems which are being both research-oriented solutions and commercial products and also attempts to classify the different systems into different groups based on the technology used. We categorized all 11 sighted wireless indoor positioning systems into 6 distinct technologies namely Infrared signals, radio frequency, ultrasound waves, vision-based analysis, electromagnetic waves, and audible sound and explains the measuring principles of each. These approaches are characterized and their key performance parameters are quantified individually. For a better understanding, these parameters are briefly compared in table form for each system so as to outline the trade-offs from the viewpoint of a user.

Habib Mohammed Hussien, Yalemzewed Negash Shiferaw, Negassa Basha Teshale

Comparative Study of the Performances of Peak-to-Average Power Ratio (PAPR) Reduction Techniques for Orthogonal Frequency Division Multiplexing (OFDM) Signals

In this paper, two distortionless PAPR reduction techniques, Selected Mapping (SLM) and Partial Transmit Sequences (PTS), are compared in terms of PAPR reduction capability and computational complexity for equal number of candidate OFDM symbols. Using MATLAB simulation, it is shown that SLM outperforms PTS in PAPR reduction capability. For small values of the number of subblock partitions, the overall computational complexity of PTS is less than SLM. However, the required PAPR reduction level may not be achieved using small values of number of subblock partitions. Hence, for large values of number of subblock partitions used in PTS, the overall computational complexity of PTS is greater than SLM. In that case, SLM outperforms PTS both in PAPR reduction capability and computational complexity.

Workineh Gebeye Abera

A Distributed Multi-hop Clustering Algorithm for Infrastructure-Less Vehicular Ad-Hoc Networks

Vehicular Ad-hoc Networks (VANETs) aim to improve travailing safety, comfort and efficiency via enabling communication between vehicles and between vehicles and infrastructure. Clustering is proposed as a promising technique to efficiently manage and deal with highly dynamic and dense features of vehicular topology. However, clustering generates a high number of control messages to manage and maintain the clustering structure. In this paper, we present our work that aims to facilitate the management of the disconnected infrastructure-less VANET areas by organizing the network topology using a distributed multi-hop clustering algorithm. The proposed algorithm is an enhanced version of the distributed version of LTE for V2X communications (LTE4V2X-D) [7] framework for the infrastructure-less VANET zone. We are able to improve the performance of LTE4V2X-D to better support clustering stability while decreasing clustering overhead. This is made possible due to a judicious choice of metrics for the selection of cluster heads and maintenance of clusters. Our algorithm uses a combination of three metrics, vehicle direction, velocity and position, in order to select a cluster-head that will have the longest lifetime in the cluster. The simulation comparison results of the proposed algorithm with LTE4V2X-D demonstrate the effectiveness of the novel enhanced clustering algorithm through the considerable improvement in the cluster stability and overhead.

Ahmed Alioua, Sidi-Mohammed Senouci, Samira Moussaoui, Esubalew Alemneh, Med-Ahmed-Amine Derradji, Fella Benaziza

Radar Human Gait Signal Analysis Using Short Time Fourier Transform

Human gait detection and identification by using radar signal is one of the recent subject of increased research area in signal processing. It has been indicated human gait information/signal is highly unusual which can be used for human detection and identification from one person to another. Most previous works related to this area extraction of features from the pace of pedestrians is only depending on the motions rhythm signal analysis and synthesis. Then Fourier transform and more recently time-frequency transforms are used to analyze the time shift/delay and identify the different parts of the human body playing part during the human movement. The analysis of the time/frequency shift usually needs to observe the process by taking a bit long time, at least long enough to get the gait signal cycle. However, the presence of several people simultaneously in the radar field of sight could involve interferences. Hence, in this paper we have been trying to use one of a powerful tool short time Fourier transform for the analysis of time-varying signals among the time frequency methods to extract some feature of human gait.

Negasa B. Teshale, Dinkisa A. Bulti, Habib M. Hussien

Classification of Mammograms Using Convolutional Neural Network Based Feature Extraction

Breast cancer is the most common cause of death among women in the entire world and the second cause of death after lung cancer. The use of automatic breast cancer detection and classification might possibly enhance the survival rate of the patients through starting early treatment. In this paper, the convolutional Neural Networks (CNN) based feature extraction method is proposed. The features dimensionality was reduced using Principal Component Analysis (PCA). The reduced features are given to the K-Nearest Neighbors (KNN) to classify mammograms as normal or abnormal using 10-fold cross-validation. The experimental result of the proposed approach performed on Mammography Image Analysis Society (MIAS) and Digital Database for Screening Mammography (DDSM) datasets were found to be promising compared to previous studies in the area of image processing, artificial intelligence and CNN with an accuracy of 98.75$$\%$$ and 98.90$$\%$$ on MIAS and DDSM dataset respectively.

Taye Girma Debelee, Mohammadreza Amirian, Achim Ibenthal, Günther Palm, Friedhelm Schwenker

Exploring the Use of Global Positioning System (GPS) for Identifying Customer Location in M-Commerce Adoption in Developing Countries

M-commerce in Kenya has seen tremendous growth over the last few years due to the availability of mobile payments, mobile internet access, and expansion of mobile banking systems. A critical factor to the success of m-commerce is timely delivery of purchased items to the customers’ premises. Timely delivery is highly dependent on the courier’s ability to locate the buyer’s physical location. To do this, the courier requires a reliable physical addressing system. However, like most developing countries, Kenya lacks a National Addressing System to provide properly registered physical identity of buildings, streets, and roads. The study explored the use of GPS in identifying customer’s location as an alternative to named physical addresses. This paper describes the study design and discusses the findings concerning the use of GPS tracking application among six retailers and thirty customers. The study reveals that geolocation can substitute physical addresses in m-commerce home deliveries.

Patrick Kanyi Wamuyu

Developing Knowledge Based Recommender System for Tourist Attraction Area Selection in Ethiopia: A Case Based Reasoning Approach

A knowledge based recommender reasons about the fit between a user’s need and the features of available products and it uses knowledge about users and products to pursue knowledge based approach to generate a recommendation, reasoning about what products/services meet the user’s requirements. Providing an effective service in the Tourism sector of Ethiopia is critical to attract more foreign and local tourists. However, there are major problems that need immediate solution. First, the difficulty of getting fast, reliable, and consistent expert advice in the sector that is suitable to each visitor’s characteristics and capabilities. Second, inadequacy of the number of experienced experts and consulting individuals who can give advice on tourism issues in the country. Therefore, this paper aims to design a recommender system for tourist attraction area and visiting time selection that can assist experts and tourists to make timely decisions that helps them to get fast and consistent advisory service so that visitors can identify tourist attraction areas that have the highest potential of success/satisfaction and that match their personal characteristics. For the development of case based recommender system, essential knowledge was acquired through semi-structured interview and document analysis. Domain experts and visitors were interviewed to elicit the required knowledge about the selection process of attraction area. The acquired knowledge was modeled using hierarchical tree structure and it was represented using feature value case representation. At the end, jCOLIBRI programming tool was used to implement the system. The main data source (case base) used to develop case based recommender system for tourist attraction area selection is previous tourist cases collected from national tour operation and ministry of culture and tourism. As a retrieval algorithm, nearest neighbor retrieval algorithm is used to measure the similarity of new case (query) with cases in the case base. Accordingly, if there is a similarity between the new case and the existing case, the system assigns the solution (recommended attraction area and visiting time) of previous case as a solution to new case. To decide the applicability of the prototype system in the domain area, the system has been evaluated by involving domain experts and visitors through visual interaction using the criteria of easiness to use, time efficiency, applicability in the domain area and providing correct recommendation. Based on prototype user acceptance testing, the average performance of the system is 80% and 82% by domain experts and visitors respectively. The performance of the system is also measured using the standard measure of relevance (IR system) recall, precision and accuracy measures, where the system registers 83% recall, 61% precision and 85.4% accuracy.

Tamir Anteneh Alemu, Alemu Kumilachew Tegegne, Adane Nega Tarekegn

A Corpus for Amharic-English Speech Translation: The Case of Tourism Domain

Speech translation research for the major languages like English, Japanese and Spanish has been conducted since the 1980’s. But no attempt were made in speech translation to/from the under-resourced language like Amharic. These activities suffered from the lack of Amharic speech and Amharic-English text corpus suited for the development of speech translation between the two languages. In this paper, therefore, an attempt has been made to collect, translate and record speech data from resourced language (English) to under-resourced language (Amharic) taking a Basic Traveler Expression Corpus (BTEC) as domain. Since there is no any Amharic text and speech corpus readily available for speech translation purposes, first, 7.43 h of Amharic read-speech has been prepared from 8,112 sentences, and second, 19,972 parallel Amharic-English corpus has been prepared taking tourism as an application domain. The Amharic speech data is recorded using smart-phone based application tool, LIG-Aikuma under a normal working environment. With the availability of such standard speech and text corpus, researcher will find a ground to further explore speech translation to/from under resourced languages.

Michael Melese Woldeyohannis, Laurent Besacier, Million Meshesha

Experimenting Statistical Machine Translation for Ethiopic Semitic Languages: The Case of Amharic-Tigrigna

In this research an attempt have been made to experiment on Amharic-Tigrigna machine translation for promoting information sharing. Since there is no Amharic-Tigrigna parallel text corpus, we prepared a parallel text corpus for Amharic-Tigrigna machine translation system from religious domain specifically from bible. Consequently, the data preparation involves sentence alignment, sentence splitting, tokenization, normalization of Amharic-Tigrigna parallel corpora and then splitting the dataset into training, tuning and testing data. Then, Amharic-Tigrigna translation model have been constructed using training data and further tuned for better translation. Finally, given target language model, the Amharic-Tigrigna translation system generates a target output with reference to translation model using word and morpheme as a unit. The result we found from the experiment is promising to design Amharic-Tigrigna machine translation system between resource deficient languages. We are now working on post-editing to enhance the performance of the bi-lingual Amharic-Tigrigna translator.

Michael Melese Woldeyohannis, Million Meshesha

Synchronized Video and Motion Capture Dataset and Quantitative Evaluation of Vision Based Skeleton Tracking Methods for Robotic Action Imitation

Marker-less skeleton tracking methods are being widely used for applications such as computer animation, human action recognition, human robot collaboration and humanoid robot motion control. Regarding robot motion control, using the humanoid’s 3D camera and a robust and accurate tracking algorithm, vision based tracking could be a wise solution. In this paper we quantitatively evaluate two vision based marker-less skeleton tracking algorithms (the first, Igalia’s Skeltrack skeleton tracking and the second, an adaptable and customizable method which combines color and depth information from the Kinect.) and perform comparative analysis on upper body tracking results. We have generated a common dataset of human motions by synchronizing an XSENS 3D Motion Capture System, which is used as a ground truth data and a video recording from a 3D sensor device. The dataset, could also be used to evaluate other full body skeleton tracking algorithms. In addition, sets of evaluation metrics are presented.

Selamawet Atnafu, Conci Nicola

Ethiopian Public Universities’ Web Site Usability

Usability is the vital aspect of any system for having quality products. There are various models for testing usability. The objective of the study is to assess the applicability of the USE (Usefulness (Usability), Satisfaction, Ease of use, and Ease of learning) model for Ethiopian Universities’ web site context. The USE model is evaluated using PLS-SEM method. The study uses three university web sites to test the model. The result found is encouraging. It has found that usability is affected by 53.3% jointly by the considered usability factors. Independently, usability is influenced by user satisfaction, ease of use, and ease of learning by 53.1%, 20.1%, and 6.1% respectively. Based on the result found, user satisfaction is the major determinant for web site usability and the “USE” model is convenient to study web site usability of Ethiopian universities.

Worku Kelemework, Abinew Ali

Comparative Analysis of Moving Object Detection Algorithms

Moving object detection plays a key role in surveillance systems, vehicle and robot guidance, regardless of it is a very troublesome task. Detecting as well as tracking objects in the video so as to distinguish motion features has been rising as a concerning research/study area in image processing/computer vision fields. One of the current demanding study area in computer/machine vision domain are humans and vehicles motion video surveillance system in a dynamic environment. It is considered as a big challenge for researchers to design a good detection technique which is computationally efficient and consuming less time. Moving object detection algorithms must be fast, reliable and vigorous to make video surveillance systems so as to avoid terrorism, crime and etc. This paper presents comparison of different detection schemes for segmenting/detecting moving objects from the background environment. The algorithms are adequate for adapting to dynamic scene condition, removing shadowing, and distinguishing/identifying removed objects both in complex indoor and outdoor. These algorithms are frame/temporal differencing (FD), simple adaptive background subtraction (BS), Mixture of Gaussian Model (MoG) and approximate median filter. These algorithms are appropriate for real time surveillance applications and each of them have their own advantages and drawbacks.

Habib Mohammed Hussien, Sultan Feisso Meko, Negassa Basha Teshale

Multiple Antenna (MA) for Cognitive Radio Based Wireless Mesh Networks (CRWMNs): Spectrum Sensing (SS)

The concept of cognitive radio (CR) rings a big paradigm shift to the wireless communication domain. Extending this concept in to wireless mesh networks (WMN) results a CRWMN which alleviates the pragmatic spectrum congestion in the ISM bands. The assimilation of MAs technology in to CRWMN brings an astonishing system performance improvement. The use of MAs in WMN improves system capacity and reliability, increases coverage area and spectrum usage efficiency; and result in lower power consumption, better interference cancellation, efficient spectrum sensing, and spectrum sharing. In spite of the significant advantages, the use of multiple antennas has considerable limitations. In this paper, we investigate the challenges, opportunities, and the possible research directions that the cognitive radio network (CRN) in general and the CRWMN in particular experience while incorporating MAs to the system and its effect on spectrum sensing.

Mulugeta Atlabachew, Jordi Casademont, Yalemzewd Negash

The Design and the Use of Knowledge Management System as a Boundary Object

Agricultural knowledge management system (KMS) involves various members coming from different social groups who possess their own knowledge which need to be combined in the system development. However, the current development of the technology ignored the indigenous knowledge of the local communities. The multi-methodological approach to KMS research in action research perspective was employed to understand the design and use of KMS for knowledge integration. Primary qualitative data were acquired through semi-structured interviews and observations. The research shall have theoretical contribution in addressing the incorporation of variety of knowledge in agriculture and practical implication to provide management understanding in developing strategies for the potential of a shared KMS as a boundary object for knowledge integration to support marginalized smallholder farmers.

Dejen Alemu, Murray E. Jennex, Temtem Assefa

Autonomous Flyer Delivery Robot

In this study, we developed a socially interactive service robot with an innovative autonomous flyer distribution function. This robot is equipped with innovative flyer storage and delivery system and could store numerous A5 to A7 flyers sizes and tissue packs at a time. Each flyer passes through an internal channel to reach the palm of the robot, which is configured at a commonly reachable height for the majority of people. Every time a flyer or tissue pack is taken from the palm of the robot, the next flyer autonomously arrives at the robot’s palm every 8 s. The developed robot was designed to have autonomous cassette and battery swapping mechanisms and could work exclusively within a localized working zone. Furthermore, it is equipped with strategies for localizing and avoiding obstacles. Thus, the robot was observed to perform flyer delivery without human intervention. The developed robot was displayed in various exhibitions held in Taiwan. The robot was seen to perform the expected task of flyer delivery which proves the robots full commercial value and a huge potential of becoming a product in the intelligent service robot market.

Tesfaye Wakessa Gussu, Chyi-Yeu Lin

Minimal Dependency Translation: A Framework for Computer-Assisted Translation for Under-Resourced Languages

This paper introduces Minimal Dependency Translation (MDT), an ongoing project to develop a rule-based framework for the creation of rudimentary bilingual lexicon-grammars for machine translation and computer-assisted translation into and out of under-resourced languages as well as initial steps towards an implementation of MDT for English-to-Amharic translation. The basic units in MDT, called groups, are headed multi-item sequences. In addition to wordforms, groups may contain lexemes, syntactic-semantic categories, and grammatical features. Each group is associated with one or more translations, each of which is a group in a target language. During translation, constraint satisfaction is used to select a set of source-language groups for the input sentence and to sequence the words in the associated target-language groups.

Michael Gasser

Massive MIMO for 5G Cellular Networks: Potential Benefits and Challenges

The concept of deploying multiple antenna arrays in the base station (i.e. massive MIMO) among other technologies, such as millimeter wave communication and network densification, that it is one of the key enabling methods in the design and development of future cellular networks. Massive MIMO is a disruptive technology; it is considered as a cornerstone in the design of future cellular networks. In this paper, we investigate the benefits of massive MIMO in terms of capacity and energy efficiency. Performance evaluation of massive MIMO is also presented with respect to spectral efficiency and energy efficiency. Moreover, the major challenges for practical deployment of massive MIMO are discussed in details.

Bekele Mulu Zerihun, Yihenew Wondie

Mathematical Modeling and Dynamic Simulation of Gantry Robot Using Bond Graph

This paper presents an initial mathematical modeling and dynamic simulation of gantry robot for the application of printing circuit on board. The classical modeling methods such as Newton-Euler, Kirchoff’s law and Lagrangian fails to unify both electrical and mechanical system models. Here, bond graph approach with robust trajectory planning which uses a blend of quadratic equations on triangular velocity profile is modeled in order to virtually simulate it. In this paper, the algebric mathematical models are developed using maple software. For the sake of simulation, the model is tested on matlab by integrating robot models which are developed by using Solidwork.

Tadele Belay Tuli

Web Usage Characterization for System Performance Improvement

Web usage mining discovers patterns of user behaviors from web log files. In this study web usage mining is employed to identify business-critical and non-business critical web traffics in University of Gondar. Apriori and FP tree algorithms are applied to extract the web browsing behavior in terms of frequently accessed sites along with their web traffics. Our research findings can be used as an input for bandwidth management and system performance improvement.

Alehegn Kindie, Adane Mamuye, Biniyam Tilahun

Critical Success Factors and Key Performance Indicators for e-Government Projects- Towards Untethered Public Services: The Case of Ethiopia

The road to digital transformation of the governance system of developing countries has been a steep uphill climb- in addition to massive investment in ICT infrastructure and applications e-Government projects often require changes in legislation, major policy decisions and restructuring of the public sectors. In this paper, pertinent issues that stand in the way of digital government ascent of Ethiopia have been investigated. To give the issues addressed a context, an e-Service that enabled consumers to pay utilities bill for water, telephone and electricity in one-stop service center has been analysed with respect to metrics developed as well as on the basis of consumers rating of the service. The main findings from consumers response of the service ratings are that, e-Government projects success cannot be judged solely by monetary return of investments which could be obtained among other things by cutting down the work force required to run the service and improvements achieved compared to how the service was delivered in manual settings before it went electronics. In fact as evidenced by the findings, monetary return of investments might not be achieved. However, other metrics such as time to process transactions, service responsiveness, availability of government services or the ability to conduct transactions anytime and anywhere, and costs associated to getting the service weighs more significance.

Dessalegn Mequanint Yehuala

Intelligent License Plate Recognition

Road traffic accident is the leading cause of deaths and injuries in the world according to the World Health Organization (WHO). Every year many deaths and injuries are reported and most of them are in developing countries; the problem has great impact in Africa. Intelligent License Plate Recognition and reporting (ILPR) plays an important role in minimizing traffic accidents by implementing traffic monitoring and management systems. Since the number of vehicles are increasing, breaking traffic rules, entering restricted areas are becoming a trend. So, to control these actions, a system which can recognize vehicles by their License Plate (LP) is crucial. In this paper, we have developed ILPR system, which aims at reducing traffic accidents by processing an input image of a vehicle and reporting on its legality status. The ILPR starts with preprocessing and then extracts the LP using edge detection and vertical projection algorithms. To identify the License Plate Number (LPN), characters found on the LP are extracted and recognized by Artificial Neural Networks (ANN), which we trained with sample characters. If the recognized LPN is found to be a suspect after cross checking it with a pre-stored database, it will be sent to a person in charge via Short Message Service (SMS). In the recognition part, different papers use template matching, but is sensitive to noise. In order to mitigate the noise problem, our system uses ANN. We have also added SMS module. The system is implemented using MATLAB and Java.

Yaecob Girmay Gezahegn, Misgina Tsighe Hagos, Dereje H. Mariam W. Gebreal, Zeferu Teklay Gebreslassie, G. agziabher Ngusse G. Tekle, Yakob Kiros T. Haimanot

Comparison of Moving Object Segmentation Techniques

Moving object segmentation is the extraction of meaningful features from series of images. In this paper, different types of moving object segmentation techniques such as Principal Component Analysis (PCA), K-Means clustering (KM), Genetic Algorithm (GA) and Genetic Algorithm Initialized K-means clustering (GAIK) have been compared. From our analysis we have observed that PCA reduces dimension or size of data for further processing, which in return reduces the computational time. However, the segmentation quality sometimes becomes unacceptable. On the other hand, due to random initialization of its centroids, KM clustering sometimes converges to local minimum which results in bad segmentation. Another algorithm which has been considered in this study is GA, which searches all the feature space and results in a global optimum clustering. Although the segmentation quality is good, it is computationally expensive. To mitigate these problems, KM and GA are merged to form GAIK, where GA helps to initialize the centroids of the clustering. From our study, it has been found out that GAIK is superior to GA in both the quality of segmentation and computational time. Therefore, in general, the analyses of the four algorithms shows that GAIK is optimal for segmenting a moving object.

Yaecob Girmay Gezahegn, Abrham Kahsay Gebreselasie, Dereje H. Mariam W. Gebreal, Maarig Aregawi Hagos

ICT4DA Workshops

Frontmatter

Towards Group Fuzzy Analytical Hierarchy Process

Group decision making takes place in almost all domains. In building construction domain, a team of contractors with disparate specializations collaborate. Little research has been done to propose group decision making technique for this domain. As such, specific teams’ competitiveness enhancements are minimal as it takes more time for individual evaluators to choose the right partners. Qualitative and quantitative methods were used. Themes and categorizations were based on deductive approach. Subsequently, Group Fuzzy Analytical Hierarchy Process (GFAHP), Multi-Criteria Decision Making (MCDM) algorithm, was designed and applied. It uses all evaluation criteria unlike Fuzzy AHP (FAHP) which excludes some criteria that are assigned zero weights. GFAHP reduces the number of pairwise comparisons required when a large number of attributes are to be compared. Validation of the technique carried out by five case studies, show that GFAHP is approximately 98.7% accurate in the selection of partners.

George W. Musumba, Ruth D. Wario

Overview of Spectrum Sharing Models: A Path Towards 5G Spectrum Toolboxes

In this paper three spectrum sharing models are studied and their relative merits are outlined to allow dynamic spectrum sharing in all bands of interest. The main criterion is to improve the availability of underutilized spectrum for secondary wireless broadband networks. The three database-assisted spectrum sharing models studied in this paper are the Licensed Shared Access (LSA), Spectrum Access System (SAS) and Television White Space (TVWS). This paper proposes a unified spectrum sharing database-assisted model as solution for improving the broadband connectivity of underserved communities and improving spectrum availability for high bandwidth services in the fifth generation (5G) networks.

Gcina Dludla, Luzango Mfupe, Fisseha Mekuria

Towards Affordable Broadband Communication: A Quantitative Assessment of TV White Space in Tanzania

A quantitative assessment of TV White Space in Tanzania was conducted to assess the level of spectrum utilization as well as a key milestone towards the use of white space for affordable broadband communication. Two approaches have been used; pollution and protection viewpoints and experimental spectrum measurements based on energy detection principle. The study focused on 470–694 MHz UHF spectrum band which is used for digital terrestrial television in Tanzania. It was found that, more than 120 MHz is available as white space in various locations in Tanzania when pollution and protection view point was used and about 184 MHz are available as white space in Dodoma urban using experimental spectrum measurements and almost 100% of the available frequencies are not used in Dodoma rural. Both approaches revealed that there is low spectrum utilization and therefore presents a best case towards development of dynamic spectrum access technologies in Tanzania.

Jabhera Matogoro, Nerey H. Mvungi, Anatory Justinian, Abhay Karandikar, Jaspreet Singh

An Evaluation of the Performance of the University of Limpopo TVWS Trial Network

A comparative study of the performance of the TV White space (TVWS) network and WiFi is presented. The Software Defined Radios and Cognitive Network Technology have presented many opportunities and possibilities, which advances the wireless technology. This paper investigates the effectiveness of the TVWS technology in comparison with the legacy broadband WiFi technology. The TVWS technology utilizes the spectrum holes in the broadcasting frequency bands. The vacant frequencies can be used opportunistically by the TVWS technology in providing broadband solutions to rural areas among other areas. It is therefore imperative to investigate the effectiveness of this technology and its performance in relation to existing technologies. The comparative study of TVWS and WiFi is therefore presented. For this study, performance-monitoring techniques were employed to analyze the basic performance metrics such as throughput and latency of the TVWS and WiFi technologies. To evaluate the performance of the technologies, two-performance analysis tools were used, Internet performance open group (Iperf) and Java Performance/Scalability Testing Framework (jperf) tool. Speedtest was also used as an additional tool. The performance data were gathered and analyzed. The results show that the performance of the TVWS of the University of Limpopo trial network still requires significant improvement for it to at least match the performance of the legacy technologies such as WiFi.

Bongani Fenzile Mkhabela, Mthulisi Velempini

ICT4DA Demos & Exhibits

Review on Cognitive Radio Technology for Machine to Machine Communication

Recently, due to the rapid and fast ever growth number of connected devices/machines starting from our house hold appliances to the large industrial machines connected through wired/wireless communication network is becoming greater or larger. Hence, the electromagnetic undesirable state and interference become very critical issue as the spectrum resources we have is limited. However, a cognitive radio technology which automatically detects available channel spectrum, then accordingly changes its transmission or reception parameters to allow more occurring or operating at the same time in wireless communications in a given spectrum band at one location is a promising method for the challenges machine to machine communication facing this days. In this paper work, some detail survey of machine to machine communication and cognitive radio technology is introduced. Moreover, the challenges and advantages we get from combining cognitive radio and Machine to machine will be discussed.

Negasa B. Teshale, Habib M. Hussien

Backmatter

Additional information

Premium Partner

    Image Credits