| 
  
   | Article of the Month - 
	  November 2014 |  
		A Prototype of RFID-Based Cadastral Boundary Mark System (RCBMS) in 
		Malaysia Tajul Ariffin MUSA, Abdullah Hisam OMAR, Ivin Amri 
		MUSLIMAN, Siti Syukriah KHAMDAN, Yip Kit MENG and Kamaludin OMAR, 
		Malaysia
		1)  This paper is peer 
		reviewed and presented at the 2014 FIG Congress in Kuala Lumpur, 
		Malaysia. The paper discuss an RFID-based cadastral boundary mark system 
		(RCMBS). The aim of the RCBMS is to modernize the conventional cadastral 
		boundary marks with marks that are lighter, robust, and easy to locate, 
		and which offer the capability to perform spatial/non-spatial cadastral 
		information on site. SUMMARY  The cadastral system in Malaysia needs to utilize 
		appropriate technology such as innovation in Information & Communication 
		Technology (ICT) to efficiently support a modern cadastral system and 
		infrastructure. Ubiquitous positioning by integrating a multi-sensor and 
		mobile database management system is an ICT innovation, which can 
		provide benefits to the cadastral surveying community, such as aiding 
		users in finding and/or updating information on the cadastral boundary 
		mark on site. In this paper, an RFID-based cadastral boundary mark 
		system (RCMBS) is discussed. The main aim of the RCBMS is to modernize 
		the conventional cadastral boundary marks with marks that are lighter, 
		robust, and easy to locate, and which offer the capability to perform 
		spatial/non-spatial cadastral information on site. The RCMBS contains 
		few subsystems and each component of the subsystem needs to be developed 
		in order to execute the system. A prototype platform of the RCBMS has 
		been produced to gather more information, demonstrate the functionality 
		to help solidify requirements, and technically understand the problems 
		of the system. It is expected that the RCBMS will provide a valuable 
		support for cadastral practice in the country. 
 
 1. INTRODUCTION  The technical future of cadastre was explored in the 
		1990s and resulted in ‘Cadastre 2014’ – a blueprint of technical 
		advancement which is a very useful guide to plan for future development 
		of the land administration system. The vision of Cadastre 2014 is to 
		present a comprehensive land recording system in order to replace the 
		traditional institution of ‘Cadastre’ and ‘Land Registration’ (Kaufman & 
		Steudler, 1998). The Malaysian government intends to implement Cadastre 
		2014 in its land registration system and cadastral data management. In 
		Malaysia, there are two organizations that are responsible for managing 
		and maintaining the cadastre system: (1) The Department of Survey and 
		Mapping Malaysia (DSMM) – responsible for preparing, producing and 
		managing the cadastral spatial information including the surveying and 
		mapping of the cadastre parcels; and (2) The Land Office (LO) – 
		responsible for managing cadastral attribute information. To move toward 
		Cadastre 2014, the Malaysian government has introduced and implemented 
		two systems; the land registration system called e-Tanah by the LO and 
		e-Kadaster by the DSMM. These two systems are considered as 
		stepping-stones toward the modernization of the cadastre system in 
		Malaysia. In the case of e-Kadaster, all the six (6) statements of 
		Cadastre 2014 have been explicitly executed (Hua, et al., 2012).  The cadastral system in Malaysia needs to make use of 
		appropriate technology such as innovation in ICT to efficiently support 
		the modern cadastral system and infrastructure. Ubiquitous positioning 
		by integrating a multi-sensor and mobile database management system is 
		an ICT innovation, which can provide benefits to the cadastral surveying 
		community, such as aiding users in finding and/or updating information 
		on cadastral boundary marks on site. Tcha (2006) proposed the 
		possibility to acquire cadastral information by wireless network 
		utilizing Radio Frequency Identification (RFID), which can be installed 
		in the boundary mark. However, superior Internet service and coverage 
		must be ensured to capture the information for the application of this 
		proposed approach. Musliman et al. (2012) use an approach of a low-cost 
		ubiquitous positioning by integrating a multi-sensor and online database 
		in a single system in order to locate and retrieve information of 
		cadastral boundary marks on site, in real-time. Their research focuses 
		on the Malaysia cadastral system with the possibility to apply a 
		cadastral cell-based concept for the proposed system. In this paper, we 
		extend the above work on the prototype of RFID-based cadastral boundary 
		mark system, or simply RCMBS.  This paper is organized into five sections. In 
		Section 2, the concept of the RCBMS is highlighted. In Section 3, 
		prototypes of the RFID-based boundary mark and the system applications 
		are emphasized. Discussion on the needs for a pilot study and the 
		possibility to test the RCBMS is provided in Section 4. Finally, Section 
		5 highlights the overall idea and concluding remarks.
 
 2. RCBMS: THE CONCEPT  The main aim of the RCBMS is to modernize the 
		conventional cadastral boundary mark with marks that are lighter, 
		robust, and easy to locate, and which offer the capability to perform 
		spatial/ non-spatial cadastral information on site.  2.1 The Approach  The RCBMS design matters for this study were 
		conducted in order to achieve the following criteria: (a) ability to 
		acquire spatial and attributes information of a cadastral boundary mark 
		directly on site; (b) a low-cost system and minimized time to locate the 
		cadastral boundary mark and retrieval of cadastral information; (c) 
		ability to utilize the National Digital Cadastral Database (NDCDB), and 
		(d) a single system to manage cadastral boundary mark information. The 
		common approach to achieve the above criteria therefore will consist of 
		technology update, input of cadastral information, effort on system 
		development and contribution from the potential user (see Figure 1).  
		 Figure 1: Common approach of the RCBMS.
 
 
 The RCBMS is adopted as a combination of Global 
		Positioning System (GPS) and the RFID. Both technologies are 
		straightforward in terms of application that is to effectively support 
		the RCBMS user to navigate to the location of the boundary mark and to 
		retrieve its cadastral information, respectively. The focus is also 
		given on the materials used in making of the boundary marks where 
		recyclable materials are preferred, and a technology in material 
		engineering is applied to produce strong and durable boundary marks at 
		low-cost.
 The DSMM has successfully developed the NDCDB for 
		Peninsular Malaysia. This national database contains the digital 
		cadastral information as the backbone for the e-Kadaster by DSMM which 
		serves cadastral activities in the country. The NDCDB is to be utilized 
		as the primary data source to facilitate the RCBMS. The challenge is 
		therefore to synchronize this large database of NDCDB with the mobile 
		database within the RCBMS.  The development of RCBMS consists of Subsystem A – 
		multi-sensors integration and design of the RFID-based boundary mark; 
		Subsystem B – the key to success in the RCBMS is the cadastral boundary 
		mark management system that deals with cell-based structure to optimize 
		data query and retrieval from the mobile database; and Subsystem C – to 
		design mobile applications of RCBMS within the RFID reader processing 
		unit. Figure 2 illustrates these three subsystems within the RCBMS.  
  Figure 2: Subsystem of the RCBMS.
 The final approach in the RCBMS is to attain the 
		cooperation amongst the government agency, university and private 
		sector. The government agencies, specifically the DSMM, work on the 
		policies, requirements and guidelines to ensure best practice of the 
		RCBMS amongst the private sector. The university is responsible for 
		conducting research and providing training for the RCBMS, both for the 
		government agencies and private sector. The private sector will benefit 
		from the RCBMS in its cadastral activities, and will also provide 
		valuable inputs to improve the system. This approach of combining the 
		interests of the above three parties will ensure best practice and 
		flexibility of the RCBMS.  2.2 Cadastral Cell-based Concept  Musliman et al. (2012) have proposed a cell-based 
		concept in order to manage the cadastral data for the RCBMS. The 
		cell-based concept has been used in the telecommunication industry which 
		is also referred to as the Global System for Mobile Communication (GSM). 
		The GSM network is made up of geographic areas. As shown in Figure 3, 
		these areas include cells, location areas (LAs), mobile services 
		switching center (MSC) or visitor location register (VLR) service areas 
		and public land mobile network (PLMN) areas.
 
 
		 Figure 3: Network Areas.
 The cell is the area giving radio coverage by one 
		base transceiver station. The GSM network identifies each cell via the 
		cell global identity (CGI) number assigned to each cell. The location 
		area is a group of cells which is the area in which the subscriber is 
		paged. Each LA is served by one or more base station controllers, yet 
		only a single MSC/VLR (refer Figure 4). Each LA is assigned with a 
		location area identity (LAI) number.  
		 Figure 4: Location Areas.
 An MSC/VLR service area represents the part of the 
		GSM network that is covered by one MSC and which is reachable, as it is 
		registered in the VLR of the MSC (refer Figure 5).  
		 Figure 5: MSC/VLR Service Areas.
 Similarly, the spatial cell-based concept can be 
		applied for the RCBMS with the RFID tagged marker being placed at spots 
		acting as the MSC/VLR. It is determined that the shape of the cell is 
		square, grid or hexagon (refer Figure 6). Jeong et al. (2011) have also 
		utilized the same concept of applying the RFID technology in their 
		prototype system of a 3D cadastre in Seoul, Republic of Korea. 
 
 
		 Figure 6: Grid type of spatial cell-based concept.
 2.3 Mobile Database for RCBMS  As aforementioned, the Subsystem B of the RCBMS deals 
		with the cadastral boundary mark management system. In this subsystem 
		lies the conceptual and physical development of the database in standard 
		structured query language (SQL) format for mobile database support. 
		Since a mobile device has storage and processing power limitation, the 
		use of a mobile database is proposed for committing data changes or 
		updates as seen in the field. The application also will send RFID tag 
		information at the current location to retrieve existing NDCDB 
		information at the server farm in real-time. It is not applicable to 
		store the whole NDCDB in a mobile device such as in the RFID reader. 
		Therefore the use of server side scripting is proposed and will be 
		embedded within the application system architecture. The data format 
		returned from this server side scripting for NDCDB information is in the 
		form of JavaScript Object Notation (JSON), i.e. a text-based open 
		standard design for human-readable data interchange across a 
		multi-platform. It is derived from the JavaScript scripting language for 
		representing simple data structures and associative arrays, called 
		objects.  In the RCBMS, a secondary database engine (MySQL) was 
		used to support the server side tasks and scripting. The MySQL database 
		is used to store RFID tag information of boundary marks and its unique 
		public identifier (UPI) key, which will virtually connect to the 
		existing NDCDB. The user will send a request to the server via the 
		application in C# environment. At the server side, each request will be 
		processed to perform standard structured query language (SQL). The 
		returned results are the information of an RFID-tagged cadastral lot 
		with its associated boundary mark and attribute information. The concept 
		of this application is shown in Figure 7.  
		 Figure 7: The concept of RCBMS mobile database and query system.
 A hypertext preprocessor (PHP) is used to perform the 
		server side task. Once a user sends a request to the server, the SQL 
		query will be executed and outputted as JSON data format (as shown in 
		Figure 8). Figure 9 shows the SQL select function code written in PHP.
		 
		 Figure 8: Sample of PHP syntax for data query.
 
 
 
 
		 Figure 9: Sample of JSON syntax within PHP for output from database 
		server.
 
 
 3. PROTOTYPE COMPONENTS OF RCBMS  As indicated in Section 2, the RCMBS contains several 
		subsystems and each component needs to be developed in order to execute 
		the system. The idea is to develop a prototype platform which can gather 
		more information, demonstrate the functionality to help solidify 
		requirements and technically understand the problem of the system.  3.1 Prototype 1: RFID-based Boundary Mark  In Prototype 1, the focus is in producing the 
		cadastral boundary mark itself. For many years, the conventional 
		cadastral boundary mark has been used in Malaysia. It is made from 
		concrete in the form of a cylindrical shape with a dimension of 70 mm in 
		diameter and 600 mm, with a weight of approximately 7 kg. This type of 
		boundary mark to a certain extent is heavy to transport, brittle and 
		does not carry any information on site. The main ingredients in 
		Prototype 1 are chosen from recyclable materials to ensure low-cost 
		production compared to the conventional concrete type boundary mark. 
		These recyclable materials consist of palm oil ash (POFA) and pulverized 
		fuel ash (PFA) (see Figure 10; left).  
  Figure 10: POFA and PFA (left) and compression test (right).
 
		 Table 1: Compression test results on three samples of Prototype 1.
 Code Sample A Sample B Sample CWeight of Specimen (kg) 0.83 0.85 0.82
 Diameter (mm) 70 70 70
 Area of Section (mm2) 3850 3850 3850
 Crushing Load (kN) 13.3 13.9 13.1
 Strength (N/mm) 3.5 3.6 3.4
 
 
 The compression test was performed on the three 
		samples of Prototype 1 (Figure 10; right) by the Civil Engineering 
		Testing Unit (CETU), Universiti Teknologi Malaysia. The compression test 
		indicates that all samples have almost similar results (see Table 1), 
		although Sample B has shown better impact on crushing load and strength. 
		This could be the reason of the weight of the specimen in Sample B which 
		is slightly heavier than the other two samples.  Several steps in producing the model of the 
		RFID-based cadastral boundary mark are shown in Figure 11. While most 
		steps can be undertaken in between 1 and 2 days, the curing process 
		requires a longer period of time. In the case of concrete, the duration 
		of curing depends on the grade and type of materials, mix proportion, 
		desired strength, shape and size of the concrete member, and 
		environmental and exposure conditions. The duration may vary from a few 
		days to a month (Kulkarni & Pereira, 2011). Experience in Prototype 1 
		has suggested that the curing process is between 7 and 28 days in order 
		to develop strength and durability of the boundary mark. The final step 
		in Prototype 1 is to equip the model with the RFID tag. A hard plastic 
		cover was used to protect the chipped RFID tag and it allows a signal 
		from the RFID reader to access the tag. The complete model of the 
		RFID-based cadastral boundary mark with its standard dimension is also 
		shown in Figure 11.
		
 
 
		 Figure 11: Steps in producing the RFID-based Boundary Mark.
 3.2 Prototype 2: The Applications  The RCMBS consists of several applications to 
		function as a complete system. In Prototype 2, these applications were 
		developed within the RFID reader by using C# Microsoft Visual Studio 
		programming language (Figure 12). The source codes of two RCBMS 
		applications for integration of spatial-attribute cadastral data and 
		user RFID interface were coded in the RFID reader.  
		 Figure 12: The source code for RCBMS applications.
 In the case of the spatial-attribute data 
		integration, the main task of the source code is to grab the cadastral 
		attributes once the RFID reader is able to scan the tag ID of the 
		RFID-based boundary mark (see also Section 2.3). These attributes were 
		arranged in each individual data table that requires a fast data search 
		engine through the database. Next, the source code will instruct the 
		RFID reader to display these attributes and the spatial data 
		accordingly. Figure 13 demonstrates this application as developed in 
		Prototype 2 of the RCBMS.  The source code for the user RFID interface is a 
		command or menu through which a user communicates with the RCBMS via the 
		RFID reader. The user interface in the RCBMS is a menu-driven interface 
		in which a user can select command choices from various menus displayed 
		on the screen of the RFID reader. The user only needs to scan the RFID 
		tag of the boundary mark in order to activate various menus and to 
		retrieve the cadastral information provided through the user interface. 
		Figure 14 shows the user interface of the RCBMS which was developed in 
		Prototype 2.  
		 Figure 13: Spatial-attribute data integration and retrieval.
 
		 Figure 14: RCBMS user interface.
 
 
 4. THE NEXT STEPS There are a few aspects of the RCBMS that require 
		further investigation. As aforementioned, the main aim of the RCBMS is 
		to modernize the conventional cadastral boundary mark with marks that 
		are lighter, robust and easy to locate, and which offer the capability 
		to perform spatial/non-spatial cadastral information on site. In the 
		next step of this study, a consideration is given to integrate the GPS 
		sensor with the RCBMS. The GPS sensor may utilize the coordinates in the 
		database to navigate the RCBMS user to the location of the boundary 
		mark. However, the navigation solution as estimated by the GPS sensor 
		may result in horizontal accuracy up to 17 m at 95% confidence level 
		(NavCen, 2008). As the location of the current boundary mark is not 
		necessarily with clear sky viewing conditions, this is a challenging 
		task for the RCBMS. 
 Next, it is suggested to improve the crushing load 
		and strength of the cadastral boundary mark in Prototype 1 (see Table 1) 
		in order to manage with the condition of the survey work. The 
		possibility is to mix the recyclable materials with some amount of 
		composite material such as coarse granular imbedded with cement. It is 
		predicted that the strength of the materials will be increased but will 
		be slightly heavier than Prototype 1. Subsequently, the focus must be 
		given on the dimension of the spatial cell-based grid (see Section 2.2). 
		The cell dimension is crucial to ensure the effectiveness in data 
		retrieval from the RCBMS database. A possible solution is to classify 
		the cadastral lots in that certain area, for example there are different 
		numbers of cadastral lots in rural and urban areas.  Another important issue in this study is to quantify 
		the benefits of the RCBMS, and then compare these with the costs 
		involved, i.e. cost-benefits analysis. In this way, the decision can be 
		made to proceed or carry out several modifications in each prototype of 
		the RCBMS. In addition, a small-scale experiment or pilot study could be 
		highly beneficial during this cost-benefits analysis with a reasonable 
		sample size of cadastral lots. 
 
 5. CONCLUDING REMARKS This paper briefly summarizes the concept and initial 
		prototype components of the RFID-based cadastral boundary mark system or 
		RCMBS. The design approach of the system has taken into account the 
		technology update, input of cadastral information, and the effort on 
		system development and contribution from the potential RCBMS users. The 
		concept of spatial cell-based grid is also explained to manage the 
		cadastral data in the RCBMS. Moreover, the use of a mobile database is 
		proposed for committing data changes or updates as seen in the field. 
		The mobile database is also suitable for the RFID reader device that 
		usually has storage and processing power limitation. In Prototype 1 of 
		the RCBMS, the recyclable materials have been used as the main 
		ingredients in producing the boundary mark. The compression test of 
		Prototype 1 shows that the sample can deliver up to 13.9 kN and 3.6 N/mm 
		on crushing load and strength, respectively. In Prototype 2, the source 
		codes of two RCBMS applications for integration of spatial-attribute 
		cadastral data and user RFID interface were coded in the RFID reader. 
		The RCBMS is expected to speed up the time taken that is usually used to 
		find and/or update the cadastral information on site. Moreover, the 
		RCBMS can generate revenue from service fees where the user may be able 
		to retrieve the online cadastral information.
 
 REFERENCES  Hua, T. C., Lim, C. K. and Abdul Halim, N. Z. (2012) 
		Menanda Aras Kemajuan Sistem Ukur Kadaster Berasaskan Kadaster 2014 - 
		Persediaan JUPEM ke arah menjayakan Kadaster 2.0. Persidangan 
		Pengarah-pengarah Ukur 2012, access via http://www.jupem.gov.my/.  Jeong, D.H., Kim, T.J., Nam, D.H., Li, H.S. and Cho, 
		H.K. (2011) A Review of 3D Cadastre Pilot Project and the Policy of 3D 
		NSDI in the Republic of Korea. The 2nd International Workshop on 3D 
		Cadastres, 16th-18th November 2011, Delft, the Netherlands.  Kaufmann, J. and Steudler, D. (1998) Cadastre 2014 – 
		A Vision for a Future Cadastral System. Working Group 01, Commission 7, 
		International Federation of Surveyors (FIG).  Kulkarni, S. B. and Pereira, C. (2011) Significance 
		of Curing of Concrete for Durability of Structures. NBM Construction 
		Information Portal, access via http://www.nbmcw.com/.  Musliman, I. A., Musa, T. A. and Omar, K. (2012) 
		Integration of Multi-Sensor for Modern Cadastral Boundary Mark: First 
		Experience. Buletin Geospatial, MaCGDI, Edisi 2/2012, pp. 6-11.  NavCen, (2008) Global Positioning System Standard 
		Positioning Service Performance Standard. 4th Ed. US Department of 
		Defense: Position, Navigation, and Timing Executive Committee. 
		Washington, DC.  Tcha, D. (2006) A Study on the U-Cadastral Space Data 
		Modeling in Korea. XXIII International Federation of Surveyors (FIG) 
		Congress, Munich, Germany, Oct 8-13.
 CONTACTS  Dr Tajul Ariffin Musa, Dr Abdullah Hisam OmarGNSS & Geodynamics Research Group
 Faculty of Geoinformation & Real Estate
 81310 Universiti Teknologi Malaysia
 Johor Bahru, MALAYSIA.
 Email: tajul.fksg@gmail.com,
		abd_hisham@yahoo.com
 Web site:
		
		http://www.geoinfo.utm.my/Research_Group/gng/aboutus.html
   |