Provincial Electricity Authority in Thailand has devised its own low cost external antenna to improve signal reception capability of GPS receivers to ensure accuracy of spatial data and enhance the process of spatial data collection
Provincial Electricity Authority (PEA) is the public-sector power utility responsible for supplying electricity to all regions of Thailand outside of metropolitan Bangkok. In the past few years, PEA has embarked on a project to modernise and enhance its existing GIS to ensure that it is in line with the digital business transformation activities being implemented at PEA. Several enterprise information system projects have been initiated by PEA including the AM/FM/GIS project. The AM/FM/GIS project aims to support the other projects at PEA in order to deliver the best services to its customers. Spatial data capturing is one of the major aspects of the AM/FM/GIS project since it involves a building of large spatial database (i.e. base map, electricity poles). Although this project has been subcontracted to a private company, the QC procedure is an unavoidable task. GPS technology was chosen as the main tool for the QC procedure. Spatial data specifications, especially in rural areas, require accuracy of better than 5 metres. As a result, a total of 152 sets of geodetic-graded Leica GS20 GPS receivers were purchased and used for spatial data checking. However, it was found that the initialisation time of the LeicaGS20 receiver was relatively slow and in some cases this could take up to 30 minutes. It was suggested by the manufacturer that an external Leica AT501 antenna should be used with Leica GS20 receiver to reduce signal acquisition time. Nevertheless, the cost of external Leica AT501 antenna was rather expensive. To solve the problem, PEA has built a smart bag equipped with a low-cost external antenna, a re-radiation kit and external battery. Results obtained from various testing environments indicate that the use of low-cost external antenna can significantly improve the GPS signal acquisition time while still maintaining the same level of positional accuracy.
Introduction
Provincial Electricity Authority (PEA) is a power utility state enterprise responsible for supplying electricity for all regions of Thailand outside of metropolitan Bangkok. With its duty to produce, acquire and distribute electric power to all households, businesses and industrial units in 73 provinces countrywide, PEA has given the highest attention to developing various enterprise-wide IT systems to strengthen its operational competence. This includes establishing the geospatial AM/FM/GIS system. In the past few years, PEA has embarked on a project to modernise and enhance its existing GIS system to ensure that it is in line with the digital business transformation activities being implemented at PEA. The AM/FM/GIS project aims to support the other projects at PEA in order to deliver the best services to its customers. Spatial data capturing and maintenance is one of the major aspects under the AM/FM/GIS project since it involves building and updating a large spatial database (i.e. base map, electricity poles, cables, transformers and metres). More than 15 million poles and about the same number of customer metres, together with transformers and all other network devices, were surveyed and captured into the PEA’s GIS database within the project’s three years period. Data in the municipal areas were surveyed at positional accuracy of 1 metre while accuracy specification was reduced to 5 metres in the other areas.
In the process of initial data surveying executed by a private contractor, PEA has an unavoidable task of data product accuracy conformance checking and QC processes. After the system and data is handed over to PEA, users in its provincial offices have to continuously survey position of the new or repositioned electric poles from the field in order to update the GIS database. GPS technology was chosen as a main tool for the positional data surveying procedure. A total of 152 sets of geodetic-graded LeicaGS20 GPS receivers were purchased and distributed to every provincial and district office of PEA. These GPS receivers are now being used to survey position of new poles and also new base-map features such as road centerlines and points-of-interests.
With its proclaimed positional accuracy specification of less than 5 metres in single point surveying, Leica GS20 receiver has meet the requirement of PEA’s role in surveying GIS data in area outside municipalities. After a few months of use however, it was found that the initialization time of some receivers was relatively slow and in some cases this could take up to 30 minutes. Many users also expressed their dissatisfaction with the speed of signal acquisition in the surveying operation. These complaints continued even after all the GPS receivers were sent back to the manufacturer for internal software and the hardware upgraded.
Although the equipment had passed the test for its performance to be at a normal level, it could not satisfy the expectations set by most PEA users. To solve the problem, it was suggested by the manufacturer that an external Leica AT501 antenna be used with the Leica GS20 receiver to reduce a signal acquisition time. This would also support the uses of the GPS equipment in continuous survey of road centerline without taking the receiver out of the vehicles. Nevertheless, the cost of the external Leica AT501 antenna was rather expensive while the entire phase II project budget was already exhausted. Trying to relieve the situation with limited resources, the PEA engineering team came up with the idea of using a low-cost GPS antenna to increase the GS 20’s signal tracking performance. Two types of devices, “Smart Bag” and “Smart Hat” were developed as complementary peripherals for the Leica GS20 receivers. Both the devices are equipped with a low-cost external antenna and re-radiation kit (Holux AR-10 model). Although they are quite similar, the “Smart Bag” device was configured to be more enduring and stable. It was then chosen to be further studied for its capability and performance.
In the following sections, the components of the Smart Bag are briefly introduced. Next, testing procedure and results are explained. Discussion of results is presented in the next section. Finally, some concluding remarks are given.
Development of Smart Bag
Previous tests have confirmed that the use of external geodetic-graded GPS antenna can significantly reduce the signal acquisition time and produce more accurate positioning results as compared to the use of internal GPS antenna (Satirapod et al., 2009; Tinnachote et al., 2011). However, the cost for purchasing 152 external geodetic-graded GPS antenna was unfeasible for the AM/FM/GIS project. In order to overcome the situation, the PEA engineering team developed a SmartBag consisting of a low-cost GPS antenna, a re-radiation kit (Holux AR-10) and a signal protection bag. The SmartBag can then be used to replace the external GPS antenna. Moreover, the Smart Bagsystem costs only one-twentieth of the cost of an external geodetic-graded GPS antenna (i.e. Leica AT501). A photo of the Smart Bag is shown in Figure 1.
Figure 1 The Smart Bagsystem developed by PEA
Testing procedure and results
To assess the quality of the Smart Bag, PEA set up two tests, signal acquisition time and data quality. Both tests are explained and results are given below. Signal acquisition time
This test aimed to check whether the use of the Smart Bag can yield similar signal acquisition time as the use of external geodetic-graded GPS antenna. The test involved the use of two Leica GS20 receivers, the first receiver connected with the Smart Bag while the second receiver connected with the external geodetic-graded GPS antenna. Test architecture is illustrated in Figure 2. In the first attempt, both the receivers were switched on at the same time and the period used recorded until each receiver yielded the first 3D coordinate solution, so-called ‘initialization time’. This test was repeated 10 times. Table 1 is a summary of results.
Figure 2 Signal acquisition time test architecture
Table 1 Comparison of initialisation time between the Smart Bag and the external antenna without deleting an almanac file
For the second attempt, we checked for the period used to produce the first 3D coordinate solution after deleting the receiver’s memory. We reset the receiver’s memory to clear up the almanac file before turning the receiver on. Then, we recorded the initialisation time. Similarly, this test was repeated 10 times and results are summarised in Table 2. Table 2 Comparison of initialization time between the Smart Bag and the external antenna by deleting an almanac file 
Data quality
In this test, we conducted two tests to check for the quality of both carrier phase and pseudo-range observations.
Carrier phase measurement test
The first test aimed to check whether the Smart Bag system can produce reliable carrier phase measurements. We therefore set up the low-cost antenna on the CU01 point located at Chulalongkorn University, Bangkok, Thailand. The coordinates of CU01 site are known and accurate to millimeter level (Satirapod, 2005). The data was collected in static mode for approximately 5 hours at a 15-second sampling rate (Figure 3). The data was converted to RINEX files. The reference station data, the IGS CUSV station which is located nearby the CU01 site, was downloaded and processed together with the data from CU01 site using the Leica Geomatic Office (LGO) software. After the data was processed, the solutions were compared with the reference coordinates. The difference in the north and east components was 1.31 cm and 1.33 cm, respectively.
Figure 3 Data collection at CU01 site, Chulalongkorn University, Bangkok, Thailand
Pseudo-range measurement test
The second test aims to check whether the pseudo-range measurements obtained from the Smart Bag system can produce reliable coordinate solutions. Similarly, we collected the data at CU01 point for approximately 3 hours using the stream mode. The stream mode will automatically store instantaneous coordinate solution in a log file at 15-second rate. The solutions were compared with the reference coordinates. Table 3 summarises the statistics.
Table 3 A summary of error statistics obtained from the1st set of pseudo-range measurements using the Smart Bag system
It can be clearly seen from Table 3 that unexpected large discrepancies were observed. We suspected a problem with the field data collection procedure. We then checked the photo taken during the observation and found that the box containing the receiver and re-radiation kit was half opened. There might be a signal intervention from outside and this could lead to unexpected errors in GPS pseudo-range observations. Therefore, we decided to run a similar test for approximately 5 hours with two SMART bag systems, one set up in a close box and another one set up in an open box. Table 4 shows the new results. Table 4 A summary of error statistics obtained from the 2nd set of pseudo-range measurements using the Smart Bag system
Discussion of results
Previous test results on the signal acquisition time with the use of internal antenna showed that the averaged initialisation time was 238 seconds without deleting the almanac file and 882 seconds on deleting the almanac file (Tinnachote et al., 2011). The use of both Smart bag and external antenna can significantly improve the signal acquisition time as compared to the use of internal antenna. Referring to tables 1 and 2, the average initialisation time between the use of Smart Bag and the external antenna were not statistically different. Thus, it can be concluded that the signal tracking performance of the Smart Bag is the same level as the external geodetic-graded GPS antenna.
Further tests were carried out to assess the quality of carrier phase and pseudo-range observations. For the carrier phase measurement test as described earlier, it can be clearly seen that the obtainable accuracy is at centimetre accuracy. This therefore confirms that the quality of carrier phase measurements obtained from the Smart Bag is high and the Smart Bag can be used for high accuracy applications (i.e. surveying and mapping). In case of pseudo-range measurements, large errors were found when the box was half opened during the field data collection. It might be caused by an intervention of bouncing signal into the receiver. This is confirmed by the second test which compares a performance of the Smart Bag in an open box and a close box. In relation to the Table 4, it is evident that the signal intervention can degrade an obtainable accuracy of the Smart Bag. Nevertheless, the quality of pseudo-range measurements is considered as the same quality as the use of external antenna.
Concluding remarks
This paper presents the test results of the developed Smart Bag. Results obtained from the signal acquisition time test confirmed that both the Smart Bag and the external geodetic-graded antenna yield similar performance. In addition, the use of Smart Bag still maintains positional accuracy at centimetre level and metre level with the use of carrier phase and pseudo-range observations, respectively. However, it is strongly recommended that when using the Smart Bag system for data collection, the box has to be closed in order to prevent any possible signal intervention. Therefore, the Smart Bag can be used to replace the external antenna without deteriorating an initialisation time and an obtainable accuracy.
References
- PEA (2011) Development of SMART bag and its performance, Plan and operation division, Network Operation Department, Provincial Electricity Authority (PEA), 3 July 2011, 38pp.
- Satirapod, C. (2005) The Use of Handheld GPS Receiver for High Accuracy Positioning Applications, Proceeding of the 10th National Convention on Civil Engineering (NCCE), Chonburi, Thailand, 2-4 May 2005, SIE1-5.
- Satirapod, C., Tinnachote, C. and Kriengkraiwasin, S. (2009) Obtainable Accuracy from GPS Surveying Under Real Environment for Positional Assessment of PEA”s Electricity Pole, Proceeding of the 14th National Convention on Civil Engineering (NCCE), Nokornratchasima, Thailand, 13-15 May 2009, 1077-1084.
- Tinnachote, C., Satirapod, C., Kriengkraiwasin, S. and Skawrattananont, S. (2011) Monthly report of the PEA AM/FM/GIS (Phase II) consultant project, August 2011, 75 pp.
