Quality Assessment of Various CHC NAV GNSS Receiver Models

DOI: https://doi.org/10.52939/ijg.v19i5.2655

Global Navigation Satellite System or GNSS is used in many sectors and applications in obtaining the position or navigation of a location. Because it can easily provide exact position, including latitude, longitude, ellipsoidal height and time without the need to establish the angles and distance between different points, GNSS technology was soon adopted for surveying. It is an essential component of mapping and surveying procedures used all over the world. By using geodetic instruments, the highest probable accuracy of the data can be achieved in obtaining position, but it also depends on the quality provided by the instrument itself. There are two major forms of pseudo-range measurements from satellites that GNSS receivers provide which is code and carrier phase measurements. The code measurement is often noisy, and the carrier phase measurement is more exact than the former, however there is an ambiguity problem that arises from carrier tracking cycle slip [1].

An ideal GNSS receiver for geodetic and other surveying applications must be able to receive and track both code pseudo ranges and carrier phase signals, including the Y-codeless signal, in order to produce valid data [2].

Nowadays, even smartphone can provide positioning and navigation by using a low-cost GNSS chipsets [3] and [4]. This shows that regular user can use GNSS technology to obtain a position. However, in survey works, high accuracy and precision is required to provide the best data to be used. There are many manufacturers that produced GNSS receiver providing the same aim to offer instrument that can be used to obtain positioning. As GNSS receivers come from various brands, the quality of GNSS data provided from the receivers may be same but may also be different. The brands that usually used are Trimble, Topcon, Leica, CHC, South etc.

A variety of error factors influence the accuracy of GPS positioning. Users must first determine the primary error causes affecting the quality of GPS sensing data in order to acquire high-precision positioning outputs (i.e., GPS observations). Cycle slips, multi-path, quasi-random errors, and atmospheric delays are the major drivers of degradation in GPS data processing that can degrade the quality of observations and, as a result, the accuracy of positioning outputs [5] [6] [7] and [8]. In order to achieve consistent high-precision positioning findings with the GPS carrier-phase analysis, errors that are not defined in a functional or statistical model must be reliably found, eliminated, and managed in data processing. One of the most difficult aspects of GPS data quality control is reliability, which refers to the ability to recognize such flaws and foresee the implications on a solution [9]. Since the deployment of GPS/GNSS in life-critical applications, integrity has become a serious problem. This is due to the fact that restricted satellite visibility, noise, multipath effect, and other undesirable limitations such as environmental degradation can drastically reduce GNSS positioning performance [10] and [11]. Integrity is one of the factors used to assess GNSS performance where it measures the confidence in the accuracy of the information provided by the overall process. Integrity is a general performance characteristic that refers to a user's ability to trust the provided value of a specific location or velocity quantity (e.g., horizontal and vertical position) [12] and [13].

2. Problem Statement

GNSS observation is applied in most survey works as it provides the highest accuracy compared to previous method that used EDM equipment such as total stations. As GNSS technology is used widely, there are many manufacturers that offer different types of GNSS receiver models where each may not provide the best data quality for positioning as it still contains errors that need to be reduced. To obtain good quality of data, choosing the right equipment is necessary to ensure that the data is reliable to be used for positioning. But, due to certain budget limit, surveyor may choose low-cost equipment. Based on previous research, it tested one type low-cost GNSS receivers’ performance and concludes that low cost receiver can provide good performance as the survey grade receiver, and that their practical application was feasible and reliable [1] [2] [14] and [15]. Since this study is conducted in Malaysia, the brands that is used regularly in survey work or for education purposes are Trimble, Topcon, CHCNAV, Leica, South and others. There are no previous studies on quality assessment for these brands.

Thus, quality assessment is required for these brands to evaluate does the data acquired from various brands of GNSS receiver provide the same data quality or vice versa. Experimentation of various GNSS model is conducted by performing suitable GPS observation method then detection of vital parameters from the data obtained is required for data integrity checking. The quality checking can be done by using free software or commercial such as Spider QC, RTKLIB, CHC Geomatics Office 2, Anubis, etc. [10] and [14].

3. Materials and Methods

3.1 Study Area

This study requires data that is obtained from the GNSS observation. To perform GNSS observation, site selection is important to avoid any obstacles and reduce the error recorded in each observation. The site to be chosen must meet criteria such as 10° - 15° cut-off elevation angle and must be at open space that is far from hazard or obstruction. GNSS receiver must be situated at places where it cannot be disturbed so that observation can be done for a long time. The study area selected is at the Stadium Hoki, UiTM Shah Alam as shown in the >Figure 1. It is suitable as it is located at open space which will not disturb any passer-by there. Meanwhile, Figure 2 shows the GNSS setup during the observation at the study area.

3.2 Data Collection

Static method is chosen for data collection of this study. For this study, the observation was done for 2 days continuously for all of the instrument simultaneously to see if each model tracks the same error, same data throughout the observation.

Figure 1: Study area

Figure 2: Instrument placed on bleachers of the stadium

Each receiver used the same setting during the start of observation, which is 1HZ for sampling interval, 10° for elevation mask, 1440 minutes (24 hours) for logging time and antenna height which differs for every receiver but within 1.600 meter to 1.700 meter. There will be 5 different receivers to be used where only two tripods will be used to set up the receivers. Each 4 of the receivers will be put on a custom-made platform as shown in the Figure 3. Throughout the observation, the receivers are monitored every hour to ensure that the observation is continuing. The observation starts at 0000 of 23^rd April 2022 until 0000 of 25^thApril 2022 (GPST). As the logging time maximum is 1440 minutes which is 24 hours, the receivers static setting was relog on 0000 hours of everyday of observation to guarantee that the observation is ongoing.

Figure 3: A custom-made platform to put 4 receivers on one tripod

Table 1: Configuration for Quality Checking in CGO2

Table 3: Results for i83

Figure 6: Sky plot for 23 April (left) and 24 April (right) of i83

Figure 7: Distribution of coordinate for 23 April (left) and 24 April (right) of i83

Table 4: Results for i80

Figure 8: Sky plot for 23 April (left) and 24 April (right) day of i80

From the Table 4 above it can be seen that the receiver’s data completeness does not reach 80% for both days. The value of data completeness for this receiver is highly effected because there is no completed observation on R06, R10, R23, C19 – C30 and all Galileo satellites for both days. There is no data completeness with less than 80% in the satellites of the first day but there are two satellites that does not reach 80% data completeness which is G60 and G20 on the second day. The L1 and L2 frequency suffers the same multipath effects compared with only 0.01m difference between MP2 of both days. Even with small difference, the satellites with unpassed value of multipath is different for each day on each frequency. As example, most unpassed multipath value detected on the MP1 for both days. MP1’s highest value is 0.99m on R20 for the first day and 0.84m on R17 for the second day. As for MP2, there are only few unpassed multipath values where it is detected on G09, G12 on both days and additional of R04 and R24 on the first day of observation. MP2’s highest value is 0.83m and 0.84m for first and second day on G12.In terms of signal-to-noise ratio, the signal strength for both L1 and L2 is strong as the value recorded starts from 41dB to 43dB for each day where it only varies about 0.03dB for both frequencies.

Table 5: Results for i73

Figure 9: Distribution of coordinate for 23 April (left) and 24 April (right) of i80

This receiver shows a big cycle slip ratio for both days where it is 3833 for first day and 490 for second day. The value of cycle slip ratio obtained is large because the percentage of data completeness highly effecting this part. As the number of complete observations is low, combined with the ionospheric delay, the cycle slip ratio will obtain increased. Although the data completeness of first day is better than the second day, the ionospheric delay obtained were low on the first day resulting a high number of cycle slip ratio. The constellation with zero cycle slip detected on the first and second days are the same which is on R06, R09, R23, C19 – C30, all E constellations. The sky plot for i80 shows there is 10 GPS satellites visible on the first and second day as shown in the Figure 9. The precision of coordinates shown in the single point map for two days shows different pattern and there is a 0.19m of vector displacement between two days average coordinates.

The Table 5 above shows that the receiver's data completeness is different for each day. it is still acceptable as it surpasses 80%. On the first day, there are more satellites that have 100% of data completeness than there were on the second day but there are similarities on satellites that does not have any completed observation which is on R06, R10, R23 on both days. There is one different satellite with zero completed observation which is C11 for first day and C14 for second day. The multipath value of both days displays small different. When compared to MP2, the MP1 frequency suffers the least from multipath effects but they only differ around 0.02m maximum for both days. The difference can be seen with details in the satellites’ value of multipath where mostly unpassed MP1 occurred on GLONASS satellites with 1.27m as the highest multipath value recorded on R24 for the first day’s observation and 1.35m on R05 for the second day. As for MP2, the unpassed multipath value occurred mostly on GPS and GLONASS satellites with highest value which is 1.68m on G12 for the first day and 1.60m on G09. In terms of signal-to-noise ratio, the signal intensity for is approximately around 44dB per day for both L1 and L2 where the value of SNR on L2 is the same for both days. It appears that the cycle slip ratio for each day is the same, at 6. As they are satellites with 0 complete observation, there is no cycle slip detected on R06, R10, R23, C11 and C14. There are more satellites with 0 cycle slip ratio on the first day compared to second day. The satellites involved are C06, C07, C08, C09, C13, C23, C28, C30, C38, C39, C40 for first day and R20, C27, C43 for second day.

Table 6: Results for i70

The sky plot for the first and second day shows the same pattern of satellite visibility and have tracked 10 GPS satellites for both days as shown in the Figure 10. The precision of the single point map's coordinates illustrated in the Figure 11 over the course of two days displays a distinct trend, with a 0.28m of vector displacement in two days of average coordinate. From the Table 6, the data completeness for each day obtained more than 80% which is 86.9% and 87.2% respectively. There are total of 13 satellites (R06, R10, R23 and C21 – C30) that have 0 complete observation which cause the overall data completeness to be reduced on both days. It shows that the multipath is very small and less than 0.5m on both days. Clearly, there is small difference of about 0.02m maximum between the multipath error for L1 while multipath error for L2 differ around 0.05m maximum during the two-day observation period. In the quality check report, there is one unpassed multipath recorded on MP1 of R04 with 0.57m (only on the first day). MP2 only appears unpassed on the second day’s observation which is highest on R22 with 1.09m. In terms of signal-to-noise ratio, the signal intensity of L1 and L2 for each day does not vary that much where the difference between each L1 and L2 is 0.04dB.

Table 7: Multipath effect

Figure 12: Sky plot for 23 April (left) and 24 April (right) of i70

Figure 13: Distribution of coordinate for 23 April (left) and 24 April (right) of i70

The cycle slip ratio for each day demonstrates the same value, which is 5. Other than the mentioned satellites that have 0 complete observation earlier, there are other few satellites with 0 cycle slip can be seen such as C01, C02, C03, C05, C06, C07, C09, C10 and C16 for the first day together with G07, C13 and C20 for the second day. The pattern of sky plot as shown in the Figure 12 for each day shows almost the same and the number of visible satellites is 10 for the first and second day. A distinctive pattern can be seen in the precision of the coordinates for the single point map in the Figure 13 over the period of two days, with a 0.25m of vector displacement for average coordinates.

4.2 Data Consistency for Each Parameter

The interference between multiple reflected singles and direct GNSS signals received by a receiver's antenna causes multipath error. The reflected signal has a significantly lesser amplitude than the direct signal, and the superposition of the signals causes an interference delay in the received signal, causing the observation value to diverge from the true value. Based on the table above, each GNSS receiver suffers differently in multipath effect even though each receiver is situated at one place where it should obtain the same multipath effect. Table 7 above shows the multipath error for all models. Overall, the multipath value for MP1 for both days starts from 0.14m – 0.46m where the highest receiver effected by the multipath is i80. While for MP2, it starts from 0.15m – 0.46m for both days where the receiver with highest multipath value is i73.

Table 8: Cycle slip ratio