Analysis of IRNSS over Indian Subcontinent Vyasaraj Guru Rao1, Gérard Lachapelle 1, Vijay Kumar S B2 1 Position, Location And Navigation (PLAN) Group http://plan.geomatics.ucalgary.ca Department of Geomatics Engineering University of Calgary 2 Accord Software & Systems Pvt. Ltd Bangalore, India
BIOGRAPHIES Vyasaraj Guru Rao is a PhD student in the Plan Group, Geomatics Engineering, University of Calgary. He has been working in the area of GNSS navigation systems for the past fourteen years with Accord Software & Systems Pvt Ltd, Bangalore, India. His research areas are Multi constellation GNSS Receiver design, Algorithm development, High Dynamics Receiver development and Pseudolite based applications. He holds a Masters Degree in Software and Systems from BITS Pilani, India and a Bachelor of Engineering Degree in Electronics and Communications from Bangalore University, India.
Prof. Gérard Lachapelle is a professor of Geomatics Engineering at the University of Calgary where he is responsible for teaching and research related to location, positioning and navigation. He has been involved with GPS developments and applications since 1980. More information is available on http://PLAN.geomatics.ucalgary.ca
Vijaykumar S.B. is a senior project leader at Accord software and systems Pvt. Ltd., Bangalore, India. He has been working in the area of GNSS navigation systems for the past eight years. His area of work includes multi constellation GNSS receiver design (GPS and GLONASS), High dynamics receiver development and pseudolite-based applications. He holds a Bachelor degree in Electronics & Communications from Gulbarga University, India.
ABSTRACT India is planning to deploy an autonomous regional satellite based navigation system to cover its territorial footprint and surrounding areas. The purpose of this system as any other existing GNSS is to cater to the needs of both specific users (Precision Service (PS)) and civilian users (Standard Positioning Service (SPS)) (Kibe & Gowrishankar 2008). The overall constellation consists of seven satellites, of which three will be in GEO stationary and four in GEO synchronous orbits. This system will be used for surveying, telecommunication, ION ITM 2011, Session B5, San Diego, CA, 24-26 January 2011
transport, identifying disaster locations and public safety, along with host of other applications. In this research work a deeper understanding on parameters specifically of interest to the receiver community such as availability, accuracy and reliability from IRNSS are explored. The impact of augmenting IRNSS with other GNSS such as GPS and GLONASS is also investigated. Finally, a discussion on how the proposed system lends itself to carrier phase positioning applications is detailed.
1
INTRODUCTION
With the advent of any new constellation brings the curiosity as to what it might offer to the users under various conditions. The Indian Regional Satellite Navigation System (IRNSS) (Kibe & Gowrishankar 2008), still being in its nascent stages of design gives ample opportunities for various research and algorithm developmental activities. With the announcement of any new constellation, an initial assessment of the signal characteristics, standalone availability and achievable accuracies is of main interest. This research attempts to focus on some of the key aspects like availability, accuracy and reliability, understand and bring out its characteristics to users within and around India. Availability is a measure of the number and geometry of satellites available for computing navigation solutions. In the case of IRNSS, assuming clear sky conditions, satellites in the GEO stationary orbit will always be visible to users on Indian subcontinent (Kibe & Gowrishankar 2008). An advantage of GEO stationary satellites is that a larger signal coverage footprint is achieved with a minimum number of satellites. However, one needs to estimate the visibility of the satellites in GEO synchronous orbit to access the overall availability. Following availability, a thorough investigation is then performed on accuracy, which gives a measure of how close a computed navigation solution is w.r.t user’s truth (Alves et al 2001). This is further decomposed into two 1/13
components: user equivalent range error (UERE) and geometric dilution of precision (GDOP). UERE is obtained by mapping all the system errors as one lumped parameter in measured range (O’Keefe 2001). DOP is a satellite geometry dependent (based on availability) parameter, which along with UERE determines user accuracy (in position domain). GDOP can further be decomposed into PDOP, HDOP, VDOP and TDOP representing position, horizontal position, vertical position and time dilution of precision respectively (Langley 1999). The DOP values provide a measure of system’s availability as they represent the geometric strength of solutions. It can also be used to represent system accuracy under the assumption that all range measurements have a similar UERE (Alves et al 2001). Reliability is the next important parameter of interest, which is a measure of system’s ability to detect faults and to estimate the effect of an undetected fault on overall solution (Ryan & Lachapelle 2000). Reliability can further be subdivided into internal reliability and external reliability (Alves et al 2001). Internal Reliability refers to system’s ability to detect a blunder through statistical testing of residuals on an epoch-by-epoch basis. The smallest such blunder is called marginally detectable blunder (MDB). The external reliability of a system is quantified by the magnitude of error in navigation solution caused by MDB (Ryan & Lachapelle 1999). Further, analysis with user defined elevation angle is performed. The availability under reduced coverage and corresponding accuracies are analyzed. Simulations for open sky conditions (base station receivers) and urban canyon scenarios reveal the satellite distribution and their accuracies. An analysis to evaluate the benefits of combining IRNSS with GPS and GLONASS was conducted. The fact that IRNSS is a regional system can use other GNSS such as GPS and GLONASS as leverage. Thus, performance parameters such as availability, accuracy and reliability are analyzed by augmenting IRNSS with GPS and GLONASS.
results for a standalone IRNSS system and Section 5 provides results of IRNSS augmented with GPS and GLONASS systems. Section 6 covers a discussion on carrier phase positioning with IRNSS and finally conclusions and future work are enumerated in Sectio n 7.
2
IRNSS SIGNAL CHARCTERISTICS
The proposed IRNSS system operation will be in two frequency bands, namely L5 and S1. On each of these frequencies, two services are available, one for civilian and the other exclusively for specific users. The SPS will have BPSK modulation and is available for the civilian community. Table 1 summarizes the IRNSS signal frequencies (Kibe & Gowrishankar 2008).
Table 1: IRNSS services and frequency of operation Service
Standard Positioning Service (SPS) Restricted Service (RS)
Frequency of Operation
Modulation
L5 (1176.45 MHz)
S (2492.08 MHz)
YES
YES
BPSK
YES
YES
BOC(5,2)
IRNSS will consist of seven satellites, with 3 in GEO and 4 in GSO orbits. Figure 1 gives the sky plot of the IRNSS satellites. The three GEO stationary satellites will be stationed at 34º, 83º and 132º longitudes, respectively. The four GSO’s are expected to be in orbits of 24,000 km apogee and 250 km perigee inclined at 29º. Two of the GSO’s will cross the equator at 55º and the other two at 111.5º longitude respectively (Kibe & Gowrishankar 2008).
Finally, the paper touches upon some of the existing methods of carrier phase positioning and how the proposed IRNSS signal structure (Kibe & Gowrishankar 2008) lends itself to this application. Three of the seven satellites are in GEO stationary orbits. The potential challenges of ambiguity fixing are detailed. The L5 and S1 bands are separated by more than 1.2 GHz. Related challenges on carrier phase positioning are discussed. The rest of the paper is organized as follows: IRNSS signal characteristics are provided in Section 2 and the simulation methodology followed by various analyses is detailed in Section 3. Section 4 provides the analysis ION ITM 2011, Session B5, San Diego, CA, 24-26 January 2011
Figure 1: IRNSS Satellite Ground Trace For simulation and subsequent analysis, parameters like time, user position and the approximate satellite positions are required. The mission planning or pre-analysis can be 2/13
carried out with these parameters. The following section details the simulation analysis tool and its modules.
3
SIMULATION METHODOLOGY
This section highlights major modules of the software analysis tool. The functionalities of each module and interconnects amongst various software components are explained in detail. Finally, values of some of the parameters considered for simulation along with rationale behind their selection are explained. In order to carry out the analysis, a software tool was developed in Matlab. This tool can generate the IRNSS, GPS and GLONASS constellations. A high level design abstraction of various modules of the simulation software is given in Figure 2. The simulations can be performed at two different user levels. First, single user input mode, in which the analysis corresponds to a particular user position (fixed latitude and longitude). Second mode of operation is designated as Sweep mode. Here, the analysis is performed on an extended area of latitude and longitude, i.e. as a grid. Based on functionalities, the simulation software at a top level can be divided into two major blocks as simulator and receiver modules. The simulator and receiver modules together provide flexibility to generate different satellite profiles and a platform to conduct analysis of various performance parameters. Functionalities of each of these modules are explained in the following section. The three basic parameters required for simulation purpose are time and approximate user and satellite positions (GPS ICD 200C). Effectively these three become the sub functions of the simulator module. The output parameters are visible satellites and their corresponding position. The true range to each satellite is computed using user and satellite positions. The default mask angle for simulation is set to 5º. However, user mask angle can be configured in the range of 5º to 90º. Finally, in order to bring in the effect of errors, i.e. pseudoranges, errors corresponding to each of these satellites are computed separately.
σ i2
=
σ z2
(1)
sin 2 ( E i )
For simulation purpose, a lumped error corresponding to each satellite is formulated based on Eq. (1), where σ z is the apriori error estimate (UERE), σ i is the standard th
deviation of measurement of i satellite and E is the elevation angle (Petovello 2010). Table 2 gives the simulation parameters and the corresponding default values considered as part of the analysis.
ION ITM 2011, Session B5, San Diego, CA, 24-26 January 2011
Following are sub-tasks of the receiver module: DOP computation, user position computation in constraint modes, reliability performance and the analysis with configured mask angles. The subtasks work on data output from the simulator module, processes and generates the corresponding results. Constraint mode and reliability analysis are user configurable as shown in Figure 2. The functionalities of the receiver form the core of any pre-analysis or mission planning. The rationale behind choosing the simulation parameters are as follows: To get a deeper insight on satellite distribution, wider grid spacing was used. However, for the analysis of specific parameters such as availability, accuracy and reliability, smaller grid spacing is configured. The system error budget is assumed to be similar to GPS. The standard deviations of errors assumed for analysis are in Table 3 (Kaplan & Hegarty 2006).
Table 2: Simulation parameters Parameter
Description
Grid Spacing (Latitude x Longitude) UERE Elevation Mask Angle
For Visibility: 10º x 10º Others: 5º x 5º 6m Default: 5º User Configured: 25º 30 min, 24 h 3m
Sample rate, duration Initial clock error for clock constraints Reliability Analysis
Indian sub continent region, where all the satellites are visible
For optimal performance, the receiver is configured with a default elevation mask angle of 5º (Alves et al 2001). The reason behind this is to reduce errors due to multipath and high ionosphere errors associated with low elevation satellites (Kaplan & Hegarty (2006), Alves et al (2001)). As such, a 5º mask angle is used for all the analysis. IRNSS satellites are in GEO and GSO orbits and the system design is such that their visibility is always ensured over India (Kibe & Gowrishankar 2008). With this design input, thirty minutes is a reasonable approximation. In most hand held receiver designs, 10PPM TCXO or even a crystal is used as the frequency source. An initial clock estimate of 3 m is a reasonable value, which is used as initial value of clock in constraint mode. The UERE is:
Table 3: 1-sigma error assumed in UERE calculation Description Satellite Clock Satellite ephemeris Ionosphere Troposphere Multipath Receiver noise
Error (m) 1 1 5 1 2 1 3/13
Figure 2: Flo chart of Simulation Software Modules
UERE = 12 + 12
2
2
2
2
+ 5 + 1 + 2 +1
6m (2) Thus a UERE of 6 m has been used for the simulation purpose as in Eq. (2). With the above information and the analysis tool, the following section present s the results of various individual analyses.
4
slightly beyond (Kibe Bhaskarayana (2008)).
&
Gowrishankar
(2008),
STANDALONE IRNSS ANALYSIS
This section details the analysis carried out using only the IRNSS satellites.
4.1
Availability
This section discusses the satellite visibility over a given region. Availability is the period of time s stem is usable or alternatively it is the ability of the syst em to provide solution over a specified region (Alves et al 2001). The intended region of operation of IRNS S is primarily over the Indian subcontinent (Kibe & Gowrishankar 2008). However, to get a deeper insight on the distribution of satellites, visibility has been computed in the region of -40º to 60º latitude and 0º to 140º longitude. Accordingly, the IRNSS satellite distrib tion is as in Figure 3. From Figure 3, it is evident that all seven I NSS satellites are assured to be visible over Indian su continent and ION ITM 2011, Session B5, San Diego, CA, 24-26 January 2011
Figure 3: IRNSS satellites availability However, based on local phen menas (in-between tall building) and antenna characteris tics, the effective visible satellites may be different. Figur 3 shows that standalone positioning is achievable with I NSS satellites over the Indian subcontinent. Unlike existing systems such as GPS or GLONASS or the proposed G LILEO and COMPASS which are global, IRNSS is a regi onal system. As such the visibility gradually decreases away from the Indian subcontinent region and thu has an impact on positioning. For the latter areas, positioning can be 4/13
accomplished by augmenting (IRNSS) with existing systems, which is explained in Section 5.
4.2
Accuracy
Dilution of precision (DOP) can be computed with satellite and user positions (Langley 1999). DOP maps the UERE into the position domain (Alves et al 2001). Figure 4 gives the position accuracy achievable with IRNSS satellites, which is of the order of 20 m over the Indian subcontinent (Kibe & Gowrishankar 2008). An assumption made here is that the user’s antenna has an unobstructed view to all visible satellites. As with other operational systems, the IRNSS satellites are evenly spread in the azimuth plane over the Indian sub continent. Thus the geometry is relatively better in the horizontal than the vertical dimension (Figure 4). Further, position accuracy is computed using (Langley 1999) a ccu ra cy
=
4.3
P D O P *U ER E
=
(
H D O P
2
+
VDOP
(3) 2
) * U E R E
external reliability. Internal reliability gives a measure of error magnitude which can be detected by the system whereas external reliability gives the measure of such an error on the overall solution (Ryan & Lachapelle 2000). IRNSS has three GEO and four GSO satellites. For the purpose of analysis, a representative of each orbit (GEO-1 and GSO-4) is chosen. The assumption here is that at any given instant only one satellite has blunder. The impact of blunder, Marginally Detectable Blunder (MDB), (O'Keefe 2001) on the horizontal and vertical positions is analyzed. The analysis was carried out using least squares residuals and the main equations used for computation are (Petovello 2010)
r
=
Cr R −1 z
(4)
where C r is the covariance of residuals, R is the covariance matrix of measurements and z is the vector of measurements.
Reliability
Statistical reliability is a method of identifying blunders or outliers. Broadly, it is categorized as internal and
Figure 4: IRNSS DOP and Position accuracy over and around Indian subcontinent
ION ITM 2011, Session B5, San Diego, CA, 24-26 January 2011
5/13
Figure 5: External Re iability of IRNSS, GEO-1 and GSO-4 satellites With blunder on either a GSO or a GE measurement vector can be written as
Z
=
satellite, the
i
Hx + M ∇ + v
(5)
Where H is the design matrix, M is vect r representing the satellite with blunder (i.e. either GEO r GSO) and v is that of measurement errors (with ze ro mean and Gaussian distribution).
(C r )ii
N
Rii
∇i ,1
(6)
Or,
(C r )ii
∆ri
(C r )ii Here
= N
Rii
(C r )ii Rii
is
(9)
∆ xi = ( H
T
R −1 H ) −1 H T R −1M i ∇ MDBi
(10)
More specifically,
HPEi
=
sqrt ( ∆xi (i,1) 2
+ ∆xi
i , 2) 2 )
(11)
= ∆xi ( i, 3) 3)
(12)
Eq. (11) & (12) are used in the computation of the horizontal position error (HPE) and the vertical position error (VPE) respectively.
(7)
to
as
the
non-centrality
parameter. This for the analysis is assum d to be 3.24, which represents a shift in the normal istribution of standardized residuals. The regions f incorrectly accepting (Type 2) or wrongly rejecting (Type-1) null hypothesis are as shown in Figure 6. T he probability associated with Type1 and Type2 errors is denoted by α and β , respectively. The non-centrality parameter is given as
δ 0 = N1−α / 2 + N 1− β
δ 0
∇ i ,1
referred
(C r )ii
. With the above equation, the ex ernal reliability is given as
VPEi
th
The residual on the i satellite (with error) is ∆ri =
( R)ii
∇ MDB =
(8)
ION ITM 2011, Session B5, San Diego, CA, 24-26 January 2011
Figure 6: Type 1/2 errors and Non Centrality parameter From Figure 5, an observable difference is the analysis wind w on longitudinal axis, which runs from 50º to 90º. A common region where both GEO-1 and GSO-4 are visible is the criterion considered 6/13
for this analysis. This region happened to be the best fit for blunder analysis on a representative GSO and GEO satellite. Figure gives 5 position errors obtained independently with blunders on GEO-1 and GSO-4 respectively. From Figure 5, it is obvious that the position error with a blunder on GEO-1 is smaller compared to a blunder on GSO-4. The reason behind this is in the simulator module, errors are introduced on channels (satellites) in accordance with Eq. (1). In this region, the elevation of GEO-1 is better than that of GSO-4. This implies that a relatively higher magnitude of error is introduced on GSO-4 as compared to GEO-1. The geometry (DOP) being the same, the relative error magnitude dictates the overall position error (Eq. 3), which is higher in the case of GSO-4.
The observation model with measurement errors is (Petovello 2010)
z = Hx + v .
With apriori information (i.e. either height or clock or both), the augmented measurement vector is given as
Z
+
=
Z H x = I x + v 0
(14)
The corresponding least squares solution is obtained as
xˆ = ( H T R −1 H 4.4
(13)
+
P−1 ) −1 H T R−1 z
(15)
Constraint Mode
In any GNSS, satellites are distributed around the azimuth plane w.r.t. the user’s antenna. However, in the vertical dimension, satellites are available only above the antenna plane. Added to this is the strong correlation that exists between vertical component and clock, making it difficult to resolve their individual contributions (Ryan & Lachapelle (2000), Langley (1999)). The result is the VDOP being relatively poor compared to HDOP, which in turn affects overall position accuracy (Langley 1999). If either the height or clock estimate is available, then the correlation between VDOP and TDOP states is reduced. This results in an improved position solution. The benefits in case of IRNSS having either or both estimates towards the final solution are described herein.
where P is the covariance matrix of the states. Figure 7, Figure 8, Figure 9, gives the position errors obtained with height and clock constraints separately and as a combination thereof, respectively. It is evident that an appreciable improvement in position accuracy is achieved in comparison with the constraint-free approach (Figure 4). With apriori information, strong correlation between clock and height states are decoupled resulting in improved position accuracy. Given that IRNSS satellites are in GEO and GSO orbits and at a high elevation, the VDOP will be relatively inferior compared with other existing GNSS. As expected, the position accuracy of the combined constraint approach is better than height, which is better than the clock constraint mode.
Figure 7: IRNSS DOP and Position accuracy with Height Constraint ION ITM 2011, Session B5, San Diego, CA, 24-26 January 2011
7/13
Figure 8: IRNSS DOP and Position accuracy with clock constraints
Figure 9: IRNSS DOP and position accuracy with height and clock constraints 4.5
Elevation Angle
In this section, analysis focuses on availability and accuracy achievable with reduced coverage due to higher mask angles. This analysis gives a good insight of the proposed system for applications such as indoors or urban canyon environments. The mask angle is changed to 25º for this particular analysis. Figure 10 and Figure 11 give the plots of availability and position accuracy ION ITM 2011, Session B5, San Diego, CA, 24-26 January 2011
respectively. An immediate inference is that there is a drop in number of satellites over the Indian subcontinent in comparison to Figure 3. Additionally, the satellite count is more biased. This is attributed to the four satellites located beyond 80º longitude. As such the area under this region will have visibility to all four satellites and few of those on the other side of 80º longitude (30º to 80º longitude, 2 GSO 8/13
and 1 GEO). Figure 10, when analyzed in conjunction with Figure 1, results in another interesting finding. IRNSS has a dedicated service for specific users in Restricted service, which will have vehicles with attitude maneuvers. The distribution of three satellites on either side (for vehicular applications assuming three on each side (Figure 10)) may necessitate additional antenna (Vyasaraj & Vijay 2007) or augmentation with other systems in order to obtain positioning with IRNSS. Figure 11 gives the position error during reduced visibility. With an increase in DOP (inferior geometry due to availability), a proportional increase in error is observed.
Figure 10: IRNSS satellite visibility with a 25º mask angle
Figure 11: IRNSS DOP and Position accuracy with a 25˚ mask angle 5
ANALYSIS OF IRNSS+GPS+GLONASS
As explained in sections 4.1 and 4.5, combining IRNSS with existing GNSS systems has some potential benefits. This section highlights the improvement in availability, accuracy and reliability due to a combined IRNSS+GPS+GLONASS constellation. The last part of this section presents the relative benefits during reduced coverage. The methodology and explanations as in Section 4 applies here as well. Only the major differences are presented.
5.1
subcontinent has drastically improved by combining IRNSS with GPS and GLONASS, which is as depicted in Figure 12.
Availability
Several works in the past have demonstrated the benefits of combining GPS and GLONASS (e.g., Ryan & Lachapelle 2000). Availability over the Indian ION ITM 2011, Session B5, San Diego, CA, 24-26 January 2011
Figure 12: IRNSS+GPS+GLONASS satellite visibility 9/13
5.2
Accuracy
5.3
IRNSS satellites which are in GEO stationary and GSO orbits will benefit from combining with GPS and GLONASS satellites from a low elevation point of view. The drawback highlighted in Section 4.5 can be circumvented with this augmentation. Figure 13 shows the improvement achievable in position accuracy. The relative geometry has improved with increased observations (satellites), and thus the overall position accuracy.
Reliability
The number of satellites and thus observations has increased with augmentation. Due to this, the relative geometry has improved as compared to the standalone case. Bracketed terms of Eq.(10), gives the measure of DOP. The improvement in DOP has resulted in a relatively lower overall external reliability as depicted in Figure 14. The HPE and VPE show one order improvement in the combined scenario.
Figure 13: IRNSS+GPS+GLONASS DOP and Position accuracy
Figure 14: HPE, VPE for GEO-1 and GSO-4 with combined IRNSS+GPS+GLONASS
ION ITM 2011, Session B5, San Diego, CA, 24-26 January 2011
10/13
Figure 15: IRNSS+GPS+GLONASS satellites with 25º mask angle
Figure16: IRNSS+GPS+GLONASS DOP and Position accuracy with 25º mask angle
5.4
Elevation Mask Angle
The number of satellites is reduced in standalone IRNSS positioning with a mask angle of 25º (Figure 10). Combining with GPS and GLONASS, a favorable scenario is observed. The satellite availability increases as shown in Figure 15. The corresponding DOP and position accuracy are given in Figure16.
ION ITM 2011, Session B5, San Diego, CA, 24-26 January 2011
6
CARRIER PHASE POSITIONING
Carrier phase positioning has been one of the major successes of the GPS program though it was not visualized at the system design phase. Several techniques have evolved over the years in the receiver architecture enabling sub-cm accuracy (e.g. Liu et al 2003). GLONASS, being an FDMA system has few limitations when it comes to carrier phase positioning (Walsh et al 1996). One of the major tasks involved in the carrier phase positioning is ambiguity resolution. In case of GPS, 11/13
one of the methods of achieving reliable ambiguity fixing is by means of linear combination of observations. It also makes ambiguity resolution easier than stand alone L1 or L2 fixing. Linear combinations are formed in several ways with relative merits and demerits. A detailed study and analysis of this is available in the literature. Table 4 gives the wavelengths corresponding to different modes of operation in GPS, with a widelane phase observable wavelength of 86.2 cm. From a carrier phase ambiguity resolution stand point, a higher dynamic range is achieved, which aids in reliable ambiguity fixing compared to standalone L1 or L2.
Table 4: Carrier phase wavelengths in different modes of operation - GPS Mode of Operation L1 L2 Wide-lane (L1-L2)
19.04 24.06 86.25
Table 5: Carrier phase wavelengths w avelengths in different modes of operation - IRNSS L5 S1 Wide-lane (S1-L5)
7
CONCLUSIONS AND FUTURE WORK
In this paper, a standalone IRNSS analysis w.r.t availability, accuracy, reliability and reduced coverage was presented. The main result result from the analysis analysis is that the seven satellites are always assured over India (for 5º mask angle). The advantages of augmenting IRNSS receivers with apriori information be it either clock or height information were presented.
Wavelength (cm)
Table 5 gives the wavelengths corresponding to IRNSS. The widelane wavelength of IRNSS is approximately equal to that of standalone L5. The advantage cited with GPS is no longer applicable in case of IRNSS. Also, the wavelength of S1 is approximately 12 cm, which is almost half of L5. With the increase in frequency, though the ionosphere errors are reduced (Kaplan & Hegarty 2006), the margins for carrier phase errors are relatively small on S1. This might pose an issue on the stand alone S1 carrier phase positioning.
Mode of Operation
In addition, for GEO satellites and static user, Doppler will be mainly due to receiver clock (Mishra & Enge 2006). Thus the observability of ambiguities is reduced for these GEO satellites. With only 4 GSO satellites, observations during the ambiguity resolution process are inadequate (Lachapelle 2010).
Wavelength (cm) 25.52 12.03 22.80
Table 1 gives the proposed IRNSS frequencies and services offered. The L5 band is located within the Distance Measuring equipment (DME) frequency of operation. DME frequency band spans from 960 MHz to 1215 MHz (Rohde & Schwarz 2009). The working principle of DME is by exchange of pulses between DME station and aircrafts. The power level of these pulses can be as high as 1 kW with a coverage range of 400 km. The ON period of these pulses is of the order of 3.5 µs. With multiple DME’s, the effective number of pulses will be higher with an overall reduction in received signal energy. This may have an adverse impact on phase tracking unless some advanced pulse mitigation techniques are adopted in the receiver (Hearty et al 2000). Thus, standalone carrier phase positioning with L5 and widelaning with S1 may pose some challenges. ION ITM 2011, Session B5, San Diego, CA, 24-26 January 2011
The paper also discussed the advantages of combining IRNSS receivers with the existing GPS or GLONASS system. As expected, significant improvements were observed in the case of availability, accuracy and reliability. A discussion on IRNSS signal w.r.t carrier phase positioning was also presented. An additional frequency close to either of the proposed frequencies in addition to an additional satellite in GSO orbit would enhance IRNSS carrier phase positioning. Adding Galileo and GAGAN is left for future work.
Acknowledgements The first author would like to thank Accord Software & Systems Pvt. Ltd for the partial financial assistance towards his Doctoral studies. The assistance of Sashidharan MAA, Systems Engineer, Accord Software Pvt Ltd. is highly appreciated towards the analysis and software development. Shashank Satyanarayana and Pratiba B Anatharamu, PhD candidates in the PLAN Group of the University of Calgary are appreciated for the useful discussions during the paper writing.
Disclaimer The analysis and results are sole findings of work carried out by primary author. The work is purely based on open source literature, is subjective to changes w.r.t ICD and the constellation itself. The results are in no way final and binding. However, with the final ICD and the almanac of IRNSS, it would be a minimal change to regenerate the above results.
12/13
REFERENCES Ryan, S. and G. Lachapelle (2000) “Impact of GPS/GALILEO Integration on Marine Navigation” , IAIN World Congress – ION Annual Meeting 2000. 721-731 Lachapelle, G., M.E.Cannon, K.O’Keefe, and P. Alves (2001) Technical Benefit Analysis of Galileo for Canada , Contract Report Prepared for Canadian Space Agency, Ottawa, (http://plan.geomatics.ucalgary.ca/project_info. php?pid=24) Ryan, S. and G. Lachapelle (1999) “Augmentation of DGNSS With Dynamic Constraints For Marine Navigation” Proceedings of ION GPS99 Conference, Nashville, September 14-17, 1999. O'Keefe, K. (2001) “Availability and Reliability Advantages of GPS/Galileo Integration”. (Session C4), Salt Lake City, 11-14 September), The Institute of Navigation, Alexandria, VA.,
Mishra, P. and P. Enge (2006), Global Positioning System: Signals, Measurements, and Performance , Second Edition, Ganga – Jamuna Press Leick, A. (1995), GPS Satellite Surveying Second Edition, John Wiley & Sons, Inc. Walsh, David, Daly and Peter (1996) “GPS and GLONASS Carrier Phase Ambiguity Resolution” Proceedings of the 9th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GPS 1996), Kansas City, MO, September 1996, pp. 899907. Fenton, P., Falkenberg, B., Ford, T., Ng, K., and Van Dierendonck, A. J., “NovAtel’s GPS Receiver-The High Performance OEM Sensor of the Future ”, Proceedings of ION GPS-91, Fourth International Technical Meeting of the Satellite Division of The Institute of Navigation, Albuquerque, NM, 11-13 September 1991, pp. 49-58.
O'Keefe, K., S. Ryan and G. Lachapelle (2002) “Global Availability and Reliability Assessment of the GPS and Galileo Global Navigation Satellite Systems”. Canadian Aeronautics and Space Journal, Canadian Aeronautics and Space Institute, 48, 2, 123-132
Bhaskaranarayana, A . (July 15th 2008) Indian IRNSS & GAGAN , Presentation to COSPAR Meeting, Montreal.
Langley, R.B. (1999) “Dilution of Precision” GPS World, vol. May, pp. 52-59
Abidin. H, Z. (1994), On the fly Ambiguity Resolution, GPS World
Lachapelle, G. (2010) Advanced GNSS Theory and Applications, ENGO625Course Notes Department of Geomatics Engineering, University of Calgary, Canada
Direndonck, A.J.V, P Fenton, B Falkenberg, T Ford, and N G Keith (1991) “The High Performance OEM Sensor of the future” Proceedings of ION GPS-91, Fourth International Technical Meeting of the Satellite Division of The Institute of Navigation, Albuquerque, NM, 11-13 September 1991, pp. 49-58
Petovello, M. (2010) Estimation for Navigation, ENGO 699.36 Course Notes Department of Geomatics Engineering, University of Calgary, Canada Hegarty, C.J, A.J Van Dierendonck, D. Bobyn, M.Tran, and T.Kim, J, “Suppressing of Pulsed Interference through Blanking” Proceedings of the IAIN World Congress, San Diego, CA, June 2000 KAPLAN, E. D. and C.J Hegarty (2006), Understanding GPS: Principles and Applications, Second edition Artech House ARINC, Navstar GPS Space Segment / Navigation User Interfaces Interface Control Document (ICD), Revision 200C, prepared and published for GPS JPO, revised September 1997, ION, Salt Lake City, USA Liu, J., M.E.Cannon, P.Alves, M.Petovello, G.Lachapelle, G. MacGougan, and L. deGroot, “A performance comparison of single and dual frequency GPS ambiguity resolution strategies” Springer-verlag, 28 th June2003. ION ITM 2011, Session B5, San Diego, CA, 24-26 January 2011
Kibe, S, V and Gowrishankar. D, APRSAF -15: Space for Sustainable Development , December 10th 2008, Vietnam
Direndonck, A.J.V, P Fenton and T Ford (1992) “Theory and performance of narrow correlator spacing in a GPS receiver” Journal of Institute of Navigation, Volume 39, No 3, U.S.A Rohde & Schwarz (2009) DME Transponder Testing, Application Hand Out G. X. Gao (2007), “DME/TACAN Interference and its Mitigation in L5/E5 Bands” in ION Institute of Navigation Global Navigation Satellite Systems Conference, 2007 Vyasaraj, G. and Vijay, S.B. (2007), “An Innovative Approach To Overcome GPS Signal Masking During Maneuvers in Aircraft or Satellite”, Proceedings of the 20th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS 2007), Fort Worth, TX, September 2007, pp. 2999-3007 13/13
