Memory and Hosting Constraints.
In developing the technology, we were required to accommodate a range of memory and hosting constraints. Using the u-Nav chipset, we have implemented an MS-Assisted solution in which the firmware is serially booted into the on-chip static random access memory (SRAM) from the cell baseband chip. This required all of the AGPS firmware to run in 64 Kbytes of program memory and 60 Kbytes of data memory. We also have implemented MS-Based and assisted conventional solutions utilizing external flash memory for the program but internal data memory only.
Utilizing the flexible correlator hardware mentioned earlier, a firmware solution running on a host processor performs all of the subATTO processing to produce code-phase and Doppler measurements or location.
The firmware supports MS-Assisted, MS-Based, and multi-mode operation. It runs on an ARM 9 microprocessor but can easily be ported to others.
It has been uniquely structured to facilitate sharing of the processing resources with foreign, high priority applications as might be found on a cell baseband processor. The firmware adapts gracefully to the loss of processing capacity and loss of signal samples. It is independent of the platform's real-time operating system (RTOS) and can be collapsed into a single task.
Conclusions
GPS operation can be enhanced in its performance through the provision of assistance data wirelessly.
This assistance can be supplied via the user plane or control plane. It can consist of satellite data gathered from remote GPS receivers with direct line of sight to the satellites, coarse receiver position, precise or coarse time, or more specific data derived from these primary elements.
The wireless network may also perform the location solution, although the benefit of this is debatable. The use of assistance can result in much faster acquisition of weaker signals, and can facilitate navigation solutions that would not otherwise be possible.
The benefit of the time assistance is dependent in a complicated way on its uncertainty. Both the search strategy and the navigation solution have to be adapted to the level of this uncertainty.
CDMA inherently facilitates precise time assistance. GSM networks, however, have to be augmented to provide it and generally will not be.
On the other hand, CDMA cell sizes can be much larger than GSM cell sizes, resulting, in principle, in slower signal acquisition and/or a need for more GPS processing capacity. However, this is not a significant factor in practice.
GSM uses short time slots for transmission and this facilitates interference mitigation via signal blanking. A CDMA handset may employ "antenna switching," but with potentially serious impact on both GPS and voice communication. RF design for concurrent operation is far preferable.
There are some differences between the standards developed for CDMA and GSM but, in general, the assistance supplied and the performance requirements are quite similar. However, the sensitivity required by the standards is well under that required to provide reliable performance under a range of indoor and urban canyon environments.
Hardware-mechanized search engines are useful but typically are not flexible enough to permit the potential of the hardware to be fully realized. subATTO signal processing allows similar performance to be achieved with far less dedicated hardware support, but requires much more general purpose processing capacity. A hybrid of multiple smaller search engines and correlators provide an ideal compromise where such processing capacity is not available.
AGPS solutions require a range of different hardware and software solutions. Our team has experience with hardware employing both correlator fingers and search engines, and with flexible correlator hardware designed to work with a host processor. Our firmware solutions have ranged from MS-Assisted firmware running in 64 KBytes of internal SRAM to taskless, RTOS–independent multi-mode firmware running on a host processor that also supports high priority foreign tasks.
Source
This assistance can be supplied via the user plane or control plane. It can consist of satellite data gathered from remote GPS receivers with direct line of sight to the satellites, coarse receiver position, precise or coarse time, or more specific data derived from these primary elements.
The wireless network may also perform the location solution, although the benefit of this is debatable. The use of assistance can result in much faster acquisition of weaker signals, and can facilitate navigation solutions that would not otherwise be possible.
The benefit of the time assistance is dependent in a complicated way on its uncertainty. Both the search strategy and the navigation solution have to be adapted to the level of this uncertainty.
CDMA inherently facilitates precise time assistance. GSM networks, however, have to be augmented to provide it and generally will not be.
On the other hand, CDMA cell sizes can be much larger than GSM cell sizes, resulting, in principle, in slower signal acquisition and/or a need for more GPS processing capacity. However, this is not a significant factor in practice.
GSM uses short time slots for transmission and this facilitates interference mitigation via signal blanking. A CDMA handset may employ "antenna switching," but with potentially serious impact on both GPS and voice communication. RF design for concurrent operation is far preferable.
There are some differences between the standards developed for CDMA and GSM but, in general, the assistance supplied and the performance requirements are quite similar. However, the sensitivity required by the standards is well under that required to provide reliable performance under a range of indoor and urban canyon environments.
Hardware-mechanized search engines are useful but typically are not flexible enough to permit the potential of the hardware to be fully realized. subATTO signal processing allows similar performance to be achieved with far less dedicated hardware support, but requires much more general purpose processing capacity. A hybrid of multiple smaller search engines and correlators provide an ideal compromise where such processing capacity is not available.
AGPS solutions require a range of different hardware and software solutions. Our team has experience with hardware employing both correlator fingers and search engines, and with flexible correlator hardware designed to work with a host processor. Our firmware solutions have ranged from MS-Assisted firmware running in 64 KBytes of internal SRAM to taskless, RTOS–independent multi-mode firmware running on a host processor that also supports high priority foreign tasks.
Acknowledgments
The entire engineering team at SigNav contributed to this work through their innovation and tireless dedication. In particular, I wish to thank Eamonn Glennon, who has made major algorithmic contributions. Our colleagues at u-Nav and its partners and customers contributed to our growing collection of insights, as have those of our other partner who cannot yet be named. This article is based on the paper "Lessons Learnt in Assisted GPS" presented at GNSS 2004, the 2004 International Symposium on GNSS/GPS, held in Sydney, Australia, December 6-8, 2004.
Source
0 comments:
Posting Komentar