Analysis and Testing of Optimal Navigation Message Rate for IMES

S. Pullen, D. Manandhar, H. Torimoto

Abstract:	The IMES (Indoor MEssaging System) augments GNSS by providing 3-D location information to indoor users, including floor numbers. Besides position data, it can also provide unique or user-defined identification information to users. The IMES signal is based on the GPS L1 C/A-code signal structure so that GPS receivers can also acquire the IMES signal. The PRN code length and navigation message rate are the same as GPS L1 C/A. The differences are only the PRN of IMES transmitters and a 8.2 kHz offset in the L1 C/A carrier center frequency. GPS L1 C/A and IMES both use a 50 bps navigation message rate, which provides very stable acquisition and tracking in the case of noisy and weak signal environments. A 50 bps message rate provides a 20-ms-long data period. This makes it possible to perform coherent integration over 10 ms during acquisition while avoiding message bit flips. Coherent integration over 10 ms ideally provides 10 dB of gain in SNR. Thus, a lower data rate (and longer data period) on the order of 50 bps has the merit of improving processing of weak signals during acquisition. This message rate has a strong advantage in the case of GPS, since GPS was designed for use outdoors, and repeated decoding of navigation message bits is not necessary once the necessary time and ephemeris data have been obtained. Once the ephemeris data are known, the receiver only needs to compute pseudorange based on L1 C/A code phase and the navigation message frame header. This is why a GPS receiver can output position data at 1 Hz (every second) or even at higher rates like 100 Hz (every 10 ms) even if the navigation message frame is six seconds long. Theoretically, position outputs at 1000 Hz are possible but would be too noisy. The time-to-first-fix (TTFF) for GPS is about 36 seconds under ideal conditions. Unlike GPS, IMES does not use pseudorange to compute position. IMES transmits position data (Latitude, Longitude, Height, Floor ID, etc.) embedded in the navigation message itself. The navigation message structure of IMES is the same as GPS L1 C/A signal. In order for the IMES receiver to obtain position data, the receiver has to decode all navigation message bits. Thus, the time required for position data, or TTRM (“True Time to Read Message”) is proportional to the navigation message length. The TTRM is 1.8 seconds to provide 2-D position data with Floor ID from IMES Message Type 0 (3 words, 90 bits) and 2.4 seconds to provide 3-D position data with floor ID from IMES Message Type 1 (4 words, 120 bits). There is no way to reduce TTRM in IMES without increasing the navigation message data rate. However, increasing the navigation message rate has other impacts on signal processing. In particular, it limits the period over which coherent integration can take place. For example, a navigation message rate of 250 bps allows coherent integration over only 2 ms without a navigation message bit flip. For these reasons, the GPS message rate of 50 bps is considered slow for IMES. IMES solutions are targeted on mobile phone users who may be walking briskly or running. In such cases, users may have entered another room or moved to a different area before the receiver finishes reading the transmitted position data . Theoretically, higher message rates provide faster position output. But this is not always true depending upon the signal strength of the acquired signal by the receiver. With faster message rates and shorter data periods, the probability of bit errors increases. Thus, although a higher data rate is better for TTRM, the resulting solutions are more likely to be invalid due to a higher bit error rate. A tradeoff is thus present between message rate and signal level (SNR). In order to study the impact of higher navigation message rates, we have conducted experiments by transmitting IMES signals at message rates of 50, 250 and 500 bps. These signals are logged by a static IMES receiver that is moved in between trials to vary the distance from the transmitter, and the resulting TTRM are recorded. As the distance between the transmitter and receiver increases, the signal becomes weaker, leading to a higher bit error rate. When the signal is weaker than a certain level, the TTRM values for 50 bps, 250 bps, and 500 bps may be significantly different. We will also conduct experiments for a receiver in dynamic mode (i.e., at normal walking speed). Dynamic mode is quite different from static mode due to multipath and fading in indoor environment. Although the time to read the navigation message is shorter for higher data rates, it takes longer to get error-free navigation message frames from the tracking output due to the resulting weaker signal levels. These experiments will be conducted at different transmitter power levels and at different radial distances from transmitter to receiver. The results of these experiments will be used to assess the tradeoff between speed and accuracy and to recommend the best message data rates for particular conditions. The use of higher message rate can be application specific and specific to a given IMES deployment location. These results will also be used as a guide to designing new message types that reduce TTRM without suffering significant bit-error penalties.
Published in:	Proceedings of the 26th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS+ 2013) September 16 - 20, 2013 Nashville Convention Center, Nashville, Tennessee Nashville, TN
Pages:	408 - 415
Cite this article:	Pullen, S., Manandhar, D., Torimoto, H., "Analysis and Testing of Optimal Navigation Message Rate for IMES," Proceedings of the 26th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS+ 2013), Nashville, TN, September 2013, pp. 408-415.
Full Paper:	ION Members/Non-Members: 1 Download Credit Sign In