Access Type
Open Access Dissertation
Date of Award
January 2020
Degree Type
Dissertation
Degree Name
Ph.D.
Department
Electrical and Computer Engineering
First Advisor
John Liu
Abstract
One benefit of data communication over the voice channel of the cellular network is to reliably transmit real-time high priority data in case of life critical situations. An important implementation of this use-case is the pan-European eCall automotive standard, which has already been deployed since 2018. This is the first international standard for mobile emergency call that was adopted by multiple regions in Europe and the world. Other countries in the world are currently working on deploying a similar emergency communication system, such as in Russia and China. Moreover, many experiments and road tests are conducted yearly to validate and improve the requirements of the system. The results have proven that the requirements are unachievable thus far, with a success rate of emergency data delivery of only 70%.
The eCall in-band modem transmits emergency information from the in-vehicle system (IVS) over the voice channel of the circuit switch real time communication system to the public safety answering point (PSAP) in case of a collision. The voice channel is characterized by the non-linear vocoder which is designed to compress speech waveforms. In addition, multipath fading, caused by the surrounding buildings and hills, results in severe signal distortion and causes delays in the transmission of the emergency information. Therefore, to reliably transmit data over the voice channels, the in-band modem modulates the data into speech-like (SL) waveforms, and employs a powerful forward error correcting (FEC) code to secure the real-time transmission.
In this dissertation, the Turbo coded performance of the eCall in-band modem is first evaluated through the adaptive white Gaussian noise (AWGN) channel and the adaptive multi-rate (AMR) voice channel. The modulation used is biorthogonal pulse position modulation (BPPM). Simulations are conducted for both the fast and robust eCall modem. The results show that the distortion added by the vocoder is significantly large and degrades the system performance. In addition, the robust modem performs better than the fast modem. For instance, to achieve a bit error rate (BER) of 10^{-6} using the AMR compression rate of 7.4 kbps, the signal-to-noise ratio (SNR) required is 5.5 dB for the robust modem while a SNR of 7.5 dB is required for the fast modem.
On the other hand, the fading effect is studied in the eCall channel. It was shown that the fading distribution does not follow a Rayleigh distribution. The performance of the in-band modem is evaluated through the AWGN, AMR and fading channel. The results are compared with a Rayleigh fading channel. The analysis shows that strong fading still exists in the voice channel after power control. The results explain the large delays and failure of the emergency data transmission to the PSAP. Thus, the eCall standard needs to re-evaluate their requirements in order to consider the impact of fading on the transmission of the modulated signals. The results can be directly applied to design real-time emergency communication systems, including modulation and coding.
Recommended Citation
Zakhem, Samer, "Error Correction For Automotive Telematics Systems" (2020). Wayne State University Dissertations. 2516.
https://digitalcommons.wayne.edu/oa_dissertations/2516