DocumentCode
3734473
Title
Performance evaluation of 802.11ah Viterbi decoder for IoT applications
Author
Thi Hong Tran;Hiromasa Kato;Shinya Takamaeda-Yamazaki;Yasuhiko Nakashima
Author_Institution
Graduation School of Information Science, Nara Institute of Science and Technology, Takayama 8916-5, Ikoma, Nara, Japan
fYear
2015
Firstpage
320
Lastpage
325
Abstract
This research is our first step on the purpose of developing low-complex Viterbi decoder for IoT applications. We evaluate how the values of Viterbi decoder´s parameters such as trace back length (L), input data bit-width (D), and LLR truncated value (E), affects to BER and PER of a communication system. The IEEE 802.11ah simulator is used with AWGN channel and BPSK modulation. Our simulation results show that both BER and PER performance are improved if L value increased. However, if L is large enough, i.e., L ≥ 60, the performance improvement becomes insignificant. In addition, both BER and PER performance are continuously improved if D value increases from 1 to 4 bits. The best improvement of both BER performance (i.e., 1.8dB) and PER performance (i.e., 1.5dB) can be seen when D increases from 2 to 3 bits. Thus D = 3 is proposed for Viterbi decoder´s hardware development. When D ≥ 4 both BER and PER performance are almost unimproved although D continues to increase. Finally, our research shows that truncating the LLR value to E = 1.75 helps the Viterbi decoder achieves the best BER and PER performance in case D = 3 bits. Base on these research results, we come to the conclude of using 20 ≤ L ≤ 40, D = 3 bits, and E ≈ 1.75 for future development of low complex Viterbi decoder circuit for IoT applications. Furthermore, in this paper we introduce some mathematics equations for simplifying the Viterbi decoder process.
Keywords
"Decoding","Viterbi algorithm","Measurement","Bit error rate","Mathematical model","Transceivers","Standards"
Publisher
ieee
Conference_Titel
Advanced Technologies for Communications (ATC), 2015 International Conference on
ISSN
2162-1020
Print_ISBN
978-1-4673-8372-1
Type
conf
DOI
10.1109/ATC.2015.7388343
Filename
7388343
Link To Document