Abstract
V2V channel models with moving scatterers produce a Doppler effect up to 12 times that of channel modeling that fits the movement of a single vehicle. As a result, the estimation of the channel always needs to be updated quickly, and the duration of the data symbol on frame 802.11p will be limited. Furthermore, there is no data symbol that can be carried in the Coherence Time (CT) range, which causes estimation errors. To overcome this, a method is needed to detect the value of CT from the channel. We improved the CT detection method on the V2V channel from our previous study with a correlation coefficient > 0.5 so that according to the reference theory. Furthermore, the CT detection results are used as the limit of long training symbol (LTS) duration and the symbol data in one frame to be sent. The duration of data symbols and LTS were set to always be within the CT range, and after the CT duration ended, the next LTS was sent to detect the next channel response and the new CT. The CT detection results that we performed resulted in an average difference in CT detection results with the CT theory of 0.0023 ms. The performance of the proposed method was up to 63% that of the spectral temporal average (STA) method and the known coherence time (KCT) method for the scenario where the data symbol length was still in the CT period. This makes the proposed approach very practical for use according to standardization 802.11p in the vehicular ad hoc network (VANET) communications system.
Original language | English |
---|---|
Pages (from-to) | 332-343 |
Number of pages | 12 |
Journal | International Journal of Wireless Information Networks |
Volume | 28 |
Issue number | 3 |
DOIs | |
Publication status | Published - Sept 2021 |
Keywords
- Autocorrelation
- Coherence time
- Doppler effect
- Moving scatterers
- V2V channel
- VANET