In this paper, we propose a channel estimation method based on a complex valued regression of the neural network for the IEEE 802.11p standard. It consists of the complex weighted summation optimized by feedforward neural network with backpropagation algorithm using initial estimated channel of the pilot and the long preamble. It also exploits the shift matrix in order to mitigate the effect from a systemic problems in IEEE 802.11p standards. The major problems of IEEE 802.11p standard are wide bandwidth of 10 MHz consisting of 64 subcarriers and relatively insufficient four pilot subcarriers at single ODFM symbol, which are unsuitable for a channel of vehicular environment. Despite these problems, the proposed method performs better than the conventional channel estimation methods. The performance of proposed scheme is provided with the comparison between constructed data pilots (CDP), Spectral Temporal Averaging (STA), and proposed scheme. The proposed channel estimation scheme has low mean square error (MSE) and bit error rate (BER) throughout the whole SNR region. It is the result from properly trained weight. At the low SNR region, especially, the performance of proposed scheme is much better than CDP and STA scheme. It is because of the noise suppression effect caused by a weighted summation algorithm.
|Title of host publication||2019 IEEE 90th Vehicular Technology Conference, VTC 2019 Fall - Proceedings|
|Publisher||Institute of Electrical and Electronics Engineers Inc.|
|Publication status||Published - 2019 Sep|
|Event||90th IEEE Vehicular Technology Conference, VTC 2019 Fall - Honolulu, United States|
Duration: 2019 Sep 22 → 2019 Sep 25
|Name||IEEE Vehicular Technology Conference|
|Conference||90th IEEE Vehicular Technology Conference, VTC 2019 Fall|
|Period||19/9/22 → 19/9/25|
Bibliographical noteFunding Information:
This research was supported by a grant (19CTAP-C151968-01) from Technology Advancement Research Program funded by Ministry of Land, Infrastructure and Transport of Korean government
© 2019 IEEE.
All Science Journal Classification (ASJC) codes
- Computer Science Applications
- Electrical and Electronic Engineering
- Applied Mathematics