| Presentation | 2020-06-24 Investigation of Adaptive Transmission Rate Control Based on Machine Learning Using Only ACK/NAK Feedback Takumi Endo, Kenichi Higuchi, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | This paper investigates an adaptive transmission rate control based on machine learning using only ACK/NAK feedback information. In the conventional approaches, the transmission rate is determined at the transmitter based on the feedback of channel quality information (CQI) from the receiver. However, as the number of terminal increases in the future radio access, the overhead of radio resource required for the CQI feedback may impact on the overall system performance. To address this issue, we consider an adaptive transmission rate control relaying only on the one-bit ACK/NAK feedback which is essentially used for hybrid automatic repeat request (HARQ) to indicating whether the received packet is correctly decoded or not. It should be noted that in the real system, the online adaptation of the transmission rate control for accommodating the changes of the channel conditions in time such as signal-to-noise ratio (SNR) and channel fading rate due to the mobility of the user terminal. Considering these issues, the proposed method is based on the long short-term memory (LSTM) which is categorized to the recurrent neural network (RNN) with additional use of probabilistic random transmission rate selection or transmission rate correction when successive ACK receptions are observed for reducing the negative effect of uncertainty in online learning based on ACK/NAK feedback. The computer simulation results show that the proposed method using only one-bit ACK/NAK feedback can achieve quite comparable throughput performance as in the conventional CQI-based method. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Transmission rate control / machine learning / LSTM / ARQ / feedback information / online control |
| Paper # | RCS2020-25 |
| Date of Issue | 2020-06-17 (RCS) |
| Conference Information | |
| Committee | RCS |
|---|---|
| Conference Date | 2020/6/24(3days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | Online |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | First Presentation in IEICE Technical Committee, Resource Control, Scheduling, Wireless Communications, etc. |
| Chair | Eiji Okamoto(Nagoya Inst. of Tech.) |
| Vice Chair | Fumiaki Maehara(Waseda Univ.) / Toshihiko Nishimura(Hokkaido Univ.) / Tomoya Tandai(Toshiba) |
| Secretary | Fumiaki Maehara(Kyushu Univ.) / Toshihiko Nishimura(NEC) / Tomoya Tandai |
| Assistant | Koichi Adachi(Univ. of Electro-Comm.) / Osamu Nakamura(Sharp) / Manabu Sakai(Mitsubishi Electric) / Masashi Iwabuchi(NTT) / Tatsuki Okuyama(NTT DOCOMO) |
| Paper Information | |
| Registration To | Technical Committee on Radio Communication Systems |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | Investigation of Adaptive Transmission Rate Control Based on Machine Learning Using Only ACK/NAK Feedback |
| Sub Title (in English) | |
| Keyword(1) | Transmission rate control |
| Keyword(2) | machine learning |
| Keyword(3) | LSTM |
| Keyword(4) | ARQ |
| Keyword(5) | feedback information |
| Keyword(6) | online control |
| 1st Author's Name | Takumi Endo |
| 1st Author's Affiliation | Tokyo University of Science(Tokyo Univ. of Science) |
| 2nd Author's Name | Kenichi Higuchi |
| 2nd Author's Affiliation | Tokyo University of Science(Tokyo Univ. of Science) |
| Date | 2020-06-24 |
| Paper # | RCS2020-25 |
| Volume (vol) | vol.120 |
| Number (no) | RCS-74 |
| Page | pp.pp.13-18(RCS), |
| #Pages | 6 |
| Date of Issue | 2020-06-17 (RCS) |