Reaserch projects
Our laboratory conducts research on modulation and demodulation techniques and network technologies based on statistical signal processing to realize next-generation wireless communications. Recently, we are also focusing on the application of machine learning on mobile communications.
Wireless network control using machine learning
Densely deployed small cell multiple-input multiple-output (MIMO) systems can potentially improve the system capacity. However, their overlapping and neighboring cells lead to an increase in inter-cell interference (ICI) and thus decrease the system capacity.
To suppress such ICI, base stations (BSs) can perform the exhaustive search (ES) in order to find the optimal combination of transmit power levels and beamforming vectors from a pre-defined codebook. However, ES requires a large amount of computational complexity that grows exponentially with the number of BSs.
To reduce the complexity, we proposed a unsuper-vised learning and deep Q-learning scheme for neural networks (NNs) that can approximate ES with much less complexity.
To suppress such ICI, base stations (BSs) can perform the exhaustive search (ES) in order to find the optimal combination of transmit power levels and beamforming vectors from a pre-defined codebook. However, ES requires a large amount of computational complexity that grows exponentially with the number of BSs.
To reduce the complexity, we proposed a unsuper-vised learning and deep Q-learning scheme for neural networks (NNs) that can approximate ES with much less complexity.
Polar code in OFDM system
We research on the code construction of polar codes on OFDM transmission over frequency selective fading channels. Since the code construction of polar codes is channel-dependent, it has been great difficulties on optimal construction of polar codes over different kinds of channels. To achieve reliable communications, our research mainly proposes an estimation-feedback system in order to use channel state information to modify frozen bits positions adapting to the specific channel, which was demonstrated by simulations that the block error rate can be reduced effectively.
Others
- Constant-amplitude OFDM modulation schemes
- Parameter estimation and beam control for hybrid beamforming MU-MIMO Systems
- Improvement of pilot contamination problem in Massive MIMO systems
- Time-to-digital converter (TDC) with real zero and its application in OFDM systems
- Underwater acoustic MIMO-OFDM communication system
In addition, we have many joint research achievements with companies (NTT Docomo, Rakuten Mobile, JAXA, etc.)