Séminaire

With 50 billion Internet-of-Things (IoT) devices foreseen for the 2020s, an exponential growth in the amount of mobile data traffic is expected, while the available spectral resource is severely limited. To face this problem, future wireless access networks, in particular Fifth Generation (5G) systems are currently under intensive research worldwide. I am mainly investigating the design of radio resource allocation and interference management technologies enabling high spectral and energy efficiency, for a wide range of wireless systems such as Small Cell Networks, Cloud Radio Access Networks (CRAN) and IoT/Wireless Sensor Networks (WSN). In this talk, I will focus on our recent works on spectrally-efficient radio access schemes for RFID and Energy Harvesting (EH) WSNs, which are crucial to support the upcoming expansion of IoT mobile data traffic. Our proposed algorithms leverage on Compressed Sensing (CS) theory given the sparsity of the set of active sensors, but also on the emerging approach of Graph Signal Processing (GSP) given the spatial "smoothness" inherent to the sensed physical data. Simulation results show that our proposed schemes can achieve very low signal reconstruction errors and outperform conventional schemes.

Selected References :

[1] M. Kaneko, G. Cheung, W.T. Su and C.W. Lin, “Graph-based Joint Signal/Power Restoration for Energy Harvesting Wireless Sensor Networks”, IEEE Globecom, Dec. 2017

[2] M. Kaneko, “Throughput Analysis of CSMA With Imperfect Collision Detection in Full Duplex-Enabled WLAN”, in IEEE Wireless Communication Letters, vol.6, no.4, pp.490-493, Aug. 2017

[3] M. Kaneko, T. Nakano, K. Hayashi, T. Kamenosono and H. Sakai, “Distributed Resource Allocation with Local CSI Overhearing and Scheduling Prediction for OFDMA Heterogeneous Networks”, in IEEE Transactions on Vehicular Technology, vol.66, no.2, pp.1186-1199, Feb. 2017

[4] M. Kaneko, W. Hu, K. Hayashi and H. Sakai, "Compressed Sensing-Based Tag Identification Protocol for a Passive RFID System", in IEEE Communication Letters, vol.18, no. 11, pp. 2023-2026, Nov. 2014