FeelMaTyC (Feedback-less Machine-Type Communication)

• Applicants: Professor Dr.-Ing. Markus Fidler; Professor Dr. Jürgen Peissig
• Subject Area: Electronic Semiconductors, Components and Circuits, Integrated Systems, Sensor Technology, Theoretical Electrical Engineering
• Term: from 2017 to 2022
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 329885056

For many machine-type communication (MTC) based applications, e.g. transmission of pictures, metering or sensor information, bi-directional communication is not necessary, as it is merely used to coordinate and assure the transmission and to perform medium access control. By application of unidirectional (feedback-less) communication from a machine-type-device to a base station, the costs for an increased resource requirement as well as the energy consumption and the integration of a receiver into the MTC-node can be saved. Thus, pure MTC-transmitter networks can be produced on low price and operated on good terms. This is a prerequisite to allow for mass applications of MTC-concepts for the wireless Internet-of-Things.On the contrary, these savings are obtained by the fact, that an MTC node cannot guarantee the target quality-of-service for its data transmission. Furthermore, the network control has no influence on the medium access methods of the individual node. To ensure the application required quality-of-service with a certain reliability, it is important to know the performance limits with respect to the MTC node density as well as their quality-of-service requirements in such an uncoordinated and feedback-less network. Additionally, it is important to optimize these limits with suitably parameterized designs.Thus, the target of this proposal is to determine the performance limits in a feedback-less network based on the joint investigation of the physical layer and medium access control layer. In addition, it targets to design suitable frame structures and mechanisms for service prioritization as well as graceful degradation in overload scenarios.Thereby, the investigation of packet collision effects on the physical layer and their resolution by mechanisms like frame design, channel coding and interleaving, usage of well defined information superposition, multiple transmission with different modes, distributed allocation of the transmit energy in time and frequency domain on transmitter side, as well as single- and multistage successive interference cancelation at receiver side will be investigated. Therefore, on the physical layer the favorable properties of a filterbank-multicarrier (FBMC) modulation scheme with respect to the excellent temporal and spectral transmission energy containment will be used in order to optimally detect and separate unsynchronized packets in the multi-access-channel.The results obtained on the physical layer will be integrated into a network simulator to obtain an assessment of the overall network performance. With these network layer simulations the different quality of service requirements, protocols, node configurations, spatial node distribution as well as the overload behavior will be assessed.Finally a software defined radio testbed will be used to validate the simulation results in a measurement based environment.

• DFG Programme: Research Grants