Opportunistic radio spectrum access: Collaborative spectrum sensing using custom RFICs and digital signal processing
Final Report Abstract
A capability to reliably identify available radio resources, referred to as spectrum sensing is one of the key challenges of cognitive wireless communication systems. The deployment of future opportunistic radio networks will depend on the reliability of sensing nodes detecting users in the shared frequency bands. The latter has been also the research topic of this activity. The main objective of this research project has been to design, develop and experimentally demonstrate a highly reconfigurable spectrum sensing platform suitable for future spectrum sensing nodes in a collaborative sensing network. Collaborative sensing network can significantly improve primary/secondary user detection by means of site diversity. The need for flexibility is underlined here as the sensing platform should operate in various environments, for example operating as a stationary indoor node, mobile or fixed outdoor node, etc. As a result, the capability to offer different sensing strategies needs to be assured. The latter flexibility is related not only to the capability to tune across a given range of frequencies but also the capability to perform detection of different types of signals, which might be even below the noise level. Design, implementation and deployment of a flexible sensing frontend based on customised radio-frequency integrated circuits and a flexible baseband digital signal processing has been successfully demonstrated in this project. The realised spectrum sensing platform offers automated spectrum scanning capability and wideband signal detection and decision scheme based on a novel sensing algorithm. The first achievement of this project is the successful implementation and demonstration of a new customised analog RF frontend. The RF frontend is built as a multi-chip hybrid system with customised SiGe integrated circuits (balanced LNA, up and down conversion mixers, variable IF filters and two integrated frequency synthesisers for up/down conversion controlled by FPGA). To achieve high image suppression and suppression of out of band interferers, offchip microstrip filters are used. DC regulators as well as FPGA interface are integrated on the same board. The second achievement of this project is successful demonstration of a novel edge detection algorithm where a special case of Hough transform has been used for the first time. The proposed signal detection approach has been implemented on a Xilinx Virtex-6 FPGA and tested in real-time. Beside successful implementation of the most critical subsystems of the sensing platform (analog frontend and baseband processor), this research activity has also opened up new horizons for future investigations: having a flexible sensing platform allows for hybrid sensing which can switch between different algorithms. An example could be turning the threshold based energy detection as frequency domain based energy detection and then switching to edge detection at predetermined time intervals. The developed flexible sensing platform is made for implementation of these and other kind of spectrum sensing experiments.
Publications
- Highly flexible cognitive radio spectrum sensing front-end. In Radio and Wireless Symposium, 2014
P. Lohmiller, A. Elsokary, S. Chartier, and H. Schumacher
(See online at https://doi.org/10.1109/RWS.2014.6830121) - A hardware prototype of a flexible spectrum sensing node for smart sensing networks. In Cognitive Radio Oriented Wireless Networks and Communications, 2015
A. Elsokary, P. Lohmiller, V. Valenta, and H. Schumacher
(See online at https://doi.org/10.1007/978-3-319-24540-9_32) - A hardware prototype of a flexible spectrum sensing node for smart sensing networks. In EAI Endorsed Transactions on Cognitive Communication, 2015
A. Elsokary, P. Lohmiller, V. Valenta, and H. Schumacher
(See online at https://doi.org/10.4108/eai.5-4-2016.151143) - Hardware Design of Spectrum Sensing Nodes for Collaborative Sensing Networks. 47. Treffen der VDE/ITG-Fachgruppe 5.2.4, New Radio Resource Management Concepts, Alcatel-Lucent Bell Labs, Stuttgart, 2015
V. Valenta, P. Lohmiller, A. Elsokary
- Implementation of cyclostationary based detection. Conference Kohutka, Czech Republic, 2015
O. Maňas, R. Maršálek. A. Elsokary, V. Valenta
