CorteXlab will foster significant scientific progress by allowing users to design, benchmark, and tune their cognitive radio protocols. It will enable evaluation of different aspects of cognitive radio in a real environment. It will help the development of application-driven research aimed at validating promising theoretical concepts. Compared to the other existing testbeds, CorteXlab will offer several particularities in order to address complex and evolved scenarios of next generation networks.
CorteXlab will be composed of a mix of radio nodes, including Wireless Sensor Network (WSN) nodes with lower flexibility (IoT-lab nodes), Single-Input Single-Output (SISO) SDR nodes, and MIMO SDR nodes. All SDR nodes (SISO or MIMO) will present a high level of RF flexibility, enabling to easily tune the operating frequency, channel bandwidth, emitted power and waveform. Last but not least, the testbed will be installed in a large (180 m²) shielded room (isolated from any external interference) and also partly covered with EM absorbing material. Depending on the set of enabled frequencies, the design of the room will enable to control the radio channel characteristics (number of paths, delays, etc …) and to ensure a high level of reproducibility of experiments.
CorteXlab will be composed of nodes uniformly distributed in the experimentation room (1.8m between each node). These nodes will be able to accept PHY layer implementations on both hardware, i.e. Field Programmable Gate Array (FPGA), and software on general purpose CPUs. Furthermore, they will be capable of outputing performance metrics, such as throughput, Bit Error Rate (BER) and power consumption. The nodes are interconnected through a high speed ethernet link, in order to allow for cooperation and sharing of information. A unified server will also be available for starting, coordinating and collecting the results of experiments. The experiments themselves can be from PHY layer up, including the possibility of cross-layer interactions.