Title: Cognitive Joint Mimo Radar Mimo Communications (C-Mrmc) Prototype

Cognitive radars optimize both transmit and receive processing to adjust to the dynamic target environment while aiming to enhance their behavioral agility by learning through experience in sensing and processing. In classical radar systems, the adaptability is limited to the receiver only. We present a Software Defined Radio (SDR) prototype that demonstrates cognitive optimization of radar transmit waveform, for the first time, to mitigate the interference to a spectrally coexisting communication link. The radar operates a multiple-input multiple-output (MIMO) array while the communication is a downlink multi-user MIMO system. Both the systems have independent transmit-receive (Tx/Rx) units and the radar system adapts its transmission to reduce its interference to the communication link. Further, on the Rx, it estimates the interference covariance matrix to enhance the detection performance. The radar achieves this adaptation by periodically acquiring the channel state information (CSI), which includes the frequencies and level of interference. The CSI is then used to design novel radar waveforms to enable coexistence. We have used Ettus USRP B210 and NI-USRP 2944R (70 MHz – 6 GHz) SDR. Baseband processing for 10 MHz, MIMO radar and MIMO communications is performed in CPU using Labview NXG 3.1 and GNU Radio 3.7 respectively. Our prototype incorporates the flexibility to update the radar waveform on-the-fly based on cognitive optimization under different constraints e.g., discrete phase. Our radar toolbox is capable of detecting target range, Doppler, angle of arrival. The MIMO communication implements an OFDM (IEEE 802.11p) Tx/Rx link with appropriate MIMO receiver processing in real-time and decodes layer-2 frames. Preliminary lab validation indicates our C-MRMC prototype achieves performance enhancements in both radar metrics (better detection at lower false alarm) and communication performance (lower synchronization failure and frame error rate) compared to non-adaptive radar transmission. Our C-MRMC prototype greatly expands the scope of the emerging signal processing research on joint sensing-communications to multiple antenna systems, thereby fitting well with the theme and audience of ICASSP.