Efficient Preprocessing of Site-Specific Radio Channels for Virtual Drive Testing in Hardware Emulators

• Published in: IEEE transactions on aerospace and electronic systems Vol. 59; no. 2; pp. 1787 - 1799
• Main Authors: Mbugua, Allan Wainaina, Chen, Yun, Raschkowski, Leszek, Ji, Yilin, Gharba, Mohamed, Fan, Wei
• Format: Journal Article
• Language: English
• Published: New York IEEE 01.04.2023; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Alignment; Antennas; Channels; Convex optimization; Delays; Emulation; Emulators; Fading channels; Filter design (mathematics); Filtering theory; finite impulse response (FIR) filter; Finite impulse response filters; FIR filters; Frequency response; Hardware; Hardware-in-the-loop simulation; Optimization; Preprocessing; Radio; radio channel emulation; Ray tracing; ray tracing (RT); Real time operation; sparse approximation; virtual drive testing (VDT)
• ISSN: 0018-9251; 1557-9603
• DOI: 10.1109/TAES.2022.3205289

Performance testing under realistic propagation channel conditions is essential for virtual drive testing, where radio channel emulators are typically employed in the laboratory for such applications. Optimal allocation of tap resources in the channel emulator is critical in hardware-in-the-loop emulation of radio channels due to the constraint of real-time operation requirements, hardware complexity, and cost. As a result, replaying arbitrary site-specific, e.g., measured or ray tracing (RT) simulated radio channels in channel emulators requires delay alignment and reduction of the number of multipath components in the channel to match the available hardware resources. However, such operations would essentially introduce inaccuracies in the emulated channel. In this article, a framework for preprocessing site-specific radio channels for hardware emulation is proposed. The delay alignment problem is formulated as a finite impulse response filter design problem, whereas the subsequent tap reduction and selection process is formulated as a sparse approximation problem. This approach enables maximization of the accuracy of the reproduced channel frequency response and Doppler profile in the hardware emulator using a limited number of taps. The efficiency of the proposed framework is demonstrated experimentally with a dynamic vehicular RT simulated channel, which is replayed on a state-of-the-art radio channel emulator.