|
Autonomous vehicles have always been a field of considerable research interest. Past research have demonstrated achievements assuring that self-driving cars are, in fact the future of mobility. Self-driving cars have been made possible by sensor fusion technique, which incorporates sensors, including camera and radar. Cameras have the best resolution. Nevertheless, their ability to sense may be affected in extreme weather or night conditions. Radars are not affected by these conditions but lack the resolution when compared with radar. Most of the automotive radars are Frequency Modulated Continuous Wave (FMCW) radars whose range resolution depends on the bandwidth of the FMCW chirp, and spatial resolution depends upon the number of the receiving antennas. Having a higher number of receiving antenna elements will improve the angular resolution. Instead of increasing physical receiving antennas, it is possible to generate virtual receiving antennas by adding transmitting antennas, commonly known as the Multiple Input Multiple Output (MIMO) technique. MIMO requires orthogonal signals in multiple transmitting antennas. Commercial automotive radars have implemented the capability of MIMO using Time Division Multiplexing (TDM) and Binary Phase Modulation (BPM) in 2Tx and 4Rx systems. Although the angular resolution is improved, the maximum unambiguous velocity is reduced by half. This paper proposes the Frequency Division Multiplexing (FDM) Technique to achieve orthogonality. A full radar system has been simulated in MATLAB environment, which shows the possibility of using FDM in automotive radars without compromising the maximum unambiguous velocity. Frequency modulated signal with different starting frequencies for two Tx antenna is used to create 8 Rx virtual channels. FDM usually requires an increment in sampling frequency of Analog to Digital Converter (ADC). In this paper, the two starting frequencies are chosen, such that the requirement of higher sampling rate has been eliminated.
|
