Autonomous driving is a new emerging technology that will enhance traffic safety. Automotive radars are essential to attaining autonomous driving since they can function in adverse weather conditions and are used for detection, tracking, and classification in traffic settings. However, the dramatic growth in the number of radar sensors used for automotive radars has raised concerns about spectral congestion and the coexistence of radar sensors. The mutual interference between multiple radar sensors downgrades the sensing performance of automotive radar and needs to be mitigated. Moreover, automotive radars have limited processing power, preventing them from using computationally heavy techniques to countermeasure interference. This thesis aims at developing, evaluating and verifying a robust waveform with required processing steps suitable for automotive radars to boost the coexistence of multiple radar sensors. To achieve this task, phase-coded frequency modulated continuous wave (PC-FMCW) and necessary processing steps are studied.
The first step is taken by investigating the sensing properties of the PC-FMCW waveforms and possible receiver strategies in Chapter 2. It is demonstrated that the ambiguity function of the code is sheared after frequency modulation. Moreover, different binary phase codes are examined with the PC-FMCW waveforms, and their sensing performance is compared in terms of integrated sidelobe level. Subsequently, two receiver approaches based on the dechirping process to decrease the sampling demands of the PC-FMCW waveforms are examined. The sensing performance of the investigated receiver approaches is compared, and the trade-offs between the sensing performance and the code bandwidth are analyzed. Moreover, the PC-FMCW waveform is applied to a real scenario, and the sensing performance of the investigated receiver structures is validated experimentally.
Chapter 3 investigates the beat signal spectrum widening due to coding and explores the smoothed phase-coded frequency modulated continuous wave (SPC-FMCW) to improve the sensing performance in the limited receiver analogue bandwidth. The abrupt phase changes seen in binary phase-coded signal is analyzed, and a phase smoothing operation to reduce the spectral broadening of the coded beat signals is proposed. The introduced SPC-FMCW waveforms are analyzed in different domains and compared with the binary phase coding. It is shown that the proposed smoothing operation decreases the spectral broadening of the coded beat signal and improves the sensing performance of the waveform.
In Chapter 4, the limitation in the group delay filter receiver approach is investigated, and the appropriate receiver strategy with low computational complexity is designed to process the PC-FMCW waveforms. The impact of the group delay filter on the coded beat signal is examined in detail, and a phase lag compensation is proposed to enhance decoding performance. It is demonstrated that performing phase lag compensation on the transmitted code eliminates the undesired effects of the group delay filter, and the beat signal is recovered properly after decoding. Then, the properties of the resulting waveforms are theoretically examined, and the sensing performance improvement over the existing approach is demonstrated. Moreover, both sensing and cross-isolation performance of the introduced waveforms with proposed processing steps are validated experimentally.
Chapter 5 studies the PC-FMCW waveforms for a coherent multiple-input-multiple-output (MIMO) radar. To this end, the MIMO ambiguity functions of the PC-FMCW waveform with different code families are investigated for their separation capability and compared with the PMCW waveform. It is illustrated that the PC-FMCW ambiguity function outperforms the PMCW one in terms of range resolution, Doppler tolerance, and sidelobe level for the identical types of codes. Afterwards, the developed phase lag compensated waveform with a single transmitter-receiver approach is performed to a coherent MIMO radar, and a novel PC-FMCW MIMO structure is proposed in Chapter 5. The introduced MIMO structure jointly utilizes phase coding in both fast-time and slow-time to achieve low sidelobe levels in the range-Doppler-azimuth domains while maintaining high range resolution, unambiguous velocity, good Doppler tolerance and low sampling requirements. The sensing performance of the introduced MIMO structure is evaluated and compared with the state-of-the-art techniques. Moreover, the proposed MIMO structure's practical limitations are investigated and demonstrated. In addition, the sensing performance of the developed approach with the simultaneous transmission is verified experimentally.
Finally, the interference resilience and communication capabilities of the developed PC-FMCW radar have been studied in Chapter 6. First, the automotive radar interference problem between various types of continuous waveforms is examined. The interference analysis formulation is extended to PC-FMCW waveforms, and a generalised radar-to-radar interference equation is proposed. The introduced equation can be utilised to quickly and accurately derive the numerous interference scenarios discussed in the literature. In addition, the proposed equation's validity to characterise the victim radar's time-frequency distribution is demonstrated experimentally using the commercially available off-the-shelf automotive radar transceivers. Afterwards, the robustness of the developed PC-FMCW radar against different types of FMCW interference cases is examined, and an improvement in the sensing performance over the conventional FMCW waveform is demonstrated. Moreover, the communication performance of the PC-FMCW with dechirping receivers is compared, and the trade-off between the bit error rate and the code bandwidth is investigated.
This thesis shows that the developed PC-FMCW radar structure can provide high mutual orthogonality to enhance the functioning of multiple radars within the same frequency bandwidth while sustaining the low sampling demand and good sensing performance. Consequently, the introduced approach can be effectively utilized by automotive radars to mitigate mutual interference between multiple radar sensors and improve the sensing performance of simultaneous MIMO transmission. Although the focus is on the application in an automotive radar context, the developed approach can also be used in other radar fields.