研究目的
To improve the degraded range resolution in FMCW radar caused by non-ideal factors without increasing the signal bandwidth.
研究成果
The proposed post-processing method effectively improves range resolution from 160 cm to 70 cm in FMCW radar without increasing bandwidth, as validated by simulations and measurements. It enables distinguishing objects at 70 cm intervals, demonstrating practical applicability for radar systems.
研究不足
The method assumes stationary objects and may not account for dynamic scenarios. It relies on accurate quantification of non-ideal factors, which could be affected by measurement errors. The improvement is limited by the inherent constraints of the LSM and radar hardware.
1：Experimental Design and Method Selection:
The study uses a post-processing method based on the non-negative least-squares method (LSM) to compensate for non-ideal factors such as reduced effective modulation bandwidth, nonlinearity in frequency sweeps, and FFT window effects. The method involves solving an equation to decompose range FFT results into peak range bins.
2：Sample Selection and Data Sources:
Simulations and measurements are conducted using radar parameters (center frequency 76.5 GHz, bandwidth 200 MHz). Measurements involve two corner reflectors with a radar cross section of 10 dBsm placed at varying range intervals.
3：5 GHz, bandwidth 200 MHz). Measurements involve two corner reflectors with a radar cross section of 10 dBsm placed at varying range intervals.
List of Experimental Equipment and Materials:
3. List of Experimental Equipment and Materials: A 77 GHz radar module implemented with NXP transceiver chipset (MR2001TX/RX/VC), NXP MCU (MPC 5775K), planar antennas, and corner reflectors. MATLAB is used for signal processing.
4：Experimental Procedures and Operational Workflow:
Beat signals are generated and processed in MATLAB. For simulations, range intervals and SNR are varied. Measurements involve placing reflectors at specific distances, collecting beat signals, applying range FFT, and then post-processing with LSM.
5：Data Analysis Methods:
Detection probabilities are calculated for distinguishing two objects. Statistical analysis includes setting thresholds for probability of false alarm (PFA) of 10^-3. Results are compared between conventional range FFT and the proposed method.
独家科研数据包，助您复现前沿成果，加速创新突破
获取完整内容
