FMCW-FMCW Interference Analysis in mmWave Automotive Radars
Software Tools: MATLAB and mmWave Studio
- R. Amar, M. Alaee-Kerahroodi and M. R. Bhavani Shankar, “FMCW-FMCW Interference Analysis in mm-Wave Radars; An indoor case study and validation by measurements,” 2021 21st International Radar Symposium (IRS), 2021, pp. 1-11, doi: 10.23919/IRS51887.2021.9466178.
- Amar, Robin and Alaee-Kerahroodi, Mohammad and Babu, Prabhu and R, Bhavani Shankar M., “Designing Interference-Immune Doppler-TolerantWaveforms for Automotive Radar Applications,” submitted to TAES (2022).
Radar sensing has various inherent advantages in different scenarios and its usage in civilian applications has been proliferating in addition to the traditional defense applications. Some of these applications include automotive, indoor sensing, airport, harbor, and highway traffic control, wave forecast, and marine climatology to name a few. With different applications and a dearth of standardization, the presence of multiple radar systems with uncoordinated transmissions has become a reality. Considering an example of an automotive application, radar sensors are used for cruise control and collision avoidance systems in cars. In the absence of coordination among different cars, scarcity of spectrum, and the need for high bandwidth to enable high-resolution sensing, it is inevitable that different cars use identical bands simultaneously, leading to interference. Similarly, its usage in Indoor Sensing for Target detection and classification with multiple sensors has gained a lot of attention. Few major applications are People counting, Flash and run-through body scanners, automotive interior monitoring, Robot positioning, industrial machine lock/cut-off, etc.
Multi-sensor setup leads to mutual interference from neighboring radars operating simultaneously in each other’s field of view (FoV). Analyses and test results involving multiple radar sensors indicate that mutual interference can be substantial unless suitable mitigation techniques are employed. For example, two 77GHz long-range radars with 18dB antenna gains and 10dBm transmit powers, facing each other, would cause -45dBm interference to each other. Interference becomes inevitable from other sensors when the application insists to have the best possible resolution for better classification of targets which leads to utilizing full bandwidth available in an operating band which is the case in many indoor and outdoor sensing scenarios. For a given system, its interference quantification (maximum possible rise in Signal-to-Interference Ratio (SIR)) and robustness is required to understand the target detection and classification capability of a sensor in worst scenarios. In this paper, an effort has been made to explore the interference occurring from “close proximity” and direct Line of Sight (LOS) sensors which are arranged in a multi-sensor arrangement for Indoor Sensing applications. The investigation has been conducted using mm-Wave Frequency Modulated Continuous Wave (FMCW) radar sensor measurement set-up supported by an elaborate simulation. Further, the backscatter and interference characteristics similarity between indoor sensing and outdoor sensing has also been discussed.
Types of Interference
Interference in automotive radars can be categorized to:
1. Same Slope Interference: This typically arises through the Multipath reflections that have a strong return (i.e. do not suffer from the severe path and backscatter losses). These types of interference are generally observed in both outdoor and indoor scenarios. In outdoor scenarios, it is observed in Tunnels and highways with high guard rails on the road borders in rainy conditions. In indoor scenarios, it is generally observed in underground parking and rooms with high reflecting walls and floor.
2. Similar Slope Interference: This type of interference occurs when the interfering sensor is operating with similar bandwidth and chirp time. The redefinition was required because practically the probability of another sensor to operate with exactly the same slope chirp is very low because the waveform generators operating on every hardware have their own linearity properties which will never yield absolutely parallel slope chirp.
3. Sweeping Interference: This type of interference occurs when the interfering sensor is operating at a significantly high or low bandwidth and different chirp time.
Parameters to derive the Link Budget:
Parameter Value: Victim / Interferer Power 12dBm , TBP 40.1773dB , Processing Gain 21.0721dB , Noise Power -68.83dB Noise Power + CFAR -57.83dB
See a nice video of simultaneous operation of 3 mmWave radars below