Automotive and Industrial RF Advanced Automotive RF Informational

What is the digital signal processing pipeline for an automotive MIMO radar from raw ADC data to target list?

The digital signal processing pipeline for an automotive MIMO radar transforms the raw ADC data from each receiver channel into a target list containing the range, velocity, and angle of each detected object. The pipeline consists of: range FFT (for each receive channel, perform an FFT along the fast-time (sample) dimension of each chirp; this converts the beat frequency signal to range bins; the FFT size equals the number of ADC samples per chirp, typically 256-1024 points; each FFT bin corresponds to a range cell of width c/(2B)), Doppler FFT (for each range bin and each receiver, perform an FFT along the slow-time (chirp) dimension across the coherent processing interval (CPI); this converts the phase progression across chirps to velocity bins; the FFT size equals the number of chirps per CPI, typically 64-256; each bin corresponds to a velocity cell of width lambda/(2 x T_CPI)), MIMO virtual array formation (arrange the range-Doppler maps from all TX/RX combinations into a virtual array; for a 3 TX, 4 RX system: 12 virtual receive elements; the virtual array geometry determines the angular resolution and field of view), angle FFT or beamforming (for each range-Doppler cell, perform an FFT or beamforming operation across the virtual array elements to estimate the angle of arrival; alternatively, use super-resolution algorithms such as MUSIC or Capon beamforming for finer angular resolution than the FFT limit), CFAR detection (apply a constant false alarm rate detector to the 3D range-Doppler-angle cube to identify cells that exceed the noise floor by a detection threshold; this produces a list of detected cells with range, velocity, and angle), and target clustering and tracking (group adjacent detected cells into target clusters, estimate each target's centroid position and velocity, and feed the detections into a multi-target tracker such as an extended Kalman filter or particle filter).
Category: Automotive and Industrial RF
Updated: April 2026
Product Tie-In: Radar ICs, PCB Materials, Antennas

Automotive MIMO Radar DSP Pipeline

The radar DSP pipeline must process millions of samples per second in real time, producing updated target lists at rates of 20-50 Hz (frame rates matching the ADAS decision cycle). The processing is typically partitioned between the radar SoC's on-chip DSP and an external automotive processor.

ParameterOption AOption BOption C
PerformanceHighMediumLow
CostHighLowMedium
ComplexityHighLowMedium
BandwidthNarrowWideModerate
Typical UseLab/militaryConsumerIndustrial

Technical Considerations

When evaluating the digital signal processing pipeline for an automotive mimo radar from raw adc data to target list?, engineers must account for the specific requirements of their target application. The optimal choice depends on the frequency range, power level, environmental conditions, and cost constraints of the overall system design.

Performance Analysis

When evaluating the digital signal processing pipeline for an automotive mimo radar from raw adc data to target list?, engineers must account for the specific requirements of their target application. The optimal choice depends on the frequency range, power level, environmental conditions, and cost constraints of the overall system design.

  • Performance verification: confirm specifications against the application requirements before finalizing the design
  • Environmental factors: temperature range, humidity, and vibration affect long-term reliability and parameter drift
  • Cost vs. performance: evaluate whether the application demands premium components or standard commercial grades
  1. Interface compatibility: verify impedance, connector type, and mechanical form factor match the system architecture
  2. Margin allocation: include sufficient design margin to account for manufacturing tolerances and aging effects

Design Guidelines

When evaluating the digital signal processing pipeline for an automotive mimo radar from raw adc data to target list?, engineers must account for the specific requirements of their target application. The optimal choice depends on the frequency range, power level, environmental conditions, and cost constraints of the overall system design.

Common Questions

Frequently Asked Questions

Where is the processing done?

Partitioned approach: the radar SoC performs the 2D FFT (range + Doppler) and CFAR detection on-chip, outputting a point cloud (list of detected range-Doppler-angle cells) to the external processor. The external processor (typically an automotive-grade SoC such as NXP S32R or TI TDA4) performs: target clustering, tracking, classification, and sensor fusion with camera and lidar. Alternatively: some radar SoCs (NXP TEF82xx, Infineon) perform the complete pipeline on-chip, outputting a tracked target list directly.

What is the MIMO virtual array concept?

In MIMO radar: each TX antenna transmits a unique waveform (orthogonal chirps, time-division, or code-division), allowing the receiver to separate the returns from each transmitter. The result is: each TX-RX pair acts as a virtual receive element at the midpoint of the TX and RX element positions. For 3 TX at positions [0, d, 2d] and 4 RX at positions [0, 3d, 6d, 9d]: the virtual array has 12 elements at positions [0, d, 2d, 3d, 4d, 5d, 6d, 7d, 8d, 9d, 10d, 11d], creating a filled 12-element uniform array from only 7 physical antennas. This gives the angular resolution of a 12-element array.

How fast must the processing be?

For a typical 77 GHz FMCW radar: chirp duration = 50 us, 128 chirps per frame, frame time = 6.4 ms + overhead = approximately 20-50 ms per frame (20-50 Hz update rate). All range FFTs must complete within the frame time: 512 FFTs x 256 points in 20 ms is very achievable on the SoC's DSP. For real-time tracking: the target list must be delivered to the ADAS controller within 50 ms total latency (including processing and data transfer). Modern automotive radar SoCs achieve end-to-end latency of 10-30 ms.

Need expert RF components?

Request a Quote

RF Essentials supplies precision components for noise-critical, high-linearity, and impedance-matched systems.

Get in Touch