• 状态 TI Thinks Resolved
  • 已锁定 已锁定
  • 回复 4 回复
  • 答案 1 答案
  • 订户 9 订户
  • 视图 265 视图
  • 用户 0 会员在此处
选项
选项
相关

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

MMWCAS-RF-EVM: Does the sensor design meet my usage scenario?

XINCHENG WANG
Prodigy 10 points
Part Number: MMWCAS-RF-EVM
Other Parts Discussed in Thread: MMWCAS-DSP-EVM, AWR1642BOOST, DCA1000EVM, AWR1642, IWR6843, IWR6843ISK, MMWAVEICBOOST

I would like to assemble a millimeter-wave radar transceiver system with the goal of collecting micro-Doppler signature signals generated by the rotation of rotors on multirotor UAVs. I have two proposed assembly schemes:

  1. Scheme 1: Connect the MMWCAS-RF-EVM to the MMWCAS-DSP-EVM, then to a PC. Two WR-12 25 dBi horn antennas are used to achieve signal gain. The horn antennas are connected to the MMWCAS-RF-EVM via a 1.0 mm-f to WR-12 waveguide transition, a 1.85 mm to 1.0 mm adapter, and a SMPM-m to 1.85 mm-f 5 cm cable. This configuration enables signal enhancement without modifying the original structure of the MMWCAS-RF-EVM.

  2. Scheme 2: Connect the AWR1642BOOST to the DCA1000EVM, then to a PC. Similarly, two WR-12 25 dBi horn antennas are used for gain, connected using a 1.0 mm-f to WR-12 waveguide transition, a 1.85 mm to 1.0 mm adapter, and a SMPM-m to 1.85 mm-f 5 cm cable, again without modifying the original structure of the AWR1642BOOST.

I would like to ask:

  • Do either of these two assembly schemes have any technical issues or compatibility concerns?

  • Which setup is more recommended?

  • Will the signal resolution from either system, under open outdoor conditions, be sufficient to clearly display the micro-Doppler frequency shifts caused by different UAV models (i.e., variations in rotor length and rotational speed)?

  • My UAV parameters are:

    • Rotor length: 20–80 cm

    • Propeller rotational speed: 2800–6000 RPM

    • RCS: 0.02–0.2 m²

If possible, I would also like to know the theoretical detection range of these systems under the given conditions, and whether the collected micro-Doppler features would be sufficient for UAV classification based on their signature differences.

  • 0
    TI__Mastermind 29315 points

    您好，

    已经收到了您的案例，调查需要些时间，感谢您的耐心等待。

  • 0
    TI__Expert 5290 points

    Hi 

    Typically, AWR1642 could detect and track vehicles within range of 150m and MMWCAS-RF-EVM could detect within the range of about 350m.

    You can compare your target RCS and material with car to understand possible range to detect a UAV.

    MMWCAS-RF-EVM is preferred as it is always of best performance.

    We did not try to detect UAV before, but from we already know, u-doppler map together with AI on IWR6843 could help us distinguish tree, walking male, dog ,fan, and AGV,  the feasibility is proved by our early stage experiment.  

    Ken

  • 0 回复 Ken Zhang
    Prodigy 10 points

    Hi,

    Thank you very much for your response, it has provided me with an initial understanding and judgment.

    We have indeed considered using the IWR6843ISK connected to the MMWAVEICBOOST, DCA1000EVM, and a PC as a solution. However, on one hand, the IWR6843ISK uses a patch antenna and lacks some interfaces, which seems to make it impossible to connect a signal gain antenna through the data interface without altering the sensor's original design. On the other hand, the IWR6843ISK has a center frequency of 60GHz, which, compared to the 77GHz center frequency of the AWR1642 and MMWCAS-RF-EVM, may result in a difference in micro-Doppler resolution. I wonder if these two points are correct in your opinion? If there are any mistakes in what I said, I would appreciate it if you could point them out and correct them.

    Additionally, I would like to ask about the distance of the objects in the experiment you mentioned with the IWR6843. How far are the objects from the sensor approximately?

    I aim to distinguish different drone models from the micro-Doppler images of different drones, and use AI for classification, rather than just detecting them as a broad category like vehicles. Therefore, I may need a higher signal frequency resolution to achieve this goal. What I mean is that my purpose is to conduct experiments, so conditions like distance can be adjusted. If the signal strength is insufficient, I can move the target closer to solve the issue. However, I am not sure if, even with a closer distance, its signal resolution will be able to detect the Doppler frequency shift caused by the rotating propellers in the hovering state of the drone. In your opinion, can the MMWCAS-RF-EVM theoretically help me achieve my goal?

  • 0 回复 XINCHENG WANG
    TI__Expert 5290 points

    Hi 

    The center frequency would not make much difference in micro-doppler resolution, 

    the resolution of doppler is mostly contributed by the actual chirp configuration inside the device, below document is essential before your experiment is conducted.  www.ti.com/.../swra553a.pdf    you can see the main factors include wavelength (center frequency), chirp number and chirp time. 

    As for interfaces, 6843 and 1642 are almost the same, sorry I don't get your point that 6843 "lacks some interfaces".

    It is possible to distinguish models with the help of micro doppler map, but pls take care of antenna gain and chirp configurations and it defines your max range/doppler and resolution. Once your KPI is determined, it will take little effort to convert it into actual RF parameters and then you will find out if your proposal is feasible or not to be implemented on TI device.

    Ken