Projects:2014S1-15 Inexpensive Portable Radar System
This project is an on going project from last year which was to build an Inexpensive Portable Radar System from the MIT design. Last year was able to have a working radar system that used two Milo cans as antennas, Laptop for processing, RF and electrical components for wave construction and mixing. As the world is being more technology advance people are turning to radar to do more everyday task e.g. helping detecting objects behind a car. With everyone trying to save designing/building a Inexpensive radar is idea and making it portable increases its functionality. Another end goal for this project is to build a radar system which can be used in level 3 RF practical's to give students a more fun project in learning about RF.
This project is an on going project from last year which was to build an Inexpensive Portable Radar System from the MIT design. Last year was able to have a working radar system that used two Milo cans as antennas, Laptop for processing, RF and electrical components for wave construction and mixing. As the world is being more technology advance people are turning to radar to do more everyday task e.g. helping detecting objects behind a car. With everyone trying to save designing/building a Inexpensive radar is idea and making it portable increases its functionality. Another end goal for this project is to build a radar system which can be used in level 3 RF practical's to give students a more fun project in learning about RF. The problems of last years radar system is that it needs a laptop connected to be able to do the processing which makes it less portable, the radar used a hardware made waveform which makes the radar restrictive to the ability of doing more complex sensing and the hardware was easily damaged which lead to issues with making it portable. One of the main issue was that the radar was not in real time which made it almost useless as a radar system which can be used for everyday tasks. This year we will try and fix most of these issues. The end goal for this project this year is to have the radar in real time with ability to produce different signals and be more portable by producing hardware onto a PCB and having a better casing for the whole radar system.
Previous Radar System
The Radar System that last year was designed from the MIT Can Radar System with some modifications done by last year’s group. The software of the system is two matlab functions which take in a wave file that is produced from an audio recording software (Audacity). This limits the radar to only being able to record data and progress it later (not real time). The hardware of the system was improve by last year by designing a PCB for the electrical components. They produced a PCB that allowed the radar system to work but the PCB had some faults. They produced a new PCB but were not able to get it working with the radar system.
- Aim 1 : Finish what last year started by removing the breadboard and replacing it with a working PCB.
- Milestone 1: Be able to produce a Range and Doppler shift image using unchanged radar
- Milestone 2: Produce same image as from Milestone1 but using the PCB in the new radar structure
- Milestone 3: Have the breadboard removed and PCB installed
- Aim 2.1 : Make the radar process in real time using MATLAB.
- Milestone 1: Read in a 30second signal from the audio jack into MATLAB
- Milestone 2: Process pre-recorded data by only using MATLAB
- Milestone 3: optimal code to produce image in expectable time for a real time system
- Milestone 4: Process data from an object being 10m away in real time
- Aim 2.2 : Make the radar process in real time using Simulink.
- Milestone 1: Produce code that will split a .wav live recording into right and left channels and produce a spectrogram of a basic signal
- Milestone 2: Implement a MATLAB function block into the Simulink model
- Milestone 3: Produce a spectrogram of a real radar signal
- Milestone 4: Produce a graphical user interface (GUI) for Simulink
- Aim 3.1 : Use an embedded processor (e.g. raspberry pi or arduino) capable of controlling the waveform digitally.
- Milestone 1: Obtain an embedded processor
- Milestone 2: Obtain all components for the waveform construction circuit
- Milestone 3: Have working code that produces a triangle wave of 20ms period constantly
- Milestone 4: Have the embedded processors produce a signal out of one of its outputs
- Milestone 5: Have the embedded processors producing the signal for the VOC in Doppler mode
- Milestone 6: Have the embedded processors producing the signal for the VOC in Range mode
- Aim 3.2 : Use the raspberry pi to be able to do the processing of the signal eliminating the need for a laptop
- Milestone 1: Feasibility study on raspberry pi real time
- Milestone 2: Convert the processing from MATLAB into the software decided for the raspberry pi.
- Aim 4 : Redesign the physical casing of the radar to improve functionality and portability
- Milestone 1: Improve the power supply circuit
- Milestone 2: Produce a PCB with the wave construction circuit for the raspberry pi
- Milestone 3: Produce a PCB with just the video amplifier circuit on it
- Milestone 4: Build new casing
- Milestone 5: Add screen to the radar system
- Aim 5 : Build an auto-system which moves the radar allowing it to do SAR imaging
- Milestone 1: Have an auto-system design
- Milestone 2: Build the auto-system
- Milestone 3: Produce SAR image using auto-system
This Year's Work
This year has been able to meet their first four aims to produce a more reliable well built radar system that is able to do real time processing and uses a Raspberry Pi to produce the signals needed to operate the radar system. The new hardware is also using new batteries that are smaller and light weight which helps making the radar system more light and portable. The new radar system weighs 3.6kg with the casing.We were also able to get last years design working with their PCB. Both radar systems still need the use of a laptop to do the processing. The new screen of the radar system is able to work with the Raspberry Pi but the Raspberry Pi is not able to do the processing of the signal in a good amount of time but also doesn't have MATLAB to be able to do the real time processing, which is way the laptop is needed still.
Future Goals of the Project
We believe there are more improvements that are able to be added to our design to make it more functional or better performance, A few ideas are:
- Turn Simulink code into c++ code which will allow Raspberry Pi to do the processing of real time
- Improve Performance of Raspberry Pi by improving code or format of the Raspberry Pi
- Finish SAR imaging auto-system
- Improve display of real time
- Try different signals into VCO to add new functions of radar or improve system
- Improve video amplifier e.g. better gain, only need one voltage input
- Improve interface and display of radar system
- Mr Andrew Reed (Project Manager)
- Mr Aaron Milne (Hardware Manager)
- Mr Nathan Uebergang ( Research Manager)
- Mr Pavlo Potykan (software Manager
- Dr Matthew Sorell
- Dr Brian W. Ng
- Bench 16 in Projects Lab
- ML Server in EM211
- Can Radar from last year
We would also like to thank my two supervisors Brian Ng and Matthew Sorell for the support and advice they have given us throughout this year without their help we would not have been able to complete are goals. We would like to give a very special thanks to Danny Di Giacomo who has been a great help throughout this project. He has gone up and beyond in the help and advice he has given my team but also the time he has given us to ask questions and order parts. Another two people that have been a great help with this project this year have been Pavel Simcik helping with PCB designs and Aubrey Benjamin Slater from the workshop who was a great help in designing and building the new cases of the radar system. Finally we would like to thank Chris Neller who worked on this project last year for the times he has come to help us understand the radar but also the ideas his given to help improve the radar.