In our blog we would like to give you valuable contributions to topics such as ADS-B, Mode-S, MLAT, flight tracking, antennas and 1090 MHz.
If you are interesed in a special subject, please let us know at: firstname.lastname@example.org
Radar isn't a very old technology. Developed during WW2, it is 70 years old. Early radar was just to detect objects in the air. This was done by transmitting a powerful signal and waiting for the reflection from the flying object. The time between transmission and reception was twice the distance of the object. The position of the antenna provided the angular position. Something we still display even on modern systems. Such traditional systems are called primary radar.
However, due to geometry the altitude resolution on primary radar was not provided (or very inaccurate in systems with several beams). Also, if two aircraft were flying nearby, there was no identification provided for those bright spots on the screen. Due this demand a system was introduced permitting the aircraft to modify the received pulse: Using 12 bits, it returns either altitude or the squawk identification. The radar station has two request patterns which it transmits towards the aircraft, and the aircraft responds with either a Mode-A pattern for squawk or a Mode-C pattern for altitude. The pattern itself does not contain any information about its kind, so only the one who has requested that info will know how to post process. As a passive receiver such as we are, it is that we only hear a number between zero and 4095, but we don't know the question.
With growing air traffic, air traffic controllers required more information about the aircraft. Also, they wanted to get control over the transmissions in a crowded airspace. This is when Mode-S became introduced. Not at least the "S" stands for "Selective". Mode-S transmissions contains some more information like higher resolution altitude, aircraft capabilities and identification. As a passive listener we now got the capability to distinguish between several formats and collect information about each single aircraft.
Last, one special frame format within the Mode-S protocol, so called DF-17, became introduced to indicate the position of the aircraft and some more information. In fact ADS-B is only one message within Mode-S. It requires 2 DF-17 frames to calculate the position unambiguous, the so called even and odd formats.
Our Radarcape and the Mode-S Beast can receive Mode-AC messages and output them on the binary ports. But these messsages do not have position information within. They cannot be used for localisation within the processing capabilities of a Radarcape.
Also Mode-S does not contain location information, but as they have enougth information to distinguish among each other and combining data of at least 3 Radarcape with a precision time stamp and a process called Multilateration (MLAT) we are able to estimate the position of the aircraft from Mode-S transmissions.
Finally, using ADS-B transmissions, the aircraft directly tells us where it believes to be or where it wants us to think it currenly is.
My motivation of this project comes from a more than 25 years ham radio experience, mainly weak signal and GHz operation, a lot of interest in signal processing, and a professional background as an electronic development engineer.
I managed to combine all my knowledge in this project, and was able to fill some gaps with the special knowledge of some friends with special know-how in microwave and digital electronics.
It all began around 1995 when I mainly participated in radio contests on 144 MHz up to 2320 MHz with always good success. However, on the upper bands there was always one that was better. I heard that part of his success was his efficiency in aircraft scatter, a method when on 1296 MHz and 2320 MHz an aircraft is used as passive reflector. I also used this mode, mainly to the Netherlands, but wasted a lot of time waiting for aircraft passing the hot spot. A flight tracker called SBS-1 was the key for my competitor, which showed that there are aircraft in the scatter zone. However, for that time, this unit was far out of reach within my budget. Also, the thought of a closed system never attracted me very much.
Around 2007 I found more about ADS-B Receiver, 1090 MHz antennas and Software like Planeplotter, the miniADSB receiver and the PIC decoder and brought it into work. Back to ham radio thinking, I was surprised about the bad performance and started tweaking. The first result was the floating comparator, which adjusted the digitial comparator automatically to the signal level, thus avoiding the awful donut effect of this set-up. Driven by an "is that all?" thinking, I asked how reception could be done without the bottlenecks of a PIC, using an FPGA as decoder. Not knowing anything about FPGA, I started teaching myself. I avoided some weaks of the existing designs and brought in my experience from Earth-Moon-Earth communication. After some time, the ADS-B Mode-S Beast was born, which was even at the early stages outperforming all existing units.
By the way, why is it called Beast? Well, at that time Planeplotter only had some devices like SBS-1 and PIC decoders as frontend, and their weak performance resulted in only a few frames per second. With the Beast, Planeplotter suddenly had to handle much more messages and went into a stall. "Hey, what a little ugly beast", I thought, and so the name was born. Those early devices did have so less performance that even a hexdump "AVR format" or CSV protocol (Port 30003) was enough for data transmission, and I introduced the Beast-Binary protocol.
A few years later I sat together with some friends, talking about aircraft not yet localized and had the idea to catch them with multilateration (MLAT). At around the same time little Linux based boards grew on the market, mainly Raspberry Pi and Beaglebone. This matched to our need of remote operation and LAN interface, transporting the data to a central server - the Mode-S Beast only had USB and a not handsome and expensive LAN solution based on Xport. I decided to use the Beaglebone because it has connectors that allow an easy integration into a metal case and because of power supply problems of the Raspberry Pi. For easy and accurate multilateration I choose a GPS module as timestamp. The add-on boards on the Beaglebone are named "capes", and so my ADS-B flight decoder was named "Radarcape".
Flightradar24 asked for the Radarcape as backbone for their own network with their proprietary software in order to provide updates for their flight tracking app. They showed me an ADS-B antenna which was hardly worth the name. Due to my ham history, I found a manufacturer in Italy who makes very good antennas which really proof the given technical data. Our portfolio widened up and so the web shop warehouse increased to a whole set up.
In 2014 I joined the activities of existing jetvision, who had some solutions with RTL-SDR and USB DVB-T dongles, and my meanwhile founded company took over the brand "jetvision". In May 2016 my company took over the whole business of these and Planevision Systems offers rack solutions based on the Radarcape only. We now are a team of around 8 people working for the development, marketing and sales with a great support by friends and other companies in all themes.