COMRAY (COMmunications arRAY) is a globally distributed array of satellite antennas linked together through the Internet. This deep space communications array is being designed to overcome two of the most significant obstacles encountered in satellite communications:
Line of sight
Since the network will include nodes located all around the globe, there will always be some nodes capable of receiving signals from satellites operating anywhere within line-of-sight of the Earth itself.
With the aide of advanced signal processing algorithms, signals received from multiple nodes may be combined to improve the signal quality and overall bandwidth of the communications channels.
Normally, we communicate with an object in space withe large single dish antennas of large diameters, sometimes more than 100 feet. The cost and effort involved as well as the spatial requirements can only be dealt with by large organizations. Such systems can, of course, be rented, sometimes even free of charge. Yet their limited availability, plus the bureaucracy of large organizations all too often result in long waits and, therefore, very low flexibility. Another shortcoming is the fact that these antennas can only cover a portion of the sky: due to the rotation of Earth, the line of sight to the object – a prerequisite for radio transmission – can only be maintained for a limited time span.
COMRAY is different: it doesn’t involve large single dish antennas, but instead utilizes the power of many smaller ones distributed all across the globe. Linked together via the Internet, forming several large communication arrays. Individual antennas can be added or disconnected as required. Hence, these arrays can easily travel around the Earth while maintaining the line of sight to the object in question. The bandwidth of these smaller antennas may be lower, but this can be compensated for by the distribution of data to multiple frequencies and antennas. In this way, high data rates can be obtained even with antennas of less than 30 feet in diameter. Cost and spatial requirements are obviously lower. The flexibility of the system is enhanced according to the total number of antennas, thus making it attractive for smaller projects, too. Participating in COMRAY is an interesting and efficient solution for any private and commercial project as well as for colleges and universities.
2012 – Technical design and implementation, First prototype link nodes, Proof of concept demonstrations
2013 – First large test array, consisting primarily of link nodes located on the European continent
2014 – First global array, with nodes located on all continents
In order to build a communications array capable of meeting the objectives of the COMRAY project, we will need to push our current level of technology to its limits. Obtaining the necessary amounts of correlated communication data requires the use of extensive digital signal processing (DSP). The hardware required to perform this processing has only recently become technically feasible, and represents the cutting edge of DSP technology.
The link node is the technological heart of each antenna. The embedded electronics are based on one of the worlds fastest Xilinx FPGA (Field Programmable Gate Array) chip. These chips are capable of meeting the rigorous demands for real-time digital signal processing power. The FPGA based system will be able to process and correlate a full 500Mhz Band in less than half a second. The only purpose of the operating systems running on the link node is to maintain higher-level inter-node connectivity over an IP (Internet Protocol) stack. Each link node will also possess a high-precision clock and a GPS receiver to facilitate the further integration of multiple signals from geographically distributed nodes.
Link Station (recv)
A link-station-recv consists of a electronically controlled antenna attached to a link-node. A typical station will be based on a 120cm satellite dish equipped with RF (Radio Frequency) components for receiving signals only. Although the dishes are somewhat small, the signals obtained from multiple stations can be correlated to improve the signal quality and overall bandwidth. It is, therefore, very important to have a large number of down-link stations. In addition, the uplink station will typically be almost three times as expensive as a receiving unit. Therefore, the vast majority of units being deployed will be of link-stations-recv type.
Link Station (send)
The configuration of the link-station-send node is almost identical to the standard link-station-recv described above. The primary difference is the inclusion of additional RF components for accomplishing signal transmission.
The processing power of link-nodes and the receiving power of link-stations are integrated together into a pool, which then provides a higher-level interface to the end-user. The pool is logical description of an abstract communications device. The pool keeps track of the trajectory of an object in space, the frequency being used to communicate with this object, and a list of the link-nodes and link-stations that are currently being (or scheduled to be) utilized for bandwidth aggregation.
Link-nodes and link-stations are only temporarily bound to a pool on an as need basis. For example, consider the case where a pool is created to communicate with a spacecraft orbiting Mars. The user specifies how many dishes will be needed, or alternatively, the desired up-link or down-link bandwidth. The pool determines which dishes are currently within line-of-sight of Mars and are available for use. If there are a sufficient number of dishes available, then they are allocated to the pool. The pool then directs each of the dishes to point at the appropriate piece of sky to begin collecting a signal. The signal collected by each link-station is processed by link-nodes. The processed data is passed on to the pool where it is correlated with data from other link-nodes before being passed on to the user. As the Earth rotates, additional link-stations are allocated, while other are deallocated as they loose line-of-sight of Mars.
There may be multiple pools active at any given time. The only constraint is the amount of available resources. Responsibility for allocating resources to each pool lies with the central gateway server.