ZACube-2: The successor to Africa s first nanosatellite

Transcription

1 ZACube-2: The successor to Africa s first nanosatellite Daniel de Villiers and Robert van Zyl French South African Institute of Technology (F SATI), Cape Peninsula University of Technology (CPUT), Cape Town, South Africa. Abstract Following on the successes of ZACube-1, a.k.a. TshepisoSat, ZACube-2 is the second instalment in the F SATI mission series. The satellite will serve as technology demonstrator for essential subsystems and form the basis on which an innovative Software Defined Radio (SDR) platform will be developed as primary payload. The SDR is highly flexible to address a wide range of communication needs and will be a test bed to validate vessel detection. Additionally, the satellite will feature a medium resolution imager as secondary payload to demonstrate the feasibility of future remote sensing applications such as ocean colour monitoring and large fire tracking. This paper details the conceptual design and highlights the choices made around the proposed development. I. INTRODUCTION The CPUT nanosatellite programme involves a visionary approach to space exploration and the associated development of technology used for this. With the successful launch of TshepisoSAT on 21 November 2013 and still operational, CPUT gained international visibility and recognition to have launched and operated the first nanosatellite developed on the African continent. The ZACube-i space missions are a series of South African nanosatellites developed by CPUT in collaboration with a growing community of research institutions in South Africa, Africa and internationally. ZACube-2 is currently in development as an advanced nanosatellite platform with a primary focus on maritime domain awareness (MDA) applications. II. ZACUBE-2: A TECHNOLOGY BASELINE FOR MDA ZACube-2 will be the second satellite in F SATI s ZACube-i nanosatellite mission series. These missions are developed at the French South African Institute of Technology (F SATI) and the Africa Space Innovation Centre (ASIC) at CPUT with funding principally from the Department of Science and Technology (DST) and the National Research Foundation (NRF). Development of some subsystems has been ongoing for a number of years and has yielded a suite of commercial CubeSat parts that is ready for use in the satellite. ZACube-2 will serve as a technology demonstrator for essential subsystems required in operational satellite missions for very specific applications that are deemed especially suitable for implementation with nanosatellites. The mission will further grow the core expertise of CPUT and its technology partners and validate the technology innovations that result from it. ZACube-2 will be a test bed for a shiptracking payload and will be used to validate the use case of employing nanosatellites in ocean vessel detection through the Automatic Identification System (AIS) protocol [1]. Additionally, the satellite will carry medium resolution imagers as a secondary payload to demonstrate the feasibility of using a nanosatellite for imaging applications, such as ocean colour monitoring and fire tracking. III. FUNCTIONAL REQUIREMENT The mission s performance will be measured against the effective tracking of vessels within the borders of the South African continental shelf. There are an estimated 6000 ships in South African waters, a number that is small enough to sustain successful reception of AIS messages. Coverage, Orbit and Responsiveness Although this mission is a single satellite mission, it is informative to determine the coverage of a constellation of satellites similar to ZACube-2. The revisit interval is the time delay between consecutive visits of satellites in the constellation over the same area. It is calculated by considering orbital height, antenna pattern, the number of satellites in the constellation, and the ability to control the orbit. Since orbit control is not within the scope of this project, the satellites will drift with respect to one another and revisit intervals will vary with time. Under precise placement conditions, preliminary simulations reveal a nominal revisit time of 30 minutes over Southern Africa with a 9-satellite constellation covering the globe in 3 separate sun-synchronous orbit planes consisting of 3 satellites evenly spread out per plane. Conceptually this configuration could enable vessel tracking with an enhanced update rate. Due to the practical

2 launch logistics and cost implications, such a constellation of satellites will most likely be a secondary payload on another mission, thus control of precise constellation placement may be limited. This will have an impact on the actual revisit time. For the sake of limiting orbital lifetime, it is preferable that the orbital height does not exceed 600 km, but the available deorbiting devices are very effective at reducing orbital life significantly and may allow a higher orbit than 600 km to be used. Limiting the orbit height to 600 km may limit the availability of launch opportunities. Responsiveness of the mission to provide users with data depends on the data capturing process, processing delays, and concept of operations. To provide users with the quickest possible access to data, AIS information can be captured by the satellite and downlinked immediately while both the AIS ship beacon and the ground station are within the satellite s footprint. If immediate access to the data is not required, it can be stored in on-board mass memory and downlinked when convenient. Ground infrastructure is vital to the responsiveness of the system. If serving Southern Africa only, ground stations in South Africa will be sufficient to support the data capture and dissemination process. In order to realise a true global tracking solution, international partners would be required to downlink, process and disseminate data. Physical Size Preliminary design work shows that the entire SDR payload will fit into one nominal sized CubeSat module. Some additional RF front-ends and antenna switches are required for the SDR to interface with the antennas. Depending on the level of redundancy added to the system, either a 2U or a 3U CubeSat form factor could accommodate all the subsystems. A 2U structure may present more risk in that deployable solar panels may be required to satisfy the power budget. There is also less space for the deorbiting mechanism, which is regarded as mandatory. A 3U structure may have enough surface area so that fixed body mounted solar panels capture enough solar energy and the increased spacing of the internal modules will reduce the likelihood of thermal hot spots. It also allows for up to 0.8U of volume for the imager payload. The larger form factor will accommodate additional RF front-ends and allow the SDR payload to operate in more communication modes using the available antennas. It is proposed to use a nanosatellite structure already developed by Clyde Space as baseline for the concept design. The fully qualified structure was designed with adaptability and ease of integration in mind. Command and Data Handling (C&DH) The spacecraft will house a number of processors for various purposes. The ADCS, developed by Stellenbosch University as part of the QB50 mission, will have a powerful processor to perform attitude estimation and correction [3]. The SDR payload also has its own processor to enable real time reception and demodulation of RF signals. The primary OBC may not require the same level of processing power as the SDR and ADCS and may be based on the Texas Instruments MSP430 class of microcontrollers (as used in TshepisoSat). This will allow the reuse of TshepisoSat s flight software as baseline for this mission. In the event of OBC failure, other processors mentioned above must be able to perform basic OBC functions. All of the above processors will be accessible from ground operations via watchdog and back door mechanisms. End-of-life In order to comply with the UN policy on the sustainable use of outer space to deorbit satellites within 25 years, it is imperative that the spacecraft features a deorbiting device once end-of-mission life is reached. Various analytical tools and techniques can be used to estimate orbital life of a satellite. Further analysis will determine what deorbiting device would be a good fit and at what time during the mission it should be deployed. The orbital height has an exponential effect on the deorbiting lifetime of the satellite. Operating in orbits below 650 km should be sufficient to rely on atmospheric drag to deorbit a typical 3U satellite within 25 years; however, it is beneficial and encouraged to deorbit in the least amount of time post mission life. A very low orbit with significant atmospheric drag may shorten orbit life too much. It is proposed to baseline the concept design with the Terminator Tape developed by Tethers Unlimited due to the ease of use and compact form factor. Unlike a deorbiting sails, the Terminator Tape does not require custom cutouts with modifications to the structure and solar panels and only weighs 83 g with a dimension of 100 x 83 x 6.5mm. Upon completion of the mission, the satellite commands deployment of the Terminator Tape. The module utilises a novel shape memory alloy (SMA) actuator to initiate deployment of a conducting tape, which generates neutral particle drag and passive electrodynamic drag to hasten the decay of the satellite s orbit. Once deployed, the orbital life of a typical 3U satellite at 600 km orbital height will reduce to an estimated 1.5 years. IV. CONCEPTUAL DESIGN The concept layout of the satellite, as highlighted in Figure 1, has been designed with the required subsystems and payload in mind, as well as the location of deployables such as the antennas and solar panels. The form factor baseline will be a 3U CubeSat. The nanosatellite framework, shown in Figure 2, can be divided into two main system areas: Platform and

3 Payload. PLATFORM The platform contains all the components that play a supporting role to operate the nanosatellite. The various subsections in the Platform can be defined as follows: feasible in terms of payload data rate. Antennas - The HSTX will interface with an in-house developed circular patch antenna mounted on the nadir facing side of the satellite. Additionally, two antennas configured for circular polarisation, one tuned for VHF (SDR payload) and the other for UHF (UTRX). The VHF antenna is needed to receive AIS messages as required for the primary objective. The AIS frequency bands are already defined, thus the antenna will be designed for this band. It is important to note that an antenna is designed for a specific centre frequency and bandwidth. Even though the SDR platform is very flexible, its frequency range of operation will be mostly be determined by the specific antenna design. PAYLOAD The payload contains the components that are directly related to the specific mission of the nanosatellite. As defined before, the primary mission objective relates to the monitoring of ocean traffic using AIS; the secondary objective is to use the payload to support imaging applications. The various subsections in the Payload can be defined as follows: Figure 1 ZACube-2 Conceptual Layout OBC (On-Board Computer ) - The OBC is responsible for the basic operation and management of the satellite subsystems. Further investigation into a feasible OBC needs to be done, although various commercially available solutions exist. EPS (Electrical Power Supply) Clyde Space offers various EPS modules coupled with deployable panel configurations for 3U nanosatellites capable of generating a peak power of up to 30 W [2]. ADCS (Attitude Determination and Control System) - In order to have a known antenna pattern pointed at the ground at all times as well as keeping the cameras nadir pointing, the satellite will be 3-axis stabilised and controlled. The baseline ADCS is ESL s ADCS suite [3]. UHF transceiver (UTRX) - The UTRX is an in-house developed product similar to the CMC flying on TshepisoSat but operates a telemetry downlink and telecommand uplink in half- duplex mode in the UHF band. S-band high data rate transmitter (HSTX) - A high data rate downlink is required to transmit all the gathered AIS and collected image data back to the ground station. Further investigation needs to be done into whether the current inhouse developed STX featuring 2Mbps downlink will be RF front-ends - The RF receiver front-end interfaces with an antenna to receive analogue signals. It contains various RF modules to amplify, filter and isolate the specific signals of importance. After isolating the required signal it is digitised using an analogue-to-digital converter. From here all processing is done digitally. The RF front-end is also designed for a specific frequency range and performance, thus there will be an AIS RF front-end used for the primary objective and an alternative pair of RF receiver and transmitter front-ends to interface with the UHF antenna. This will enable the SDR to operate in the UHF band as a redundant telemetry and control radio. Payload processor - The payload processor performs all processing related to the specific mission payload. In the case of ZACube-2, it performs SDR processing for AIS communication and image processing. For SDR processing it receives a digital stream of sampled analogue data from multiple RF front-ends and can perform various operations on this data stream. It can both demodulate and extract radio messages or it can store the raw signal for post processing. The payload processor will also interface with the imagers onboard. In most cases it would just store images to be downloaded at a later stage, but if needed it can also perform limited image processing on-board the spacecraft. Extensive research went into the selection of a suitable payload processor. The Xilinx Zynq processor has been identified as a good candidate [4]. The processor contains both an FPGA with a dual core ARM processor, thus combining the world of high performance processing with the flexibility of

4 the processing you would find on a smartphone. This device is suitable for both SDR and image processing applications. The FPGA dedicates a part of hardware for specific tasks. This is ideal for handling the processing intensive and continuous tasks, like filtering, demodulation and extracting the input signal. The ARM is a generic processor that executes code from memory, thus it is good at adapting and changing between different tasks. The functionality of the payload processor can be upgraded in flight. Updating the ARM is similar to installing an App on your smartphone, whilst an FPGA upgrade is similar to a firmware upgrade. Ensuring secure operation and management of the upgrade procedure will form an important part of the system design. flexible than one that has its RF functionality permanently hard-wired in hardware. This flexibility is a key benefit: it gives technology the ability to adapt and evolve to changing environments and it results in a single-technology platform that can address a wide range of applications. We see an opportunity emerging where the abundant supply of miniaturised low power processing components, combined with the flexibility achieved using SDR concepts, makes it possible to produce nanosatellites that are generic enough to benefit from economies of scale in production, but are also flexible enough to satisfy a broad range of applications. This increases the likelihood of adapting to new opportunities to serve Government requirements or commercial markets as they arise. This is a substantial benefit in an industry where existing deployments are extremely hard to modify and are rigid in their application. The adaptability of the SDR payload adds benefit to satellites already in service, allowing in-flight reconfiguration, as well as to satellites being developed for future constellation replenishment missions where changes to the payload can be rapidly implemented without any hardware changes. Furthermore, the use of SDR technology can assist with improving the stability of a constellation of satellites: when problems are discovered, the issue can be fixed and a patch uploaded to the constellation to update the nanosatellites with the improved software. VI. SDR AS PRIMARY PAYLOAD The primary objective of the SDR payload is to effectively track vessels within the satellite s footprint through AIS. The primary focus is on the South African Maritime Domain. But future missions may expand to global monitoring. Figure 2 ZACube-2 subsystem block diagram Storage - The payload will contain sufficient storage to store both AIS and imaging data of importance. The processor payload alone features 64GB flash storage. When ZACube-2 is in range of a ground station, it will download payload data to preserve on-board storage. Additional solid-state storage will be investigated should the physical space permit it. V. SOFTWARE DEFINED RADIO (SDR) For radio frequency (RF) communications electronics, the performance of recent processor chipsets (thanks to the revolution in mobile technology) allows for more RF signal processing to be shifted from dedicated hardware chipsets to software implementations. The result is Software Defined Radio (SDR) - a platform for RF solutions that is far more In such a case, the estimated data volume may be calculated from a maximum of 60,000 vessels globally, of which 45,000 are active daily. The biggest challenge in receiving AIS messages from so many vessels is that the channel experiences signal collisions that make the reception of individual messages very difficult. Advanced signal processing techniques and directional antennas can be used to decrease the number of signals received simultaneously by effectively filtering according to the varying amounts of Doppler shift with which the signals arrive at the satellite. The CubeSat AAUSat developed by Aalborg University in Denmark received 13,000 messages per hour over the Mediterranean [5]. Triton-1 demonstrated that it could decode up to 4 AIS messages received simultaneously using a proprietary de-collision algorithm [6].

5 geographically to overcome signal contention, a directional beam pattern can be beneficial. While the satellite does not provide enough space for a highly directional VHF antenna to be mounted, some configurations could provide some gain. A turnstile antenna with crossed dipole elements in the pitch and yaw axes (as in the concept design in Figure 1) have a gain pattern with higher gain in the plane of the elements and decreased gain in the roll axis (velocity vector). Figure 3 RF transmit/receive front-end bank Frequency bands covered by the SDR payload The number of frequency bands that the SDR payload can operate in are primarily constrained by the number of antennas on the satellite. The concept layout of the satellite allows for a total of three antennas. Transmit/receive RF front-ends will cover the extended AIS band from about MHz. An additional SDR uplink/downlink pair will be implemented in the UHF band so that it can be used as a backup telemetry and command transceiver. As a secondary use case the UHF link can be beneficial for potential store and forward applications. An S-band link via a patch antenna for payload data downlink will be implemented with the SDR also having access to the S- band front-end. The RF front-end bank is highlighted in Figure 3. VII. CONCLUSION Although the mission of ThepisoSat is still ongoing with exceeded expectations, we turn our focus to the development of ZACube-2. The students and staff involved have definitely learned a tremendous amount from the experience and also forged relationships with other institutions, industry partners and organisations that might not have come about otherwise. The successes and experience gained from designing, building and operating ThepisoSat are invalueble and have provided us a solid foundation for building its successor. [1] International Maritime Organisation, Automatic Identification Systems - Last visited 28 May [2] Clyde Space, Standalone 30Wh Battery, no. C3-USM-5016-CS-BAT-30Wh, Issue 1, 28 April [3] Electronic Systems Laboratoy, QB50 ADCS, brosure, 27 January [4] Xilinx, Zync All Programmable SoC, Last visited 28 May [5] Larsen, J., Analysis of Received AIS Data from LEO Cubesat, Aalborg University, Department of Electronic Systems, Denmark, 24 November [6] Triton-1 LEOPS, blog, - Last visited 28 May Pointing requirements For a stable antenna pattern, a nadirpointing platform is required. To aid in separating AIS signals