No abstract

Software-defined-radio (SDR) becomes an attractive technique for the development of GNSS receivers due to universal hardware and high flexibility. However, the performance of signal processing can be a challenging task. Real-time mode implementation requires fast floating point calculations in several threads, not available for most part of embedded systems. This paper describes the system-on-chip based device drastically increasing computational performance. A summary of computational complexity of each stage of GNSS receiver is provided and several particular solutions are proposed.

This paper presents the results of theoretical and experimental research of a prototype of a propulsion system for periodical low-Earth orbit correction of research-and-educational nanosatellites. For that purpose, the prototype is to provide at least 20 m/s relative velocity for a 3U CubeSat with a mass not exceeding 4.5 kg. The personnel and environment safety were taken into account during testing and operation along with the ability to be launched as an associated payload by a “Soyuz”-series launch vehicle or from the ISS. An electro-thermal propulsion system (ResistoJet) was designed with “nonfreezing” mixture of ethanol and distilled water used as the working fluid. It is shown that a standard vehicle power system is capable of initiating one corrective thrust impulse per orbit with flight velocity change of about 0.1 m/s by introducing pulse energy storage units and allocating sufficient time for their charging in the flight profile. The propulsion system prototype was tested in atmospheric conditions. For that purpose shortened “atmospheric” nozzles were used. Testing was carried out using a zero-torque test-bench with high-speed cameras. The measured thrust value was in agreement with the calculated value of 44 mN.

Software-defined radio (SDR) - based GNSS receivers are flexible. SDR technology allows changing processing algorithms and adding new signals without any hardware modifications. The main drawback of such a receiver is high computation load. Signal tracking module takes most part of CPU time. This paper considers distributed receiver architectures where most of the processing is performed in a remote PC or a server. The amount of raw data in the SDR receiver is too large for complete transfer. Two methods of reduction are discussed: truncating epochs and disabling tracking for a certain period. The limitations of each method are estimated using simulations. Epochs can be truncated by 50-75% depending on signal-to-noise ratio. The maximum disabling time is about 0.5s for a stationary receiver and 0.1s for a car receiver. Using both methods in combination reduce data volume by 95.4% for a stationary receiver and 83.3% for a car receiver.

No abstract