Multi-Objective Reinforcement Learning for Cognitive Radio-Based Satellite Communications

Ferreira, Paulo Victor R.; Paffenroth, Randy; Wyglinski, Alexander M.; Hackett, Timothy M.; Bilen, Sven G.; Reinhart, Richard C.; Mortensen, Dale J.

NASA NTRS · 20170001407

Multi-Objective Reinforcement Learning for Cognitive Radio-Based Satellite Communications

Ferreira, Paulo Victor R.·Paffenroth, Randy·Wyglinski, Alexander M.·Hackett, Timothy M.·Bilen, Sven G.·Reinhart, Richard C.·Mortensen, Dale J.

View original source ↗Export metadata ↓

Abstract

Previous research on cognitive radios has addressed the performance of various machine-learning and optimization techniques for decision making of terrestrial link properties. In this paper, we present our recent investigations with respect to reinforcement learning that potentially can be employed by future cognitive radios installed onboard satellite communications systems specifically tasked with radio resource management. This work analyzes the performance of learning, reasoning, and decision making while considering multiple objectives for time-varying communications channels, as well as different cross-layer requirements. Based on the urgent demand for increased bandwidth, which is being addressed by the next generation of high-throughput satellites, the performance of cognitive radio is assessed considering links between a geostationary satellite and a fixed ground station operating at Ka-band (26 GHz). Simulation results show multiple objective performance improvements of more than 3.5 times for clear sky conditions and 6.8 times for rain conditions.

Explore related subjects

software defined radio cognitive radio ionosphere

Keep this discovery

Explore connections, maps & timelines

BibTeX RIS

Ferreira, Paulo Victor R., Paffenroth, Randy, Wyglinski, Alexander M., Hackett, Timothy M., Bilen, Sven G., Reinhart, Richard C., Mortensen, Dale J.. 2016-10-18. Multi-Objective Reinforcement Learning for Cognitive Radio-Based Satellite Communications. https://ntrs.nasa.gov/citations/20170001407

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related reports

NASA NTRS2019

Characterization of a Photon Counting Test Bed for Space to Ground Optical Pulse Position Modulation Communications Links

The National Aeronautics and Space Administration (NASA) Glenn Research Center (GRC) has developed a laboratory transmitter and receiver prototype of a space-to-ground optical communications link. The system is meant to emulate future deep space optical communication links, such as the first crewed flight of Orion, in which the transmitted laser is modulated using pulse position modulation and the receiver is capable of detecting single photons. The transmitter prototype consists of a software defined radio, a high extinction ratio electro-optic modulator system, and a 1550 nm laser. The receiver is a scalable concept and utilizes a single-pixel array of fiber coupled superconducting nanowire single photon detectors. The transmit and receive waveforms follow the Consultative Committee for Space Data Systems (CCSDS) Optical Communications Coding and Synchronization Standard. A software model of the optical transmitter and receiver has also been implemented to predict performance of the optical test bed. This paper describes the transmitter and receiver prototypes as well as the system test configuration. System level tests results are presented and shown to align with predictions from software simulations. The validated software model can be used to in the future to reduce the design cycle of optical communications systems.

software defined radio↗

NASA NTRS2016

Link Adaptation for Mitigating Earth-To-Space Propagation Effects on the NASA SCaN Testbed

In Earth-to-Space communications, well-known propagation effects such as path loss and atmospheric loss can lead to fluctuations in the strength of the communications link between a satellite and its ground station. Additionally, the typically unconsidered effect of shadowing due to the geometry of the satellite and its solar panels can also lead to link degradation. As a result of these anticipated channel impairments, NASA's communication links have been traditionally designed to handle the worst-case impact of these effects through high link margins and static, lower rate, modulation formats. The work presented in this paper aims to relax these constraints by providing an improved trade-off between data rate and link margin through utilizing link adaptation. More specifically, this work provides a simulation study on the propagation effects impacting NASA's SCaN Testbed flight software-defined radio (SDR) as well as proposes a link adaptation algorithm that varies the modulation format of a communications link as its signal-to-noise ratio fluctuates. Ultimately, the models developed in this work will be utilized to conduct real-time flight experiments on-board the NASA SCaN Testbed.

software defined radio↗

NASA NTRS2015

Earth-Facing Antenna Characterization in a Complex Ground Plane/Multipath Rich Environment

The Space Communications and Navigation (SCAN) Testbed was a Software Defined Radio (SDR)-based payload launched to the International Space Station (ISS) in July of 2012. The purpose of the SCAN Testbed payload was to investigate the applicability of SDRs to NASA space missions in an operational space environment, which means that a proper model for system performance in said operational space environment is a necessary condition. The SCAN Testbed has line-of-sight connections to various ground stations with its S-Band Earth-facing Near-Earth Network Low Gain Antenna (NEN-LGA). Any previous efforts to characterize the NEN-LGA proved difficult, therefore, the NASA Glenn Research Center built its own S-Band ground station, which became operational in 2015, and has been successfully used to characterize the NEN-LGAs in-situ pattern measurements. This methodology allows for a more realistic characterization of the antenna performance, where the pattern oscillation induced by the complex ISS ground plane, as well as shadowing effects due to ISS structural blockage are included into the final performance model. This paper describes the challenges of characterizing an antenna pattern in this environment. It will also discuss the data processing, present the final antenna pattern measurements and derived model, as well as discuss various lessons learned.

software defined radio↗