Future spacecraft communication subsystems will potentially benefit from software-defined radios controlled by artificial intelligence algorithms. In this paper, we propose a novel radio resource allocation algorithm leveraging multiobjective reinforcement learning and artificial neural network ensembles able to manage available resources and conflicting mission-based goals. The uncertainty in the performance of thousands of possible radio parameter combinations and the dynamic behavior of the radio channel over time producing a continuous multidimensional state-action space requires a fixed-size memory continuous state-action mapping instead of the traditional discrete mapping. In addition, actions need to be decoupled from states in order to allow for online learning, performance monitoring, and resource allocation prediction. The proposed approach leverages the authors' previous research on constraining decisions predicted to have poor performance through ”virtual environment exploration.” The simulation results show the performance for different communication mission profiles, and accuracy benchmarks are provided for the future research reference. The proposed approach constitutes part of the core cognitive engine proof-of-concept delivered to the NASA John H. Glenn Research Center's SCaN Testbed radios on-board the International Space Station.
26 October 2018 in IEEE Transactions on Cognitive Communications and Networking
Cognitive algorithms for communications systems have been presented in literature, but very few have been integrated into a fielded system, especially space communications systems. In this paper, we describe the implementation of a multi-objective reinforcement-learning algorithm using deep artificial neural networks acting as a radio-resource-allocation controller. The developed software core is generic in nature and can be ported readily to another application. The cognitive engine algorithm implementation was characterized through a series of tests using both a ground-based system and a space-based system. The ground system comprised of engineering-model software-defined radios, commercial modems, and RF equipment emulating the targeted space-to-ground channel. The on-orbit communication system, including a space-based, remotely controlled transmitter, resides on the International Space Station and operates with a ground-based receiver at NASA Glenn Research Center. Through a series of on-orbit tests, the cognitive engine was tested in a highly dynamic channel and its performance is discussed and analyzed.
Large-scale natural disasters present complex challenges for disaster relief communications. Fixed infrastructures, such as cell towers or radio base stations, may be completely destroyed during a disaster or this infrastructure may never have existed. In a disaster situation, having unreliable communications systems can put a relief personnel's safety at risk as well as make the effort much less effective. Furthermore, emergency situations require time-sensitive communications that could mean the difference between life and death. The communications system described in this paper provides a rapidly deployable, data-centric mobile communications system for all organizations engaged in disaster relief: first responders, search-and-rescue, emergency medical and health services, etc. Utilizing the IEEE 802.11b/g standard, this system creates a mobile wireless local area network through a series of “wearable routers”. The routers provide local Wi-Fi access to all users within of their respective ranges, and then all of these routers are connected to each other through an ultra-high-frequency backhaul network. Ultimately, from the user's perspective the network appears to be a standard Wi-Fi network with enhanced range. The purpose of this network is to provide communications between both local and widespread users until more traditional communications systems are restored. A proof-of-concept prototype using commercial-off-the-shelf components has been realized, and the real-world performance of the system has been characterized in Boston, MA and Pittsburgh, PA. The results show that this system provides a viable solution, but requires further attention to antenna design and in-band interference