Project Objective

Evaluate and test navXcom algorithms for an autonomous, scalable, lunar ground-based cognitive RF network system, enabling reliable navigation, communication, increased data rates, and resilience in the unpredictable lunar environment.

Project Description

This project develops a cognitive network software for a cost-effective ground-based RF navigation and communication network on the lunar surface through support and evaluation of the navXcom software architecture. Aligned with NASA's LunaNet and SCaN programs and supporting NASA's Artemis goal of landing American astronauts on the Moon, it aims to create an autonomous and scalable system that enables precise Position, Navigation and Time (PNT) services and advanced Communication capabilities while prioritizing innovation, cost-effectiveness, and scientific research. Through AI techniques, advanced routing algorithms, and a data-centric network architecture, it optimizes resource utilization, communication efficiency, and operational simplicity. By leveraging wideband sensing, Optical communications, WiFi, and 3GPP frequencies, the network could achieve high data rates ensuring optimal performance, reliable high-speed communication, and navigation for Lunar users in the unpredictable space environment. The network architecture comprises flexible communication platforms, including a Central Communications Terminal (CCT) as the primary node for communication with Earth and overhead orbiters. The CCT also serves as the lander that carries Mobile NAVCOM Stations (MNCS) –- autonomous rovers equipped with omni-directional communication antennas and transmitting-only navigation antennas –- to the Lunar surface. MNCS autonomously deploy and can discover nodes, adjusting their position for optimal coverage. Once deployed, the MNCS form a mesh topology RF NAVCOM network, providing comprehensive communication and precise real-time navigation through triangulation to Lunar Equipment and Exploration Units (EEUs). The employment of standardized LunaNet frequencies and protocols like IP, DTN, and CCSDS ensure seamless integration with NASA Systems. The end product of this project will be a thoroughly tested, robust, and scalable RF network infrastructure software capable of cognitive communication, enabling high data rates, precise PNT, and advanced COMM capabilities.

Project Results and Conclusions

Over the course of FY2024, the NASA and navXcom worked together to develop a testing approach and scenario to allow for evaluation of the developed algorithms This network utilizes a CCT to coordinate and define the routing across the network to allow for data to be shared between remote users across the operational area.

To support this, the team will be performing field testing of the software at the Lunar Regolith Field on-site at Marshall Space Flight Center. In this demonstration, the navXcom software will be loaded on existing Lunar Harpoon platforms (which host the required cross-link radios and processing capability) to act as distributed surface nodes. One of these harpoons will be mounted to a small rover platform to act as an autonomous user which will navigate itself across the field to maintain data links under multiple operational scenarios. These includes a nominal traverse of an area with all ground nodes in view as well as simulation of scenarios where one or two nodes go out of view and as such, the rover has to navigate back into a position with known communication. Similarly, a mobile user will also be modeled by the connection of a ground node to a portable computer. They will similarly traverse the field and ensure that data can pass across the network even if individual nodes are unavailable. In Figure 1, this use case is presented as if data needs to be routed from User 1 to User 3, but they are not in line of communication; hence, the data must hop across multiple nodes to reach its destination.

The two main software elements developed are EdgeNode and QNN (neural network) codes that perform the local and network level routing capabilities. Similarly, a front-end GUI was developed for the TTC and portable user to allow for insight into the state of the network. We have successfully developed and delivered a comprehensive virtual network simulation as part of this milestone. The simulation includes a Central Communications Terminal (CCT) and five edge nodes, with advanced data routing using the OpenShortest Path First (OSPF) protocol, signal strength monitoring, link quality indicators, and integrated cybersecurity measures.

The network uses IPv6, and key metrics such as the Link Quality Indicator (LQI) and Received Signal Strength Indicator (RSSI) are measured in this simulation through radios modeled after XBee devices. On top of the core network components, we have implemented AI-driven neural networks that leverage reinforcement learning and genetic algorithms, along with quantum computing technologies. These additions are designed to further enhance the network's performance and scalability.

Core Network Components Overview:

Central Communications Terminal (CCT) Components:

• Mobility Management Entity (MME)
• Serving Gateway (SGW)
• Packet Data Network Gateway (PGW)
• Home Subscriber Server (HSS)
• IP Multimedia Subsystem (IMS)
• Baseband Unit (BBU)
• High-Performance RF Transceivers (Modeled as XBee Radios)
• Directional and Omnidirectional Antennas (Modeled as XBee Radios)
• Spectrum Management Tools
• IPv6
• DHCP Server
• OSPF Router with DTN (DTNOSPFRouter)
• PostgreSQL (Database)
• DTN Components: Buffer, DTN Router

Mobile NAVCOM Stations (MNCS) Components:

• MME Client
• SGW Client
• PGW Client
• HSS Client
• IMS Client
• BBU Client
• High-Performance RF Transceiver Client (Modeled as XBee Radios)
• Navigation Transmitter (Modeled as XBee Radios)
• Spectrum Management Tool Clients
• IPv6 Client
• DHCP Client
• OSPF Router with DTN (DTNOSPFRouter)
• MySQL (Database)
• DTN Components: Buffer, DTN Router

The team is utilizing prior development to support integrated rover and harpoon testing conducted previously to provide the baseline capability for this testing. The primary addition is the evaluation of the navXcom software platform for network and navigation control.

As of January 2025, all software required for testing these algorithms is on hand and is being installed on the ground platforms ahead of a test campaign in mid-January 2025. Additionally, software testing has been conducted by navXcom to validate the algorithms being implemented in simulation. The field testing was postponed from November 2024 due to personnel availability. During this test event, the NASA and navXcom teams will work together to finalize hardware and software interfaces and to perform initial evaluation of the technology. Following the test event, further tests will be conducted by MSFC personnel in software and hardware simulation to assess its performance under a range of scenarios, allowing for more complex modeling of lunar-like terrain.

This project develops and evaluates cognitive networking software and algorithms for a ground-based radio network, ensuring precise PNT, high-data-rate communication, autonomous deployment, and scalability. By combining wideband sensing and communication with AI/ML algorithms, it supports LunaNet's ground infrastructure needs and critical capability gaps such as: providing precise PNT for lunar users, implementing AI/ML on SWaP-constrained platforms, and enabling integrated wide-band sensing and narrow-band communication. This network handles complexity, enhances data return, and empowers scientific missions.