Beyond GPS: HORDE PNT for contested environments

Story

December 04, 2025

Evan Alexander

Microchip Technology

Military technology development is at an inflection point, where the surging deployment of uncrewed aerial vehicle (UAV) technology collides with the total denial of the GPS signal commonly used for position, navigation, and timing (PNT) in such platforms. These trends necessitate a transformation in the underlying PNT network architecture in military platforms. Harmonized Order of Registered and Distributed Elements (HORDE) PNT (a software tool for distributed computing) rejects the commonly deployed hierarchical PNT network architecture, opting for a decentralized approach under which PNT assets can be dispersed across a network, yet leveraged by any platform. Unlike current architectures, HORDE PNT is fundamentally designed for operation in contested environments, enabling challenging missions like coordinating a decentralized UAV swarm for coherent sensing.

In the modern theater of war, the evolution of military operations has been marked by the development of advanced technologies that require precise position, navigation, and timing (PNT). Platforms like uncrewed aerial vehicles (UAVs) enable users to leverage unprecedented capabilities in intelligence, surveillance, and reconnaissance (ISR), necessitating heightened levels of precision for operational coordination. In swarm operation, each platform uses GPS by continuously receiving position and time data to update relative position within the swarm. Platforms using military GPS receivers are capable of meter-level position accuracy and timing accuracy within 10 nanoseconds (1E-9) of Coordinated Universal Time (UTC). The dependency on GPS for synchronization within and between platforms has proven to be a double-edged sword, as the concerns of GPS vulnerabilities – particularly in conflict zones – including jamming, can disrupt operations for hours or days, rendering traditional PNT systems ineffective.

GPS and the limits of hierarchical PNT architectures

PNT systems have an inherent requirement for reference frame information in order to be useful. The very nature of PNT data is that it describes the relative position, motion, and time between objects. Unless a system can determine what the PNT data is relative to, PNT data is therefore meaningless. The appearance of PNT objects operating in a globally coordinated reference frame results from a PNT object aligning its local reference frame to a coordinated reference frame, such as GPS. Many PNT systems today operate in a hierarchical architecture in which GPS is at the top of the hierarchy, and downstream or adjacent systems rely on the PNT information and reference frames disseminated from GPS (UTC and WGS-84). This leader-follower architecture works well when GPS is available, but ultimately limits a platform’s ability to operate when the global reference frame is not continuously available, for example, during periods of GPS denial. (Figure 1.)

[Figure 1 ǀ A figure contrasts the hierarchical GPS model (left) with the HORDE [Harmonized Order of Registered and Distributed Elements] decentralized approach (right).]

In Figure 1, the left graphic is the default method of converting PNT data to a coordinated reference frame, which simplifies the process of exchanging PNT data between systems, but levies a requirement for global coherency between PNT objects. This setup requires all PNT objects to align to the same coordinated reference frame before collaborating with one another. The right graphic is an alternative approach to enforcing global coherency. HORDE PNT is a method of sharing PNT data without enforcing the requirement to subscribe to a global reference frame like WGS-84 or UTC. GPS is extremely useful when it is available, but systems should be designed to operate whether it is present or not.

Theory of decentralized HORDE PNT

The HORDE protocol was developed to leverage harmonization as a method of synchronization and provides a mechanism for grouping systems into a decentralized horde. The harmonization process enables synchronization through the measurement of relative time and frequency offsets. All HORDE elements act as peers, with no organizational structure or hierarchy enforced for synchronization or PNT distribution. This lack of organization makes HORDE inherently resilient to disruption while still supporting the traditional method of synchronization using the distribution of coherent references as GPS does.

In this architecture, each individual network node or platform establishes and uses its own local reference frame (position, orientation, and time) independent of GPS or any other authoritative source. Nodes in a network share PNT data across the network but are only locally aware based on measurements with their neighbors. Data must be translated from a neighbor’s reference frame into the local reference frame in order to be used.

The main hurdle in implementing this model involves establishing a local reference frame for each node and then translating PNT data across the varying reference frames as data is exchanged between nodes. Methods for instantiating a local PNT reference frame using only two-way ranging measurements between platforms and the ability to translate timestamps between asynchronous platforms must be developed to overcome this hurdle. The ability to do so enables a swarm of UAVs to cold-start without GPS, enhancing its capability as each node in the swarm discovers its neighbors and begins to share PNT data across the network and execute coordinated missions.

Translating PNT data between asynchronous platforms

Asynchronous two-way time transfer (ATWTT) enables a system to translate data between platforms operating asynchronously, meaning their clocks are not aligned to a common reference frame and do not necessarily transfer data at the same time or rate. ATWTT works by measuring relative time offsets between local clocks and translating the time stamps into the local reference frame. PNT data then can be shared across a network and utilized by any node on the network. The HORDE protocol translation method optimally pairs PNT data and a relative time-offset measurement to maximize simultaneity. PNT data time stamp translation uses linear interpolation of the relative time-offset measurements, which are selected to maximize simultaneity. (Figure 2.)

[Figure 2 ǀ A diagram displays the measurements necessary to calculate clock offset and simultaneity. HORDE PNT also deploys methods for translating relative position and orientation across nodes, but these methodologies are not described in this figure.]

Requirements for swarm deployment

The requirements to effectively share PNT data across a network for a swarm deployment include the accurate measurement and time stamping of the data, as well as near instantaneous (within milliseconds) sharing of that PNT data over an RF data link. PNT and reference frame data must be time stamped with sub-100-picosecond accuracy and the data link between platforms needs to share that data in near-real-time. HORDE PNT scales to dozens of nodes but requires line-of-sight for optimal ranging accuracy in complex environments. With this architectural framework established, each node in the swarm can begin an operation, even in a completely GNSS-denied environment.

The steps to operation include:

Establish a local reference frame.
Discover the other nodes within the network.
Map the topology and relative positions of other nodes in the network.
Share PNT data across the network.
Cooperate on coordinate missions.

The ability to cold-start a swarm in a completely GNSS-denied environment is not suitable with the hierarchical architecture that is deployed in many military platforms today. HORDE PNT is the architectural framework required in the new age of combat where GPS is commonly unavailable and drones roam the skies.

Evan Alexander is a Product Manager for Microchip Technology, Government Systems Group. Evan’s team develops advanced PNT systems in support of the U.S. Department of Defense (DoD). He holds an M.S. in engineering and technology management and a B.S. in mechanical engineering from Colorado School of Mines.

Microchip Technology https://www.microchip.com/