Wireless Picosecond Time Synchronization for Distributed Antenna Arrays with Dynamic Connectivity

Naim Shandi,Jason M. Merlo,Jeffrey A. Nanzer

2024-10-23

Abstract:Phase, time, and frequency coordination are crucial for the coherent operation of distributed antenna arrays. This paper demonstrates a high accuracy decentralized time synchronization method for arrays with dynamic connectivity. To overcome challenges such as alteration in network topology, system scaling, and single- or multiple-node failures, we present a decentralized time synchronization method implementing the average consensus algorithm and two-way time transfer. Our approach supports decentralized time synchronization with picosecond accuracy ($< 13$ ps), unbiased convergence, resilience to node failures, and robustness to other network connection adversities. A time synchronization precision of under 3 ps was achieved for a fully connected frequency syntonized network. The algorithm ensured convergence in time even for a network with only one link connecting two randomly chosen nodes at each iteration; however, the speed of convergence decreased along with the number of connected links. The method is exhibited for a six-node distributed array, by both simulation and experimental implementation using software defined radios.

Signal Processing

What problem does this paper attempt to address?

This paper attempts to solve the high - precision time synchronization problem of distributed antenna arrays in dynamic network topologies. Specifically, the paper focuses on how to achieve sub - nanosecond - level (especially picosecond - level) time synchronization accuracy in the case of dynamically changing node connections. The following are the key issues and challenges proposed in the paper:

1. **The Importance of Phase, Time and Frequency Coordination**:
- For the coherent operation of distributed antenna arrays, the coordination of phase, time and frequency is crucial. Especially in applications such as high - gain beamforming, very precise time alignment is required.

2. **Challenges of Dynamic Network Topologies**:
- Changes in network topologies (such as the increase or decrease of node connections), system expansion, and single or multiple node failures will all affect the accuracy and stability of time synchronization.

3. **Limitations of Existing Methods**:
- Traditional centralized time synchronization methods rely on reference nodes and are vulnerable to single - point failures. Therefore, researchers have turned to decentralized methods to improve the robustness and reliability of the system.

4. **The Need to Achieve Picosecond - level Time Synchronization**:
- In applications of high - bandwidth signals, relative timing must be aligned to a small fraction (usually tens of picoseconds) of the signal information duration. This requires time synchronization methods to have extremely high accuracy.

To solve these problems, the paper proposes a decentralized time synchronization method based on the average consensus algorithm and two - way time transfer technology. This method can achieve picosecond - level time synchronization accuracy in dynamically connected distributed antenna arrays, and has unbiased convergence, resistance to node failures and robustness to other adverse factors of network connections.

### Main Contributions

- **High - precision Time Synchronization**: Achieved a time synchronization error of less than 13 picoseconds, with a standard deviation (accuracy) of less than 3 picoseconds.
- **Dynamic Network Adaptability**: Can maintain time synchronization even when network connections are constantly changing.
- **Decentralized Architecture**: Does not need to rely on a single reference node, improving the reliability and robustness of the system.

Through simulation and experimental verification, this method demonstrates its effectiveness and stability under different network connection conditions, providing a reliable time synchronization solution for future distributed communication and sensing systems.