Average Consensus with Dynamic Quantization for Directed Networks

This paper addresses the challenge of achieving average consensus in distributed systems with limited communication bandwidth, a common constraint in real-world applications. The proposed algorithm, PP-ACDC, offers a communication-efficient solution by using dynamic quantization and a finite-time termination mechanism. This is significant because it allows for precise consensus with a fixed number of bits, making it suitable for resource-constrained environments.

• •Proposes PP-ACDC, a deterministic distributed algorithm for average consensus.
• •Uses dynamic quantization framing (zooming and midpoint shifting) to preserve the global average.
• •Achieves asymptotic (exact) average consensus on strongly connected digraphs.
• •Includes a fully distributed and synchronized finite-time termination mechanism.
• •Suitable for resource-constrained multi-agent systems.