Myopic Coding in Multiterminal Networks

This correspondence investigates the interplay between cooperation and achievable rates in multiterminal networks. Cooperation refers to the process of nodes working together to relay data toward the destination. There is an inherent tradeoff between achievable information transmission rates and the level of cooperation, which is determined by how many nodes are involved and how the nodes encode/decode the data. We illustrate this tradeoff by studying information-theoretic decode-forward-based coding strategies for data transmission in multiterminal networks. Decode-forward strategies are usually discussed in the context of omniscient coding, in which all nodes in the network fully cooperate with each other, both in encoding and decoding. In this correspondence, we investigate myopic coding, in which each node cooperates with only a few neighboring nodes. We show that achievable rates of myopic decode-forward can be as large as that of omniscient decode-forward in the low signal-to-noise ratio (SNR) regime. We also show that when each node has only a few cooperating neighbors, adding one node into the cooperation increases the transmission rate significantly. Furthermore, we show that myopic decode-forward can achieve nonzero rates as the network size grows without bound.