Low-Complexity AoI-Optimal Status Update Control with Partial Battery State Information in Energy Harvesting IoT Networks

TL;DR

This paper tackles AoI optimization in energy-harvesting IoT networks with partial battery information by introducing inferred pBSI and low-complexity policies. The main approach combines a CN policy for single-sensor pBSI problems and a WUGC extension to multi-sensor settings, supported by an online-offline hybrid estimation of the post-update value function and a block-based VI solver to keep complexity tractable. Key theoretical results include a universal lower bound on on-demand AoI and a near-optimality guarantee for CN across various energy arrival processes, evidenced by numerical experiments in both single- and multi-sensor scenarios. The findings demonstrate that pBSI-based strategies can substantially improve status freshness with scalable computation, enabling practical deployments in large EH-IoT networks and suggesting avenues for integration with reinforcement learning in more complex environments.

Abstract

For a two-hop IoT system consisting of multiple energy harvesting sensors, a cache-enabled edge node, and multiple monitors, the status update control at the edge node, which has partial battery state information (pBSI) of the sensors, is formulated as a pBSI problem. The concept of inferred pBSI is introduced to reduce the noiseless single-sensor pBSI problem to a Markov decision process with a moderate state-space size, enabling the optimal policy to be obtained through a value iteration algorithm. A lower bound on the expected time-average on-demand age of information performance is established for the general single-sensor status update problem. For the single-sensor pBSI problem, a semi-closed-form policy called the current-next (CN) policy is proposed, along with an efficient post-update value iteration algorithm with a per-iteration time complexity proportional to the square of the battery capacity. A weighted-update-gain-competition (WUGC) approach is further leveraged to extend the CN policy to the multi-sensor case. Numerical results in the single-sensor case demonstrate the near-optimal performance of the CN policy across various energy arrival processes. Simulations for an IoT system with $100$ sensors reveal that the WUGC-CN policy outperforms the maximum-age-first policy and the random-scheduling-based CN policy under Bernoulli energy arrival processes.

Figures (7)

• Figure 1: A two-hop IoT network consisting of multiple sensors, one edge node, and multiple monitors.
• Figure 2: An example of pBSI of sensor $k$ at time $t$, where the circled U represents the successful status update, a circled F represents a failed status update, and an empty circle represents no update commanded.
• Figure 3: The relation between the inferred pBSI and the original pBSI when the most recent status update failed.
• Figure 4: The structural properties of the CN policy for Bernoulli energy arrivals.
• Figure 5: The addtivie and multiplicative gaps of the CN, NO, OFT, and eBSI-Opt policies for $\eta=0.7$, $\xi=1,0.7,0.4$, and Bernoulli energy arrivals with $\lambda \in \{ 0.1,0.12,\dots, 0.3\}$.

Theorems & Definitions (18)

• Proposition 1
• Corollary 1
• Theorem 1
• Remark 1
• Proposition 2
• Remark 2
• Definition 1
• Theorem 2
• Proposition 3: The $i$-th-Request-Success Estimation
• Proposition 4: The $x$-th-Step-Success Estimation