Compressed Sensor Caching and Collaborative Sparse Data Recovery with Anchor Alignment

TL;DR

The paper addresses efficient data recovery in cache-enabled wireless sensor networks by exploiting spatial-temporal sparsity via a Kronecker-based CS model. It introduces CoSR-AA, a distributed recovery method that uses anchor-node alignment to couple caches with minimal communication, castable into a consensus ADMM framework. To speed up convergence, it proposes Deep CoSR-AA, a graph neural network that unfolds ADMM iterations and incorporates an embedded autoencoder for compact inter-cache messages, achieving near-centralized performance with far fewer iterations. The approach is validated through simulations showing improved reconstruction accuracy and reduced communication overhead, with strong robustness to varying anchor strategies and correlation lengths. Overall, the work provides a scalable, communication-efficient framework for collaborative data recovery in cache-rich WSNs, with practical implications for IoT sensing systems.

Abstract

This work examines the compressed sensor caching problem in wireless sensor networks and devises efficient distributed sparse data recovery algorithms to enable collaboration among multiple caches. In this problem, each cache is only allowed to access measurements from a small subset of sensors within its vicinity to reduce both cache size and data acquisition overhead. To enable reliable data recovery with limited access to measurements, we propose a distributed sparse data recovery method, called the collaborative sparse recovery by anchor alignment (CoSR-AA) algorithm, where collaboration among caches is enabled by aligning their locally recovered data at a few anchor nodes. The proposed algorithm is based on the consensus alternating direction method of multipliers (ADMM) algorithm but with message exchange that is reduced by considering the proposed anchor alignment strategy. Then, by the deep unfolding of the ADMM iterations, we further propose the Deep CoSR-AA algorithm that can be used to significantly reduce the number of iterations. We obtain a graph neural network architecture where message exchange is done more efficiently by an embedded autoencoder. Simulations are provided to demonstrate the effectiveness of the proposed collaborative recovery algorithms in terms of the improved reconstruction quality and the reduced communication overhead due to anchor alignment.

Figures (10)

• Figure 1: Illustration of the sensor caching scenario.
• Figure 2: Model of the compressed sensor caching and data recovery operations at the multiple caches.
• Figure 3: The neural network architecture at cache $c$ in stage $k$ of the Deep CoSR-AA algorithm. The green circle and the orange circle denote the neuron without and with the activation function of the neural network, respectively.
• Figure 4: Illustration of the messaging between caches over different stages of the Deep CoSR-AA algorithm. The red arrows indicate the flow of variables sent outward from cache c to its neighboring caches in ${\cal D}_c$. The blue arrows indicate the flow of variables coming inward to cache $c$ from all caches ${\cal D}_c$.
• Figure 5: NMSE versus the compression ratio $M/N$ for the case with $C = 4$.

Theorems & Definitions (1)

• proof