[Paper Review] Near-Optimal Distributed Scheduling Algorithms for Regular Wireless Sensor Networks

K. Shashi Prabh|arXiv (Cornell University)|Nov 13, 2012

Mobile Ad Hoc Networks26 references3 citations

TL;DR

This paper presents near-optimal, distributed, message-passing-free scheduling algorithms for regular multi-hop wireless sensor networks using TDMA with spatial reuse. It achieves constant schedule complexity independent of network size by leveraging a k-hop interference model, with approximation ratios of 4/3 for hexagonal and 5/4 for square-grid networks, and establishes feasibility regions in the SINR model where these bounds hold while satisfying signal-to-interference-plus-noise ratio constraints.

ABSTRACT

Wireless sensor networks are normally characterized by resource challenged nodes. Since communication costs the most in terms of energy in these networks, minimizing this overhead is important. We consider minimum length node scheduling in regular multi-hop wireless sensor networks. We present collision-free decentralized scheduling algorithms based on TDMA with spatial reuse that do not use message passing, this saving communication overhead. We develop the algorithms using graph-based k-hop interference model and show that the schedule complexity in regular networks is independent of the number of nodes and varies quadratically with k which is typically a very small number. We follow it by characterizing feasibility regions in the SINR parameter space where the constant complexity continues to hold while simultaneously satisfying the SINR criteria. Using simulation, we evaluate the efficiency of our solution on random network deployments.

Motivation & Objective

• To design distributed, collision-free scheduling algorithms for regular wireless sensor networks with minimal communication overhead.
• To achieve near-optimal schedule length in multi-hop hexagonal and square-grid networks under a k-hop interference model.
• To characterize feasible SINR parameter regions where constant schedule complexity (independent of n) is maintained while satisfying reception reliability.
• To eliminate message passing in scheduling, making the approach energy-efficient for resource-constrained sensor nodes.
• To provide exact analytical results for regular topologies that inform asymptotic behavior in general networks.

Proposed method

• Uses a graph-based k-hop interference model to define spatial reuse constraints, ensuring no interference within k hops of a receiver.
• Develops distributed scheduling algorithms based on node logical addresses without message passing, reducing communication overhead.
• Applies a 4/3-approximation algorithm for hexagonal networks and a 5/4-approximation for square-grid networks, with optimality proven for odd k in square grids.
• Derives SINR feasibility regions by analyzing signal-to-interference-plus-noise ratio constraints using a polynomial path-loss model with exponent γ ∈ [3,4].
• Employs a distance-based lattice characterization (via MAX{|x|,|y|,|x−y|}) to prove structural properties of regular topologies.
• Validates results through simulations on random deployments, measuring schedule length, interference ratio (ρ), and network scalability.

Experimental results

Research questions

• RQ1 Can distributed scheduling in regular WSNs achieve constant schedule complexity independent of network size n under a k-hop interference model?
• RQ2 What is the approximation ratio of the proposed scheduling algorithm for hexagonal and square-grid networks, and is it optimal for certain k values?
• RQ3 What are the feasible SINR parameter regions where the constant schedule complexity is preserved while satisfying physical-layer reception constraints?
• RQ4 How does the performance of the algorithm scale with network size and path-loss exponent γ?
• RQ5 To what extent can non-uniform power assignment improve feasibility and performance in the SINR model?

Key findings

• The schedule complexity in both hexagonal and square-grid networks is independent of the number of nodes n and scales quadratically with k, the interference radius.
• The proposed algorithm achieves a 4/3-approximation ratio for hexagonal networks and a 5/4-approximation for square-grid networks, with optimality proven for odd k in square grids.
• Feasibility regions in the SINR parameter space are identified where constant schedule complexity (independent of n) is maintained, covering typical operating points.
• Simulations show that both minimum and average interference ratio ρ decrease or remain stable with increasing network size, indicating scalability.
• The use of logical node addresses and absence of message passing significantly reduce communication overhead, making the scheme energy-efficient for sensor networks.
• The results suggest that regular topology analysis provides exact expressions and leading coefficients that inform asymptotic behavior in general networks.