[Paper Review] Stability Analysis of Slotted Aloha with Opportunistic RF Energy Harvesting

Abdelrahman M. Ibrahim, Özgür Erçetin+1 more|Sabanci University|Jan 27, 2015

Energy Harvesting in Wireless Networks31 references4 citations

TL;DR

This paper analyzes the stability of a slotted Aloha network with two node types: Type I (unlimited energy) and Type II (RF energy-harvesting only). It proposes an equivalent system model to derive inner bounds on the stable throughput region under half-duplex and full-duplex energy harvesting, showing that the inner bound accurately predicts system stability in simulations, especially for the full-duplex case with finite battery capacity.

ABSTRACT

Energy harvesting (EH) is a promising technology for realizing energy efficient wireless networks. In this paper, we utilize the ambient RF energy, particularly interference from neighboring transmissions, to replenish the batteries of the EH enabled nodes. However, RF energy harvesting imposes new challenges into the analysis of wireless networks. Our objective in this work is to investigate the performance of a slotted Aloha random access wireless network consisting of two types of nodes, namely Type I which has unlimited energy supply and Type II which is solely powered by an RF energy harvesting circuit. The transmissions of a Type I node are recycled by a Type II node to replenish its battery. We characterize an inner bound on the stable throughput region under half-duplex and full-duplex energy harvesting paradigms as well as for the finite capacity battery case. We present numerical results that validate our analytical results, and demonstrate their utility for the analysis of the exact system.

Motivation & Objective

• To analyze the stability of a slotted Aloha network where one node harvests RF energy from interference of another node.
• To address the challenge of modeling interacting queues in energy-harvesting random access networks with non-ideal energy harvesting and finite battery constraints.
• To develop an equivalent system model that enables analytical characterization of the stability region despite the complexity of the exact system.
• To validate the analytical inner bounds through extensive simulations of both the equivalent and exact system models.
• To provide design insights for medium access protocols in RF energy-harvesting networks with heterogeneous energy availability.

Proposed method

• Proposes an equivalent system model where the energy-harvesting node (Type II) is decoupled from its battery dynamics, enabling analytical tractability.
• Applies stochastic dominance techniques to derive an inner bound on the stability region for both half-duplex and full-duplex energy harvesting modes.
• Models the system as two interacting queues: one for the Type I node (unlimited energy) and one for the Type II node (energy-constrained), with transmission success probabilities dependent on interference.
• Derives stability conditions based on the service rates of the queues, accounting for the energy harvesting process and battery capacity constraints.
• Introduces a generalized system model (S_G) that captures the full-duplex operation and finite battery effects, enabling derivation of an inner bound on the stability region.
• Validates analytical results via simulation of both the equivalent system (S_G) and the exact system (S_O), comparing queue lengths and service rates.

Experimental results

Research questions

• RQ1 How does opportunistic RF energy harvesting from interference affect the stability of a slotted Aloha network with energy-constrained nodes?
• RQ2 What is the achievable stable throughput region in a half-duplex versus full-duplex energy harvesting setup with finite battery capacity?
• RQ3 Can an equivalent system model accurately predict the stability behavior of the exact system with interacting queues and energy harvesting dynamics?
• RQ4 How do the stability conditions derived for the equivalent system relate to the actual stability of the exact system?
• RQ5 What is the impact of energy harvesting efficiency and transmission power on the system’s stable operation region?

Key findings

• The inner bound on the stability region derived for the equivalent system (S_G) accurately captures the stable behavior of the exact system (S_O), as confirmed by simulation of average queue lengths.
• For the full-duplex case, the stability condition on the Type II node’s queue is sufficient but not necessary, indicating the inner bound is conservative but reliable.
• The stability condition on the Type I node’s queue is both necessary and sufficient in all system variants (S_D, S_G, S_O), indicating it is a tight constraint.
• Simulations show that the sum of average queue lengths remains bounded within the inner bound region, while growing unboundedly outside it, confirming the region’s validity.
• The maximum service rate of the Type II queue is achieved on the boundary of the stability region in the half-duplex case, but can be exceeded in the full-duplex case, supporting the inner bound’s conservativeness.
• Sample path simulations confirm that unstable queue behavior (unbounded growth) occurs only outside the stability region, validating the analytical thresholds.