Life-Add: Lifetime Adjustable design for WiFi networks with heterogeneous energy supplies

Shengbo Chen, Tarun Bansal, Yin Sun, Prasun Sinha, Ness B. Shroff

Abstract

Standard IEEE 802.11 WiFi protocols suffer from significant energy inefficiency due to "idle listening," where devices consume power while waiting for a clear channel. This paper introduces Life-Add (Lifetime Adjustable Design), a decentralized MAC protocol that replaces traditional backoff timers with a sleep-wake contention scheme. By modeling contention using exponentially distributed sleep intervals, Life-Add optimizes network utility while guaranteeing target operational lifetimes for devices with heterogeneous energy sources. Simulation results demonstrate that Life-Add significantly extends device lifetime and improves throughput compared to standard 802.11.

Introduction

Mobile and IoT devices often operate under strict energy constraints, yet standard WiFi protocols are not optimized for diverse energy profiles. The primary source of energy waste is idle listening: a device wanting to transmit must keep its radio active to sense the channel and count down its backoff timer. This state consumes nearly as much power as active reception. While power-saving modes exist, they often fail in high-traffic environments.

Furthermore, modern networks include devices with varying power sources—from wall-plugged units to those relying on small batteries or trickle-charge solar cells. Life-Add addresses these challenges by providing a framework that maximizes network utility (throughput and fairness) while respecting individual device lifetime requirements.

The Life-Add Protocol

The core innovation of Life-Add is the transition from a continuous-sensing backoff mechanism to a discrete Sleep-Wake Contention scheme.

Figure 1: Timing diagram of the Life-Add protocol illustrating the sensing, packet transmission, and random sleep intervals used to avoid idle listening.

When a device senses that the channel is busy, it immediately powers down its radio and enters a sleep state for a duration determined by an exponential distribution. Upon waking, the device re-senses the channel. This approach effectively eliminates the energy consumed during traditional backoff periods. To handle complex environments, Life-Add incorporates multi-AP heuristics, such as device collaboration to mitigate the near-far effect and congestion control factors that adjust sleep times upon timeouts to manage hidden terminals.

Mathematical Optimization and Control

The choice of exponentially distributed sleep times is central to the protocol's mathematical tractability. The "memoryless" property of the exponential distribution allows the authors to model asynchronous WiFi contention as a set of coupled probabilities. This leads to a non-convex proportional-fair utility maximization problem, constrained by each device's specific energy replenishment rate and target lifetime.

The Access Point (AP) computes optimal sleep parameters based on network conditions and broadcasts them to all associated devices. This centralized calculation combined with decentralized execution ensures the system remains scalable and responsive to changing network loads.

Performance Evaluation

The protocol was evaluated using the ns-3 network simulator under UDP saturation conditions, comparing Life-Add against standard IEEE 802.11b.

Figure 3: Throughput vs. actual lifetime comparison. Life-Add (lines) provides a significantly better trade-off between network performance and device longevity compared to standard 802.11 (points).

Key findings include:

  • Lifetime Extension: Life-Add increased average device lifetime from approximately 148 minutes to over 252 minutes in single-AP scenarios.
  • Throughput Gains: Despite the increased sleep time, average throughput rose from 0.95 Mbps to 1.41 Mbps. This improvement stems from a significant reduction in packet collision rates.
  • Fairness: Jain’s fairness index improved from 0.427 to 0.691, indicating a more equitable distribution of network resources.
  • Backward Compatibility: In mixed networks, the presence of Life-Add devices reduced overall collisions, benefiting both upgraded and legacy hardware.

Conclusion and Future Work

Life-Add demonstrates that energy efficiency in WiFi networks does not have to come at the cost of performance. By replacing idle listening with optimized sleep intervals, the protocol achieves superior throughput and fairness while meeting strict lifetime guarantees. Future research will focus on the interaction between Life-Add and TCP congestion control, optimization for downlink traffic, and validation through physical hardware implementations.

Summary updated

BibTeX
@inproceedings{chen2013life,
  author = {Shengbo Chen and Tarun Bansal and Yin Sun and Prasun Sinha and Ness B. Shroff},
  title = {{Life-Add: Lifetime Adjustable design for WiFi networks with heterogeneous energy supplies}},
  booktitle = {WiOpt},
  year = {2013},
  url = {https://openreview.net/forum?id=f_dUc_EmyfmQ},
}
Figure