[Paper Review] Age of Information in a Multiple Access Channel with Heterogeneous Traffic and an Energy Harvesting Node

Zheng Chen, Νικόλαος Παππάς+2 more|arXiv (Cornell University)|Mar 12, 2019

Age of Information Optimization6 references4 citations

TL;DR

This paper analyzes age of information (AoI) and delay in a multiple access channel with two heterogeneous nodes: a grid-connected source with bursty data traffic and an energy-harvesting (EH) sensor sending status updates. Using a slotted ALOHA-like random access model, it derives closed-form expressions for the delay of the grid-connected node and the AoI of the EH node, revealing that mutual interference and energy constraints create non-trivial tradeoffs in performance, especially when the EH node's transmission probability is high relative to its energy availability.

ABSTRACT

Age of Information (AoI) is a newly appeared concept and metric to characterize the freshness of data. In this work, we study the delay and AoI in a multiple access channel (MAC) with two source nodes transmitting different types of data to a common destination. The first node is grid-connected and its data packets arrive in a bursty manner, and at each time slot it transmits one packet with some probability. Another energy harvesting (EH) sensor node generates a new status update with a certain probability whenever it is charged. We derive the delay of the grid-connected node and the AoI of the EH sensor as functions of different parameters in the system. The results show that the mutual interference has a non-trivial impact on the delay and age performance of the two nodes.

Motivation & Objective

• To understand the performance tradeoffs in a multiple access channel with heterogeneous traffic: one grid-connected node with bursty data and one energy-harvesting (EH) sensor sending status updates.
• To model and quantify the impact of mutual interference and energy availability on the delay of the throughput-oriented grid-connected node and the AoI of the EH sensor.
• To derive closed-form expressions for delay and AoI as functions of transmission probabilities, data arrival rates, and energy arrival rates.
• To reveal the non-trivial interplay between interference, energy constraints, and performance metrics in interference-limited, heterogeneous networks with EH nodes.

Proposed method

• Models a time-slotted multiple access channel (MAC) with two source nodes: S1 (grid-connected, bursty data) and S2 (EH sensor, status updates).
• Assumes S1 uses slotted ALOHA with transmission probability q1 and data arrival rate λ; S2 transmits status updates with probability q2 when battery is non-empty.
• Models energy harvesting at S2 as a Bernoulli process with rate δ, assuming infinite battery capacity and one energy unit per transmission.
• Uses a Markov chain model to analyze the queueing behavior of S1 and the energy state of S2, deriving steady-state probabilities for performance evaluation.
• Applies the early departure, late arrival model to S1’s queue to compute delay, and uses the age evolution model to compute average AoI for S2.
• Derives closed-form expressions for the average delay of S1 and the average AoI of S2 in terms of system parameters (q1, q2, λ, δ, θ).

Experimental results

Research questions

• RQ1 How does the transmission probability of the EH sensor (q2) affect its own age of information (AoI), especially when energy is scarce?
• RQ2 What is the impact of the grid-connected node’s data arrival rate (λ) on the AoI of the EH sensor due to mutual interference?
• RQ3 How does the energy arrival rate (δ) at the EH node influence its AoI, and when does increasing q2 become counterproductive?
• RQ4 What performance tradeoffs emerge between AoI and delay in a MAC with heterogeneous traffic and mixed power sources?
• RQ5 How does the SINR threshold (θ) affect the AoI of the EH node, and under what conditions does AoI increase with higher thresholds?

Key findings

• The delay of the grid-connected node S1 increases with its data arrival rate λ and the transmission probability q2 of the EH node, and becomes infinite when λ exceeds the system’s stable throughput.
• The AoI of the EH sensor S2 first decreases and then increases with q2 when energy is scarce (δ = 0.3), due to increased collisions and wasted energy from excessive transmission attempts.
• When the EH node has sufficient energy, the AoI of S2 monotonically decreases with q2, indicating that interference is the dominant limiting factor only under low energy availability.
• The AoI of S2 decreases with increasing energy arrival rate δ, as higher δ increases the probability of successful transmission.
• The AoI of S2 increases with the SINR threshold θ, since higher thresholds reduce the successful transmission probability, leading to longer age cycles.
• When λ is low, the AoI of S2 increases with λ due to higher interference from S1; when λ is high (saturated queue), the AoI of S2 saturates, as interference becomes constant.