Quality of Information Maximization for Wireless Networks via a Fully Separable Quadratic Policy

Abstract

Abstract-An information collection problem in a wireless network with random events is considered. Wireless devices report on each event using one of multiple reporting formats. Each format has a different quality and uses different data lengths. Delivering all data in the highest quality format can overload system resources. The goal is to make intelligent format selection and routing decisions to maximize time-averaged information quality subject to network stability. Lyapunov optimization theory can be used to solve such a problem by repeatedly minimizing the linear terms of a quadratic drift-plus-penalty expression. To reduce delays, this paper proposes a novel extension of this technique that preserves the quadratic nature of the drift minimization while maintaining a fully separable structure. In addition, to avoid high queuing delay, paths are restricted to at most two hops. The resulting algorithm can push average information quality arbitrarily close to optimum, with a trade-off in queue backlog. The algorithm compares favorably to the basic driftplus-penalty scheme in terms of backlog and delay. Furthermore, the technique is generalized to solve linear programs and yields smoother results than the standard drift-plus-penalty scheme.

Achieving utility-delay-reliability tradeoff in stochastic network optimization with finite buffers

“…The inequality (4) implies that the drift-pluspenalty policy achieves cost within O(1/V ) of the optimal cost, which can be made as small as desired by choosing a sufficiently large value of V . The equality (5) implies that average queue backlog grows linearly with V . Applying Little's law gives the [O(1/V ), O(V )] utility-delay tradeoff (see [17] for a standard description of Little's law).…”

“…This general framework has been used to solve several network optimization problems such as network routing [2], throughput maximization [3], dynamic power allocation [4], quality of information maximization [5]. The framework yields low-complexity algorithms which do not require any statistical knowledge of the network states.…”

Energy‐aware quality of information maximisation for wireless sensor networks

“…We adopt the metric QoI defined in [3, 16] to quantify the value of aggregated information at the FC. Hence, the time averaged QoI can be expressed as where u ( x ) is a non‐decreasing, continuously differentiable, and strictly concave function in x with the maximum first derivative being β and u (0) = 0.…”

“…Extensive researches have devoted to improve the quality of information (QoI) to satisfy the application‐oriented requirements. The authors in [2, 3] defined the QoI from information accuracy and timeliness to characterise the quality of initial information content at SNs and to reflect the delay of packet through the network, respectively, and then proposed two adaptive information collection algorithms to maximise the time averaged QoI. In [4], with rate weight assignment and flow routing determined, a data rate allocation framework was developed to improve the QoI, instead of the quantity of collected information.…”

“…By exploiting the strong spatial–temporal correlations among measurements at SNs, Tang and Yuan [6] developed a set of distributed sensing schedules to maximise the QoI under a fixed duty cycle. However, all the approaches presented in [2–6] fail to evaluate how efficiently the energy is consumed and thus insufficiently deal with the energy efficiency (EE) requirements exerted by power scarcity in WSNs.…”