Choong Ming Chin

A Dynamic Channel Assignment Strategy via Power Control for Ad-Hoc Network Systems

The increasing demand for wireless network services have resulted a plethora of studies on the efficient management of radio resources to improve network capacity. As radio spectrum is a scarce resource, sharing of radio frequency has to be considered among wireless network nodes and by doing so introduces interference among users, which in turn limit the network capacity. Our study addresses the problem of dynamically assigning channels in ad-hoc wireless networks via power control in order to satisfy their minimum QoS requirements. The objective then is to maximize the number of co-channel links subject to some stability conditions. In order to assign the optimal number of co-channel nodes that can co-exist over a wide range of operating SINR values, we propose two novel ways to find the optimal combination of co-channel links so that a feasible power vector can be found within its power limits.

Neural network demand models and evolutionary optimisers for dynamic pricing

Dynamic pricing is a pricing strategy where price for the product changes according to the expected demand for it. Some work on using neural network for dynamic pricing have been previously reported, such as for forecasting the demand and modelling consumer choices. However, little work has been done in using them for optimising pricing policies. In this paper, we describe how neural networks and evolutionary algorithms can be combined together to optimise pricing policies. Particularly, we build a neural network based demand model and use evolutionary algorithms to optimise policy over build model. There are two key benefits of this approach. Use of neural network makes it flexible enough to model a range of different demand scenarios occurring within different products and services, and the use of evolutionary algorithm makes it versatile enough to solve very complex models. We also evaluate the pricing policies found by neural network based model to that found by other widely used demand models. Our results show that proposed model is more consistent, adapts well in a range of different scenarios, and in general, finds more accurate pricing policy than other three compared models.

An AI-based system for pricing diverse products and services

This paper describes an applied research work that looks at different ways to effectively manage resources. Particularly, it describes how revenue management techniques can be used to balance demand against capacity, and describes a system that uses different OR and AI techniques to intelligently price diverse products and services. This system can produce pricing policies for wide range of products and services regardless of the model of demand used. The system incorporates a model specification layer, which provides flexibility in defining the demand model for different products. It also incorporates an optimisation layer, which takes the specified model as an input and produces the pricing and production guidelines for the product. The system can be either used as a stand alone system or can be incorporated as a generic modelling and optimisation component within a larger revenue management system.

Call Admission Control with Adaptive Active Link Protection for Wireless Systems

In wireless communication systems, a call admission control mechanism is usually deployed to determine whether a new communication node can be admitted into the network system. In wireless systems that implement power control, the maximum achievable signal-to-interference plus noise ratio (SINR) for a new node depends on the link gains amongst all the co-channel interfering nodes involved and the white noise strength. Thus, one of the challenges in call admission control (CAC) in a wireless system with power control is the estimation of maximum achievable SINR when information about global link gains is not available. In this paper, by ignoring the white noise factor we present a predictor for the maximum achievable signal-to-interference ratio (SIR) of a new node trying to get admission into a wireless system. Using the SIR predictor we then calculate an optimal active link protection margin, which together with the SINR or SIR threshold would constitute an enhanced threshold value for the new node to attain. By doing so current active links would be protected from performance degradation should the maximum achievable SIR value common to all the nodes be lower than the SIR threshold. For each nodal topology ranging from low density to high density values, the accuracy of the predictor is evaluated by means of simulation in terms of mean error and root mean square error. Together with finding the corresponding optimal active link protection margin, efficient CAC mechanism to ensure stability of the feasible system can be maintained over a wide range of operating SIR values

Trading strategies in radio spectrum management

Radio resource management (RRM) is one of the most challenging and one of the most important aspects in the provisioning of quality of service (QoS) for wireless communication systems. With the ever increasing size of wireless mobile community and its demand for high-speed multimedia communications, efficient resource management becomes a paramount importance due to limited resources available such as spectrum and power availability. As radio spectrum is a finite resource, new approaches are therefore needed to maximize the existing spectrum to ensure the wireless userspsila and network providerspsila QoS requirements are met. In this paper we propose a novel radio spectrum trading via derivatives contracts as a means to address short-term demands for spectrum whilst ensuring end-to-end QoS is met.

Dynamic Pricing with Neural Network Demand Models and Evolutionary Algorithms

The use of neural networks for demand forecasting has been previously explored in dynamic pricing literatures. However, not much has been done in its use for optimising pricing policies. In this paper, we build a neural network based demand model and show how evolutionary algorithms can be used to optimise the pricing policy based on this model. There are two key benefits of this approach. Use of neural network makes it flexible enough to model range of different demand scenarios occurring within different products and services, and the use of evolutionary algorithm makes it versatile enough to solve very complex models. We also compare the pricing policies found by neural network model to that found by using other widely used demand models. Our results show that proposed model is more consistent, adapts well in a range of different scenarios, and in general, finds more accurate pricing policy than the other three compared models.