arXiv subjects
Publications and source records attributed to Douglas J. Leith.
A novel fast algorithm for finding quasi identifiers in large datasets is presented. Performance measurements on a broad range of datasets demonstrate substantial reductions in run-time relative to the state of the art and the scalability of the algorithm to realistically-sized datasets up to several million records.
We consider the proportional fair rate allocation in an 802.11 WLAN that supports multi-user MIMO (MU-MIMO) transmission by one or more stations. We characterise, for the first time, the proportional fair allocation of MU-MIMO spatial streams and station transmission opportunities. While a number of features carry over from the case without MU-MIMO, in general neither flows nor stations need to be allocated equal airtime when MU-MIMO is available.
The 802.11e standard enables user configuration of several MAC parameters, making WLANs vulnerable to users that selfishly configure these parameters to gain throughput. In this paper we propose a novel distributed algorithm to thwart such selfish behavior. The key idea of the algorithm is for honest stations to react, upon detecting a selfish station, by using a more aggressive configuration that penalizes this station. We show that the proposed algorithm guarantees global stability while providing good response times. By conducting a game theoretic analysis of the algorithm based on repeated games, we also show its effectiveness against selfish stations. Simulation results confirm that the proposed algorithm optimizes throughput performance while discouraging selfish behavior. We also present an experimental prototype of the proposed algorithm demonstrating that it can be implemented on commodity hardware.
We formulate adaptation of antenna tilt angle as a utility fair optimisation task. This optimisation problem is non-convex, but in this paper we show that under reasonable conditions it can be reformulated as a convex optimisation. Using this insight, we develop a lightweight method for finding the optimal antenna tilt angles, making use of measurements which are already available at base stations, and suited to distributed implementation.
In this paper we characterise the maximal convex subsets of the (non-convex) rate region in 802.11 WLANs. In addition to being of intrinsic interest as a fundamental property of 802.11 WLANs, this characterisation can be exploited to allow the wealth of convex optimisation approaches to be applied to 802.11 WLANs.
Wireless 802.11 links operate in unlicensed spectrum and so must accommodate other unlicensed transmitters which generate pulsed interference. We propose a new approach for detecting the presence of pulsed interference affecting 802.11 links, and for estimating temporal statistics of this interference. This approach builds on recent work on distinguishing collision losses from noise losses in 802.11 links. When the intervals between interference pulses are i.i.d., the approach is not confined to estimating the mean and variance of these intervals but can recover the complete probability distribution. The approach is a transmitter-side technique that provides per-link information and is compatible with standard hardware. We demonstrate the effectiveness of the proposed approach using extensive experimental measurements. In addition to applications to monitoring, management and diagnostics, the fundamental information provided by our approach can potentially be used to adapt the frame durations used in a network so as to increase capacity in the presence of pulsed interference.
We consider multi--hop networks comprising Binary Symmetric Channels ($\mathsf{BSC}$s). The network carries unicast flows for multiple users. The utility of the network is the sum of the utilities of the flows, where the utility of each flow is a concave function of its throughput. Given that the network capacity is shared by the flows, there is a contention for network resources like coding rate (at the physical layer), scheduling time (at the MAC layer), etc., among the flows. We propose a proportional fair transmission scheme that maximises the sum utility of flow throughputs subject to the rate and the scheduling constraints. This is achieved by {\em jointly optimising the packet coding rates of all the flows through the network}.
We define a class of multi--hop erasure networks that approximates a wireless multi--hop network. The network carries unicast flows for multiple users, and each information packet within a flow is required to be decoded at the flow destination within a specified delay deadline. The allocation of coding rates amongst flows/users is constrained by network capacity. We propose a proportional fair transmission scheme that maximises the sum utility of flow throughputs. This is achieved by {\em jointly optimising the packet coding rates and the allocation of bits of coded packets across transmission slots.}
By combining the features of CSMA and TDMA, fully decentralised WLAN MAC schemes have recently been proposed that converge to collision-free schedules. In this paper we describe a MAC with optimal long-run throughput that is almost decentralised. We then design two \changed{schemes} that are practically realisable, decentralised approximations of this optimal scheme and operate with different amounts of sensing information. We achieve this by (1) introducing learning algorithms that can substantially speed up convergence to collision free operation; (2) developing a decentralised schedule length adaptation scheme that provides long-run fair (uniform) access to the medium while maintaining collision-free access for arbitrary numbers of stations.
In this paper we establish the log-convexity of the rate region in 802.11 WLANs. This generalises previous results for Aloha networks and has immediate implications for optimisation based approaches to the analysis and design of 802.11 wireless networks.
In this paper we build upon the recent observation that the 802.11 rate region is log-convex and, for the first time, characterise max-min fair rate allocations for a large class of 802.11 wireless mesh networks. By exploiting features of the 802.11e/n MAC, in particular TXOP packet bursting, we are able to use this characterisation to establish a straightforward, practically implementable approach for achieving max-min throughput fairness. We demonstrate that this approach can be readily extended to encompass time-based fairness in multi-rate 802.11 mesh networks.