Tutorial 1, Friday 9:00 –11:10
Title: Cross-layer protocol optimization for wireless network systems (by Prof. Swades De) [Slides]

Abstract: Wireless network systems are constrained by limited processing, band- width, and energy resources. Widespread adoption to untethered network connectivity implies that the users look for seamless connectivity and quality-of-service/quality-of-experience guarantee on their low-cost wireless/mobile devices. This trend has called for cross-layer adaptive protocol solutions and network system level optimization research. From system design and performance optimization viewpoints it is important that the protocol performances are quantifiable as functions of various device and network system dependent stochastic parameters. In this presentation, through a few example cases, we will discuss how cross-layer interactions are analytically modelled and optimized. Specif- ically, we will look into how the constraints of specific physical communication mediums, device constraints, along with the user interests in some cases are accounted in more ef- ficient design strategies at the medium access control, error/flow control, packet routing, as well as source coding stages. The network systems of interest in this discussion include low power ad hoc sensor networks as well as broadband wireless networks.

Tutorial 2, Friday 14:00 –16:10
Title: An introduction to point processes (by Prof. D. Yogeshwaran) [Notes]

Abstract: Point processes are an important tool in modeling and study of geometric networks. In this talk, I shall introduce some of the basics of point processes. The aim shall be to familiarize the audience with many examples of point processes as well as the refined Campbell theorem–the key computational tool. I shall also illustrate its applications. I shall also discuss in detail the two popular point processes–Poisson and Binomial point processes.

Tutorial 3, Saturday 9:00 –12:30
Title: Percolation, Connectivity and Coverage in Random Networks (by Prof. Srikanth Iyer) [Notes]

Abstract: We will study three problems of fundamental interest in random geometric graphs, namely percolation, connectivity and coverage. A random geometric graph consists of n vertices distributed in space with an edge between any two points provided they are located within distance r(n) of each other. The above three problems can be stated alternatively as follows. As n becomes large, how should r(n) scale so that the graph contains a giant connected component, is connected with high probability, or the union of balls of radius r(n) around each vertex covers a pre-specified fraction of the entire space.

Tutorial 4, Saturday 14:00 –16:10
Title: Determinantal point processes (by Prof. Manjunath Krishnapur) [Slides]

Abstract: The most well-known point process is the Poisson process. It is widely used in many models (in mathematics and in wireless communication) because of the ease of analysis although in many contexts the extent of independence present in a Poisson process is unrealistic. Determinantal point processes are a class of point processes that lack independence (in fact the points tend to ’repel’ each other) but nevertheless are amenable to analysis - mathematically and computationally. It may be of interest to investigate the behaviour of the usual models replacing a Poisson process by a determi- nantal process. We shall try to give an overview of what is known about determinantal processes. Basic probability is required, but no prior knowledge of determinantal pro- cesses is needed. Anyone who wishes to read before the lecture may consult the survey article: Hough, J. Ben, Krishnapur, Manjunath, Peres, Yuval, Virag, Balint, Determi- nantal processes and independence, Probability Surveys, 3, (2006), 206-229 (electronic).

Invited talk 1, Friday 12:00 –13:00
Title: Achieving Non-Zero Information Velocity in Wireless Networks (by Prof. Rahul Vaze) [Slides]

Abstract: In wireless networks, where each node transmits independently of other nodes in the network (the ALOHA protocol), the expected delay experienced by a packet until it is successfully received at any other node is known to be infinite for signal-to-interference-plus-noise-ratio (SINR) model with node locations distributed according to a Poisson point process. Consequently, the information velocity, defined as the limit of the ratio of the distance to the destination and the time taken for a packet to successfully reach the destination over multiple hops, is zero, as the distance tends to infinity. A nearest neighbor distance based power control policy is proposed to show that the expected delay required for a packet to be successfully received at the nearest neighbor can be made finite. Moreover, the information velocity is also shown to be non-zero with the proposed power control policy. The condition under which these results hold does not depend on the intensity of the underlying Poisson point process.

Invited talk 2, Friday 17:00 –18:00
Title: Transmission capacity of wireless ad hoc networks (by Prof. Radhakrishna Ganti) [Slides]

Abstract: Transmission capacity metric, introduced by Weber et. al., quantifies the density of simultaneous packings in a bi-polar interference limited ad hoc network. This metric was successfully used in quantifying the impact of physical layer processing on the overall throughput of an ad hoc network. In this talk, we will look at the existing results on transmission capacity when the nodes are distributed as a Poisson point process for different TX-RX strategies. We will also look at the transmission capacity for different MAC schemes, which induce correlations among the node locations.

Invited talk 3, Saturday 17:00 –18:00
Title: A Framework for Designing Energy-Harvesting Multihop Wireless Network (by Prof. Joy Kuri) [Slides]