Greg Kuperman

Analysis and algorithms for partial protection in mesh networks

This paper develops a novel mesh network protection scheme that guarantees a quantifiable minimum grade of service upon a failure within the network using multipath routing. Typically, networks fully guarantee service after a single-link failure, which is often an over-provisioning of resources to maintain essential traffic for the infrequent event of a failure. Our scheme guarantees that a fraction q of each demand remains after any single-link failure, at a fraction of the price of full protection. A linear program is developed to find the minimum-cost capacity allocation to meet both demand and protection requirements. For q≤12, an exact algorithmic solution for the minimum-cost routing and capacity allocation is developed using multiple shortest paths. For q>12, an algorithm is developed based on disjoint path routing that performs, on average, within 1.4% of optimal, and runs four orders of magnitude faster than the minimum-cost solution achieved via the linear program. Moreover, the partial protection strategies developed achieve reductions of up to 83% over traditional full protection schemes.

Uncoordinated MAC for Adaptive Multi-Beam Directional Networks: Analysis and Evaluation

In this paper, we consider medium access control (MAC) policies for emerging systems that are equipped with fully digital antenna arrays which are capable of adaptive multi-beam directional communications. With this technology, a user can form multiple simultaneous transmit or receive beams, allowing for greater spatial reuse and higher network throughput. The enabling technology that we consider is the ability to use digital post-processing to form multiple receive beams in real-time without a priori knowledge of the time and angle-of-arrival of the transmission. We present a novel unslotted, uncoordinated ALOHA-like random access MAC policy for multi-beam directional systems that asymptotically achieves the capacity of the network. Such an approach is particularly useful for systems where propagation delay makes the overhead associated with any sort of coordination prohibitive. We also consider the impact of numerous practical considerations including power constraints, latency, and beamwidth on the performance of our MAC policy.

Partial protection in networks with backup capacity sharing

We develop a novel network protection scheme that guarantees a minimum grade of service upon a link failure by utilizing backup capacity sharing between demands. Our scheme achieves up to 83% reductions in protection resources.

A high-fidelity statistical model of frequency hopping interference for fast simulation

Testing existing or developing new military waveforms requires computationally efficient simulations to allow for large scenarios that would not be feasible with hardware. However, fast running simulations have typically not represented wireless interference, especially frequency hopping interference, accurately. We develop the first major step toward a high-fidelity statistical model of frequency hopping interference that allows for much faster simulation times than high-fidelity deterministic models and even allows for real-time emulations of relatively large networks. We analyze the accuracy of our model by comparing it to the deterministic model it is based on, and we show some performance results by measuring computation times for a wide range of number of interferers.

Practical Capacity Benchmarking for Wireless Networks

Recent advances in the optimal design of cross-layer congestion control, routing and scheduling algorithms allow us to calculate wireless network capacity [3]. However, for large networks, the algorithm is computationally intractable. In this paper, we provide an upper bound on the wireless network capacity that is computationally efficient to implement. Our upper bound calculations consider multiple different wireless interference models. Our approach draws on results from multicommodity flow over the sparsest cut. We develop a polynomial time randomized algorithm to approximate the sparsest cut in general wireless networks. To ascertain the performance of our upper bound, we use the cross-layer approach in [3] to develop a lower bound by considering only a subset of independent sets. The lower bound is shown to be within 95 percent of our upper bound on average for the primary interference model. For 802.11, 802.16 and the 2-hop interference models, the lower bound is within 70 percent of the upper bound. By applying the sparsest cut recursively in the network, we also develop a capacity heat map that allows us to visualize the regional capacity and identify network bottlenecks.

A comparison of OLSR and OSPF-MDR for large-scale airborne mobile ad-hoc networks

In this paper, we compare the proactive MANET routing schemes of OLSR and OSPF-MDR via high-fidelity simulation, and consider their suitability for large-scale airborne networks. A successful MANET routing scheme must be bandwidth efficient and robust to frequent topology changes. To assess the two protocols, we simulate them in networks with up to 400 mobile nodes, under a variety of network densities. We evaluate them on the basis of the amount of routing overhead generated, the rate of successful packet delivery, and the time it takes until all of the routing tables converge. We find that OLSR requires up to an order magnitude higher router overhead than OSPF-MDR, while providing only a marginal benefit in packet delivery success rates. The largest difference between the two protocols is the time it takes for their routing tables to converge in the presence of packet loss. OLSR has consistent convergence times for networks of all sizes, while the convergence time of OSPF-MDR increases with network size.

Characterizing deficiencies of path-based routing for wireless multi-hop networks

With the emergence of the Internet of Things, there is a renewed focus on multi-hop wireless networking to connect these systems of smart-devices. Many of the proposals to support this new networking paradigm continue to use the concept of routing: a path between users is formed via a series of point-to-point links. We believe that the characteristics of the wireless environment inherently make path-based routing unsuitable for wireless networking, and that new approaches need to be considered. In this paper, we examine the effectiveness of routing for reliably and efficiently delivering data in a wireless network. Specifically, we demonstrate that path-based routing (1) experiences high packet loss due to the inherently unreliable nature of control information, (2) is unable to ensure reliable message delivery in a lossy environment, and (3) incurs a high cost for route maintenance and repair. We compare path-based routing to newer “smart-flooding” protocols, and show that these alternative approaches provide better delivery at lower cost.

Large scale over-the-air testing of Group Centric Networking

This paper presents experimental verification of the performance of Group Centric Networking (GCN), a networking protocol developed for robust and scalable communications in lossy networks where users are localized to geographic areas, such as military tactical networks. Initial simulations in NS3 showed that GCN offers high delivery with low network overhead in the presence of high packet loss and high mobility. We extend the investigation to verify GCNs performance in actual over-the-air experimentation. In the experiments, we deployed GCN on a 90-node Android phone test bed distributed across an office building, allowing us to evaluate its performance over-the-air on real-world hardware. GCNs performance is compared against multiple popular wireless routing protocols, which we also run over-the-air. These tests yield two notable results: (1) the seemingly benign environment of an office is in fact quite lossy, with high packet error rates between users that are geographically close to one another, and (2) that GCN does indeed offer high delivery with low network overhead, which is in contrast to traditional wireless routing schemes that offer either high delivery or low overhead, or sometimes neither.

Group centric networking: A new approach for wireless multi-hop networking

Abstract : ABSTRACTIn this paper, we introduce a new networking architecturecalled Group Centric Networking (GCN) that is designed tosupport the large number of devices expected with the emergenceof the Internet of Things. GCN is designed to enablethese devices to operate collaboratively in a highly efficientand resilient fashion, while not sacrificing the users? abilityto communicate with one another. We do a full protocolimplementation of GCN in NS3, and verify GCN in emulationand on real hardware communicating over an RF channel.We show that GCN utilizes up to an order of magnitudefewer network resources than traditional wireless routingschemes, while providing superior connectivity and reliability.

Subband bandwidth considerations for multiple-antenna multicarrier communications

We outline guiding principles for choosing the subband bandwidth of a multicarrier system with multiple antennas. Initially, we review considerations for the choice of bandwidth in fast-fading doubly dispersive channels. We append to these considerations those relevant to the use of multiple antennas for intereference rejection in the wideband regime. To weigh the benefits of narrow subbands, which permit narrowband beamforming for better interference rejection and signal gain, against the cost of a multiplicity of subbands, which is principally that of a high peak-to-average power ratio, we introduce an achievable rate metric.