Radim Bartos

Mobile video delivery with HTTP

Expansion in 3G cellular coverage and the emergence of more powerful mobile devices has increased demand for massively scalable mobile video delivery. The rapid adoption of the third screen as a primary screen for video has highlighted inefficiencies in the mobile delivery ecosystem and scalability issues in the mobile delivery infrastructure. This article provides an overview of the current mobile content delivery ecosystem and discusses the expanding role of HTTP-based mobile video delivery. A new class of HTTP-based mobile delivery protocols seeks to address existing quality and scalability issues by simplifying and standardizing mobile video delivery. This article shows how segment-based delivery has enabled HTTP-based live streaming and dynamic bitrate adaptation while increasing scalability through the use of existing CDN infrastructure.

Analysis of the Gated IPACT Scheme for EPONs

Interleaved Polling with Adaptive Cycle Time (IPACT) is one of the earliest proposed polling schemes for dynamic bandwidth allocation in Ethernet Passive Optical Networks (EPONs) and has been extensively used as a benchmark by many subsequent allocation schemes. In this paper, we attempt to construct a mathematical model of the IPACT scheme under the gated service discipline. For N = 1 ONU, we derive closed-form expression for the steady state grant size. For N > 1 ONUs, we need to consider separately a small and a large load-distance ratio. For the former case, the N = 1 ONU model holds even for N > 1. For the latter case, we find a closed form expression for the grant size. Our model shows a reasonable match with the values obtained from simulation for the steady state queue size and hence the throughput and delay.

A survey of protocols for Intermittently Connected Delay-Tolerant Wireless Sensor Networks

Abstract Intermittently Connected Delay-Tolerant Wireless Sensor Networks (ICDT-WSNs), a branch of Wireless Sensor Networks (WSNs), have features of WSNs and the intermittent connectivity of Delay-Tolerant Networks (DTNs). The applications of ICDT-WSNs are increasing in recent years, however, the communication protocols suitable for this category of networks often fall short. Most of the existing communication protocols are designed for either WSNs or DTNs and tend to be inadequate for direct use in ICDT-WSNs. This survey summarizes characteristics of ICDT-WSNs and their communication protocol requirements, and examines the communication protocols designed for WSNs and DTNs in recent years from the perspective of ICDT-WSNs. Opportunities for future research in ICDT-WSNs are also outlined.

STWnet: a high bandwidth space-time-wavelength multiplexed optical switching network

We propose STWnet, a self-routing high bandwidth optical network architecture for interconnecting users, grouped together as g groups with w users per group. STWnet uses the three dimensions of space, time, and wavelength by combining the advantages of space and temporal switching with the benefits of wavelength parallel data transmissions. Technologically difficult switching of individual wavelengths is avoided by prearranging transmissions in a way that they can be switched in a wavelength insensitive manner. Wavelengths are reused within the network thus allowing for a larger switching fabric. The proposed architecture can be internally expanded either in the spatial or temporal dimension to allow for multiple packets to be delivered to the same destination group. The expansion factor is determined based on the group knockout principle and given typical traffic patterns is a small number. STWnet allows easy group to group multicasting and broadcasting while system-wide multicasts and broadcasts can be achieved through repetitive group-to-group transmissions. The network uses readily available components such as opto-electronic directional couplers, fixed wavelength transmitters, and diffraction based parallel receivers while avoiding the use of relatively slow and expensive tunable components.

The impact of network latency on the synchronization of real-world IEEE 1588-2008 devices

Precision Time Protocol (PTP) is a high precision time synchronization protocol designed to run over a local area network. PTP, often referred to as 1588, is defined by the IEEE Standard 1588TM-2008. The protocol theoretically allows synchronization at the nanosecond level. In this project we study the performance of the protocol in an environment where multiple 1588 devices are connected via a network in which impairments that are typically observed in real networks are introduced and non-1588 devices are present. The performance was assessed by observing the impact on the clock synchronization of the 1588 devices. The results provide valuable insight into the real-world accuracy and robustness of the protocol.

IPACT with Smallest Available Report First: A New DBA Algorithm for EPON

Dynamic Bandwidth allocation in Ethernet Passive Optical Networks (EPONs) has been an area of intense research in recent years. Most of the proposed solutions offer clever methods for fair grant sizing, traffic prediction, and prioritized, differentiated services. Barring some work by Kamal et al. and some elements in the scheme proposed by Ma et al., no work has been done on exploring the order of granting (i.e., ONU sequencing) in an EPON. In this paper, we propose an unexplored heuristic for improving the performance of the IPACT scheme with respect to the most important metric: packet delay. In this heuristic, the OLT always grants that ONU which has the Smallest (Available) Reported queue length, First (SARF). Our simulations indicate that our heuristic can improve the delay performance of IPACT by 10-20% (when tested under the gated allocation policy).

HTTP Live Streaming Bandwidth Management Using Intelligent Segment Selection

HTTP-based segmented delivery has become a popular choice for video distribution. The ability for clients to explicitly perform dynamic bitrate adaptation on segment boundaries has allowed clients to optimize the end user viewing experience. Client-based rate adaptation schemes, though, are typically greedy in nature and within the context of the network as a whole, these schemes may be sub-optimal. In this paper we propose a congestion-aware rate adaptation scheme that works on top of existing segmented delivery protocols to provide network bandwidth management and fairness. We examine the HTTP Live Streaming approach and compare our community-based approach with the native Apple® implementation, though the framework is extendable to other protocols. We show the improved user experience with our scheme through higher quality video delivery and lower bitrate selection thrashing.

SXmin: a self-routing, high-performance ATM packet switch based on group-knockout principle

We propose SXmin: a self-routing, group-knockout principle based asynchronous transfer mode (ATM) packet switch which provides comparable delay-throughput performance and packet loss probabilities at significantly reduced hardware requirements compared to earlier switches. The M/spl times/N SXmin consists of an N/spl times/N Batcher sorter followed by log/sub 2/N-1 stages of sort-expander (SX) modules arranged in the form of a complete binary tree. Each SX module consists of a column of 2/spl times/2 switches with a wraparound-unshuffle input-output interconnection. This enables the hierarchical utilization of the group-knockout principle to expand the number of inputs by a small factor at each stage, resulting in a significant reduction in overall hardware complexity. Routing at each switch is controlled by a single bit. However, in case of contention, a dual bit resolution algorithm is used locally which drops excess packets in a predetermined manner while ensuring global randomness of packet loss over the entire switching network. There are no internal buffers at the individual stages and therefore the internal delay is constant and proportional to the number of stages. The use of simple hardware components and regular interconnections in the SX modules makes the network suitable for optical implementation.

Transparent Clocks vs. Enterprise Ethernet switches

Precision Time Protocol (PTP) is a high-precision time synchronization protocol designed to operate over a local area network. PTP, typically referred to as 1588, is defined by the IEEE Standard 1588™-2008 and provides clock synchronization at the nanosecond level. Even though devices with support for 1588 are widely available, there have been few performance studies of such devices. This paper investigates the impact of latency and high bandwidth background traffic on 1588 clock synchronization in a network consisting of both 1588 and non-1588 aware switches. As expected, we found that 1588-aware switches provide higher precision time synchronization in smaller-scale networks. In larger networks with congestion, 1588-aware switches were unable to maintain high accuracy clock synchronization due to the lack of traffic prioritization. Our results also show that having cut-through Enterprise Ethernet switches with prioritization enabled is adequate for maintaining sub-microsecond synchronization.

Closed-form expression for the collision probability in the IEEE Ethernet Passive Optical Network registration scheme

We derive a closed-form expression for the message collision probability in the IEEE 802.3ah Ethernet Passive Optical Network (EPON) registration scheme. The expression obtained, although based on an approximation, shows a good match with simulation results. We use the results of our analysis to compute the size of the most efficient contention window and the most efficient number of nodes serviced by a given window size.