Vassilis Tsaoussidis

Deep-Space Transport Protocol: A novel transport scheme for Space DTNs

The Delay-/Disruption-Tolerant Networking Architecture calls for new design principles that will govern data transmission and retransmission scheduling over challenged environments. In that context, novel routing, transport and application layer algorithms have to be established in order to achieve efficient and reliable communication between DTN-nodes. In this study, we focus on the evolution of the terrestrial Internet into the Interplanetary or Space Internet and propose adoption of the Deep-Space Transport Protocol (DS-TP) as the transport layer scheme of choice for the space networking protocol stack. We present DS-TPs basic design principles and we evaluate its performance both theoretically and experimentally. We verify that practice conforms with theory and observe great performance boost, in terms of file delivery time between DTN-nodes, in case of DS-TP. In particular, the gain of DS-TP against conventional proposals for deep-space communications increases with the link error rate; under conditions DS-TP can improve the performance of the transport layer protocol by a factor of two (i.e., DS-TP can become two times faster than conventional protocols).

The dynamics of responsiveness and smoothness in heterogeneous networks

Additive increase/multiplicative decrease-based protocols, including transmission control protocol (TCP), TCP-friendly, and a new generation of rate-based protocols, attempt to control the tradeoff of responsiveness and smoothness. Traditionally, smoothness has not been a main concern since it does not impact the performance of regular Internet applications such as the Web, FTP, or e-mail. However, multimedia-driven protocols attempt to favor smoothness at the cost of responsiveness. In general, smoothness and responsiveness constitute a tradeoff; however, we uncover undesirable dynamics of the protocols in the context of wireless/mobile networks with high-error rate or frequent handoffs: low responsiveness is not counterbalanced by gains in smoothness, but instead, produces a conservative behavior that degrades protocol performance with both delay-tolerant and -sensitive applications. Based on our observations, as well as on further analysis of the impact of the bottleneck queue on channel utilization, we seek an alternative strategy for smooth window adjustments. We introduce a new parameter /spl gamma/, which implements a congestion avoidance tactic and reaches better smoothness without damaging responsiveness.

Packet size and DTN transport service: Evaluation on a DTN Testbed

We present a study of DTN transport layer service, evaluating its performance during a file transmission from a Mars relay satellite towards Earth. The experiments are being held in a DTN Testbed designed and implemented for emulating and evaluating deep-space communications. Bundle Protocol truncates application data into bundles of different size and Licklider Transmission Protocol is used as the convergence layer protocol, encapsulating bundles into blocks and fragmenting blocks into segments. Bundle size, LTP block size and segment size are under consideration to achieve better transmission in terms of delivery time, overhead and memory occupancy at the sending node. Our most interesting conclusion is that memory is released faster when using small bundles and LTP blocks. We also conclude that segments should be close to the maximum boundary of underlying layers MTU, and that ignoring these dependencies, file delivery time in a channel of steady bit error rate has little dependence from packet sizes.

SSVP: A congestion control scheme for real-time video streaming

In this paper, we present a new end-to-end protocol, namely Scalable Streaming Video Protocol (SSVP), which operates on top of UDP and is optimized for unicast video streaming applications. SSVP employs Additive Increase Multiplicative Decrease (AIMD)-based congestion control and adapts the sending rate by properly adjusting the inter-packet-gap (IPG). The smoothness-oriented modulation of AIMD parameters and IPG adjustments reduce the magnitude of AIMD oscillation and allow for smooth transmission patterns, while TCP-friendliness is maintained. Our experimental results demonstrate that SSVP eventually adapts to the vagaries of the network and achieves remarkable performance on real-time video delivery. In the event where awkward network conditions impair the perceptual video quality, we investigate the potential improvement via a layered adaptation mechanism that utilizes receiver buffering and adapts video quality along with long-term variations in the available bandwidth. The adaptation mechanism sends a new layer based on explicit criteria that consider both the available bandwidth and the amount of buffering at the receiver, preventing wasteful layer changes that have an adverse effect on user-perceived quality. Quantifying the interactions of SSVP with the specific adaptation scheme, we identify notable gains in terms of video delivery, especially in the presence of limited bandwidth.

Delivery Time Estimation for Space Bundles

We present a method for predicting delivery time of bundles in space internetworks. The bundle delivery time estimation (BDTE) tool exploits contact graph routing (CGR), predicts bundle route, and calculates plausible arrival times along with the corresponding probabilities. Latency forecasts are performed in an administrative node with access to an instrumentation database (DB) appropriate for statistical processing. Through both analysis and experimentation, we demonstrate that estimates of bundle earliest plausible delivery time and destination arrival probabilities can be provided.

DS-TP: Deep-Space Transport Protocol

We present deep-space transport protocol (DS- TP), a new reliable protocol for deep-space communication links. DS-TPs main advantage is its ability to complete file transfers faster than conventional TCP, SCPS-TP and Saratoga. Therefore, missions with small connectivity time are greatly favored. Deep space communication links are characterized by long propagation delays, high BERs, intermittent connectivity (i.e., blackouts) and bandwidth asymmetries. Common approaches to deal with the above unique characteristics are: rate-based, open-loop protocols to deal with huge propagation delays; regular retransmissions to deal with high BERs; transmission suspension to deal with blackouts; SNACKs to deal with bandwidth asymmetries. We adopt some of the above approaches, namely, the open-loop, rate-based transmission and the SNACKs and focus on the optimization of the rest, namely, the retransmission strategy of the transport protocol to deal either with high BERs or with blackouts. More precisely, DS-TP includes the Double Automatic Retransmission (DAR) technique. DAR sends each packet twice, importing some intentional delay (Rd) between the original transmission and the retransmission. Therefore, in the presence of communication gaps (i.e., errors or blackouts), corrupted packets will eventually be replaced by the same correct packets that arrive with delay Rd. Rd, however, is much smaller than the traditional TCP-RTO value. Our theoretical performance evaluation results reveal that DS-TP presents high potential for deploy ability. In particular, we show that for PER=50%, DS-TP completes a file transfer in half time of a conventional protocol.

The TCP minimum RTO revisited

We re-examine the two reasons for the conservative 1-second Minimum TCP-RTO to protect against spurious timeouts: i) the OS clock granularity and ii) the Delayed ACKs. We find that reason (i) is canceled in modern OSs; we carefully design a mechanism to deal with reason (ii). Simulation results show that in next generations high-speed, wireless-access networks, TCP-RTO should not be limited by a fixed, conservative lower bound.

Why TCP timers (still) don't work well

We argue that the design principles of the TCP timeout algorithm are based solely on RTT estimations and may lead to flow synchronization, unnecessary retransmission effort and unfair resource allocation. We present a new Window-Based Retransmission Timeout algorithm (WB-RTO) for TCP, which exhibits two major properties: (i) it cancels retransmission synchronization, which dominates when resource demand exceeds resource supply and (ii) it reschedules flows on the basis of their contribution to congestion. WB-RTO achieves better fairness and slightly better goodput with significant less retransmission effort.

Contact Graph Routing enhancements for delay tolerant space communications

In this paper we present two enhancements to Contact Graph Routing (CGR), a Delay-/ Disruption- Tolerant Networking routing algorithm developed by NASA JPL for space environments with predetermined connectivity schedules, such as Interplanetary communications and LEO satellite systems. The first enhancement, CGR-ETO, aims to improve the accuracy of predicted bundle delivery time by considering the available information on queueing delay. The second, the Overbooking Management, aims to proactively handle contact oversubscription, which occurs when high priority bundles are forwarded for transmission on a contact that is already fully subscribed by lower priority bundles. Both enhancements have been inserted as optional features in the Interplanetary Overlay Network CGR implementation in order to comparatively evaluate their performance on a GNU/Linux testbed running the full protocol stack. The results show that the two enhancements are complementary and can significantly improve routing decisions compared to standard CGR.

A Rate Control Scheme for Adaptive Video Streaming Over the Internet

In this paper, we propose a new streaming protocol, namely dynamic video rate control (DVRC), which enables adaptive video delivery over the Internet. DVRC operates on top of HDP providing a congestion-controlled flow of unreliable datagrams. The proposed rate control scheme is able to interact with new and existing video streaming applications which are capable of adjusting their rate based on congestion feedback. DVRC attempts to optimize the performance of video delivery with concern to friendliness with interfering traffic. Exploring DVRCs potential through extensive simulations, we identify notable gains in terms of bandwidth utilization and smooth video delivery. Furthermore, our results indicate that the protocol allocates a well-balanced amount of network resources maintaining friendliness with coexisting flows.