Bundle protocol (BP) is the main protocol of delay/disruption-tolerant networking which is developed to provide reliable data delivery services in a stressed communications environment. A typical application of BP is for data delivery in a challenging deep-space communication environment. However, little work has been done in analyzing the performance of BP in deep-space communications, especially in presence of highly channel-rate asymmetry. In this paper, we present analytical modeling of the transmission performance of BP over deep-space communication channels characterized by highly asymmetric channel rates. The model is built to estimate the expected file delivery time (and goodput) of the protocol for reliable data delivery in a deep-space transmission scenario. The model is validated by conducting data transfer experiments using a PC-based experimental test bed.

Bundle Protocol Over Highly Asymmetric Deep-Space Channels

The delay/disruption-tolerant network (DTN) was proposed as an end-to-end networking architecture for file delivery service in a stressed communication environment. A typical application of DTN technology is for reliable data delivery in deep-space vehicle communications. The main protocol of DTN, i.e., bundle protocol (BP), utilizes a store-and-forward mechanism and a custody transfer option for transmission reliability. However, little work has been done on the performance evaluation of DTN's BP in deep-space communications. In this paper, we present a study of the retransmission timeout (RTO) timer setting for the use of BP for reliable file (or data) transmission over a relay-based deep-space vehicle communication infrastructure characterized by extremely long latency, lossy data links, and highly asymmetric channel rates. Unlike most studies proposed for terrestrial Internet for which the RTO timer is set equal to or longer than the round-trip time (RTT), we propose to set the RTO timer shorter than the RTT and expect that it will increase delivery efficiency per unit of time. We intend to find an RTO timer setting for maximum data delivery performance of BP, which is considered to be an optimal RTO timer, for deep-space missions. For the first time, analytical models are built for delivery performance involving the RTO timer. Using the built models, we tune the RTO timer toward the highest data delivery performance. The models are validated by running text file transfer experiments using a PC-based testbed and tuning the RTO timer toward the best performance of BP. A performance comparison between the optimal RTO timer proposed in this study and the existing commonly used RTO timers is also presented.

Optimal RTO Timer for Best Transmission Efficiency of DTN Protocol in Deep-Space Vehicle Communications

As the indispensable supplement of terrestrial communications, Low Earth Orbit (LEO) satellite network is the crucial part in future space-terrestrial integrated networks because of its unique advantages. However, the effective and reliable routing for LEO satellite network is an intractable task due to time-varying topology, frequent link handover, and imbalanced communication load. An Extreme Learning Machine (ELM) based distributed routing (ELMDR) strategy was put forward in this paper. Considering the traffic distribution density on the surface of the earth, ELMDR strategy makes routing decision based on traffic prediction. For traffic prediction, ELM, which is a fast and efficient machine learning algorithm, is adopted to forecast the traffic at satellite node. For the routing decision, mobile agents (MAs) are introduced to simultaneously and independently search for LEO satellite network and determine routing information. Simulation results demonstrate that, in comparison to the conventional Ant Colony Optimization (ACO) algorithm, ELMDR not only sufficiently uses underutilized link, but also reduces delay.

/pdf/distributed-routing-strategy-based-on-machine-learning-for-1rw80mrt91.pdf

Distributed Routing Strategy Based on Machine Learning for LEO Satellite Network

Delay/disruption tolerant networking (DTN) was proposed for reliable data transfer despite frequent link disruptions and long propagation delay that are typical of the deep-space communication environment. DTN communications rely heavily on its core bundle protocol (BP) for reliable data delivery. To date, little work has been seen in theoretical analysis of the performance of BP in deep-space communications. In particular, an analytical understanding of the performance of BP for reliable data delivery in a deep-space communication environment in the presence of link disruptions is quickly needed. In this paper, we present a study of the effect of link disruption on transmission performance of BP, in both analytical and experimental manners, in a deep-space communications system characterized by link disruptions accompanied by an extremely long propagation delay and lossy data links. For the first time, an analytical framework is built to estimate the effect of link disruption on the total bundle delivery time of BP depending on its starting time with respect to bundle delivery. The analytical models are validated by running experiments using a test bed.

Analytical Framework for Effect of Link Disruption on Bundle Protocol in Deep-Space Communications

Delay-/disruption-tolerant networking is recognized by the National Aeronautics and Space Administration as the only candidate network communication technology that approaches the level of maturity required for deep-space vehicle communications. Extensive performance evaluation and development of efficient data transmission mechanisms for its main protocol, bundle protocol (BP), are needed. In this paper, a new transmission approach based on “proactive” retransmission is proposed for BP for highly efficient data delivery in deep-space communications characterized by a high rate of data loss, long signal propagation delay, and highly asymmetric channel rate, given ample transmission bandwidth. The objective is to ensure successful and reliable deep-space file transfer within a single communication round-trip interval. The expected total number of transmission attempts needed for successful delivery of a file is first calculated based on the available space channel quality and the configured bundle size. Then, the time-out interval for custodial retransmission is set according to the expected number of transmission attempts required, such that all these attempts will be undertaken within the interval of a single round-trip. The performance evaluations based on both theoretical modeling and file transfer experiments indicate that the proposed proactive retransmission approach achieves the objective and has significant performance advantages over the existing reactive retransmission regardless of the transmission conditions.

Proactive Retransmission in Delay-/Disruption-Tolerant Networking for Reliable Deep-Space Vehicle Communications

Delay/disruption-tolerant networking was developed to enable automated network communications despite the long link delay and frequent link disruptions that generally characterize deep-space communications. It uses the well-known approach of store-and-forward with optional custody transfer, in which a node agrees to hold a file in memory (or storage) until its successful reception is acknowledged by the next node. The performance and memory consumption of delay/disruption-tolerant networking's Licklider transmission protocol (LTP) and bundle protocol in deep space will bear on decisions to adopt this technology. There is currently an urgent need to evaluate the performance and memory dynamics for file transmission by LTP and bundle protocol. In this paper, we present a study of memory dynamics for LTP-based transmission in a typical relay-based deep-space communication system characterized by an extremely long signal-propagation delay, lossy data links, and asymmetric data rates. Analytical models are built to quantify the dynamics of memory occupancy/release and memory release latency imposed by the use of LTP for reliable and complete file delivery in deep-space missions. File-transfer experiments are conducted using a test bed to validate the models.

Modeling memory-variation dynamics for the Licklider transmission protocol in deep-space communications

Delay/disruption tolerant networking (DTN) was proposed as an overlay-network technology for reliable data delivery in a stressed communication environment. A typical application of DTN is for reliable data delivery in deep-space communications, which is characterized by an extremely long signal propagation delay, random and lengthy link disruptions, lossy data links, and asymmetric channel rates. At present, the National Aeronautics and Space Administration (NASA) and several other space agencies consider DTN to be the most effective technology to handle the lengthy link delays and link disconnection inherent in deep-space communications. Bundle protocol (BP), developed as the main protocol of DTN, is used to build an overlay network and to provide custody-based, store-and-forward message transmission.

Mingjian Qiu

Papers

Modeling memory-variation dynamics for the Licklider transmission protocol in deep-space communications

Optimal RTO Timer for Best Transmission Efficiency of DTN Protocol in Deep-Space Vehicle Communications

RTO timer for best transmission efficiency of bundle protocol in deep-space communications