Search

Leveraging a data set of almost half a billion packets with high-precision packet times and sizes, we process it to extract characteristics of the bursts emitted over Starlink's Ethernet interface. The structure of these bursts directly reflect the physical layer receipt of OFDMA frames on the satellite link. We study these bursts by analyzing their rates, and by proxy the transition between different physical layer rates. The results highlight that there is definitive structure in the transition behavior, and we note specific behaviors such as particular transitionsteps associated with rate switching, and that rate switching occurs mainly to neighboring rates. We also study the joint burst rate and burst duration transitions, noting that transitions occur mainly within the same rate, and that changes in burst duration are often performed with an intermediate short burst in-between.Finally, we examine the configurations of the three factors burst rate, burst duration, and inter-burst silent time, which together determine the effective throughput of a Starlink connection., DRIVE

Although Starlink has been deployed for several years, a detailed understanding of system internals is still lacking. In this work we employ precise per-packet timestamps obtained from a hardware-timestamp capable NIC connected to a Starlink terminal.We find that Starlink frame timing details are readily observable at the network layer by analyzing the packet timing patterns.Based on a one-week measurement campaign we collect around half a billion of packet size and timing observations. Processing these observations yields 2.3 million transmission bursts. To learn details on the radio resource management we develop a methodology to infer the effective physical layer sending rate. Our findings show that although Starlink throughput can vary widely over multiple time-scales, there are a small number of fundamental physical layer transmission rates. We employ Gaussian Mixture Modeling to determine 14 such fundamental transmission rates, and relate the obtained rates to previous knowledge of the Starlink OFDMA frame structure. Our empirical observations provide an excellent match for a radio resource configuration where a Starlink frame employs 1000 subcarriers and 287 symbols per frame for user traffic transmission, which for uniform 4-QAM modulation yields a base rate of 430.5 Mbps. This physical layer base rate appears to mostly be varied by multiples of 27 Mbps, in several instances likely by modifying the modulation of a subset of the symbols in multiples of 18 symbols., DRIVE

Low Earth Orbit (LEO) satellite networks such as Starlink can provide broadband connectivity in sparsely populated regions, including those located in arctic and subarctic regions. However, only satellites in a subset of inclination planes can serve these northern regions. In order to evaluate Starlink connectivity in these geographically extreme regions, we perform a drive test across the arctic circle measuring Starlink and cellular connectivity. We analyze the results with regards to Starlink throughput characteristics including the effects of the 15 second reconfiguration interval, and compare the throughput and connectivity. Starlink generally provide better throughput, with around 10x better median throughput than our comparison cellular network. However, the low tail of the distribution is less impressive with more frequent connectivity losses. Our results also show that the combination of Starlink and cellular has the potential to provide considerable gains in aggregated connection reliability and tail throughput. We additionally discuss policy implications relating to state subsidy policies designed to ensure that the whole population has access to broadband connectivity., Part of ISBN 9798350317886QC 20240718

Although Starlink has been rolled-out for several years, there is still a lack of knowledge regarding system details and performance characteristics. To address this, we perform a network layer measurement campaign utilizing precise times-Tamping to analyze throughput variations at multiple time scales, and infer system timing details. On larger timescales we quantify the diurnal variations where the throughput varies with the time of day. On the medium timescales we establish the likely frequency allocation and beam switching period to be 15 seconds. The associated connectivity disturbances contribute to severe link underutilization for single long-lived TCP flows, which typically reach only 46% of the estimated link capacity. On the sub-millisecond timescale our network layer measurements corroborate recent physical layer investigations of the Starlink frame timing, which is confirmed to be 1.33 ms. © 2023 Owner/Author(s).

With a growing user base and the deployment of new systems, such as 5G and Starlink, a deep understanding of the varying link throughput for wireless systems is highly important. While detailed analysis of link throughput can be done on packet traces, collecting extensive packet traces often faces storage and privacy challenges. Instead, we propose using traces of link-wide inter-packet delay (IPD) to enable highly granular link throughput characterization on a wider scale. To this end, we present an eBPF-based tool designed to capture IPDs, and evaluate the accuracy of captured IPDs with the IPD tool and tcpdump, both with and without access to hardware timestamps. While hardware provided timestamps provide accurate IPDs, we find that software based timestamps lead to IPD values which are very inaccurate, but still useful in aggregate form to characterize throughput at millisecond timescales. Furthermore, we show that concurrent packet processing incurs a significant amount of packet reordering, which necessitates the consideration of several previous packets when computing the link-wide IPD. Finally we present an example use case of IPD collection, characterizing frequent silent periods during a speedtest over a 5G link.

We examine the connection reliability of LTE cellular infrastructure for supporting train signaling systems. In particular, the impact of simultaneous use of multiple networks on reliability is considered, along with failure correlation effects. We present a tailored reliability model, and report on data collected from many train-mounted cellular routers. Connection reliability reaches 99.994% when aggregation is used, compared to 99.953% for the best single link. Both modeling and measurement results show greatly improved reliability when aggregating over multiple links, thus indicating that commercial cellular networks may be useful for providing connectivity to future train signaling systems., QC 20230324

Multipath connectivity and aggregation of multiple communication links is actively being researched with the aim to achieve higher throughput and lower latency. In this work we perform an emulation-based evaluation of the relative goodput of MPTCP and TCP in a 5G usage context. A large range of path capacity and delay conditions is explored, for both the primary and secondary paths, with over 2000 different configurations evaluated. Evaluations are performed over eight combinations of MPTCP schedulers and congestion controls. The results show that MPTCP running over two links provide lower goodput than TCP over a single link for the majority of cases. Asymmetry in link conditions is in many cases a major complication for the MPTCP scheduler. To examine the predictability of poor performance, and to obtain further insight on the structure of this phenomena, we perform regression modeling of the relative good put. In addition to the traditional approaches of Linear Regression and Random Forest, we also employ Sym-bolic Regression to obtain mathematical expressions capable of providing insight on the path conditions most contributing to poor MPTCP performance. Such regression expressions can be informative when evaluating different schedulers or link aggregation approaches.

The detailed performance characteristics of networking equipment is to a large extent a function of the software that controls the underlying hardware components. Most networking equipment is regularly updated with new software versions. By studying performance changes related to such changes in software, it is possible to identify particular software versions that affect the performance of the system. Consequently, having automated methods for detecting changes in network equipment performance is crucial. In this work we study the change point detection problem arising when the placement in time of software updates is known a priori, but the presence of any performance implications on any of the thousands of performance indicators that can be collected is unknown. The ability to improve the automated detection of such change points by clustering the monitored systems according to the set of collected indicators has not been fully evaluated. We here report our experience with employing clustering, together with a bootstrap-based change point detection, across a range of performance indicators. We evaluate four variations of clustering approaches, and demonstrate the resulting improvement in change point detection sensitivity.

Efficient operation of networking systems is important from resource utilization, OPEX, and energy consumption perspectives. A major factor in efficient operations is the underlying software that controls the networking hardware or virtualized network functions. Most software in hardware-based networking devices is periodically updated, which may or may not have impact on various aspects of the performance of the device. We consider the issue of change point detection in network performance indicators, aiming to detect when such software updates co-occur with changes to any subset of collected performance metrics. In particular, we study the change point detection problem that arises when the placement in time of firmware changes is known a priori, but the presence of any performance implications is unknown. We focus on evaluating change point detection in operational network equipment log data, and consider diurnal variation suppression approaches. We propose the use of periodicity filtering to remove anomalous data sources, and apply a resampling technique using bootstrapping to determine when a software update has performance implications. Our results show that this automated change point detection approach can locate performance-related changes, and that load normalization appears to be the most sensitive approach to diurnal variation suppression.

While complex machine learning methods can achieve great performance, human-interpretable details of their internal reasoning is to a large extent unavailable. Interpretable machine learning can remedy the lack of access to model reasoning but remains an elusive feat to fully achieve. Here we propose ranked selectors as a method for post-hoc explainability of classification outcomes from arbitrary classification models, with an initial emphasis on tabular data of moderate dimensions. The method is based on constructing a set of selectors, or rules, delimiting a local class consistent domain with maximal cover around the item of interest. The extended adjacent feature space is probed to achieve a ranking of the selectors. The method supports the use of an explicit foil class Q, allowing the formulation of contrastive queries in the form 'Why inference P instead of alternative inference Q?'. The answer is given as a short list of disjoint rules, a format previously shown to be amenable to human interpretation. We demonstrate the proposed method in open datasets, and elaborate on its stability aspects relative to other comparable methods.

Important network functions such as traffic classification and intrusion detection often depend on high-throughput regular expression matching. To achieve high performance, regular expressions can be represented as state machines, which are then merged. However, determining which individual state machines should ideally be merged together is a challenging optimization problem. We address this problem by using genetic algorithms with novel problem-specific operators. To allow large scale evaluation of the new operators, we devise two ML-based surrogate models for the expensive fitness evaluation function. Our results from a set of production scale regular expressions show that using the most appropriate operations provides large gains over a naive baseline, but also that no universal best combination of operators exist. We provide some insights into which operators perform best for different objectives, and show the variation between TCP- and UDP-specific regular expressions.