Ericsson (United Kingdom)

Would slightly better performance be significantly more valuable from a market perspective? Would significantly better performance be just slightly more expensive to implement? When dealing with performance, usability, reliability, and so on, you often end up in difficult trade-off analysis. You must take into account aspects such as release targets, end-user experience, and business opportunities. At the same time, you must consider what is feasible with the evolving system architecture and the available development resources.Quality requirements are of major importance in the development of systems for software-intensive products. To be successful, a company must find the right balance among competing quality attributes. How should you balance, for example, investments for improved usability of a mobile phone's phone book and better mobile positioning? In the context of quality requirements, decision making typically combines market considerations and design issues in activities such as roadmapping, release planning, and platform scoping. Models that address requirements prioritization in a market-driven context often emphasize functional aspects. (For a comparison of other relevant techniques with Quper, see the sidebar.) Quper provides concepts for reasoning about quality in relation to cost and value and can be used in combination with existing prioritization approaches.

In modern (Intel) processors, Last Level Cache (LLC) is divided into multiple slices and an undocumented hashing algorithm (aka Complex Addressing) maps different parts of memory address space among these slices to increase the effective memory bandwidth. After a careful study of Intel's Complex Addressing, we introduce a slice-aware memory management scheme, wherein frequently used data can be accessed faster via the LLC. Using our proposed scheme, we show that a key-value store can potentially improve its average performance ~12.2% and ~11.4% for 100% & 95% GET workloads, respectively. Furthermore, we propose CacheDirector, a network I/O solution which extends Direct Data I/O (DDIO) and places the packet's header in the slice of the LLC that is closest to the relevant processing core. We implemented CacheDirector as an extension to DPDK and evaluated our proposed solution for latency-critical applications in Network Function Virtualization (NFV) systems. Evaluation results show that CacheDirector makes packet processing faster by reducing tail latencies (90-99th percentiles) by up to 119 μs (~21.5%) for optimized NFV service chains that are running at 100 Gbps. Finally, we analyze the effectiveness of slice-aware memory management to realize cache isolation.

An experimental validation is presented of a novel method for usability testing that entails the playback of dynamic eyetracking data to cue the elicitation of retrospective verbal reports. Participants in our study produced: (1) think-aloud reports during an online search task, and (2) retrospective reports during another online search task, with reports being cued by the playback of either the screen capture of events or the participant’s own eye-movements. Task-completion times and response rates were recorded for all reporting methods. Fewer participants completed the search task whilst thinking aloud, indicating the reactivity of this technique. Verbal transcripts were coded for instances of usability problems. The eye-cued method identified more usability problems than the think-aloud or screen-cued methods. A significant interaction between search engine type and retrospective cue type suggests that the value of the eye-cue method for eliciting usability problems may be greatest with more complex search environments. Our results demonstrate that when cued appropriately, retrospective reports may be less reactive and more informative than other verbalisation techniques.

A Bluetooth radio in 0.18 /spl mu/m CMOS technology works on 2.5-3.0 V, dissipating 75 mW in RX and 90 mW in TX. RX uses a 2 MHz IF with an active poly-phase bandpass filter. The fractional-N PLL uses a VCO running at 5 GHz. TX uses IQ modulation. Special attention is paid to Si-crosstalk because this radio is combined with baseband circuitry. Silicon area is 5.5 mm/sup 2/.

The wide adoption of mobile broadband services by users of smartphones and other mobile terminals is being enabled by radio access technologies with better performance than many fixed residential broadband lines. Mobile network operators are upgrading their networks to deliver growing packet traffic cost effectively, while maintaining critical operational functions such as base station synchronization and resilience to faults. Carrier Ethernet technology is figuring strongly in the upgrade from the existing TDM mobile backhaul between radio base stations and mobile core networks.

Large and hyperdiverse marine ecosystems pose significant challenges to biodiversity monitoring. While environmental DNA (eDNA) promises to meet many of these challenges, recent studies suggested that sponges, as "natural samplers" of eDNA, could further streamline the workflow for detecting marine vertebrates. However, beyond pilot studies demonstrating the ability of sponges to capture eDNA, little is known about the dynamics of eDNA particles in sponge tissue, and the effectiveness of the latter compared to water samples. Here, we present the results of a controlled aquarium experiment to examine the persistence and detectability of eDNA captured by three encrusting sponge species and compare the sponge's eDNA capturing ability with established water filtration techniques. Our results indicate that sponges and water samples have highly similar detectability for fish of different sizes and abundances, but different sponge species exhibit considerable variance in performance. Interestingly, one sponge appeared to mirror the eDNA degradation profile of water samples, while another sponge retained eDNA throughout the experiment. A third sponge yielded virtually no DNA sequences at all. Overall, our study suggests that some sponges will be suitable as natural samplers, while others will introduce significant problems for laboratory processing. We suggest that an initial optimization phase will be required in any future studies aiming to employ sponges for biodiversity assessment. With time, factoring in technical and natural accessibility, it is expected that specific sponge taxa may become the "chosen" natural samplers in certain habitats and regions.

Biomedical signals are measurable time series that describe a physiological state of a biological system. Entropy algorithms have been previously used to quantify the complexity of biomedical signals, but there is a need to understand the relationship of entropy to signal processing concepts. In this study, ten synthetic signals that represent widely encountered signal structures in the field of signal processing were created to interpret permutation, modified permutation, sample, quadratic sample and fuzzy entropies. Subsequently, the entropy algorithms were applied to two different databases containing electroencephalogram (EEG) signals from epilepsy studies. Transitions from randomness to periodicity were successfully detected in the synthetic signals, while significant differences in EEG signals were observed based on different regions and states of the brain. In addition, using results from one entropy algorithm as features and the k-nearest neighbours algorithm, maximum classification accuracies in the first EEG database ranged from 63% to 73.5%, while these values increased by approximately 20% when using two different entropies as features. For the second database, maximum classification accuracy reached 62.5% using one entropy algorithm, while using two algorithms as features further increased that by 10%. Embedding entropies (sample, quadratic sample and fuzzy entropies) are found to outperform the rest of the algorithms in terms of sensitivity and show greater potential by considering the fine-tuning possibilities they offer. On the other hand, permutation and modified permutation entropies are more consistent across different input parameter values and considerably faster to calculate.

Recognizing the increasing demand for intelligent transportation systems, the initial set of the Long Term Evolution (LTE) technical enablers for vehicle-to-everything (V2X) communication services has been substantially enhanced in Release 15 (Rel-15) and will be further developed in Rel16. These enhancements are driven by the 25 use cases identified for V2X by the Third Generation Partnership Project, which are categorized as vehicle platooning, extended sensors, advanced driving, and remote driving. In this article, we provide an overview of the new V2X features supported by Rel-15 LTE systems, including carrier aggregation, higher-order modulation, low latency support, and new resource management solutions. We also discuss the possible next steps of the 3GPP V2X technology evolution in the upcoming releases.

Memory access is the major bottleneck in realizing multi-hundred-gigabit networks with commodity hardware, hence it is essential to make good use of cache memory that is a faster, but smaller memory closer to the processor. Our goal is to study the impact of cache management on the performance of I/O intensive applications. Specifically, this paper looks at one of the bottlenecks in packet processing, i.e., direct cache access (DCA). We systematically studied the current implementation of DCA in Intel processors, particularly Data Direct I/O technology (DDIO), which directly transfers data between I/O devices and the processor's cache. Our empirical study enables system designers/developers to optimize DDIO-enabled systems for I/O intensive applications. We demonstrate that optimizing DDIO could reduce the latency of I/O intensive network functions running at 100 Gbps by up to ~30%. Moreover, we show that DDIO causes a 30% increase in tail latencies when processing packets at 200 Gbps, hence it is crucial to selectively inject data into the cache or to explicitly bypass it.

Simple and cost efficient means for providing and controlling quality of service (QoS) are important for 3GPP operators. This is particularly important for operators planning to provide a wide range of IP-based services across 3GPP broadband shared channels. We point out shortcomings of today's (Rel. 5) 3GPP QoS concept and, based on these, explain why that demand is currently not met. Based on a set of requirements identified for an evolved 3GPP QoS concept, we propose four small additions to the 3GPP specifications. The resulting evolved QoS concept is a realization of DiffServ for 3GPP access networks enhanced with the integration with session admission control. Operator QoS control is exercised from the policy charging rule function (PCRF) and through pre-configuration via the management plane. The evolved QoS concept provides a 3GPP operator with capabilities beyond those found in state-of-the-art fixed broadband access networks

This paper presents feedforward compensation techniques for low voltage fully differential folded cascode amplifiers and OTAs. Fully differential folded cascode cells have been fabricated and tested with different feedforward configurations in order to demonstrate and compare the usefulness of these techniques for broadbanding the gain-bandwidth product of low voltage, low power CMOS amplifiers and for providing high quality low voltage CMOS OTA integrators.

The authors describe a procedure for the empirical determination of system dynamic characteristics in which the time of measurement may be exchanged for test signal/noise ratio. The technique hinges on the well-known result that, if white noise is applied to a linear system, the cross-correlation of input and output gives the system impulse response.The authors point out that, in practice, it is much more convenient to use a periodic ‘random’ test input rather than a true stochastic input. Furthermore, by employing binary test inputs, the multiplica tion of output and input (in the cross-correlation) becomes simply a switching process. In describing some preliminary experimental work, the authors discuss the problems of mechanization supported by analytical considerations. The errors of such measurements are examined (including the interesting case when the test signal is com parable with system noise), and some illustrative results are presented.

A previously presented method for estimation of time errors in time-interleaved A/D converter systems is here verified on measurements from a dual A/D converter system. The advantage of this estimation method, compared to other methods, is that it does not require any knowledge about the input signal. The estimation is most accurate for slowly varying input signals but the signal quality is improved even when the estimation is done for a sinusoidal signal close to the Nyquist frequency.

This paper deals with a case study of testing with a usability testing package (UTUM), which is also a tool for quality assurance, developed in cooperation between industry and research. It shows that within the studied company, there is a need to balance agility and formalism when producing and presenting results of usability testing to groups who we have called Designers and Product Owners. We have found that these groups have different needs, which can be placed on opposite sides of a scale, based on the agile manifesto. This becomes a Designer and a Product Owner Manifesto. The test package is seen as a successful hybrid method combining agility with formalism, satisfying organisational needs, and fulfilling the desire to create a closer relation between industry and research.

We report an experimental comparison between broadband fibre Bragg gratings (FBGs) and conventional dispersion compensating fibre (DCF) for a 40 x 10Gb/s DWDM system over 525km. A performance-optimised configuration using FBG compensators is presented.

A monolithic low noise high linearity LNA/mixer circuit for 2.5 GHz applications has been fabricated in IBM 47 GHz SiGe production process. The measured performance is 8 dBm input IP3, 1.6 dB NF and 12 dB Gain for a low noise amplifier (LNA), and 2.5 dBm input IP3, 7.5 dB NF and 14 dB gain for a downconversion mixer with a total current consumption of 26 mA for a 2.75 V supply. LNA matching, mixer RF and LO matching and two baluns are all integrated on chip, requiring no critical RF tuning components.

We propose a cost-effective implementation of reconfigurable add&drop functionality in Optical Cross Connects based on the concept of common add&drop in the node (OXC with per- Node Add&Drop, OXC-NAD). The proposed architecture constrains usable transponders wavelengths, but the analysis on significant mesh topologies demonstrates that this does not affect the behavior of the transparent networks. The OXC-NAD is shown to also significantly reduce the node cost irrespective of the nodal degree.

This paper describes the comparative evaluation of four methods of building software development effort models based on least squares regression, artificial neural networks, case-based reasoning and rule induction. Some deficiencies are identified in the main measurement of estimating effectiveness currently used in comparative evaluations, ‘mean magnitude of relative error’ (MMRE), and a complementary measurement, ‘mean variation from estimate’ (MVFE) is suggested as more accurately reflecting the practitioner's viewpoint. Given the current state of development of the techniques, the parallel use of least squares regression and case-based reasoning is recommended as appearing to give the most reliable results in the studied environment.

Resolving symbol references is an important part of many application areas from development environments to various static analyser tools, especially when it is used for code comprehension purposes. Different occurrences of the same program elements, like function definitions and their call sites, variable declarations and their usage, or type definitions and their applications should be connected. In case of the C++ programming language, the most current tools use mangled names to correlate symbols, e.g. when implementing actions like "go to definition" or "list all references". However, for large projects, where multiple binaries are created, symbol resolution based on mangled names can be, and usually is, ambiguous. This leads to inaccurate behaviour even in major development tools. In this paper we explore the reason of this ambiguity, and propose our clustering algorithm based on essential build information to improve the accuracy of symbol resolution. We implemented our method as part of the CodeCompass open source code comprehension tool and measured its efficiency.

Development of large and complex software intensive systems with continuous builds typically generates large volumes of information with complex patterns and relations. Systematic and automated approaches are needed for efficient handling of such large quantities of data in a comprehensible way. In this paper we present an approach and tool enabling autonomous behavior in an automated test management tool to gain efficiency in concurrent software development and test. By capturing the required quality criteria in the test specifications and automating the test execution, test management can potentially be performed to a great extent without manual intervention. This work contributes towards a more autonomous behavior within a distributed remote test strategy based on metrics for decision making in automated testing. These metrics optimize management of fault corrections and retest, giving consideration to the impact of the identified weaknesses, such as fault-prone areas in software.