Error Detection

This paper presents a technique to derive and implement error detectors to protect an application from data errors. The error detectors are derived automatically using compiler-based static analysis from the backward program slice of critical variables in the program. Critical variables are defined as those that are highly sensitive to errors, and deriving error detectors for these variables provides high coverage for errors in any data value used in the program. The error detectors take the form of checking expressions and are optimized for each control-flow path followed at runtime. The derived detectors are implemented using a combination of hardware and software and continuously monitor the application at runtime. If an error is detected at runtime, the application is stopped so as to prevent error propagation and enable a clean recovery. Experiments show that the derived detectors achieve low-overhead error detection while providing high coverage for errors that matter to the application.

Previous software-only error detection techniques have provided high-coverage, low-latency detection but suffer significant performance overheads with a large percentage of benign detections. This paper presents a FPGA hard- ware implementation of application-aware data error de- tectors. The detectors are automatically derived at compile time and executed in hardware at runtime, minimizing the performance overhead. We implement the static detectors using the Reliability and Security Engine, which provides a standard interface for developing reliability and secu- rity hardware modules. An initial, proof-of-concept model shows that there is only a 2% performance penalty when the detectors are implemented in hardware.

Machine translation (MT) has been an important field of research in the last decades and is currently playing a key role in the translation market. The variable quality of results makes it necessary to combine MT with postediting, to obtain high-quality translation. Post-editing is, however, a costly and time-consuming task. Additionally, it is possible to improve the results by integrating more information in automatic systems. In order to improve automatic systems performance, it is crucial to evaluate the quality of results produced by MT systems to identify the main errors. In this study, we assessed the results of MT using an error-annotated corpus of texts translated from English into Italian. The data collected allowed us to identify frequent and critical errors. Detecting and correcting such errors would have a major impact on the quality of translation and make the post-editing process more accurate and efficient. The errors were analyzed in order to identify patterns of er...

Modern computer systems are commonly evaluated using aggregate performance metrics such as latency, throughput, and utilization. While these metrics provide useful summaries of externally visible behavior, they do not fully capture variability in execution structure or the presence of redundant computation across repeated executions. This paper introduces a schema-based measurement framework for observing execution-level variability and redundant computation as measurable and non-causal properties of system behavior. The framework defines execution path variability, reuse failure, redundant computation, observed work, reference work, work inflation, and recompute ratio as observable concepts designed for reproducible execution analysis. The framework is implemented using Argus v1 and execution-instability-bench. Argus provides execution-level observation and schema-compliant artifact generation, while execution-instability-bench provides controlled workloads and reproducible benchmark scenarios for execution observation experiments. Experiments are conducted in a local CPU-based macOS environment under repeated execution conditions. Across two schema-compliant observation runs, Argus records measurable differences in execution path counters, reuse failures, repeated operations, and work inflation. In the observed runs, the number of distinct execution paths increases from 1 to 3, ordering changes increase from 0 to 46,904, reuse failures increase from 42,744 to 152,808, and repeated operations increase from 88,336 to 280,818. The work inflation ratio increases from 1.287552 to 1.457061, indicating measurable differences in observed work relative to reference work across execution conditions. The results also show that changes in execution work and recomputation do not necessarily correspond to monotonic changes in mean latency. The purpose of this work is not to establish universal causal claims about execution behavior, but to provide a reproducible framework for observing and characterizing execution-level variability and redundant computation across computing environments. The proposed schema and tooling provide a reproducible basis for future CPU, GPU, multi-process, and distributed execution validation.

Fi ber D i s t r i b u t e dD at a Int er f ace ( FD D I ) R aj Jain D igita l E quipment C orporatio n 5 5 0 K i n g S t . (LK G1-2/A19) Little to n , M A 0 1 460 A BST R A C T Fiber D istrib u te d D a t aI n t e r face (FD D I) i s a 1 0 0m e g a b its p e r s e c o n dber o p t iclo c a l a r e a n e t w o r k ( L A N) standard b e in g d e v e lo p e d b y t he A m eric a n N a t io n a l S t a n d a r d I n s t itu te (A N S I). We a n a ly ze the i m p a c t o f v a r io u s d e s ig nd e c isio n s o nt h e e r r o r d e t e c t io n c a p a b ility o f t h e p r o t o c o l. Inp a rtic u la r, w e q u a n t ifyfra m e e r r o r r a t e , t o k e nlo ss r a t e , a n du n d etectede r r o r r a t e . S e v e r a l c h a r a c t e r istic s o f t h e 3 2 -bitfra m e c h e c k s e q u e n c e ( F C S ) p o ly n o m ia l, w h ic h isa lso u s e di nI E E E 802 L A N p r o t o c o ls, a r e d isc u sse d . T h e s t a n d a r d u s e s a \ n o n -return t o z e r o i n v e r t o n o n e s " ( N R ZI) s ig n a l e n c o d in g a n d a 4 b it-to-5 bit(4 B / 5 B ) s y m b o l e n c o d in g i n t h e p h y s ic a l l a y e r . D u e t o t h e c o m b in a tio no f N R Z I a n d4B /5Bencodin g , m a n yn o ise e v e n t s a r e d e t e c t e db yc o d e ( o r s y m b o l)v io la tio n s. Ala rg e p e r c e n t a g e o f e r r o r s i s a lso d e t e c t e d b y f r a m in g v io la tio n s. S o m e o f t h e r e m a inin g e r r o r s a r e d e t e c t e d b y F C S v io la tio n s. T h e e r r o r s t h a t e s c a p e t h e s e t h r e e v io la tio n s re m a inu n d e te c te d . T h e p r o b a b ility o f u n d e t e c t e d e r r o r s d u e t o c r e a t io n o f f a lse s t a r t in g d e lim ite rs, f a lse e n d in g d e lim ite rs, o r m e r g in g o f t w o f r a m e s i s a n a ly z e d . It i s s h o w nt h a t e v e r y n o ise e v e n t r e s u lts i n t w o c o d e -bite rro rs, w h ic hint u r nm a y r e s u lt inu p t o f o u r d a t a -bite rro rs. T h e F C S c a n d e t e c t u p t o t w o n o ise e v e n t s . C r e a t io n o f a false s t a r t in g d e lim ite r o r e n d in g d e lim ite r o n a s y m b o l b o u n d a r y a lso r e q u ire s t w o n o ise e v e n ts. T h isa ssu m e s e n h a n c e d f r a m e v a lid ity c r ite ria . W e j u s t ifyth e e n h a n c e m e n t s b y q u a n t ify in g t h e ire ect. T hisa n a ly sish e reislim ite dto n o ise e v e n t s n o t r e s u ltin ginac h a n g e o f s y m b o l b o u n d a r ie s. E xtensio n s t o t h e c a s e o f c h a n g e ds y m b o l b o u n d a r ie s i s c o n t in u in g a n d w ill b e p r e s e n t e d a t a l a t e r t im e .

The Canadian Astronomy Data Centre at the Dominion Astrophysical Observatory in Victoria, B.C., was created in 1986 to help Canadian astronomers access data from the Hubble Space Telescope (HST). During the development of the CADC, its experts worked together with astronomers from the Space Telescope Science Institute and the European Space Agency to create archives for HST and make its data available to astronomers through the internet, which was then in its early days. The result was a notable Canadian contribution to HST and the establishment of a Canadian home for digital astronomical data. CADC played an important role in the establishment of the internet in Canada and is now involved in the major changes in astronomy relating to the availability of large quantities of astronomical data. This article discusses the challenges faced by the CADC and its staff during its first decade of operations.

Although OCR technology is now commonplace, character recognition errors are still a problem, in particular, in automated systems for information extraction from printed documents. This paper proposes a method for the automatic detection and correction of OCR errors in an information extraction system. Our algorithm uses domain-knowledge about possible misrecognition of characters to propose corrections; then it exploits knowledge about the type of the extracted information to perform syntactic and semantic checks in order to validate the proposed corrections. We assess our proposal on a realworld, highly challenging dataset composed of nearly 800 values extracted from approximately 100 commercial invoices and we obtained very good results.

Although OCR technology is now commonplace, character recognition errors are still a problem, in particular, in automated systems for information extraction from printed documents. This paper proposes a method for the automatic detection and correction of OCR errors in an information extraction system. Our algorithm uses domain-knowledge about possible misrecognition of characters to propose corrections; then it exploits knowledge about the type of the extracted information to perform syntactic and semantic checks in order to validate the proposed corrections. We assess our proposal on a realworld, highly challenging dataset composed of nearly 800 values extracted from approximately 100 commercial invoices and we obtained very good results.

HermesD is a high-rate, low-power wireless transmission system to aid research in neural prosthetic systems for motor disabilities and basic motor neuroscience. It is the third generation of our "Hermes systems" aimed at recording and transmitting neural activity from brain-implanted electrode arrays. This system supports the simultaneous transmission of 32 channels of broadband data sampled at 30 ks/s, 12 b/sample, using frequency-shift keying modulation on a carrier frequency adjustable from 3.7 to 4.1 GHz, with a link range extending over 20 m. The channel rate is 24 Mb/s and the bit stream includes synchronization and error detection mechanisms. The power consumption, approximately 142 mW, is low enough to allow the system to operate continuously for 33 h, using two 3.6-V/1200-mAh Li-batteries. The transmitter was designed using off-the-shelf components and is assembled in a stack of three 28 mm 28-mm boards that fit in a 38 mm 38 mm 51-mm aluminum enclosure, a significant size reduction over the initial version of HermesD. A 7-dBi circularly polarized patch antenna is used as the transmitter antenna, while on the receiver side, a 13-dBi circular horn antenna is employed. The advantages of using circularly polarized waves are analyzed and confirmed by indoor measurements. The receiver is a stand-alone device composed of several submodules and is interfaced to a computer for data acquisition and processing. It is based on the superheterodyne architecture and includes automatic frequency control that keeps it optimally tuned to the transmitter frequency. The HermesD communications performance is shown through bit-error rate measurements and eye-diagram plots. The Manuscript

One of the biggest obstacles in the formalisation of the Java bytecode is that the language consists of around 200 instructions. However, a rigorous handling of metatheoretic properties of a programming language requires a formalism which is compact in size. Therefore, the actual Java bytecode instruction set is never used in the context. Instead, the existing formalisations usually cover a 'representative' set of instructions. This paper describes a design of formalisation that provides a concise set of abstract, generic instructions that can be specialised to obtain any particular bytecode instruction. In this way one can work with a manageable set of instructions to prove general facts about the Java bytecode, but at the same time all the bytecode instructions are available to enable direct verification of actual bytecode programs. A considerable part of the design has been realised in Coq.

The presentation will contrast the complexity results for proving assertions in classical and intuitionistic logic. The comparison will be built upon the known results for propositional logic and predicate one. The predicate case will be based upon the Mints hierarchy in intuitionistic logic which will be contrasted with its counterpart i.e. the prenex hierarchy in classical logic. The di erences in complexity will be illustrated with examples of particular proving mechanisms that are responsible for the divergence, which should facilitate the understanding of where the mathematics is done constructively.

Vehicle-to-everything (V2X) communications impose stringent physical-layer integrity requirements, particularly under short-packet transmission and mobility-induced channel variation. This paper studies whether standard-compliant online selection of Cyclic Redundancy Check (CRC) polynomials and Quasi-Cyclic Low-Density Parity-Check (QC-LDPC) coding rates can reduce silent (undetected) errors in 5G New Radio (5G-NR) V2X links. The joint configuration problem is formulated as a lightweight Contextual Bandit (CB) with a small, discrete action space, and a discounted LinUCB policy is evaluated against greedy online adaptation and a conservative fixed baseline. A 5G-NR-compliant physical-layer simulation is developed using Sionna, modeling mobility through time-correlated Rayleigh fading, where vehicle speed governs channel correlation, and non-stationary interference via a two-state Markov process. The learning agent operates on coarse receiver feedback, including a noisy Signal-to-Noise Ratio (SNR) estimate and indicators of burst interference and deep fades, and targets minimization of the Undetected Error Probability (\(P_{UE}\)) while accounting for the Detected Error Probability (\(P_{DE}\)). Overall, our objective is to delineate the mobility regimes in which learning-assisted CRC–QC-LDPC configuration improves physical-layer integrity in 5G-NR V2X systems. Our results indicate that learning-assisted adaptation is most effective at low to moderate mobility, reducing \(P_{UE}\) by up to 50–70\% relative to greedy selection in the low-SNR regime (\(-5\) to 5~dB) and approaching the best fixed configuration at higher \(E_b/N_0\). At high mobility (\(\geq 180\)~km/h), fast channel decorrelation weakens temporal predictability, limiting the effectiveness of online learning and reducing performance differences across policies.

With the advent of high-speed broadband Internet access, the need to protect digital videos is highly recommended. The main objective of this study is to propose an adaptive algorithm for watermarked digital videos in the frequency domain based on discrete cosine transform (DCT). The watermark signature image is embedded into the whole frame of the video. The green channel of the RGB frame is selected for the embedding process using the DCT algorithm as it shows the recommended quality of the watermarked frames. The experiment results indicate that the proposed algorithm shows robustness and high quality of the watermarked videos by testing various strength values ∆ for different videos. It offers resistance against different types of attacks.

Based on strategies for on-line error detection in data and control path structures in simple microprocessors, this approach proposes techniques for the control-and component-error detection in high-performance processors. Detected errors are classified on-line with respect to their impact on the control and data flow. A compensation of detected errors is performed by micro rollback with different rollback distances according to pre-defined priority classes of error handling.

The augmented assembly language @Asm is proposed in order to transcend the fragmentation of architecture-specific dialects, to provide a unified framework for diverse processing paradigms as a universal assembly language and to function as a self-compiling bootstrap instrument adaptable to any processor system. The language augmentations include: flexible machine-language descriptions, general memory and data management directives, custom lexical identification through regular expressions, parsing facilities, generalized macroprocessing, flexible assembly control instructions, customizable encoding and code generation features, compiler-oriented abstraction mechanisms at the language level. The native abstraction augmentations enable expressive and concise high-level descriptions within assembly language for any present, emerging, or future systems.

In this study we measured Event-Related Potentials (ERPs) and evoked heart rate (EHR) to investigate performance monitoring in 10-to 12-year-old children. The children received feedback on their performance while conducting a probabilistic learning task. Error-related ERP components time-locked to the response increased in amplitude when the children had learned the task, whereas the feedback-locked components decreased. Concerning EHR, there was a general reduction in feedbackrelated heart rate deceleration when the children had learned. Moreover, a prolonged heart rate deceleration was observed at error feedback onset in comparison to positive feedback, which shifted in timing when the task progressed. Together, the ERP and EHR-measures suggest a shift from external to internal monitoring when the children are learning by performance feedback. The data suggest that error-and feedback-related EHR deceleration is a reflection of the same error monitoring system that is responsible for the emergence of the Error-Related Negativity.

A novel concurrent soft error detection and correction scheme is introduced for parallel hardware implementations of transform-based convolution. The proposed technique is based on the structure of radix-2 Fast Fourier Transforms (FFT) of length 2 n where n is an integer. The scheme can provide up to 100 percent detection and correction of isolated single soft errors in the convolution at the cost of little more than double the system area, rather than triple, as is required when using conventional Triple Modular Redundancy (TMR).

Dr. Gainor's book will be suitable for a general audience, while also being scholarly. Highly visible interactions among the general public, astronomers, engineers, government officials, and members of Congress about HST's servicing missions by Space Shuttle crews is a central theme of this history book. Beyond the glare of public attention, the evolution of HST becoming a model of supranational cooperation amongst scientists is a second central theme. Third, the decision-making behind the changes in Hubble's instrument packages on servicing missions is chronicled, along with HST's contributions to our knowledge about our solar system, our galaxy, and our universe. A fifth theme of the book covers the impact of HST and the images it produces on the public's appreciation for the universe." Provided by publisher.

In this paper the problem of unsatisfactory diagnostic efficiency of pseudorandom testing (PRT) technique used to detect faults of digital circuits in testing medical systems of critical importance is presented. The simulations have revealed a weakness of commonly used PRT technique that generally does not assure that all stuck at faults are detected. Thus, it is suggested that PRT technique should be supplemented with additional deterministic testing sequences to enhance fault detection. To design built-in selftesting (BIST) structures a genetic algorithm and digital system stochastic model were used. By employing the stochastic model it is possible to reduce considerably the computer simulation time required for BIST architecture design. In particular, an effect of proportional and ranking selections on the convergence of genetic algorithm in searching for a globally optimal solution, while maintaining the diversity of the population of individuals and limiting its premature stagn...

Grammar checking is the task of detection and correction of grammatical errors in the text. English is the dominating language in the field of science and technology. Therefore, the non-native English speakers must be able to use correct English grammar while reading, writing or speaking. This generates the need of automatic grammar checking tools. So far many approaches have been proposed and implemented. But less efforts have been made in surveying the literature in the past decade. The objective of this systematic review is to examine the existing literature, highlighting the current issues and suggesting the potential directions of future research. This systematic review is a result of analysis of 12 primary studies obtained after designing a search strategy for selecting papers found on the web. We also present a possible scheme for the classification of grammar errors. Among the main observations, we found that there is a lack of efficient and robust grammar checking tools for...

The current research aims to study the link between the type of vision experienced in a collaborative immersive virtual environment (active vs. multiple passive), the type of error one looks for during a cooperative multi-user exploration of a design project (affordance vs. perceptual violations), and the type of setting in which multi-user perform (field in Experiment 1 vs. laboratory in Experiment 2). The relevance of this link is backed by the lack of conclusive evidence on an active vs. passive vision advantage in cooperative search tasks within software based on immersive virtual reality (IVR). Using a yoking paradigm based on the mixed usage of simultaneous active and multiple passive viewings, we found that the likelihood of error detection in a complex 3D environment was characterized by an active vs. multi-passive viewing advantage depending on: (1) the degree of knowledge dependence of the type of error the passive/active observers were looking for (low for perceptual viol...

Available solutions for fault tolerance in embedded automation are often based on strong customisation, have impacts on the whole life-cycle, and require highly specialised design teams, thus making dependable embedded systems costly and difficult to develop and maintain. The TIRAN project 1 develops a framework which provides fault tolerance capabilities to automation systems, with the goal of allowing portable, reusable and cost-effective solutions. Application developers are allowed to select, configure and integrate in their own environment a variety of software-based functions for error detection, confinement and recovery provided by the framework. Market investigations with users and producers of automation systems have recognised the benefits offered by dependable systems, not only in the classical area of safetycritical tasks, but also in mission-critical ones, due to the high economic impact of failures. Moreover, dependable 1 ESPRIT Project 28620 TIRAN -"Tailorable fault tolerance frameworks for embedded applications".

We analyze the dynamic behaviors of a novel device, i.e., cascaded high-speed laser power converters (LPCs), which can detect the direct-current (dc) component of an incoming highspeed optical data stream and efficiently convert its dc component to dc electrical power. By utilizing a p-type photoabsorption layer in our LPC, the problem of slow-motion holes can be eliminated, and only the electrons act as the active carriers. We can thus achieve high-speed performance with the LPC under forwardbias operation with a small electric field inside. Furthermore, according to our modeling and measurement results, there are a significant alternating-current capacitance reduction and an electron-trapping effect at the interface between the absorption and collector layers with a significant degradation in the carrier drift velocity. These become more serious with the increase in optical pumping power and forward-bias voltage and truly limit the net optical-to-electrical (O-E) bandwidth of the device. In order to overcome such a transient-time-limited bandwidth and further increase the maximum dc output voltage of the LPC, we connect two single LPCs in series (cascade). Error-free data detection of 10-Gb/s and an O-E dc power-generation efficiency of 21.1% can be achieved simultaneously at a wavelength of 850 nm by the use of such two cascaded LPCs.

In this study we tested the hypothesis that perceptual awareness judgments are sensitive to accuracy feedback about the previous action. We used a perceptual discrimination task in which participants reported their stimulus awareness. We created two conditions: No-feedback and Feedback (discrimination accuracy feedback was provided at the end of each trial). The results showed that visual awareness judgments are related to the accuracy of current and previous responses. Participants reported lower stimulus awareness for incorrectly versus correctly discriminated stimuli in both conditions; they also reported lower stimulus awareness in trials preceded by incorrect discrimination responses, compared to trials preceded by correct discrimination responses. This difference was significantly stronger in the Feedback condition, in which we also observed post-error slowing for PAS ratings. We discuss the relation between visual awareness and the effects of performance monitoring and interp...

Electronic spreadsheets play an indispensable role in the simulation, modeling, and analysis of bioenergy systems, and their results have the ability to affect decision-making significantly. Prior research has shown that spreadsheets are highly error-prone, and that a large percentage of these errors are difficult to detect. To that end, we developed computer code (implemented in Visual Basic for Applications, running under Microsoft Excel) to detect a particularly insidious form of spreadsheet error: the hard-coding error. These errors are defined as the presence of one or more unreferenced numerical values in a cell formula. The code was used to audit six engineering spreadsheets relevant to bioenergy systems, three developed in our lab (and reported on in other sessions at the AIM), and three in the public domain. The preliminary audit results were analyzed to understand the nature and distribution of hard-coding errors. The preponderance and diversity of hard-coding errors in these spreadsheets motivated us to subcategorize them. Together, the hard-coding error detection program and sub-categorization program provide a robust and rapid means of detecting and categorizing multiple types of hard-coding errors. Use of these programs could increase the reliability of spreadsheet software used in simulation, modeling, and analysis of bioenergy systems.

The importance of the inspection process has been magnified by the requirements of the modern manufacturing environment. In electronics mass-production manufacturing facilities, an attempt is often made to achieve 100% quality assurance of all parts, subassemblies, and finished goods. A variety of approaches for automated visual inspection of printed circuits have been reported over the past two decades. In this survey, algorithms and techniques for the automated inspection of printed circuit boards are examined. A classification tree for these algorithms is presented and the algorithms are grouped according to this classification. This survey concentrates mainly on image analysis and fault detection strategies; these also include state-of-the-art techniques. A summary of the commercial PCB inspection systems is also presented.

SUMMARY A brief description of the gravity and topographic data banks of Greece is given. About 22 000 gravity stations are included in the newly compiled gravity data bank from an initially available data set of 33 000 stations, after the application of specific rejection criteria, and classified in squares of 100 km x 100 km. The data cover an area between 18" and 27"E, and 33" and 42"N (Fig. 1). A matrix of 2 km x 2 km mean elevation values was also estimated for the same area. These data constitute the topographic data bank. The preparation and structure of both data banks are also briefly discussed here. Key words: gravity, gravity anomalies, Greece, topography. INTRODUCTION The immense increase during the last few decades in the amount of all kinds of collected data has led to the need for a more systematic treatment concerning their storage and handling. Compilation of data banks has proven to be a useful method in this respect. The University of Athens wa...

Error detection and behavioral adjustment are core components of cognitive control. Numerous studies have focused on the anterior cingulate cortex (ACC) as a critical locus of this executive function. Our previous work showed greater activation in the dorsal ACC and subcortical structures during error detection, and activation in the ventrolateral prefrontal cortex (VLPFC) during posterror slowing (PES) in a stop signal task (SST). However, the extent of error-related cortical or subcortical activation across subjects was not correlated with VLPFC activity during PES. So then, what causes VLPFC activation during PES? To address this question, we employed Granger causality mapping (GCM) and identified regions that Granger caused VLPFC activation in 54 adults performing the SST during fMRI. These brain regions, including the supplementary motor area (SMA), cerebellum, a pontine region, and medial thalamus, represent potential targets responding to errors in a way that could influence VLPFC activation. In confirmation of this hypothesis, the errorrelated activity of these regions correlated with VLPFC activation during PES, with the cerebellum showing the strongest association. The finding that cerebellar activation Granger causes prefrontal activity during behavioral adjustment supports a cerebellar function in cognitive control. Furthermore, multivariate GCA described the "flow of information" across these brain regions. Through connectivity with the thalamus and SMA, the cerebellum mediates error and post-error processing in accord with known anatomical projections. Taken together, these new findings highlight the role of the cerebello-thalamo-cortical pathway in an executive function that has heretofore largely been ascribed to the anterior cingulate-prefrontal cortical circuit.

This paper explains how aviation accident investigators understand system's drifts, their main contributing factors and the way safe boundaries can be restored through concrete actions. We believe their experience can improve our knowledge about drifts into failure mechanisms and about action models to manage these drifts. After analyzing 70 french airforce accident reports and having noticed the total absence of violation's written trace, we interviewed 8 accident investigators and confronted them to the analysis of an accident report in order to assess if they have managed the risks related to violations. Our results show they know the difference between "work as imagined" and "work as actually done", they detect the reasons of drifts, establish the boundaries not to be crossed, and give guidelines based on their judgment of the drift validity. When assessing operator's interpretation of rule acceptability and flexibility, and violation's validity, accident investigators judgments are naive or too tolerant, depending on the "distance" between the analyst and the analyzed situation. What blocks violation management is the fragile analysis of operator's intention and of the socially accepted rules to which behaviour should be compared. Proactive work analysis in safety-emergent "normal" conditions and professional reflective activities about drifts into failure and safety boundaries should improve safety.

The transmission errors in a wireless or packet oriented network may dramatically decrease the performance of a distributed speech recognition (DSR) system. Error concealment has been shown to be an effective way to mantain an acceptable word error rate when dealing with error prone communication channels. In this paper we propose an extension of our previously introduced soft features approach for the case that the soft-output of the channel decoder is not available at the server side of the DSR system. We found a simple method to estimate bit reliability information which still gives good speech recognition results. It is shown that some other error concealment schemes turn out to be special cases of the method proposed here.

The decays B[J, Bj -► e+e~, are searched for in 3.5 million hadronic Z events, which constitute the full LEP I data sample collected by the L3 detector. No signals are observed, therefore upper limits at the 90% (95%) confidence levels are set on the following branching fractions: Br( B3 e+e~) < 1.4(1.8) x 1(T5; B r(B " -> e +e_ ) < 5.4(7.0) x 10~5; Br(B" -► < 1.0(1.4) x 10~5; Br(B" -► ¡m+ fx~) < 3.8(5.1) x 10~5; Br(BS ->■ e * / ^) < 1.6(2.0) x 1(T5; Br(B" -e * / ^) < 4.1 (5.3) x 10-5 . The results for B*1 -> e+e~ and B" -> e± /x:p are the first limits set on these decay modes.

A measurement of the K( S J and A inclusive production rates and momentum spectra in two-and three-jet events is presented. On the basis of about 3. J million Z decays collected with the L3 detector at LEP, we observe that the production of these particles is well modelled by string fragmentation.