| 1. |
- Eriksson, Johan, et al.
(författare)
-
Using Timber in a multi-body design environment to develop reliable embedded software
- 2008
-
Ingår i: Intelligent vehicle iniative (IVI) technology controls and navigation systems, 2008. - Warrendale, Pa. : Society of Automotive Engineers, Incorporated. - 9780768020359 - 768020352 - 0768020352
-
Konferensbidrag (refereegranskat)abstract
- A major challenge for the automotive industry is to reduce the development time while meeting quality assessments for their products. This calls for new design methodologies and tools that scale with the increasing amount and complexity of embedded systems in today's vehicles.In this paper we undertake an approach to embedded software design based on executable models expressed in the high-level modelling paradigm of Timber. In this paper we extend previous work on Timber with a multi-paradigm design environment, aiming to bridge the gap between engineering disciplines by multi-body co-simulation of vehicle dynamics, embedded electronics, and embedded executable models. Its feasibility is demonstrated on a case study of a typical automotive application (traction control), and its potential advantages are discussed, as highlighted below:shorter time to market through concurrent, co-operative distributed engineering, andreduced cost through adequate system design and dimensioning, andimproved efficiency of the design process through migration and reuse of executable software components, andreduced need for hardware testing, by specification verification on the executable model early in the design process, andimproved quality, by opening up for formal methods for verification.
|
|
| 2. |
- Eriksson, Rikard, et al.
(författare)
-
Generating Weekly Training Plans in the Style of a Professional Swimming Coach Using Genetic Algorithms and Random Trees
- 2022
-
Ingår i: Proceedings of the 9th International Performance Analysis Workshop and Conference & 5th IACSS Conference (Advances in Intelligent Systems and Computing 1426). - Cham : Springer International Publishing. - 2194-5365 .- 2194-5357. - 9783030993337 - 9783030993320 ; 1426, s. 61-68
-
Konferensbidrag (refereegranskat)abstract
- Optimal training planning is a combination of art and sci- ence, a time-consuming task that requires expert knowledge. As such, it is often exclusively available to top tier athletes. Many athletes outside the elite do not have access or cannot afford to hire a professional coach to help them create their training plans. In this study, we investigate if it is possible to use the historical training logs of elite swimmers to con- struct detailed weekly training plans similar to how a specific professional coach would have planned. We present a software system based on machine learning and genetic algorithms for generation of detailed weekly training plans based on desired volume, intensity, training frequency, and athlete characteristics. The system schedules training sessions from a library extracted from training plans written by a professional swimming coach. Results show that the proposed system is able to generate highly accurate training plans in terms of training load, types of sessions, and structure, compared to the human coach.
|
|
| 3. |
- Gorschek, Tony, et al.
(författare)
-
A controlled empirical evaluation of a requirements abstraction model
- 2007
-
Ingår i: Information and Software Technology. - Newton, MA : Elsevier BV. - 0950-5849 .- 1873-6025. ; 49:7, s. 790-805
-
Tidskriftsartikel (refereegranskat)abstract
- Requirement engineers in industry are faced with the complexity of handling large amounts of requirements as development moves from traditional bespoke projects towards market-driven development. There is a need for usable and useful models that recognize this reality and support the engineers in a continuous effort of choosing which requirements to accept and which to dismiss off hand using the goals and product strategies put forward by management. This paper presents an evaluation of such a model that is built based on needs identified in industry. The evaluation's primary goal is to test the model's usability and usefulness in a lab environment prior to large scale industry piloting, and is a part of a large technology transfer effort. The evaluation uses 179 subjects from three different Swedish Universities, which is a large portion of the university students educated in requirements engineering in Sweden during 2004 and 2005. The results provide a strong indication that the model is indeed both useful and usable and ready for industry trials. © 2006 Elsevier B.V. All rights reserved.
|
|
| 4. |
- Kolomvakis, Nikolaos, 1985, et al.
(författare)
-
Quantized Uplink Massive MIMO Systems with Linear Receivers
- 2020
-
Ingår i: IEEE International Conference on Communications. - : Institute of Electrical and Electronics Engineers Inc.. - 9781728150895 ; 2020-June
-
Konferensbidrag (refereegranskat)abstract
- This paper considers the uplink of a single-cell multi-user massive multiple-input multiple-output (MIMO) system. Each receiver antenna of the base station is assumed to be equipped with a pair of analog-to-digital converters (ADCs) to quantize the real and imaginary part of the received signal. We propose a novel Bussgang-based weighted zero-forcing (B-WZF) receiver, which distinguishes the clipping and granular distortion. Numerical results demonstrate that for sufficiently high SNR and users that do not experience deep large-scale fading, the B-WZF brings significant performance gain over existing linear receivers in the literature, when the training sequence length is higher than the number of users. © 2020 IEEE.
|
|
| 5. |
- Kolomvakis, Nikolaos, 1985, et al.
(författare)
-
Reconstruction of Clipped Signals in Quantized Uplink Massive MIMO Systems
- 2020
-
Ingår i: IEEE Transactions on Communications. - : Institute of Electrical and Electronics Engineers Inc.. - 0090-6778 .- 1558-0857. ; 68:5, s. 2891-2905
-
Tidskriftsartikel (refereegranskat)abstract
- This paper considers the uplink of a single-cell multiuser massive multiple-input multiple-output system. Each receiver antenna of the base station (BS) is assumed to be equipped with a pair of analog-to-digital converters to quantize the real and imaginary part of the received signal. We propose a novel clipping-aware receiver (CA-MMSE), which performs minimum mean square error (MMSE) reconstruction only on the clipped received samples, while the granular samples are left unchanged after the quantization. On this basis, we present an iterative algorithm to implement the CA-MMSE receiver and derive a sufficient condition for its geometrical convergence to a fixed point. We show that as long as the number of BS antennas or the quantization resolution is sufficiently high, then, the performance of the CA-MMSE is as good as the optimal MMSE receiver which reconstructs all quantized received symbols. Additionally, we propose a novel Bussgang-based weighted zero-forcing (B-WZF) receiver which distinguishes the clipping and granular distortion and it is shown that as long as the received training symbols per antenna are correlated, the CA-MMSE brings significant improvements compared to conventional receivers in the literature while for users that do not experience deep large-scale fading the simpler B-WZF is near to the CA-MMSE for sufficiently high signal-to-noise ratio and quantization resolution. © 1972-2012 IEEE.
|
|
| 6. |
- Lagö, Thomas L, et al.
(författare)
-
On Robust Estimation of Water Current Using a Frequency Domain Approach : The Symmiktos Method
- 1994
-
Konferensbidrag (refereegranskat)abstract
- The estimation of water current by using acoustic Doppler current meters (ADCM) is a well-known method. Determination of the water current is based on the Doppler shift caused by the movement of small particles in the water, related to the current as an average. These movements cause a Doppler shift with an assumed Gaussian-shaped distribution in time. This assumption is valid for clean water and water where one particle size dominates the scattering and thereby the Doppler shift. If many bubble size are involved in the scattering, the Gaussian-based estimators may generate errors.
|
|
| 7. |
- Lagö, Thomas L, et al.
(författare)
-
Performance of an Acoustic Doppler Current Estimation Method Based on The Symmiktos Method
- 1995
-
Konferensbidrag (refereegranskat)abstract
- The estimation of water current by using acoustic Doppler current meters is a well-known method. Determination of the water current is based on the Doppler shift caused by the movement of small particles in the water, and related to the current as an average. These movements cause a Doppler shift with an assumed Gaussian-shaped distribution in time. Due to scattering from many different bubble sizes and other scatterers, a complex Doppler signal is received with a large variance. The Symmiktos method estimates the first moment of the Doppler signal by using a frequency domain method. This method combines linear and non-linear digital signal processing. The result is a robust and accurate estimation method with a very low bias. In order to verify this performance, a mathematical model of the Doppler shift is used, where effects from different bubble sizes and other reflexes can be taken into account. This mixed statistical model is then used for synthesizing data with different levels of complexity, but with known parameters. These synthesized data makes it possible to quantify the accuracy and performance of different estimation methods such as the Symmiktos method or the Covariance method.
|
|
| 8. |
- Lagö, Thomas L, et al.
(författare)
-
The Symmiktos Method : A Robust and Accurate Estimation Method for Acoustic Doppler Current Estimation
- 1993
-
Konferensbidrag (refereegranskat)abstract
- Sea current measurements have been of great interest for many years. Mechanical current instruments are still widely used for such applications. Electronic devices, using ultrasound, can measure the current at different depths. The current at different depths, or cells, is directly proportional to the doppler shift in that cell. When the burst or ping is transmitted toward the surface, scattering occurs from entrained air bubbles and other particles in the water. This is the base for the doppler signal and the measurement principle for an acoustic doppler current meter. Due to the variance in the scattering signal, a trade-off between burst length, cell size and measurement time has to be made. Bias in the doppler estimate is often a drawback when trying to decrease the variance. Traditional estimators are based on a Gaussian assumption about the scattered signals distribution, or rather the doppler shift distribution. A robust and accurate estimate of the doppler shift consists of a bias-free estimation of the current with a low variance for various types of data. The Symmiktos method is an estimation method that has improved the accuracy of the doppler calculation, compared to standard methods. This paper contains a description of the background to acoustic doppler current meters, the scattering signal, the traditional assumptions for a doppler estimate, as well as suggestions for the improvement of models for the estimation. An introduction to the Symmiktos method, and the models that are the foundation for this new estimation method is also presented.
|
|
| 9. |
- Svahnberg, Mikael, et al.
(författare)
-
Perspectives on Requirements Understandability : for Whom Does the Teacher's Bell Toll?
- 2008
-
Konferensbidrag (refereegranskat)abstract
- Software development decision makers use many different information sources as a basis for their decisions. One of these sources is the requirements specification, which is used in a large number of processes throughout the software development cycle. In order to make good decisions, the quality and completeness of the available information is important. Hence, requirements must be written in a way that is understandable for the different decision makers. However, requirements are rarely written with an explicit perception of how to make them understandable for different target usages. In this study we investigate the implicit assumptions of current and future requirements engineers and their teachers regarding which usages they perceive as most important when creating requirements. This is contrasted with industrial viewpoints of the relative importance of different requirements usages. The results indicate that the teachers and future requirements engineers have a strong focus towards in-project perspectives, and very little in common with the perspectives of industry managers. Thus, we are training students to serve as software developers, and not software engineering managers.
|
|
