51 Mathematik
Refine
Document Type
- Bachelor Thesis (2)
- Master Thesis (1)
Language
- English (3)
Is part of the Bibliography
- no (3)
Keywords
- Chirp spread spectrum (1)
- Compensation (1)
- FMCW Radar (1)
- Human Tracking (1)
- Long–range radio module (1)
- Natural language processing (1)
- Printed circuit board design (1)
- Simulation (1)
- Text classification (1)
- Time-of-flight ranging (1)
- Transmission lines and impedance matching (1)
- Vibration (1)
Institute
The German logistics company Schenker AG categorizes its customers into so-called vertical markets. For instance; category “Automotive” is assigned to car manufacturer BMW AG. The classification allows the company to evaluate its revenue and profits on different customer segments which, in turn, has an impact on Schenker’s strategic planning.
Until now the assignment is carried out manually which means that someone from sales department should perform some research on a customer’s public profile whenever a new customer is registered in the database. With the rapid growth of global trade in recent years and Schenker’s expansion to the Asian market with thousands of new customers the manual approach is no longer sustainable in a global market.
This thesis provides an alternative solution based on scraping customer data available in the web and classification of the extracted content. We deal with three significant difficulties: find web data related to a given customer name (we do have a company name but no homepage URL in Schenker database), extract a predictive portion of the data without introducing too much noise and, finally, set up a classification algorithm. Most importantly, the whole process needs to be implemented automatically.
For the classification task, we have identified two tree-based classification algorithms random forest and extreme gradient boosting (xgboost). Random forest performs better by using package ranger with an overall 52% accuracy and 87.8% multiclass area under the curve. On the other hand, xgboost takes less time to compute, but the accuracy is poor as compared to random forest.
This thesis is an initial attempt to investigate the influence of mechanical vibrations on the 24 GHz FMCW radar developed by IMST GmbH and develop a simple vibration compensation model for the human tracking algorithm.
The project objectives are to determine the radar signal distortion caused by vibrations theoretically, conduct simulations to evaluate the effects of vibrations on the human tracking process, and develop a basic model to compensate the disturbing vibrations.
The single radar pulse simulations conclude that when the radar is vibrating at a low frequency, the signal would not be distorted if the change of phase of the intermediate frequency is less than half a resolution cell. While the results from the multiple radar pulses simulations reveal that factors like the SNR value, vibrating amplitude and the target’s velocity would influence the human tracking process. The higher the SNR value, vibrating amplitude and target’s velocity, the more the detected ghost targets when there is only one target presence. A signal demodulation method is then proposed which indicate the possibility in reducing the disturbance caused by the vibrations.
Low–power, long–range wireless solutions are needed for transmission of environmental and industrial monitoring data from remote, battery–driven sensors. Chirp spread spectrum (CSS) modulation techniques are able to meet these requirements. A CSS–based radio module was designed to operate in the 2.4 GHz frequency band, whose advantages are worldwide availability, lack of duty–cycle restrictions and a broad spectrum. The module’s power consumption is satisfactory, but its range has to be improved. The accuracy of time-of-flight (ToF) distance estimation, an extra feature of the module, has not been evaluated. Impedance matching, transmission lines and printed circuit board design are among the practical aspects discussed, while ToF ranging and orthogonality of CSS signals are considered from a theoretical point of view.