Albert Höglund

A computer host-based user anomaly detection system using the self-organizing map

Computer systems are vulnerable to abuse by insiders and to penetration by outsiders. The amount of monitoring data generated in computer networks is enormous. Tools are needed to ease the work of system operators. Anomaly detection attempts to recognize abnormal behavior to detect intrusions. A prototype UNIX anomaly detection system has been constructed. The system is host-based and monitors computer network host users. The system contains an automatic anomaly detection component. This component uses a test based on the self-organizing map to test if user behavior is anomalous. Both the test and the application are presented.

WCDMA common pilot power control for load and coverage balancing

The paper validates the feasibility of automating the setting of common pilot power in a WCDMA radio network. The pilot automation improves operability of the network and it is implemented with a control software aiming for load and coverage balancing. The control applies measurements of base station total transmission power of neighboring cells and terminal reports of received pilot signal level to determine the pilot qualification. The pilot power of a cell is periodically updated with simple heuristic rules in order to improve the load and coverage balance. The approach was validated using a dynamic WCDMA system simulator with a deployment of macro and micro cells on a city region whose measured propagation characteristics were incorporated into the model. The results showed that the proposed control method balanced load and coverage and improved the air interface performance measured as a function of packet throughput.

WCDMA common pilot power control with cost function minimization

The paper validates the feasibility of automating the setting of common pilot power in a WCDMA radio network. The pilot automation improves operability of the network and it is implemented with control software aimed at load balancing while also assuring pilot coverage. The pilot power of a cell is periodically updated by a gradient descent method that minimizes a cost function. The cost function describes differences in load between the cell and neighboring cells and pilot coverage that terminals of the cell measure. The approach is validated using a dynamic WCDMA system simulator with a deployment of macrocells and microcells on a city region whose measured propagation characteristics are incorporated into the model. The results show that the proposed control method slightly improves the air interface performance measured as a function of packet throughput.

Utilization of advanced analysis methods in UMTS networks

The scope of this paper is to introduce new analysis and visualization methods for WCDMA cellular networks. The proposed examples are mainly based on the self-organizing map (SOM) method, but also other neural and statistical methods are equally applicable. The main motivation for advanced methods is to increase the abstraction level from the raw network measurements, i.e. radio access network language, to network functional areas or a language closer to the business of network operator. Furthermore the vast amount of quality of service (QoS) and service combinations, that 3G will enable, require effective data handling procedures.

Automated optimization of key WCDMA parameters

Summary This paper validates the feasibility of automated optimization of key wideband code division multiple access (WCDMA) radio resource management parameters using control methods. The parameters are regularly adjusted in order to improve performance. The parameters examined in this study include the total cell transmission power target, the received total interference target, the downlink radio link power maximums, the handover windows and the pilot channel powers. The control was based on expert-defined rules, which applied specific trade-off policies and statistics of poor quality calls, blocking, packet queuing, power and interference levels and terminal measurements to qualify the parameter values. The approach was validated using a dynamic WCDMA system simulator with a deployment of macro and micro cells on a city region. Results on automated optimization of single parameters on cell level and results on simultaneous multi-parameter optimization on cell-cluster level are presented in this paper. The use of the automated parameter optimization methods was shown to result in a significant increase of capacity in comparison to the default parameter settings. Copyright # 2004 John Wiley & Sons, Ltd.

Quality-based tuning of cell downlink load target and link power maxima in WCDMA

The objective of the paper is to validate the feasibility of auto-tuning WCDMA link power maxima and adjust cell downlink load level targets based on quality of service. The downlink cell load level is measured using total wideband transmission power. The quality indicators used are call-blocking probability, packet queuing probability and downlink link power outage. The objective is to improve performance and operability of the network with control software aiming for a specific quality of service. The downlink link maxima in each cell are regularly adjusted with a control method in order to improve performance under different load level targets. The approach is validated using a dynamic WCDMA system simulator. The conducted simulations support the assumption that the downlink performance can be managed and improved by the proposed cell-based automated optimization.

UMTS radio network multiparameter control

Five key UMTS radio network parameters are simultaneously optimised with an automatic control method. The parameters relate to admission control, power control, handover control. and coverage. The optimization is guided by heuristic expert-defined rules, which apply specific trade-off policies and statistics of poor quality calls, blocking rates, power and interference levels, and terminal measurements to qualify the parameter values, the method was validated using a dynamic WCDMA system simulator with a deployment of 17 cells in Helsinki city center. The same parameter value was applied in each cell as the network formed of a uniform cell structure. The method was shown to produce convergence of parameters and stable operation. The obtained results showed that the method improved overall network performance in comparison to fixed planned default values The capacity of network was improved close to 20% with slightly decreased hut still acceptable quality of calls.

Quality-based auto-tuning of cell uplink load level targets in WCDMA

The objective of this paper was to validate the feasibility of auto-tuning WCDMA cell uplink load level targets based on quality of service. The uplink cell load level was measured with received wideband total power. The quality indicators used were called blocking probability, packet queuing probability and degraded block error ratio probability. The objective was to improve performance and operability of the network with control software aiming for a specific quality of service. The load level targets in each cell were regularly adjusted with a control method in order to improve performance. The approach was validated using a dynamic WCDMA system simulator. The conducted simulations support the assumption that the uplink performance can be managed and improved by the proposed cell-based automated optimization.

Auto-tuning of service-specific requirement of received EbNo in WCDMA

The paper validates the feasibility of auto-tuning of service- and bit rate-specific EbNo (E/sub b//N/sub 0/) requirements. The EbNo requirement is the level of the received bit energy to the interference and noise density that the receiver equipment requires for proper decoding of the signal. The planned EbNo values can be used e.g., to scale the powers when the service is varying. The proposed methods, one for uplink and one for downlink, are tested using a dynamic WCDMA system simulator with a deployment of macrocells on a city region whose measured propagation characteristics have been incorporated into the model. The results show that the proposed methods tune the initially incorrect planned EbNo values so that the throughput in the system increases.

Computer Network User Behaviour Visualisation Using Self Organising Maps

Computer systems are vulnerable to abuse by insiders and to penetration by outsiders. The amount of monitoring data generated in computer networks is enormous. Tools are needed to ease the work of system operators. Anomaly detection attempts to recognise abnormal behaviour to detect intrusions.