Lost Silence: An emergency response early detection service through - - PowerPoint PPT Presentation

Lost Silence:

An emergency response early detection service through continuous processing of telecommunication data streams

Qianru Zhou

Heriot-Watt University, Edinburgh, U.K. October 2017

1

2

At 21:30 on 1st June, 2015, on the Yangtze River, in Jianli, Hubei, China, the cruise ship “Eastern Star” ship capsized with 458 passengers and crew on

• board. No distress signal was sent. It was several

hours before the emergency services became aware

• f the tragedy and 442 lives were lost.

Motivation

\

3

4

When a ship capsizes, is there any way to detect this as soon as it happens? The Answer is YES! This can be carried automatically in Real Time via RDF STREAM PROCESSING…

5

What I am really doing …

Building knowledge based autonomic network management system for hybrid telecommunication networks.

6

Some Basic Telecommunication Knowledge

7

• Hybrid access technologies.
• WiFi, LTE, 3G/2G, Satellite, Mesh, etc…
• Hybrid data schemas from different service providers/vendors.
• The data in real telecommunication system are extremely dynamic.
• E.g., system log, SNMP polling, tcpdump, and configuration data, etc.
• The granularity of SNMP, Syslog and tcpdump update period is in terms
• f a few seconds, one second and microseconds, respectively

Telecommunication networks are HETEROGENOUS By annotating the data with RDF/OWL vocabularies, we can present real time information as streams of uniformly RDF encoded linked-data.

Challenges

8

Ontology Adopted in Lost Silence

9

To describe the fact that “a phone lost signal at location (329.860, 246.792)”

An Example …

10

A Uniformed Spatial Segmentation

• Defined based on the third

decimal degree of the latitude and longitude coordination

• Resolution: 0.001◦ × 0.001◦,
• r 100m × 100m.

— Geo Pixel

Simulation Scenario

11

We outline a scenario based on the cruise ship “Eastern Star” capsizing

• n the Yangtze River, in

Jianli, Hubei, China, 1st June, 2015.

12

• Approx. 286,000 phones in the city covering an area of

373.520 km2

• The mobile phones penetration rate is 96.7% in China,

Simulation Scenario

13

The C-SPARQL query string to detect lost phones

14

• Blind Zones
• Regions have no or limited signal coverage.
• could be avoided by a simple comparison to the

list before an alert is sent.

• Multiple Phones deliberately damaged
• For example, secret dealing, illegal trading, etc.
• Worth sending alarms anyway.

Possible Reasons for False Alarms …

15

Results - 5 seconds

The number of lost phones

• f the whole city found by

all threads. The total number of lost phones in the city, shown as circles.

16

Results - 20 seconds

The number of lost phones

• f the whole city found by

all threads. The total number of lost phones in the city, shown as circles.

17

Results - 30 seconds

The number of lost phones

• f the whole city found by

all threads. The total number of lost phones in the city, shown as circles.

18

Results - Pressure Test

A pressure test is also carried

• ut to simulate an extreme

scenario in which 10 ships capsize in different locations

• n the river simultaneously,

As shown in the Fig below, the abnormality of a large number

• f lost phones in all of the 10

geo-pixels are detected, and alerts are sent successfully.

• Heterogenous telecommunication data are

concentrated and formatted.

• Continuous RDF processing for phone data

streams querying in real time.

• Simulation based on a real life shipwreck incident.

19

Conclusions