Merge pull request ningli6#14 from ningliVT/web

Change introduction on the demo page

Readme.md

@@ -1,18 +1,10 @@
 ## Introduction
-This project implements techniques for protecting privacy of primary users proposed in paper: Protecting the Primary Users’ Operational Privacy in Spectrum Sharing. [2014 IEEE International Symposium on Dynamic Spectrum Access Networks (DYSPAN), p 236-47, 2014]
-### Web interface description
-User need to specify number of channels that primary users operate on. Then draw an analysis region on Google Map and select locations for primary users on different channels. The region should cover all primary users on any channel. Each channel must have at least one primary user. This web application is able to plot 3 different kinds of figures: Inaccuracy vs Queries, heat map on Google map, trade-off curve/bar. Users need to specify number of queries. Query locations are  either uniformly randomly generated or generated based on more sophisticated algorithm which selects next query location wisely.
-### Protection techniques
-We compute inaccuracy for each channel to evaluate the result of inference as a metric for our protecting algorithms.
-* First technique for protecting location privacy is perturbation with additive noise. Whenever response the server will lie to the querying client in a way that it will chooses to respond with a smaller value if possible.
+This project implements techniques for protecting privacy of primary users proposed in paper: Protecting the Primary Users’ Operational Privacy in Spectrum Sharing. 
 
-* Second technique for protecting privacy is to use transfiguration, which transfigure circular contours into convex polygon. It works pretty good for small numbers of sides, such as triangle and rectangle.
+[2014 IEEE International Symposium on Dynamic Spectrum Access Networks (DYSPAN), p 236-47, 2014](http://ieeexplore.ieee.org/xpl/abstractAuthors.jsp?arnumber=6817800)
 
-* The third technique is K-anonymity, which groups PUs into a group of k members, such that they share a larger area of protection contour, then compute a virtual primary user whose protection contour is the minimum covering circle for all other PUs in that group. Use the virtual PU as the new PU to respond to new queries. Poor grouping decision can lead to poor spectrum utility performance.
+## Installation
+For instructions on setting up and server and installation please refer to our wiki page. 
 
-* The fourth technique is K-Clustering, which uses a k-means algorithm to classify PUs into different group according to their location distribution, then compute a virtual primary user whose protection contour is the minimum covering circle for all other PUs in that cluster. Use the virtual PU as the new PU to respond to new queries.
-
-### Query options
-User can select random query and/or smart query in the process of simulation. Generally random query is much faster and works better with countermeasures. Smart query usually brings down inaccuracy really quick but it's computationally demanding so it can't operate on a larger map. Smart query sometimes won’t cooperate with countermeasures.
 ### Author
 *Ning Li([email protected])*

web/index.html

@@ -45,16 +45,12 @@
     </nav>
 
     <div class="jumbotron" style="margin-top: 30px !important;">
-        <h2>Moto demo</h2>      
+        <h2>Web-based Simulation Tool for Evaluating incumbent user's location Privacy in Spectrum sharing (STEPS)</h2>      
         <p>
-            This is a demo of techniques for protecting the Primary Users’ operational privacy in spectrum sharing 
-            proposed in the paper: 
+            Although using geolocation databases for enabling dynamic spectrum sharing has a number of practical advantages, it also raises potentially serious operational security (OPSEC) and privacy issues.  For instance, secondary users (SUs), through seemingly innocuous  queries to the database, may be able to determine or infer the operational attributes and location of incumbent user (IU) systems operating in a given area of interest.  It is quite possible for adversarial SUs to obtain knowledge beyond what is revealed directly by the database's query replies by using database inference techniques -- this is referred to as a database inference attack.  This OPSEC issue has important implications in the deployment and  operation of database-driven spectrum sharing ecosystems, particularly those that involve spectrum sharing between federal government, including military, communications/radar systems and commercial wireless systems. 
         </p>
         <p>
-            <cite>
-                <a href="http://ieeexplore.ieee.org/xpl/abstractAuthors.jsp?arnumber=6817800" target="_blank">Protecting the Primary Users’ Operational Privacy in Spectrum Sharing.</a>
-                <small>[2014 IEEE International Symposium on Dynamic Spectrum Access Networks (DYSPAN), p 236-47, 2014]</small>
-            </cite>
+            This tool, which is referred to as STEPS (Simulation Tool for Evaluating incumbent user's location Privacy in Spectrum sharing), can be used to visualize the potential impact of database inference attacks as well as generate quantitative results that can be used to evaluate and compare various countermeasures for thwarting such attacks.  The Instructions page provides detailed instructions on how to use this tool.  Examples of two different simulation scenarios are described in Example A and Example B pages.  
         </p>
     </div>
 
@@ -286,7 +282,7 @@ <h4>Email</h4>
 
     <br>
     <!-- result of passed params -->
-    <button type="submit" class="btn btn-primary btn-lg" style="float: right;" onclick="getParams();">Start demo <span class="glyphicon glyphicon-play"></span></button>
+    <button type="submit" class="btn btn-primary btn-lg" style="float: right;" onclick="getParams();">Start simulations <span class="glyphicon glyphicon-play"></span></button>
     <br><br><br><br>
 </div>