Free Trial

Safari Books Online is a digital library providing on-demand subscription access to thousands of learning resources.

Share this Page URL
Help

10.5.2 Sacrificial Node-Based Defense Ag... > Sacrificial Nodes Selection - Pg. 264

264 CHAPTER 10 Defending Against Physical Attacks in Wireless Sensor Networks R noti R noti s 7 R noti s 2 s 5 s 6 : Attacker : Sensor detects the attacker and sends an AN message to all other sensors in its notification area. The noti- fication message contains the global ID of sensor s 1 and the notification area is a circle of radius R noti centered at sensor s 1 . In our defense proto- col, we let R noti be the same as the communication range for the sensors. Recall that the attacker is generally more powerful than a sensor in terms of sensing ability, as such, sensor s 1 is quite likely to be detected by the attacker already when it detects the attacker. Thus, it is better for sensor s 1 to send out AN message instead of switching to a sensing/sleeping state. After sending out the AN message, sensor s 1 will stay in a sending state. For the recipients of the AN message sent by sensor s 1 , which are sensors s 2 , s 3 , and s 4 in Fig- ure 10-2, we assume that sensors s 2 and s 3 decide to be sacrificial nodes, whereas sensor s 4 does not. Sensors s 2 and s 3 will each send out an SN mes- sage at different times. In our protocol, we apply s 3 s 1 s 4 FIGURE 10-2 Protocol description. to sleeping state for two reasons. First, this will result in a large coverage loss, which is an overkill because the attacker will only choose to move in one direction after destroying sensor s 1 . Second, they are already in the protection area of sensor s 2 . They may be notified of the approaching of the attacker by sensor s 2 before their own passive