METHOD OF TRUST MANAGEMENT IN WIRELESS SENSOR NETWORKS

• US 20090328148A1
• Filed: 07/24/2008
• Published: 12/31/2009
• Est. Priority Date: 06/30/2008
• Status: Abandoned Application

• Current Assignee: Industry-Academic Cooperation Foundation Yonsei University, Zscaler Incorporated
• Original Assignee: Industry-Academic Cooperation Foundation Yonsei University
• Inventors: SHAIKH, Riaz Ahmed, LEE, Sung Young, LEE, Young Koo
• Application Number:
  US12/178,722
• Publication Number:
  US 20090328148A1
• Time in Patent Office: Days
• US Class Current: 726/3
• CPC Class Codes: H04L 63/12 Applying verification of the received information cryptographic mechanisms or cryptographic arrangements for data integrity or data verification H04L9/32
• CPC Class Codes: H04L 67/12 specially adapted for proprietary or special-purpose networking environments, e.g. medical networks, sensor networks, networks in vehicles or remote metering networks

Abstract

The present invention relates to Group-based trust management scheme (GTMS) of wireless sensor networks. GTMS evaluates the trust of a group of sensor nodes in contrast to traditional trust management schemes that always focused on trust values of individual nodes. This approach gives us the benefit of requiring less memory to store trust records at each sensor node in the network. It uses the clustering attributes of wireless sensor networks that drastically reduce the cost associated with trust evaluation of distant nodes. Uniquely it provides not only a mechanism to detect malicious or faulty nodes, but also provides some degree of a prevention mechanism.

Claims

1. A method of trust management in wireless sensor networks, comprising the steps of:

each node in a group calculating individual trust values for all group members, assigning one of trust states based on the trust values to other nodes in the group and forwarding the assigned trust states to a cluster-head;

the cluster-head assigning a global trust state to each node based on the relative difference in trust states for the node.

2. The method according to claim 1, wherein said trust states include trusted, un-trusted and un-certain.

3. The method according to claim 2, if a node has any past experience between other nodes, then trust value is calculated using time-based past interaction.

4. The method according to claim 3, said time-based past interaction is calculated using the equation

$\begin{array}{c}{T}_{x,y}=\end{array}$

[ 100 



( S x , y S x , y + U x , y ) 



( 1 - 1 S x , y + 1 ) ] = 

[ 100 



( S x , y ) 2 ( S x , y + U x , y ) 

( S x , y + 1 ) ]

wherein [.] is nearest integer function, S_x,y is the total number of successful interactions of node x with node y during predetermined time, U_x,y is the total number of unsuccessful interactions of node x with node y during the predetermined time.

5. The method according to claim 2, if a node does not have any past experience between nodes, then trust value is calculated using peer recommendations.

6. The method according to claim 5, node x calculates the trust value of node y using

$T_{x,y}=[\frac{\sum _{}}{}$

i ∈

R x ⋃

C x 



T x , i * T i , y 100 * j ] ;

$j=<br><br>R_x\bigcup <br><br>C_x<br><br>\le <br><br>n-2$

where, [.] is the nearest integer function, T_x,i is the trust value of recommender, and T_i,y is the trust value of node y sent by node i assuming that j nodes are trusted and uncertain in a group.

7. The method according to claim 2, each node will quantize trust values into three states using the following equation:

$\mathrm{Mp}<br><br>\left(T_{x,y}\right)=\left\{\begin{array}{cc}\mathrm{trusted}& 100-f\le <br><br>T_{x,y}\le <br><br>100\\ \mathrm{uncertain}& 50-g\le <br><br>T_{x,y}<<br><br>100-f\\ \mathrm{untrusted}& 0\le <br><br>T_{x,y}<<br><br>50-g\end{array}\right\}$

wherein f is the half of the average values of all trusted nodes, g is the one third of the average values of all untrusted nodes.

8. The method according to claim 7, f is calculated using

$\begin{array}{c}{f}_{j+1}=\{\begin{array}{c}[\frac{1}{2}\end{array}\end{array}$

( ∑

i ∈

R x 



T x , i 

R x 

) ] 0 <



R x 

≤

n - 1 f j 

R x 

= 0 



and 



g 



is 



calculated 



using g j + 1 = { [ 1 3 

( ∑

i ∈

M x 



T x , i 

M x 

) ] 0 <



M x 

≤

n - 1 g j 

M x 

= 0

wherein R_x represents the set of trusted nodes for node x, M_x is the set of un-trusted nodes for node x, and n is the total number of nodes that contains trusted.

9. The method according to claim 2, said cluster-head defines a random variable X such that

$X$

( s i , j ) = { 2 when s i , j = trusted 1 when s i , j = un 

- 

certain 0 when s i , j = un 

- 

trusted

and for m such random variables said cluster-head defines the following standard normal random variable for a node j

$Z_j=\frac{\sqrt{3}<br><br>\left(X<br><br>\left(s_{\mathrm{ch},j}\right)+\underset{i=1,i\ne <br><br>j}{\overset{m}{\sum <br><br>}}<br><br><br><br>X<br><br>\left(s_{i,j}\right)-m\right)}{\sqrt{m}}$

wherein S_ch,j represents the state of node j at cluster-head ch,

and said cluster-head assigns the global trust state to each node such that if Z_j□

[−

1, 1] then the node j is assigned to be un-certain, if Z_j>

1, trusted, and if Z_j<

−

1, un-trusted.

10. The method according to claim 2, a base station receives from each cluster-head the trust values of other cluster-heads and calculates the trust value of each group using

$T_{\mathrm{BS},G_1}=[\frac{\sum _{}^{}}{}$

i = 1 

G 

- 1 



( T BS , ch i ) 

( T G i , G 1 ) 

G 

- 1 ] , …



, 

T BS , G m = [ ∑

i = 1 

G 

- 1 



( T BS , ch i ) 

( T G i , G 

G 

) 

G 

- 1 ]

wherein T_BS,chi is the trust value of the cluster-head i at the base station, T_gi,G1 is the trust value of group G₁ at group G_i and |G| represents the number of groups in the network.

11. The method according to claim 10, the cluster-head requests the base station for the trust value of another cluster-head if it does not have any record of past interactions within a predetermined time window with said another cluster-head.

12. The method according to claim 10, the base station assigns one of the three possible states to the whole group based on the trust values.

13. The method according to claim 2, the cluster-head is a sensor node that has higher computational power and memory as compared to other sensor nodes.

14. The method according to claim 2, the cluster-head periodically broadcast the trust state request packet within the group.

15. The method according to claim 10,

the base station periodically multicast request packets to the cluster-heads, and

on request, the cluster-heads forward their trust vector related to the recommendations of other groups based on past interactions to base station as follows

T_ch=(T_ch,1,T_ch,2, . . . ,T_ch,|G|−

1)

wherein T_ch,i represents the trust value of another cluster j calculated at cluster-head ch.