TRUST MANAGEMENT SYSTEM DESIGN FOR THE INTERNET OF THINGS: A CONTEXTAWARE AND MULTI-SERVICE APPROACH Yosra Ben Saied, Alexis Olivereau, Djamal Zeghlache, Maryline Laurent Presented by Ali Asgar Sohanghpurwala INTRODUCTION Machine to Machine (M2M) and Internet of Things (IoT) architectures becoming prevalent Wireless Sensor Networks (WSNs) introduced unattended wireless topologies with resource constrained nodes IoT expands on WSN requirements Wider architectures More heterogeneous Inconstant resource capabilities Increased autonomy WHY DOES IOT NEED A TMS? Nodes expected to securely communicate with external Internet nodes, but likely don't have resources to do it alone Collaboration needs to be controlled, to protect against attacks Cr yptographic methods don't account for insider attacks Constraints such as computing power, battery life, limited bandwidth Need to collaborate to meet this goal Cooperative techniques for routing and security have been proposed in literature Cryptographically trusted nodes can lie, alter data, or selfishly refuse to collaborate Existing WSN and MANET insuf ficient for IoT HOW IS TMS DIFFERENT FOR IOT? IoT nodes providing different services assessed by same TMS Non-malicious nodes may temporarily have low capabilities IoT nodes are highly heterogeneous Node owned by multiple self-interest communities Complex malicious patterns arise with coexistence of heterogeneous and selfconcerned nodes ASSESSMENT OF PRIOR TMS WRT IOT OVERVIEW Use past behavior to determine task -specific trust levels for each node Eventually only the best partners for a specific ser vice are proposed to requesting node Fine-tune trust levels, even in presence of malicious and erroneous nodes Geographically centralized TM ser vers Multi-phase approach INITIALIZATION AND INFORMATION GATHERING Initially all nodes are assumed trustworthy Bootstrapping period is required to gather information before results are trustworthy Trust manager speeds up process by targeting nodes and inducing artificial interactions Requesting node classifies behavior of assisting node as positive or negative Evaluations are stored in trust manager Context under which evaluations are received is important Aging, resource capacity, etc. of evaluated node Execution time REPORT INFORMATION Each report R ij refers to jth report regarding QoS for assisting node P i Each report contains the following information: ENTIT Y SELECTION When a node asks for assistance, the trust manager returns a list of trustworthy assisting nodes Five steps: 1) Restrict set of proxies p i 2)Restrict the set of reports R ij for each proxy P i 3)Compute weights (w Rij) for each report R ij 4)Compute trust value T i for each proxy p i 5)Provide requestor with list of best suited proxies ENTIT Y 1: RESTRICT SET OF PROXIES P I Select candidates based on service requirements Examples: Lightweight communication may require nodes in same multi-cast group Signature delegation schemes may require nodes dispersed in specific locations May require neighbors in radio range ENTIT Y 2: RESTRICT SET OF REPORTS Find most meaningful reports for prospective nodes Ideal reports: Assisting node provided the same service Assisting node status was the same as it is now It is likely that there won't be enough ideal reports to judge the node p i in specific context We can calculate context similarity by quantifying node capabilities and service similarity ENTIT Y 2:QUANTIFY PARAMETERS Quantifying node capability is easy: Percentage of Battery, CPU power, Memory available Service similarity isn't as straightforward Estimate service similarity based on resource requirements Of measurable resources, energy consumption is recommended by authors ENTIT Y 2: CONTEXT SIMILARIT Y Report R ij sent by all nodes j, regarding interactions with node P i contains: S j – service provided by C j – capability N j – Note Try to match with target values: S target – Current service in request C target – Current P i capability ENTIT Y 2: CONTEXTUAL DISTANCE dS max , dC max - tolerance of selection mechanism for capability and ser vice measurements First term represents distance from center of (( S target ,C target ), dS max , dC max ) ellipse Node that behaves well for expensive ser vice, is likely to behave well for less demanding ser vice Second term represents distance between R ij and (S max , 0) Node per forming well for low -demand ser vice, doesn't mean it per forms well in demanding ser vice Second term represents distance between R ij and (0, C max ) ENTIT Y 2: D IJ ILLUSTRATION Positive report close to ( S max , 0) means node per formed well for expensive ser vice while near min capacity Negative Report close to (0, C max ) means that node per formed poorly for simple ser vice, while at max capacity Any report close to center of target ellipse is ver y similar Retained report R ij should have d ij such that: d ij (R ij , R Target ) < t, where: ENTIT Y 2: EXAMPLE ENTIT Y 2: EXAMPLE CONT. ENTIT Y 3: COMPUTE WEIGHT FOR EACH REPORT Weight of each report (w ij ) determined by contextual distance (d ij ) and age (t now - t j ) λ,θ are parameters in range [0,1] expressing 'memory' of the system θ (resp. λ) is adjusted according to expected rapidity of change Lower θ (resp. λ) indicates lower importance for past reports (resp. more contextually distant reports) s = ½ * (N 2 j -N j ), where N j is the score given by witness s = 1 when score is -1, and 0 when score is 0,1 weight of negative score is doubled compared to positive or neutral scores ENTIT Y 4: COMPUTE TRUST VALUE FOR EACH PROXY T i is trust value for proxy p i QR j is the quality of recommendation of witness node j trustworthiness score based on accuracy of past reports Ranges between -1 and 1 w Rij is weight from previous slide ENTIT Y 5: PROVISION BEST RATED PROXIES OF P I Securely send list of best rated nodes to requestor Finally done with Entity selection TRANSACTION AND EVALUATION Client node relies on list of trusted proxies provided by Trust Manager to select partners Sends positive or negative score for each partner to TM Evaluation technique depends on service provided Could be direct observation, or could solicit feedback from peers Received reports should take into account node credibility LEARNING Learning phase qualifies system as a cognitive process In security scenarios: Adaptive security systems dynamically react by applying new security policies in reaction to environment change Cognitive security introduces learning step. Assessment of enforced action eventually modifies system behavior so a different action may be taken next time. LEARNING STEPS 1)Update witness nodes' qualities of recommendation 2)Update of assisting nodes' reputation levels LEARNING 1: UPDATE QUALIT Y OF RECOMMENDATION Simple Concept Decrease QR score if witness gave a 'bad' score to a 'good' node Increase QR score if witness gave a 'good' score to a 'good' node Use weighted average avoids excessive variations of QR allows precise choice of extent to which QR must be oriented towards 1 (good recommender), 0 (non-usable data), and -1 (bad recommender) LEARNING 1: QR DIRECTION Let X be a witness node who evaluated node P, which later provided a ser vice to node F Node F sends report R F with score N {-1 ,0,1} TM uses F's report to update QR score of each recommender System retrieves n stored QR scores for all witness nodes X has for example QR x (QR 1 ,...,QR n-1 , Qr n ) System extracts note N from R F and retrieves w Rx corresponding to R X QR X F represents direction that QR for X should evolve r = 1 if X and F agree on rating, r= -1 when they are opposite, and r=0 when they are of f by 1 C F is weight of r, increases when weight of X's previous report is high, increases if F is a good recommender LEARNING 1: WEIGHTED AVERAGE QR i represents X's rec. history C i is respective weight for X's rec. history θ represents 'memory' factor of system Negative N_QR means X is reporting opposite of actual service quality Instead of discarding X's ratings, consider the opposite! LEARNING 2: UPDATE ASSISTING NODES' REPUTATION LEVELS Reputation distinct from trust Trust measures ability of node to perform specific task Reputation refers to overall trustworthiness of node in the system Combine QR (QR Fj )and ser vice ratings (N Fj ) with age-based weighting factor TM recalculates reputation af ter each interaction Nodes with low -reputation are added to blacklist TEST SIMULATION SIMULATION PARAMETERS QR EVOLUTION QR EVOLUTION CONT. ATTACK RESILIENCE ATTACK RESILIENCE CONT. CONCLUSION Presented generic, context-aware TMS for IoT Dynamic trust scores assigned to nodes based on node status, and required function Independent score given for Quality of Recommendation QR score is adjusted through learning phase System withstands several classes of attacks