Citation Context

...odes transmit their decisions to the fusion center. Based on the received sensor decisions, the fusion sensor makes the final decision u0 and the optimum fusion rule in this case is a k-out-of-n rule =-=[13]-=-, [14]. Letting u0 0 if the fusion sensor decides H0 and letting u0 1 if the fusion sensor decides H1, we have u0 1; u1þ þun k ð2Þ 0; u1þ þun <k; where k is an integer between 1 and n. In our as...

Citation Context

... reputation-based classifier (RBC) algorithm [12], [26]. In RBC when the network parameters (e.g., the nodes’ operating points) are known, the optimal q-out-of-L rule can be computed (see for example =-=[16]-=-, [27]). However, when the FC is not aware of all the network parameters as is the case here, majority rule has been used in [12] and is also used here for our comparisons. In addition, in (45) the th...

Citation Context

...[6]. In particular, optimal design of the fusion rule under different conditions of quantization at the individual sensors, topologies of the network, and channel conditions have been investigated in =-=[7]-=-–[11]. Mission-critical applications of WSNs involving distributed detection demand operational security [11]. On the other hand, networks must cope with insecure links. Due to the limited power and l...

Citation Context

...given by k A = ( ) θ1 ln Λ − n ln θ0 ( ) (2.3) (1 − θ1)θ0 ln (1 − θ0)θ1 On the other hand, for the TPFC, the optimum value of k is given by k TP = k A (λ, 0, 0), which the TPFC calculates as given in =-=[29]-=- because of the lack of knowledge about the presence of the stochastic cipher in each sensor. Hence the fusion rule can be generalized as follows: ⎧ n∑ ⎪⎨ 1, if zi ≥ k i=1 u0 = n∑ ⎪⎩ 0, if zi < k i=1 ...

Citation Context

...PU is not there. 978-1-4244-7213-0/10/$26.00 ©2010 IEEEThe usual data fusion rule for hard decisions is “K out of N” rule, meaning that at least K advices have to be H1, so the global decision is H1 =-=[6]-=-. The global decision is H1 by a particular CR user to access a resource when X=1. Otherwise, it is H0 when X=0. We assume that N cooperators exist. However, not all of them are necessary for cooperat...

Citation Context

...ws: q : ℜ−→{q1,q2, ··· ,qM} Where Q(X) =qi, if ti−1 ≤ X ≥ ti where t0 = −∞, tM+1 = ∞. 5output q 8 q 7 q 6 q 5 input t 1 t 2 t 3 t 5 t 6 t 7 q 4 q 3 q 2 q 1 FIGURE 1.4. Quantizer In [31], [32], [34], =-=[36]-=- and [33], the observations made at the sensors are quantized to a single bit which may be considered as the sensors local decisions. In [33] the author shows that using likelihood-ratio test as the l...

Citation Context

...smit their decisions to the fusion sensor. Based on the received sensor decisions, the fusion sensor makes the final decision u0 with the optimum fusion rule which, in this case, is a k-out-of-n rule =-=[3]-=-. Let u0 = 0 if the fusion sensor decides H0 and let u0 = 1 if the fusion sensor decides H1,wehave, { 1, u1 + ···+ un ≥ k u0 = (2) 0, u1 + ···+ un < k where k is an integer between 1 and n.For all th...

Citation Context

...as focused on a WSN with a large number of sensors which is based on a specific signal attenuation model, and the problem of designing an optimum local decision rule has been investigated. In [6] and =-=[7]-=- the problem of binary hypothesis testing using binary decisions from independently and identically distributed sensors is studied, and the optimal fusion rules are obtained. In this paper we study th...