A random-amplitude-weighting security approaches at physical layer for wireless sensor networks

In this paper, a random amplitude weighting approach is proposed to guarantee the security at physical layers for distributed wireless sensor networks. Each sensor estimates the channels by training symbols sending from the expected receiver. Then the collaborative sensors cooperate in transmission signals to the receiver with weightings. Weighting coefficients with random amplitudes and fixed phases compensating the phase offsets are picked by each sensor in each transmission. With this approach, equivalent channels between the sensors and the expected receiver are almost invariant and the phase alignment is achieved. Simultaneously channels between the sensors and the eavesdropper are changing fiercely and the phase of the received signal at the eavesdropper is confusing. Hence the eavesdropper can hardly demodulate the received signals correctly and timely. Compared to the two proposed methods, our method is more suitable for wireless distributed structure in acquisition of weighting coefficients. Simulation results shows that our method can provide a better SNR performance with corresponding lower bit error rate (BER) at expected receiver.