Optimal Adaptive Antijamming in Wireless Sensor Networks

Wireless sensor networks have been widely applied to various application domains such as environmental monitoring and surveillance. Because of reliance on the open transmission media, a sensor network may suffer from radio jamming attacks, which are easy to launch but difficult to defend. Attacked by jamming signals, a sensor network may experience corrupted packets and low network throughput. A number of defense techniques have been proposed. However, each defense technique is suitable for only a limited range of network and jamming conditions. This paper proposes an adaptive approach to antijamming for sensor networks by combining the strength of state-of-the-art antijamming techniques, which enables each node to adaptively select the optimal antijamming technique for different jamming conditions. The great challenge is that the sensor network undergoes varying jamming conditions over time. We address this challenge by formulating the antijamming problem of the sensor network as a Markov decision process and propose an efficient algorithm for computing the best antijamming strategy. By comprehensive simulation experiments, we demonstrate that a sensor network using the derived antijamming strategy can well defend from radio jamming attacks and in the meanwhile retain high energy efficiency. 1. Introduction With the appealing characteristics of low-cost, easy to deploy, and unattended operation and the ability of withstanding harsh environmental conditions, wireless sensor networks have been implemented in a wide range of applications, such as environment monitoring [1] and event detection [2]. Sensor networks transmit wireless signals over the open shared media. This leaves a sensor network vulnerable to radio jamming attacks. In [3, 4], several jamming attacks have been explored, which corrupt control packets, such as RTS (Request-to-Send) and CTS (Clear-to-Send). The jammer just keeps sending packets like RTS to prevent transmission of legitimate packets. These methods are usually based on the statistics of packet transmission history and can cause severe damage to the sensor network with only modest overhead. Thus, antijamming is enormously important for secure operation of sensor networks. As being well known, sensor nodes are typically powered by batteries and hence limited in power supply. This has been generally accepted as one of the crucial issues of the sensor network. Therefore, antijamming needs to be energy efficient. A number of antijamming methods have been proposed, such as channel surfing [5], error correction codes and transmission