Automatic Dependent Surveillance Broadcast (ADS-B) has been proposed as both an extension and an alternative to Primary and Secondary Surveillance RADAR. Although ADS-B has many advantages, security was not a key issue in its design. Packets are sent in clear text, making it vulnerable to many attacks. A main concern is that some of these attacks can be easily implemented using inexpensive ADS-B transmitters. For instance, ADS-B is vulnerable to injection attacks, which introduce ghost aircraft into an surveillance system. In this paper we propose a method that addresses these issues by detecting malicious ADS-B transmitters using a network of sensors with an associated data fusion center. A key aspect of our solution is that the data fusion process has a low message exchange overhead, enabling its applicability to current constrains of ADS-B sensors. Our method can be implemented using commercial off-the-shelf ADS-B receivers.