You are here

Exploiting Radio Propagation Reciprocity in Wireless Networks


OBJECTIVE: Develop the system components for utilizing the reciprocity characteristics of radio propagation to improve wireless network security and efficiency. DESCRIPTION: There is a critical military need for both increases in wireless system performance and spectral efficiency, and more distributed security techniques. In particular, the military is deploying wireless systems much more broadly and to lower echelons, which significantly complicates the process for distributing cryptographic key information. Creating a secret among authorized radios in order to encrypt or authenticate traffic is paramount to protecting personnel and missions. A scalable technical approach is needed to meet the evolving needs of military wireless communications. This technology is directly applicable to future systems developed at the Joint Tactical Networking Center (JTNC, formerly the Joint Program Executive Office of the Joint Tactical Radio System, JPEO JTRS). Wireless networks suffer from variations in received signal levels and delays due to motion of the transmitter or the receiver, as well as objects in the vicinity of either. One fundamental characteristic of radio wave propagation is that the delays and fades are the same (i.e. reciprocal) in both directions. In other words, the channel experience from radio 1 to radio 2 is the same as that experienced from 2 to 1. However, the radio frequency (RF) components and other electronics internal to the radios typically do not possess this reciprocity property. Consequently, wireless system designs typically assume no benefit from the reciprocity of the radio propagation. There is much to be gained if the system can take advantage of this property, both in terms of performance and potentially in security. System performance can be improved by lowering control overhead and enabling faster reaction times to time-varying fading because a radio can adapt its transmissions based on its own measurements of the channel rather than waiting for feedback from its intended receiver. The benefit can be even more pronounced in multi-antenna systems[1]. Security can be improved by leveraging the variations in the reciprocal channel as a secret shared (i.e. a key) among the two communicating radios in order to authenticate or encrypt transmissions between the two. Because the channel is dependent on the position of each of the radios, a third radio that is not collocated with either of the original two will not experience the same propagation fades and therefore will not be able to determine the secret key. A variety of work, both theoretical and experimental, has been performed on this topic, but a consistent set of analysis and experimentation has yet to lead to a recommended system approach. Clearly, the accuracy with which the devices can be calibrated determines the correlation of the channel estimate to the actual reciprocal channel. Most existing work focuses on the use of channel state information for a particular measurable statistic. For this effort, proposed solutions should assess the trade-off between accurate device calibration and the relevant performance metric. Proposed solutions should quantify the secrecy rate of a pair of radios exploiting channel reciprocity relative to a potential eavesdropper, as well as suggest approaches to share such a secret among several radios, including potential vulnerabilities of such approaches. In addition, the research should address the system impacts of maintaining calibration of the devices to produce a particular agreement in channel estimates. Analysis should be supported by experiments with actual radios, where the radios can be calibrated to varying degrees of fidelity. Agreement in channel estimates at participating radios should be quantified. The research should demonstrate the ability to generate a shared secret among desired radios, as well as to utilize the channel estimate to improve system efficiency. PHASE I: Propose a relevant channel model that incorporates time-varying fading and a model of RF component responses. The models should capture multiple antennas per radio and more than two radios in the system. Quantify the performance improvement and degree of secrecy as a function of the ability to calibrate radios and isolate propagation effects. Recommend approaches to adapt to multiple antennas and shared secrets among several participating radios. Evaluate system-level impact of effort to maintain sufficient agreement of channel estimates. Assess potential vulnerabilities of proposed solutions. Propose an approach to test solutions in a relevant environment. PHASE II: Develop a test environment to evaluate proposed solutions and assess vulnerabilities, including the ability to adjust the fidelity of channel state information available to the signal processing subsystem. Implement proposed solutions in test radios with sufficient data capture capabilities to quantify desired performance metrics. Test radios and proposed solutions in relevant environment with motion and interference. Adjust analysis from Phase I based on experimental results. Evolve proposed solutions to improve performance and secrecy in real RF propagation. PHASE III: Security of transmissions among radios is critical to military operations. As mobile systems become more widely deployed, more distributed key generation approaches will become necessary. In addition, information exchange requirements are increasing as access to radio spectrum is decreasing. The results of this research effort can enable scalability and efficiency of future military communications systems, and may potentially lead to an applique that can enhance existing systems in the near-term. Leveraging channel reciprocity has great potential benefits in commercial applications, both for licensed commercial cellular systems and for unlicensed wireless access points. The ever-growing demand for wireless services, particularly for mobile devices, is driving technology to be more spectrally efficient as well as adaptive to changing channel conditions. Moreover, the commercial world tends to lag the defense community in system security and the increased use of wireless devices for sensitive applications such as banking and medical monitoring has caused a gap between the security needs and capabilities of commercially available wireless systems. The solutions proposed as part of this effort can partially bridge that gap. REFERENCES: 1. Kaltenberger, et al.,"Relative Channel Reciprocity Calibration in MIMO/TDD Systems,"ICT Mobile Summit 2010, 19th Future Network & Mobile Summit, June 16-18, 2010, Florence, Italy. 2. Liu, H. et al.,"Collaborative Secret Key Extraction Leveraging Received Signal Strength in Mobile Wireless Networks,"in Proceedings of 2012 IEEE INFOCOM, March 25-30, 2012, Orlando, FL. 3. Wilson, R., D. Tse, and R. A. Scholtz,"Channel Identification: Secret Sharing using Reciprocity in Ultrawideband Channels,"IEEE Transactions on Information Forensics and Security, Vol. 2, No. 3, September 2007.
US Flag An Official Website of the United States Government