In the world of wireless communications, the concept of the buffer-aided relay usually generates interest because some believe it could improve connectivity and eliminate the familiar “sorry-can-you-hear-me-now?” moments resulting from weak signals. But could it also hold the key to more-secure communications?

In a paper recently published in IEEE Transactions on Wireless Communications, Jing Huang and A. Lee Swindlehurst from the University of California-Irvine use statistical signal processing theory to show that buffer-aided relays could protect not only against spotty channels, but also against eavesdroppers. Their findings could pave the way for more secure transmission at a time when wireless innovation moves faster than data security.

In the paper, Huang and Swindlehurst propose a scheme to make the half-duplex two-hop relay network – an advanced physical architecture that could soon support the popular LTE wireless network – more secure.

Since encryption keys can be vulnerable in wireless environments, we may need to consult the physical layer to address growing security concerns.

Consider this diagram, in which Alice and Bob represent two wireless devices communicating with each other, and Eve is an eavesdropper (naturally):




Alice’s data is initially transmitted to the relay, which then re-transmits it to Bob during a designated “time slot.” As anyone who has ever taken a phone call from an obstructed area knows, the signals are sometimes weak. One new solution to improving this would be the use of a buffer. The buffered relay can favor certain criteria and optimize throughput regardless of time slots, storing data until a weak channel becomes sufficiently strong for transmission.

That doesn’t do much, however, to prevent Eve from waiting for the right time to catch the data.

In order to deter eavesdropping, Huang and Swindlehurst propose programming the buffer relay to transmit when channel Yb is relatively stronger than channel Ya and Ye. Using the theory of statistical signal processing to make the case, this link selection policy could increase the likelihood that the data will get to Bob, and only Bob. Eve, or any shadowy eavesdropper trying to tap the system, will be graced with nothing more than artificial mush.

“Since encryption keys can be vulnerable in wireless environments, we may need to consult the physical layer to address growing security concerns,” said Huang. “Buffered relays can act as natural helpers, and our model could offer a reliable and relatively inexpensive solution for a more secure wireless network.”

So why hasn’t this method been considered previously? Relays are standardized in only certain types of advanced LTE networks, and they haven’t yet been commercially deployed; this is due to their early development and complexity, which currently creates overhead. If and when relays do join the network, they will no doubt be equipped with buffers against weak signals, and – if we’re lucky – security threats.

You can also find more articles about “Relay Communications” in IEEE Xplore.