Enhanced Secure Key Exchange Systems Based on the Johnson- Noise Scheme

Abstract

We introduce seven new versions of the Kirchhoff-Law-Johnson-(like)-Noise (KLJN) classical physical secure key exchange scheme and a new transient protocol for practically-perfect security. While these practical improvements offer progressively enhanced security and/or speed for non-ideal conditions, the fundamental physical laws providing the security remain the same.In the "intelligent" KLJN (iKLJN) scheme, Alice and Bob utilize the fact that they exactly know not only their own resistor value but also the stochastic time function of their own noise, which they generate before feeding it into the loop. By using this extra information, they can reduce the duration of exchanging a single bit and in this way they achieve not only higher speed but also an enhanced security because Eve's information will significantly be reduced due to smaller statistics.In the "multiple" KLJN (MKLJN) system, Alice and Bob have publicly known identical sets of different resistors with a proper, publicly known truth table about the bit-interpretation of their combination. In this new situation, for Eve to succeed, it is not enough to find out which end has the higher resistor. Eve must exactly identify the actual resistor values at both sides.In the "keyed" KLJN (KKLJN) system, by using secure communication with a formerly shared key, Alice and Bob share a proper time-dependent truth table for the bit-interpretation of the resistor situation for each secure bit exchange step during generating the next key. In this new situation, for Eve to succeed, it is not enough to find out the resistor values at the two ends. Eve must also know the former key.The remaining four KLJN schemes are the combinations of the above protocols to synergically enhance the security properties. These are: the "intelligent-multiple" (iMKLJN), the "intelligent-keyed" (iKKLJN), the "keyed-multiple" (KMKLJN) and the "intelligent-keyed-multiple" (iKMKLJN) KLJN key exchange systems.Finally, we introduce a new transient-protocol offering practically-perfect security without privacy amplification, which is not needed in practical applications but it is shown for the sake of ongoing discussions.

Random-Resistor-Random-Temperature Kirchhoff-Law-Johnson-Noise (RRRT-KLJN) Key Exchange

“…Consequently these resistor sets and temperatures are deterministic, which in accordance with Kerckhoffs's principle [23] implies that all of the parameters are known by Eve. (Note that "keying" these parameter values by randomly generating and disseminating them via secure communication, by using the formerly shared key, is of course possible, just as it is the case of the Keyed-KLJN scheme [5] and some quantum versions [24,25], but such enhancements to make Eve's job more difficult are not the topic of the present paper).…”

“…A disadvantage of the RRRT-KLJN scheme is that the Kish-Granqvist temperature-compensation defense mechanism [16] cannot be used to nullify cable resistance effects against an as yet unknown attack type of such kind. A perhaps more practical version of the RRRT-KLJN scheme is the generalization of the formerly proposed Multiple-KLJN (MKLJN) system [5] based on a random choice of a known large set of resistors by introducing a random choice of temperatures from a known large set of such data. As already mentioned above, the known sets must be properly checked, because only choices with degenerated voltage/current/power values can be considered secure-not the singular values.…”

On the “Cracking” Scheme in the Paper “A Directional Coupler Attack Against the Kish Key Distribution System” by Gunn, Allison and Abbott

“…Here it is important to realize that the cable voltage and cable current are orthogonal-i.e., uncorrelated-in order to ensure zero net power flow and satisfy the Second Law of Thermodynamics [7,[12][13][14][15], so that U(t) I(t) 0 .…”

A Static-Loop-Current Attack against the KLJN Secure Key Exchange System

“…None of them could impair the unconditional security of the scheme, since either there was a proper defense scheme against or attack was based on misconceptions [3][4][5][6][7][8][9][10][11][12][13][14][15][16][17][18][19][20][27][28][29][30], including experimental errors. On the other hand, passive attacks against practical KLJN systems deviating from the ideal assumptions of the scheme [2] were discussed in several papers [4][5][6][7][8][9][10][11][12][13]. For example, parasitic or non-ideal features such as transients, wire's resistance, cable capacitance, temperature differences, delay effects and transients, etc.…”