Physically unclonable functions using pulse width chaotic maps
Technologies are provided for clockless physically unclonable functions (PUFs) in reconfigurable devices. Embodiments of the disclosed technologies include processing circuitry configured to perform numerous operations. The operations can include receiving a challenge continuous pulse signal, and generating a response continuous pulse signal by iteratively extending the challenge continuous pulse signal in time-domain. In some configurations, the iteratively extending includes generating a next continuous pulse signal by operating on a prior continuous pulse signal according to a stretching function, and generating a second next continuous pulse width signal by operating on the next continuous pulse signal according to a folding function.
1. A method, comprising:
receiving, by a device, a challenge continuous pulse signal;
generating a response continuous pulse signal by iteratively extending the challenge continuous pulse signal in time-domain, wherein the iteratively extending comprises,
generating a next continuous pulse signal by operating on a prior continuous pulse signal according to a stretching function, the prior continuous pulse signal having a prior width in time-domain and the next continuous pulse signal having a next width in time-domain that is greater than the prior width in time-domain; and
generating a second next continuous pulse width signal by operating on the next continuous pulse signal according to a folding function, the second next continuous pulse width signal having a second next width in time-domain that is less than a threshold time interval.
2. The method of claim 1 , wherein the challenge continuous pulse signal is iteratively extended for a defined number of iterations, the method further comprising outputting the response continuous pulse signal after the defined number of iterations.
3. The method of claim 1 , wherein operating on the prior continuous pulse signal according to the stretching function and operating on the next continuous pulse signal according to the folding function correspond to operations of a chaotic map, wherein the chaotic map comprises one of a Bernoulli shift map, a tent map, or a logistic map.
4. The method of claim 1 , wherein the operating on the prior continuous pulse signal according to the stretching function comprises extending the prior width in time-domain by a multiplicative factor greater than unity.
5. The method of claim 4 , wherein the device comprises programmable processing circuitry constituting a field-programmable gate array (FPGA) or an application-specific integrated circuit (ASIC).
6. The method of claim 1 , wherein the operating on the next continuous pulse signal according to the folding function comprises,
determining that the next width in time-domain is less than the threshold time interval; and
passing through the next continuous pulse signal as the second next continuous pulse width signal, wherein the second next width in time-domain is equal to the next width in time-domain.
7. The method of claim 1 , wherein the operating on the second next continuous pulse width signal according to the folding function comprises,
determining that the second next width in time-domain is greater than the threshold time interval; and
generating a third defined width in time-domain by subtracting a defined time offset from the second next width in time-domain.
8. An apparatus, comprising:
processing circuitry configured to perform operations comprising:
receiving, by the apparatus, a challenge continuous pulse signal;
generating a response continuous pulse signal by iteratively extending the challenge continuous pulse signal in time-domain, wherein the iteratively extending comprises,
generating a next continuous pulse signal by operating on a prior continuous pulse signal according to a stretching function, the prior continuous pulse signal having a prior width in time-domain and the next continuous pulse signal having a next width in time-domain that is greater than the prior width in time-domain; and
generating a second next continuous pulse width signal by operating on the next continuous pulse signal according to a folding function, the second next continuous pulse width signal having a second next width in time-domain that is less than a threshold time interval.
9. The apparatus of claim 8 , wherein the challenge continuous pulse signal is iteratively extended for a defined number of iterations, the operations further comprising outputting the response continuous pulse signal after the defined number of iterations.
10. The apparatus of claim 8 , wherein operating on the prior continuous pulse signal according to the stretching function and operating on the next continuous pulse signal according to the folding function correspond to operations of a chaotic map, wherein the chaotic map comprises one of a Bernoulli shift map, a tent map, or a logistic map.
11. The apparatus of claim 8 , wherein the operating on the prior continuous pulse signal according to the stretching function comprises extending the prior width in time-domain by a multiplicative factor greater than unity.
12. The apparatus of claim 11 , wherein the apparatus comprises programmable processing circuitry constituting a field-programmable gate array (FPGA) or an application-specific integrated circuit (ASIC).
13. The apparatus of claim 8 , wherein the operating on the next continuous pulse signal according to the folding function comprises,
determining that the next width in time-domain is less than the threshold time interval; and
passing through the next continuous pulse signal as the second next continuous pulse width signal, wherein the second next width in time-domain is equal to the next width in time-domain.
14. The apparatus of claim 8 , wherein the operating on the second next continuous pulse width signal according to the folding function comprises,
determining that the second next width in time-domain is greater than the threshold time interval; and
generating width defined width in time-domain by subtracting a defined time offset from the second next width in time-domain.
15. A method comprising:
(i) receiving a challenge signal that defines parameters of components of a device, an input width of a pulse, or a combination of the two;
(ii) receiving an input pulse of a defined width;
(iii) performing iterative stretching and folding operations on the input pulse using the components of the device to create a chaotic map that operates on the input pulse using the components of the device;
(iv) transforming one or more output signals of the iterative stretching and folding process into a group of output binary bits;
(v) based at least on the group of output binary bits, determining bits of the group of output binary bits that are unique across nominally identical implementations of the device and its processes;
repeating (i) to (v) at least one time to construct keys or tables of the bits that can be used to identify the device among its nominally identical counterparts;
receiving, by an apparatus, a challenge continuous pulse signal;
generating a response continuous pulse signal by iteratively extending the challenge continuous pulse signal in time-domain, wherein the iteratively extending comprises,
generating a next continuous pulse signal by operating on a prior continuous pulse signal according to a stretching function, the prior continuous pulse signal having a prior width in time-domain and the next continuous pulse signal having a next width in time-domain that is greater than the prior width in time-domain; and
generating a second next continuous pulse width signal by operating on the next continuous pulse signal according to a folding function, the second next continuous pulse width signal having a second next width in time-domain that is less than a threshold time interval.
16. The method of claim 15 , wherein operating on the prior continuous pulse signal according to the stretching function and operating on the next continuous pulse signal according to the folding function correspond to operations of the chaotic map, wherein the chaotic map comprises one of a Bernoulli shift map, a tent map, or a logistic map.
17. The method of claim 15 , wherein the operating on the prior continuous pulse signal according to the stretching function comprises extending the prior width in time-domain by a multiplicative factor greater than unity.