I developed a heterogeneous routing protocol and i want to add security to it using RSA. Although, symmetric key algorithm could have been better in terms of energy consumption , yet i want to try using RSA. In my protocol, the normal sensor nodes are suppose to encrypt the sensed data, but i do not know the exact energy the nodes will use to encrypt the data if one is using RSA. so i assumed it be twice the energy required for data aggregation. i want to find out if this assumption is ok or not. what is the exact energy used for only encryption. thanks
You have to be more specific about your assumptions and setting:
1- What is the key size used in RSA, i.e., 1024 bits, 2048bits or more
2- Is RSA implemented in software or hardware
3- What is the size of data you are encrypting
Typically you won't use RSA to encrypt the actual data, you will use it to encrypt an AES key, then use AES to encrypt the actual data: https://en.wikipedia.org/wiki/Hybrid_cryptosystem
See this paper for example numbers of energy consumption of some cryptographic primitives: http://palms.ee.princeton.edu/PALMSopen/potlapally03analyzing.pdf
Mr. Karim, can i get AES in MatLab so that i can try it the way you suggested.
thanks for the suggestion.
No your assumption is not right. Encryption depends on the computational power consumption "power usage per instruction execution " data aggregation depends on sensors types, communication protocols. This shows how they defers. We do not use RSA for encryption since it is math heavy algorithm. We use it for key exchange for aes or des algorithms. I have a paper in my profile that measures power consumption of different encryption algorithm in smartphones. However if you want to calculate power consumption of encryption you need 5 variables, voltage, current, CPI of the controller unit, instructions in assembly language in the algorithm and the size of your data. It is easy to obtain all of them
- Badges
- Science topic
- Similar topics
- Methodology
- Sensors