Forward-private Sublinear DSSE (FS-DSSE)

Basic implementation of FS-DSSE. The full paper is published in IEEE ICC 2018 and available at https://eprint.iacr.org/2017/1222.pdf.

This project is built on CodeLite IDE (link: http://codelite.org). It is recommended to install CodeLite to load the full FS_DSSE workspace.

Required Libraries

1. ZeroMQ (download link: http://zeromq.org/intro:get-the-software)

2. Libtomcrypt (download link: https://github.com/libtom/libtomcrypt)

3. Google sparsehash (download link: https://github.com/sparsehash/sparsehash)

4. Intel AES-NI (optional) (download link: https://software.intel.com/en-us/articles/download-the-intel-aesni-sample-library)

Intel AES-NI installation guide (optional)

FS_DSSE leverages Intel AES-NI to accelerate cryptographic operations. The Intel-AES-NI is available in Intel® Core™ i5, Intel® Core™ i7, Intel® Xeon® 5600 series and newer processor (see https://ark.intel.com/Search/FeatureFilter?productType=processors&AESTech=true for a complete list). This functionality can be disabled to test FS_DSSE with other CPU models (see the Configuration Section below). Here the brief instruction to install Intel-AES-NI:

1. Extract the .zip file downloaded from https://software.intel.com/en-us/articles/download-the-intel-aesni-sample-library
2. Open the Terminal and go to Intel_AESNI_Sample_Library_v1.2/intel_aes_lib
3. Run ./mk_lnx_lib64.sh, which will generate the header and library files in intel_aes_lib/include and intel_aes_lib/include directories, resp.
4. Copy header files and library files to your local folders (e.g., /usr/local/include and /usr/local/lib).

Configuration

All FS_DSSE configurations are located in FS_DSSE/config.h.

Highlighted Parameters:

#define INTEL_AES_NI				-> If enabled, use Intel AES-NI library

#define DISK_STORAGE_MODE            	   	-> If enabled, encrypted index will be stored on HDD (RAM if disabled)
	
#define SEND_SEARCH_FILE_INDEX        		-> If enabled, search result will contain specific file indexes

#define PEER_ADDRESS "tcp://localhost:5555"	-> Server IP Address & Port

const std::string SERVER_PORT = "5555";		-> Server Port number

#define  MAX_NUM_OF_FILES 1024              	-> Maximum number of files (It MUST be the power of 2 and divisible by 8)

#define  MAX_NUM_KEYWORDS 12000             	-> Maximum number of keywords

Notes

The folder FS_DSSE/data as well as its structure are required to store generated FS_DSSE data structures. The database is located in FS_DSSE/data/DB. The implementation recognize DB as a set of document files so that you can copy your DB files to this location. The current DB contains a small subset of enron DB (link: https://www.cs.cmu.edu/~./enron/).

Build & Compile

Goto folder FS_DSSE/ and execute

make

, which produces the binary executable file named FS_DSSE in FS_DSSE/Debug/.

If there is an error regarding to BOOL/bool type when compiling with Intel-aes-ni

• Access the header file named iaesni.h, go to line 51, and comment that line as follows:

#ifndef bool
//#define bool BOOL 			-> line 51
#endif

If the hardware does not support Intel-aes-ni

1. Disable INTEL_AES_NI in FS_DSSE/config.h

2. Change the make file in FS_DSSE/MakeFile by removing the library linker -lintel-aes64

Usage

Run the binary executable file FS_DSSE, which will ask for either Client or Server mode. The FS_DSSE implementation can be tested using either single machine or multiple machines with network:

Local Testing:

1. Set PEER_ADDRESS in FS_DSSE/config.h to be localhost.
2. Choose SERVER_PORT identical with what indicated in PEER_ADDRESS.
3. Compile the code with make in the FS_DSSE/ folder.
4. Go to FS_DSSE/Debug and run the compiled FS_DSSE file with two different Terminals, each playing the client/server role.

Real Network Testing:

1. Set PEER_ADDRESS and SERVER_PORT in FS_DSSE/config.h with the corresponding server's IP address and port number.
2. Run make in FS_DSSE/ to compile and generate executable file FS_DSSE in FS_DSSE/Debug folder.
3. Copy the file FS_DSSE in FS_DSSE/Debug to different machines
4. Execute the file and follow the instruction on the screen.

Further Information

For any inquiries, bugs, and assistance on building and running the code, please contact Muslum Ozgur Ozmen ([email protected]) or Thang Hoang ([email protected]).