徐秋亮,唐春明

1.山東大學 軟件學院,濟南 250101

2.廣州大學 數學與信息科學學院,廣州 510006

安全多方計算研究兩個或多個持有私有輸入的參與者,聯合計算某些功能函數,各自得到他們預定的輸出,并達到正確性、隱私性、公平性等安全特性.近年來,隨著云計算、大數據、人工智能、區(qū)塊鏈等技術的出現和發(fā)展,數據安全與隱私保護問題面臨的挑戰(zhàn)日益突出,安全多方計算被寄予厚望,公認是解決這類問題的重要手段,因而得到空前重視.

為了進一步促進我國安全多方計算領域的理論和應用研究,本刊通過廣泛征稿和約稿,組織了本期“安全多方計算技術” 專欄,希望能對該領域的發(fā)展起到積極促進作用.經過同行評審,本期共收錄6 篇論文.

綜述性論文1 篇.趙川等人的論文《實用安全兩方計算及其在基因組序列比對中的應用》,總結分析了安全兩方計算近年來在實用性方面的主要研究成果,并重點介紹了實用安全兩方計算在基因組序列比對中的研究進展.作者從安全兩方計算的兩個主要構造方法(同態(tài)加密和混亂電路)出發(fā),分別給出了較為清晰的發(fā)展脈絡.此外,該文指出了現階段基于安全兩方計算的基因組序列比對研究中存在的幾點不足,并分析了未來可能的研究方向.

特定安全多方計算協(xié)議論文3 篇.杜潤萌等人的論文《矩陣與增廣矩陣秩相等問題的保密計算及應用》,在半誠實敵手模型下,設計了矩陣與增廣矩陣秩相等判定問題的安全多方計算協(xié)議,并將該協(xié)議用于解決保密判斷多項式整除、保密判定直線與直線的位置關系等問題.李占利等人的論文《云環(huán)境下多方保密計算最大值、最小值及其統(tǒng)計學應用》首先使用0-1 編碼方法,使得保密數據隱藏于所編碼的0-1 數組中,然后利用云環(huán)境下多密鑰NTRU 全同態(tài)加密算法,在半誠實敵手模型下,設計了在云計算環(huán)境下解決最大值、最小值計算的安全多方計算協(xié)議.葛雪等人的論文《直方圖與餅形圖的保密生成協(xié)議》,利用加同態(tài)加密算法,設計了數據求和的安全多方計算協(xié)議,并進一步轉化為直方圖與餅形圖,該協(xié)議也是在半誠實敵手模型下證明安全的.

安全多方計算應用協(xié)議2 篇.朱巖等人的論文《基于安全多方計算的區(qū)塊鏈智能合約執(zhí)行系統(tǒng)》,提出了基于安全多方計算的智能合約框架、面向線性秘密共享的公平安全多方計算算法設計、以及非阻塞信息傳遞接口等三方面技術,保證了智能合約執(zhí)行中的輸入隱私性和計算正確性,從而增強了區(qū)塊鏈中智能合約執(zhí)行安全.王啟正等人的論文《一種處理隱私保護數據的神經網絡》,利用基于同態(tài)加密的安全多方計算技術,設計了神經網絡計算外包協(xié)議,在保障數據隱私性的前提下保留了數據的可計算性.

安全多方計算是密碼學的一個重要研究領域,它隨著公鑰密碼技術的出現而產生,經過了長時間的發(fā)展.近年來,隨著安全多方計算協(xié)議本身效率的提升,以及各種分布式應用場景對安全和隱私保護的迫切要求,安全多方計算協(xié)議的研究成為密碼學方向最有活力的研究領域之一.通過本期專欄,希望讀者能夠管中窺豹,見安全多方計算技術最新研究動向之一斑.

In the secure multi-party computation(SMPC),two or more participants who have private inputs,want to jointly calculate some functionalities,get their outputs,and achieve some secure properties such as correctness,privacy,fairness,and so on.In recent years,with the development of cloud computing,big data,artificial intelligence,block chains,and other technologies,the challenges of data security and privacy protection have become increasingly prominent.Secure multi-party computing has been highly expected,and is recognized as an important technique to solve such problems,which has attracted unprecedented attention.

In order to promote the theoretical and applied research of SMPC in China,Journal of Cryptologic Research organized the special column on ”Secure Multi-party Computing Technology” by calling and inviting for papers widely,hoping to promote the development of this field.6 papers are selected into this special column after the peer review process.

One review is included in this special column.“Advances in Practical Secure Two-party Computation and Its Application in Genomic Sequence Comparison”by Zhao Chuan et al.reviews the advances in practical secure two-party computation in recent years,and focuses on the major research results in the field of genomic sequence comparison based on secure two-party computation.The authors start with introducing two major construction techniques(i.e.,homomorphic encryption and garbled circuit),and give a clear development trend,respectively.In addition,the existing deficiencies in this research area and some possible research directions are pointed out.

Three papers on specific secure multi-party computation protocol are included in this special column.The paper entitled“Privately Determining Equality of Ranks of Matrix and Its Augmented Ones and Applications”by Du Run-Meng et al.designs an SMPC protocol in semi-honest adversaries model to determine whether the rank of a matrix is equal to its augmented ones,and applies this protocol in privately determining the relationship between two lines,determining whether a polynomial divides another one.The paper entitled “Secure Multiparty Computation of the Maximum and the Minimum in Cloud Environment and Its Statistics Application”by Li Zhan-Li et al.adapts 0-1 encoding method to encode a private number into an array,and then using the multikey NTRU fully homomorphic encryption algorithm in cloud environment,designs the SMPC protocol in semi-honest adversaries model to compute the maximum and the minimum value.The paper entitled “Histogram and Pie Chart of Confidentiality Generation Agreement” by Ge Xue et al.uses homomorphic encryption algorithm to design an SMPC protocol to compute the summation,and further convert it to generate the histogram and pie chart.The protocol is also proved in semi-honest adversaries model.

Two papers on secure multi-party computation application protocol are included in this special column.The paper entitled “Smart Contract Execution System over Blockchain Based on Secure Multi-party Computation”by Zhu Yan et al.proposes three technologies:a smart contract framework based on secure multi-party computation(SMPC),a fair SMPC algorithm built on linear secret sharing,and a non-blocking message passing interface,to ensure the privacy of inputs and the correctness of computing result during smart contract execution.Thus,the execution security of smart contract can be enhanced by the proposed technologies in the Blockchain.The paper entitled “Neural Network for Processing Privacy-protected Data” by Wang Qi-Zheng et al.uses the homomorphic encryption based SMPC to design a neural network computation outsourcing protocol.It keeps the computability of data under the premise of ensuring data privacy.

SMPC is an important research field in cryptography.It has been developed for a long time with the emergence of public key cryptography.In recent years,with the improvement of the efficiency of secure multi-party computing protocols and the urgent requirement of security and privacy protection in various distributed application scenarios,the research of SMPC protocols has become one of the most liven research fields in cryptography.Through this special column,we hope that readers can get a glimpse of the research trends in SMPC technology.