Privacy Protection Mechanism For Spatio-temporal Crowdsourcing In Participatory Sensing Environment

With the popularization of 5G technology and the arrival of Industry 5.0,network science has provided new development opportunities and challenges,and crowdsourcing has also become a new driving force for the development of network science theory and engineering.Spatio-temporal crowdsourcing can increase the potential of crowds to perform tasks involving real-world scenarios related to physical locations.Its main feature is the existence of tasks,requiring workers to appear in specific locations within a specific time to complete the task.However,in spatio-temporal crowdsourcing system and sensing information transmission,due to the spatio-temporal sensitivity of sensing information and the insecurity of network channel,participants often face the risk of privacy leakage.Therefore,in the spatio-temporal crowdsourcing system,it is particularly important to protect the private information of participants.Privacy protection technology based on spatio-temporal crowdsourcing has also become a research hotspot in the field of Internet privacy protection.The privacy protection methods of spatio-temporal crowdsourcing are mainly divided into privacy protection methods based on spatio-temporal information and communication security methods based on cryptography.Based on the above two parts,this paper solves the difficulties and shortcomings of previous research methods,respectively designing a dynamic publishing privacy protection mechanism based on k-anonymity and l-diversity,and a privacy protection mechanism based on lightweight encryption methods.The specific research content includes the following two points:1.Most data publish methods consider static publish,which is likely to cause high time complexity and poor timeliness.In response to this problem,this paper combines the idea of dynamic clustering to improve the timeliness of dynamic publishing and increase time complexity.In addition,most researchers simply anonymize participants’ location information,and do not consider that attackers can also infer other private information based on participants’ time information.This paper simultaneously anonymizes the location attributes and time attributes of participants,effectively avoiding background knowledge attacks and homogenous attacks against location attributes.2.Existing research methods of spatio-temporal crowdsourcing privacy protection often protect participants’ private information locally in some way,without considering the security during network transmission.This paper adopts the methods of participant encryption and requester decryption to protect sensing information during transmission.More importantly,due to limited storage,computing power and battery capacity,it is often difficult for resource-constrained participants to implement encryption algorithms with high computational costs.Therefore,traditional data encryption methods cannot be well applied to resource-constrained participants.To solve this problem,this paper uses a lightweight stream cipher algorithm for data encryption,and uses chaotic mapping and product algebra to improve the randomness and periodicity of keystream.