k-Anonymity is an important solution to protecting privacy of queries in LBS (location-based service).However, it is pointed out in literatures that k-anonymity cannot protect privacy of continuous queries effectively. Acontinuous query issuing model is proposed, which incorporates a query issuing interval model and a consecutivequeries relationship model. Under this continuous query issuing model, two attacking algorithms are proposed forClique Cloaking and Non-clique Cloaking respectively. Then this paper argues that the cardinality of anonymity-setis not a good anonymity measurement under such attack and an entropy-based anonymity measurement AD(anonymity degree) is proposed. Experimental results demonstrate that the attacking algorithms have high successrate in identifying query senders when the consecutive queries have strong relationship, and that AD is a betteranonymity measurement than the cardinality of anonymity-set.