A wireless sensing system with n sensors, observing independent and identically distributed continuous random variables with a symmetric probability density function, and one non-collocated estimator acting as a fusion center is considered. The sensors transmit information to the fusion center via a limited capacity communication medium modeled by a collision channel. It is assumed that there is no communication among the sensors prior to transmission, and the collision channel allows at most k<n simultaneous transmissions. Assuming that each sensor uses a symmetric threshold communication strategy, the problem of designing a threshold that minimizes a mean-squared error criterion is considered. Theoretical analysis shows the existence and uniqueness of the optimal threshold for this optimization problem.