Heterogeneous cellular network offers a promising architecture for supporting multiservice traffic generated by Internet of thing (IoT) paradigm and human-to-human (H2H) communications. In this paper, a distributed device association algorithm is proposed to consider the diverse quality of service requirements for H2H devices and IoT devices coexisted in heterogeneous cellular network. In this context, the traffic generated by H2H devices and IoT devices is categorized into 2 main classes, ie, primary service class (PSC) requiring low outage probability and secondary service class (SSC) with a minimum long-term rate requirement. Primary service class is principally used to model H2H traffic, while SSC is mainly used to model IoT traffic. The association problem is formulated to maximize the system revenue in terms of the number of devices belonging to the SSC while minimizing the global outage probability for devices modeled to PSC under quality of service guarantee. Since the proposed optimization problem is an NP combinatorial problem, an efficient solution is derived by relaxing the association constraints and adopting the Lagrange dual decomposition method. Simulation results show that our proposed algorithm outperforms the comparable schemes especially on outage performance for PSC-modeled devices and number of accommodated SSC-modeled devices.