Frequent location updates of individual Internet of Things (IoT) devices can cause several problems (e.g., signaling overhead in networks and energy depletion of IoT devices) in massive machine type communication (mMTC) systems. To alleviate these problems, we design a distributed group location update algorithm (DGLU) in which geographically proximate IoT devices determine whether to conduct the location update in a distributed manner. To maximize the accuracy of the locations of IoT devices while maintaining a sufficiently small energy outage probability, we formulate a constrained stochastic game model. We then introduce a best response dynamics-based algorithm to obtain a multi-policy constrained Nash equilibrium. From the evaluation results, it is demonstrated that DGLU can achieve an accuracy of location information that is comparable with that of the individual location update scheme, with a sufficiently small energy outage probability.

In federated learning (FL), to minimize the convergence time while reaching a desired accuracy, it is important to select appropriate mobile devices (MDs) as participants with the considerations of their characteristics (e.g., data distribution, channel condition, and computing power). In this article, we first formulate a Markov decision process-based MD selection problem in which the increased accuracy per unit-time is maximized. To solve the formulated problem without any prior knowledge of the environment, a deep Q-network (DQN) algorithm can be exploited. However, storing previous experiences into the replay memory for DQN consumes an increased time since the FL server needs to conduct a number of actual FL procedures and observe the resulted rewards. To address this problem, we proposed a virtual experience-based MD selection algorithm (VE-MSA). In VE-MSA, the FL server generates virtual experiences (especially reward) without any actual FL procedures by using two neural networks approximating the round time and the increased accuracy in a round according to the selected MDs, respectively. Evaluation results demonstrate that the derived optimal policy can achieve a target accuracy within the shortest time among comparison schemes.

In a non-orthogonal multiple access (NOMA) environment, an Internet of Things (IoT) device achieves a high data rate by increasing its transmission power. However, excessively high transmission power can cause an energy outage of an IoT device and have a detrimental effect on the signal-to-interference-plus-noise ratio of neighbor IoT devices. In this paper, we propose a neighbor-aware NOMA scheme (NA-NOMA) where each IoT device determines whether to transmit data to the base station and the transmission power at each time epoch in a distributed manner with the consideration of its energy level and other devices’ transmission powers. To maximize the aggregated data rate of IoT devices while keeping an acceptable average energy outage probability, a constrained stochastic game model is formulated, and the solution of the model is obtained using a best response dynamics-based algorithm. Evaluation results show that NA-NOMA can increase the average data rate up to $22\%$ compared with a probability-based scheme while providing a sufficiently low energy outage probability (e.g., $0.05$).