Cyber-physical social system (CPSS) has emerged as a new paradigm to help social users share and exchange data by the close association with the cyberspace and physical world. To further improve the performance of CPSS, the incentive computing scheme to provide efficient crowd sourcing in the CPPS becomes a challenge. Therefore, in this paper we propose a novel incentive scheme for CPSS based on the reputation of social users. First, we present a framework to provide crowd sourcing service in CPSS by dividing social users into three types, which are malicious users, speculative users and honest users, respectively. Second, based on the reputation of social users, an incentive scheme is proposed to encourage users to contribute sourcing data. Next, an auction game model is developed to help CPSS select the optimal social user to obtain the needed data. Finally, simulation results show that the proposal can obtain a lower cost and higher data accuracy than other conventional methods.

Conventional optimization-driven secure relay selection relies on maximization algorithm and accurate channel state information (CSI) of both legitimate and eavesdropper channels. Particularly, estimating and collecting accurate eavesdropper CSI is a difficult task. In this paper, we exploit the benefits of machine learning in solving secure relay selection problem from a data-driven perspective. We convert secure relay selection to a multiclass-classification problem and solve it by a decision-tree-based scheme, which is composed of three phases - preparing training data, building decision tree and predicting relay selection. To meet decision tree’s requirement that input features must take discrete values, a feature extraction method is proposed to generate discrete input by quantizing the accurate CSI of legitimate and eavesdropper channels. By this means, the decision-tree-based relay selection only requires quantized CSI feedback which takes substantially fewer bits in predicting phase. For the purpose of optimizing quantization parameters and enhancing decision tree prediction, we further derive three splitting criteria, i.e. information gain, information gain ratio and Gini index. Simulation results show that if the quantization parameters are set properly, the proposed decision-tree-based scheme can achieve satisfactory performance in terms of average secrecy rate while reducing computational complexity and feedback amount.

This work is an initial attempt to exploit the benefits of machine learning in solving relay selection for dualhop networks. From data-driven perspective, we convert relay selection to multiclass-classification problem and propose a decision-tree-based selection scheme. Input features are generated by a binary quantization to the equivalent CSI of each relay, with which the dependence of relay selection on maximization algorithm is removed. To optimize quantization parameter, we derive and maximize two splitting criteria - information gain and Gini index. Experiment results show that decision-tree-based scheme achieves comparable performance to optimal selection in terms of average end-to-end (E2E) rate while reducing computational complexity and feedback amount.

The coexisting oscillations are observed with a memcapacitor‐based circuit that consists of two linear inductors, two linear resistors, and an active nonlinear charge‐controlled memcapacitor. We analyze the dynamics of this circuit and find that it owns an infinite number of equilibrium points and coexisting attractors, which means extreme multistability arises. Furthermore, we also show the stability of the infinite many equilibria and analyze the coexistence of fix point, limit cycle, and chaotic attractor in detail. Finally, an experimental result of the proposed oscillator via an analog electronic circuit is given.

Memristor and memcapacitor are new nonlinear devices with memory. We present a novel memcapacitor model that has the capability of capturing the behavior of a memcapacitor. Based on this model we also design a chaotic oscillator circuit that contains a HP memristor and the memcapacitor model for generating good pseudorandom sequences. Its dynamic behaviors, including equilibrium points, stability, and bifurcation characteristics, are analyzed in detail. It is found that the proposed oscillator can exhibit some complex phenomena, such as chaos, hyperchaos, coexisting attractors, abrupt chaos, and some novel bifurcations. Moreover, a scheme for digitally realizing this oscillator is provided by using the digital signal processor (DSP) technology. Then the random characteristics of the chaotic binary sequences generated from the oscillator are tested via the test suit of National Institute of Standards and Technology (NIST). The tested randomness definitely reaches the standards of NIST and is better than that of the well-known Lorenz system.

Conventional optimization-based relay selection for multihop networks cannot resolve the conflict between performance and cost. The optimal selection policy is centralized and requires local channel state information (CSI) of all hops, leading to high computational complexity and signaling overhead. Other optimization-based decentralized policies cause non-negligible performance loss. In this paper, we exploit the benefits of reinforcement learning in relay selection for multihop clustered networks and aim to achieve high performance with limited costs. Multihop relay selection problem is modeled as Markov decision process (MDP) and solved by a decentralized Q-learning scheme with rectified update function. Simulation results show that this scheme achieves near-optimal average end-to-end (E2E) rate. Cost analysis reveals that it also reduces computation complexity and signaling overhead compared with the optimal scheme.