This paper presents a novel approach to human gait analysis with a sensor-based technique involving a wearable inertial measurement unit (IMU). The proposed system emphasizes the detection of certain abnormal gait patterns, including hemiplegic, tiptoe, and cross-threshold gait. First, we use the dynamic step conjugate gradient algorithm to calculate the attitude of the gait data, and we then use the gait feature information location algorithm to segment the attitude data. The segmented attitude data are used as input in the classification model. In this paper, we propose an algorithm based on a long short-term memory network and convolutional neural network (LCWSnet) for diagnosis and classification of abnormal gait patterns using the leg Euler angle information, and parameters related to features can be adjusted adaptively according to the feedback of objectives and optimization functions. We optimize the convergence layer of the LSTM-CNN model and improve the classification accuracy of abnormal gait. The experimental results demonstrate that the proposed LCWSnet-based technique is able to detect gait abnormality in the data. The proposed personalized gait classification approach is accurate and reliable and can be implemented for the abnormal gait. INDEX TERMS Convolutional neural network, dynamic step conjugate gradient algorithm, gait feature information location, long short-term memory network, wireless body area network.

The widespread application of wireless mobile services and requirements of ubiquitous access have resulted in drastic growth of the mobile traffic and huge energy consumption in ultradense networks (UDNs). Therefore, energy-efficient design is very important and is becoming an inevitable trend. To improve the energy efficiency (EE) of UDNs, we present a joint optimization method considering user association and small-cell base station (SBS) on/off strategies in UDNs. The problem is formulated as a nonconvex nonlinear programming problem and is then decomposed into two subproblems: user association and SBS on/off strategies. In the user association strategy, users associate with base stations (BSs) according to their movement speeds and utility function values, under the constraints of the signal-to-interference ratio (SINR) and load balancing. In particular, taking care of user mobility, users are associated if their speed exceeds a certain threshold. The macrocell base station (MBS) considers user mobility, which prevents frequent switching between users and SBSs. In the SBS on/off strategy, SBSs are turned off according to their loads and the amount of time required for mobile users to arrive at a given SBS to further improve network energy efficiency. By turning off SBSs, negative impacts on user associations can be reduced. The simulation results show that relative to conventional algorithms, the proposed scheme achieves energy efficiency performance enhancements.