Sparse matrix multiplication (SpMM) is a critical kernel used in a wide range of applications, but irregular memory access patterns and memory bandwidth bottleneck as well as load imbalance make the efficient and scalable processing on parallel architectures a significant challenge. Motivated by the memory-bound nature of SpMM computation, we propose a cost-effective SpMM accelerator based on a DRAM processing-in-memory (PIM) approach. Our design introduces a novel dataflow to exploit high bank-level parallelism and reuse both input and output data even for highly sparse matrices. Our proposed architecture, SPIMA, features multiple input buffers for scheduling DRAM access, a output buffer and vector register files working holistically, co-designed to maximize the performance of our novel dataflow. Our experimental results using various sparse matrices demonstrate that our proposed dataflow and architecture are robust in terms of matrix size and sparsity. Compared with the state-of-the-art accelerators implemented on PIM, ASIC, and FPGA, we estimate that our PIM architecture can yield competitive performance with highly sparse matrices.

The compliant motion of humanoid robots is one of their most important characteristics for interacting with humans and various environments in the real world. During walking, compliant motion ensures stable contact between the foot and ground, but walking stability is degraded by position tracking performance and unknown disturbances. To address the issue of instability of humanoid robot walking with compliant motion control, this paper proposes a model for real-time walking pattern generation considering the motion control performance of a robot. The dynamic model of a robot with a motion controller is described as a second-order system approximating position tracking performance with a linear inverted pendulum model to determine the relationship between the zero-moment point and center of mass (CoM). The CoM trajectory is calculated using preview control based on the dynamics model and current state of the robot. Therefore, even if the robot has the low tracking performance due to compliant motion control, the walking stability can be ensured. The proposed method was implemented on our humanoid robot, DYROS-JET, and its performance was demonstrated through improved stability during walking.

Actuator modules of humanoid robots have relatively higher joint elasticity than those of industrial robots. Such joint elasticity could lead to negative effects on both the tracking performance and stability for walking. Especially, unstable contact between the foot and ground caused by joint elasticity is a critical problem, as it decreases the stability of position-controlled humanoid robots. To address this problem, this paper introduces a novel control scheme for position-controlled humanoid robots by which we can obtain not only enhance compliance capability for unknown contact but also suppress the vibration caused by joint elasticity. To estimate the disturbance caused by external forces and modeling errors between the actual system and nominal system, a disturbance observer based estimator is designed at each joint. Furthermore, a linear feedback controller for the flexible joint model and a gravity compensator is considered to reduce vibration and deflection due to the joint elasticity. The proposed control scheme was implemented on our humanoid robot, DYROS-JET, and its performance was demonstrated by improved stability during dynamic walking and stepping on objects.

Achieving the benefits of high-capacity of optical networks and the mobility feature of wireless networks leads to integrate EPON and WiMAX for a promising broadband access solution. To efficiently put both benefits together, we propose an integration architecture of EPON-WiMAX based upon a Distributed Antenna (DA) environment, where collaborative Base Stations (BSs) concurrently transmit same wireless downlink signals (specifically for multicast and broadcast services (MBSs)) to Mobile Stations (MSs) in overlapped cell coverage areas. It helps enlarge the region of the available network coverage area by increasing Signal to Interference and Noise Ratio (SINR) in overlapped cell coverage areas through cooperation between Optical Line Terminal (OLT) and Optical Network Unit (ONU)- BS. We also present an cost-efficient cell planning to optimally control the size of overlapped cell coverage areas for the proposed DA-based integration architecture with a case study under a required region of the available network coverage area in consideration of the number of ONU-BSs and the distance between ONU-BSs. Performance evaluation results show that the proposed DA-based integration architecture enhances cost efficiency compared to the Traditional Antenna (TA) (non-DA)-based integration architecture with a similar level of spectral efficiency of MSs.

A Power-Saving Mechanism (PSM) operates with sleep-mode and wake-mode, based on the receipt of requests for transition to each mode. The sleep-mode operation is manipulated by adjusting operating parameters such as the minimum sleep interval (Tmin) and the maximum sleep interval (Tmax). Interestingly, both energy consumption and the response delay of the request for a BS initiation of awakening are reciprocally affected by the relative sizes of these two operating parameters compared to the sleep duration (from the beginning of sleep-mode to that of wake-mode). To resolve this issue, this paper proposes a new PSM, called EPSM, which adaptively and simultaneously controls the operating parameters by efficiently reflecting the sleep duration. Moreover, depending on the current remaining energy state, this mechanism can also increase the available sizes of the operating parameters to be manipulated for achieving more intensive energy conservation. A numerical model is developed with a Markov chain for performance evaluation of the proposed mechanism pertaining to energy consumption and the response delay. The evaluation results substantiate that this mechanism can enhance energy conservation within reasonable response delay compared to the standard mechanism. Moreover, under an insufficient remaining energy state, the EPSM can prolong battery life while enduring an increasing response delay.

The sleep-mode operation in power management helps to increase the life of a mobile station by saving energy consumption while it increases the MAC service data unit (SDU) response delay. Due to the trade-off relationship between these two performance metrics, it is important to determine the suitable values for two operating parameters, the initial sleep interval (Tmin) and the final sleep interval (Tmax), which manipulate the sleep-mode operation in the IEEE 802.16e MAC. To reach a fuller understanding of this problem, this paper first presents the sleep-mode operation. A new power management mechanism based on the knowledge on remaining energy in a mobile subscriber station (MSS) is then proposed. Benefits of this mechanism are to reduce the MAC SDU response delay under sufficient energy and to increase the life of a station under insufficient remaining energy. An analytical model for sleep-mode operation is developed, and the proposed mechanism is validated by computer simulation.

Since channel scanning time in IEEE 802.11 handoff is dominant in total handoff delay time, attention has to be paid to the reduction of channel scanning time for effectively supporting delay sensitive applications such as voice over IP (VoIP). Therefore, this paper proposes a new channel scanning scheme that adaptively adopts full channel scanning before, and first satisfaction first reservation scanning after a required delay time obtained from the application layer. Simulation results substantiate that the proposed scheme can keep the total handoff delay as low as possible compared to the required delay while maintaining acceptable Received Signal Strength (RSS) to a Mobile Host (MH) and thus the Quality of Service (QoS) of the applications can be more guaranteed than the conventional and previous schemes.