Combining Concurrent Transmissions Scheduling and Power Allocation for Multimedia Content Distribution in 60 GHz based Indoor Networks Tra Huong Thi Le, Choong Seon Hong Department of Computer Engineering, Kyung Hee University huong_tra25@khu.ac.kr, cshong@khu.ac.kr Abstract 60 GHz millimeter wave (mmWave) technology offers the potential for multi-Gbps indoor applications. In this paper, we apply this technology for multimedia content delivery in indoor network where devices are battery-constrained. We first schedule the concurrent transmissions by using vertex coloring algorithm. Then, we consider the power allocation problem to optimize the aggregate network utility in each timeslot. Numerical results show that our proposed scheme can decrease the number of timeslots as well as improve the aggregate network utility. 1. Introduction 60 GHz is a promising technology to support applications that require gigabit data rate. However, the signal in this network decreases more dramatically than traditional sub 6 GHz band. To deal with this problem, directional antenna or beamforming is used to compensate propagation loss. Since in directional communication, transmission energy focuses toward a specific direction, idle spaces can be created. Neighbor devices can utilize these idle spaces to transmit and receive data. In other word, the utilization of directional communication permits multiple concurrent transmissions which improves network throughput. Besides, standard such as IEEE 802.15.3 [1] is designed to achieve very high throughput in mmWave WPAN networks. However, in this existing standard, each timeslot is allocated to one pair of transmitter and receiver, which does not fully exploit the potential of concurrent transmissions. In addition, how to schedule multiple transmissions in the same timeslot to improve throughput is important and challenging issue. Typically, multimedia content delivery consumes more energy than other low rate services because it is delay sensitive and requires high data rate. But, many portable devices in smart home have a limited lifetime because they depend on battery. Therefore, it is essential to manage resource such as power efficiently. In this paper, we consider a scenario where multiple pairs of users deliver multimedia content in a small room or house. Firstly, we try to schedule concurrent transmissions by using vertex coloring scheduling algorithm [2]. Then, in each timeslot, power allocation problem is considered to maximize the utility function and satisfy QoS requirements of all links scheduled to transmit in this timeslot. The rest of this paper is organized as followed. In Section II, we describe the transmission data rate. In Section III, the vertex coloring scheduling algorithm is presented. Utility functions and power allocation problem formulation are considered in Section IV. Some numerical results are presented in Section V and conclusions are drawn in Sections VI. 2. Transmission Data rate We assume that the average link throughput is affected by the path loss. d i is the distance between source device T i and destination device R i of link i. The path loss at distance d i in dB can be modeled as: 0 , 0 10 0 0 ( ), ( ) 10 log ( ), i i ii i i PL d d d h d PLd n d d d ≤ = + > (1) Where n is the path loss exponent and is usually in the range of 2 to 6 for indoor environment. PL(d 0 ) is the path loss function at the reference distance d 0 and can be calculated by Friis equation for free space transmission. With p i is the transmission power of transmitting device of link i, the achievable data rate for this link: , 2 0 , log [1 ] i ii i j ji i j ph r B NB ph η ≠ ≤ + + ∑ (2) Where B is bandwidth, [0,1] η ∈ is the coefficient describing the efficiency of the transceiver design, N 0 is the one-side noise power spectral density. 3. Scheduling based on Vertex Coloring Algorithm a. Condition for Concurrent Transmissions Concurrent transmissions may suffer from the possible interference from other transmitters. In [3], the authors introduced an exclusive region [ER] around the receiver, which is defined as the threshold distance that the mutual 1142 2015년 한국컴퓨터종합학술대회 논문집