Uplink Scheduling for LTE Video Surveillance Systems

Internet of Things, an increasing number of machine to machine (M2M) devices are being deployed. In particular, one of the M2M applications, video surveillance, has been widely discussed. Long Term Evolution (LTE), which can provide a high rate of data transmission and wide range of coverage, is a promising standard to serve as an M2M video surveillance system. In this paper, we studied a performance maximization problem in an LTE video surveillance system. Given a set of objects and a set of cameras, each camera has its own performance grade and its own coverage. The goal is to maximize the performance of the system by allocating limited resources to cameras while all objects should be monitored by the selected cameras. We propose a heuristic method to select the cameras and allocate resources to them to solve the problem. Moreover, to reduce the load of the LTE system, a dynamic adjustment method is also proposed. I. INTRODUCTION To meet the increasing demand for mobile wireless access, the Third Generation Partnership Project (3GPP) Long Term Evolution (LTE) has attracted significant attention. LTE provides high network capacity, high transmission speed, and a wide coverage range. Orthogonal Frequency-Division Multiple Access (OFDMA) is adopted in LTE downlink (DL) transmission while Single-carrier FDMA (SC-FDMA) is adopted in the LTE uplink (UL) transmission to conserve power. The scheduling algorithm of resource allocation plays an important role for LTE, since an efficient algorithm can greatly increase the LTE system throughput. In the LTE DL system, due to the multiuser diversity gain, most of the algorithms use the channel conditions as the comparison metric to maximize the aggregate throughput. Some previous works modified the well-known Proportional Fair scheduler to ensure the fairness among the users [1]. Some other works took the quality of service (QoS) of each user into consideration [2]. In such algorithms, users with higher priority will be served earlier. Differing from the LTE DL system, the allocation policy used by the scheduler in the LTE UL must consider the contiguity constraint of SC-FDMA. Contiguity constraint means one user must be allocated adjacent resource blocks (RBs). Thus, although the comparison criteria of scheduling for the LTE UL are similar to those of LTE DL, the approaches are different. Most schedulers for UL have considered information related to adjacent RBs while allocating resources to users. However, neither for DL nor for UL, there is no scheduler which takes …