`Control in WCDMA System with MIMO(cid:1)
`
`Cheol Yong Jeon and Yeong Min Jang
`
`School of Electrical Engineering,
`Kookmin University,
`861-1, Jeongneung-dong, Songbuk-gu, Seoul 136-702, Korea
`{feon77,yjang}@kookmin.ac.kr
`
`Abstract. This paper presents an efficient capacity evaluation algo-
`rithm of the WCDMA with multiple-input multiple-output (MIMO) sys-
`tem for quality of service (QoS) support. To define the capacity of the
`system, we derive the Eb/No gain taking into account MIMO concept
`and the outage probability as the QoS measure using central limit ap-
`proximation, Chernoff bound and the refined large deviation approach.
`Based on the QoS measures, we propose an efficient transient call ad-
`mission control (CAC) algorithm. Numerical results show that there is a
`substantial increment in system capacity by adopting MIMO system and
`the theory of the refined large deviation approach is a good approach for
`transient QoS support.
`
`1 Introduction
`
`Third-generation mobile wireless systems are often referred to as universal mobile
`terrestrial telecommunication systems (UMTS’s). The UMTS system intends
`to integrate all forms of mobile communications, including terrestrial, satellite,
`and indoor communications. Consequently, UMTS must support a number of
`different air interfaces [1]. One of the main air interfaces for this system is referred
`to as wideband code division multiple access (WCDMA), which is the topic of
`this paper. WCDMA is based on CDMA scheme by which multiple users are
`assigned radio resource using spread spectrum techniques. Although all users are
`transmitting in the common bandwidth, individual users are separated from each
`other via the use of orthogonal codes. If total energy from all users, however,
`is over the given threshold (signal-to-noise ratio (SNR) or energy per bit per
`bandwidth), the system can not admit any more users [2, 3].
`Here we focus on MIMO system to increase the system capacity. Wireless
`channel using multiple antennas at each ends is commonly referred to as a MIMO
`channel. Technology built around MIMO channels resolves the fundamental issue
`of having to deal with two practical realities of wireless communications: a user
`terminal of limited battery power, and a channel of limited bandwidth. Given
`fixed values of transmit power and channel bandwidth, this new technology offers
`
`(cid:1) This work was˙supported by the KOSEF through the grant No. R08-2003-000-10922-0.
`
`M. Ajmone Marsan et al. (Eds.): QoS-IP 2005, LNCS 3375, pp. 545–558, 2005.
`c(cid:1) Springer-Verlag Berlin Heidelberg 2005
`
`Bell Northern Research, LLC, Exhibit 2016, Page 1 of 2
`
`
`
`546
`
`Cheol Yong Jeon and Yeong Min Jang
`
`a sophisticated approach to exchange increased system complexity for increasing
`the capacity (i.e., the spectral efficiency of the channel, measured in bits per
`hertz) up to a value significantly higher than that attainable by any known
`method based on a single-input single-output (SISO) channel. More specifically
`the MIMO channel capacity is roughly proportional to the number of transmitter
`or receiver antennas, whichever is smaller. That is to say, we have a spectacular
`increase in spectral efficiency, with the channel capacity being roughly doubled
`by doubling the number of antennas at both ends of the link [1, 4, 5].
`Up to now, the capacity analysis was investigated individually MIMO and
`CDMA. The MIMO spectral efficiency increases linearly with antenna number
`and the capacity of CDMA system decreases by transmitted user’s power [2].
`In this paper, we focus on the MIMO technology and on the calculating simul-
`taneous user numbers in the WCDMA systems with MIMO. Then this paper
`presents how much more increase users in this system by numerical result. We
`apply the fluid flow queuing model for CAC in WCDMA system with MIMO.
`Traffic control is necessary to avoid possible congestion at each network node and
`achieve the QoS requirement by each connection. Due to real time constraints
`and the dynamic behavior in the system, preventive (e.g. predictive) control is
`more suitable then reactive control [6]. CAC is form of the preventive traffic
`control. The CAC is responsible for deciding whether or not a new call can be
`accepted while maintaining QoS in the network. We choose the transient outage
`probability as measure of QoS and desire a scheme which optimizes both the
`transient and steady state performance [7, 8]. So, we provide a complete tran-
`sient solution of the system starting from a given initial condition. This paper
`present a general theory to deal with the transient analysis of multiple types of
`traffic. In order to cope with the computational complexity, a general theory of
`approximations and bounding approaches should be explored. In this paper, we
`propose new approximations and bounds for the transient fluid model [9] to deal
`with requirement of real time computation.
`This paper is organized as follows: Section II discusses the spectral efficiency
`of WCDMA system with MIMO. We discuss CAC algorithm and derive the
`outage probability and cell loss ratio for CAC in section III. Section IV presents
`numerical result and finally come to conclusion in section V.
`
`2 Spectral Efficiency of the WCDMA System
`with MIMO
`
`2.1 MIMO in Gaussian Channel
`
`When we use nT transmit antennas and nR receive antennas, m =min(nT , nR)
`and n=max(nT , nR), the MIMO capacity increases linearly with m. The average
`SNR per nR receiver antenna [10, 11] is given by
`E[|Hx|2]
`E[|n2]
`
`SN R = g
`
`(1)
`
`Bell Northern Research, LLC, Exhibit 2016, Page 2 of 2
`
`