R. V. Rajakumar

A cooperative secondary user localization based primary user localization method for cognitive radio networks

Cognitive radio is a way of opportunistically sharing the scarce spectrum among licensed and unlicensed users of the spectrum. One of the key challenges in deploying cognitive radio networks (CRNs) is to find out the spectrum holes in the primary (licensed) wireless systems for allowing the secondary (unlicensed) users to share. In order to detect spectrum holes, CRN needs to sense the spectrum to find whether the primary user (PU) is available. In this paper, a cooperative algorithm to estimate the PU location and its coverage, based on received signal strength (RSS) at several local secondary users (SUs) locations, is presented. Maximum likelihood (ML) algorithm is used to estimate the PU locations and its transmission power. Locations of SUs are estimated using Time Difference of Arrival (TDOA) method and the cooperation among the Cognitive Radio Base Stations (CBSs) of CRN. For the convenience of analysis we take error as Euclidian distance between actual location and estimated location. The performance evaluation uses the mean error as a metric.

A Wideband RF Frontend Architecture for Software Defined Radio

This paper presents an RF front-end architecture for wideband software defined radio to function in 400 MHz to 3.5 GHz frequency range. The frequency range covers almost all the existing standards (like GSM, UMYS, CDMA, WiFi etc.) as well as future standards (like WiMax which is still to be deployed). Although the 400 MHz–3.5 GHz band is taken, the architecture allows designers to design the front-end for any range of frequencies. In wideband front-ends, devices with high linearity are used to lower the level of intermodulation products. There are different front-end architectures available for catering to wideband operation. However, some of these, particularly the wideband ones suffer with intermodulation problem. This paper presents a partitioning mechanism that uses multiple medium bands to limit the intermodulation problem. The proposed frequency-partitioning scheme makes it possible to suppress intermodulation products further for front-end using devices of given linearity. Thus this helps in achieving better performance along with relaxing the linearity requirement of devices.

Energy cost analysis of data plane and control plane protocols for 3GPP-LTE

Energy efficiency has become one of the key factors in the design of next generation wireless system. A lot of computation and network activity is done by protocols in mobile systems. Increase in the performance and capacity leads to an increase in complexity and overhead in protocols and signaling procedures. With the rapid growth in the information and communication technologies (ICT) and ICT related technologies there is a need to make energy efficient design in all aspects including the protocol and control signalling for better system design. Little has been done for power consumption for protocol processing and signalling in mobile cellular systems. In this paper, a detail comparison between L2 and L3 processing for control plane and user plane protocol is presented. We consider the analysis for specific hardware. However, it can be used for any other hardware platform. To give a total picture, we have shown the power consumption of web a browsing session for one hour with the increasing number of UEs, considering idle-connected and handover signaling protocols.

Energy impact of signalling protocols in 3GPP-LTE and guidelines for savings

Energy efficiency has become one of the key factors in the design of next generation wireless systems. A lot of computation and processing is done by protocol in mobile systems. UE battery gets unnecessary drain due to excess processing of signalling messages. With the rapid growth in the information and communication technologies (ICT) and ICT related technologies there is a need to make energy efficient design in all aspects including the control signalling for better system design. In this paper we presents a detailed analysis of processing done in L2 and L3 protocols for more frequent signalling procedure. A framework to evaluate protocol processing in signalling protocols is modelled considering processing at various interfaces air/X2. A set of guidelines are proposed to reduce processing overload and improve energy efficiency, with percentage savings obtained in UE as well as in eNodeB.

A Low Complexity PAPR Reduction Scheme without Side Information for Pilot Assisted OFDM Systems

Selected Mapping (SLM) is a well known distortionless phase control Peak-to-Average-Power-Ratio (PAPR) reduction technique for Orthogonal Frequency Division Multiplexing (OFDM) systems. A major limitation of SLM-OFDM systems is that the receiver requires exact information of the phase set used at the transmitter. Usually side information (SI) is either transmitted explicitly at the loss of data rate or avoided using Blind SLM (BSLM) techniques at the expense of computational complexity. In the present work we investigate four SLM based PAPR reduction techniques and propose a new technique for pilot based SLM-OFDM systems which avoids the requirement of SI with low complexity. The proposed scheme involves cyclic shifting of pilots for each of the independent mapping in SLM systems. We compute a minimum distance vector at the receiver to detect the exact phase set used in the transmitter and avoid explicit transmission of SI. Bit Error Rate (BER) performance of the proposed scheme is evaluated in AWGN channels.

Performance analysis of flexible reuse in cellular networks

Flexible frequency reuse schemes namely fractional frequency reuse (FFR) and soft frequency reuse (SFR) are suggested in the literature to improve the cell edge throughput over unity frequency reuse in orthogonal frequency division multiple access networks. In this study, the authors focus on real time (RT) traffic for which blocking probability and Erlang load are the important performance metrics. Both FFR and SFR have signal-tointerference-plus noise-ratio (SINR) threshold and power ratio as important design parameters, whereas FFR has a bandwidth partitioning ratio as an additional parameter. They investigate the cell capacity and the cell edge performance in terms of Erlang load with blocking probability constraint brought about by the suitable choice of these parameters. They also analyze the fairness aspect which is a critical indicator of user satisfaction. They give a complete comparison between FFR and SFR for the cell edge as well as total cell for both RT and BE traffics. For RT traffic, with proper choice of the parameters, the cell edge and the total cell performance in FFR and SFR is improved by a notable gain over the reference scheme. For BE traffic, both schemes provide significant gains at cell edge over reuse one and reuse three schemes.

Validation of a DiffServ based QoS model implementation for real-time traffic in a test bed

This paper presents a DiffServ based adaptive QoS model for real-time interactive traffic in constrained bandwidth IP network without over provisioning the users. A WRR based QoS scheduling algorithm has been implemented to allocate the network resources to the users. A tuning parameter for VBR real-time traffic is also introduced in order to provide fair performance to the non real-time traffic as well. The QoS model is implemented at the transmitter edge of the network. Edge switches and routers are configured with the proposed algorithm. The simulation (using NS2) and experimental results (using a test bed with real time network scenario) are presented for validation of the proposed model.

Analysis of impact of network activity on energy efficiency of 3GPP-LTE

The current trends in usage of wireless communication envisaged an explosive growth in terms of high data rate, number of smartphones, and high QoS. This leads to a growing concern in industry and academia for re-evaluation of design in all aspects in terms of its energy effciency. Signalling reduction is one of the ways to improve energy-effciency of mobile networks. Excessive signalling consumed battery power of UE as well as adds traffc to the network. Therefore, detail analysis of signalling load in future wireless standards like Third Generation Partnership Projects (3GPP) Long Term Evolution (LTE) is need of the hour. This paper provides a detail load analysis of various signalling procedure, whose increase in frequency leads to unwanted battery consumption as well as adds network load to other networking elements. Load due to idle-connected, handover, attach, and tracking are update (TAU) procedure is evaluated using probability of RRC states and number of transitions per second for a non real time data traffc. Finally, a quantitative analysis of energy savings opportunities by reducing the network activity in UE and eNodeB is presented.

Measurement of power consumption in multimedia mobiles for various network activities in 2G/3G networks

Due to the explosive growth in mobile communication, energy efficiency in mobile networks has become an important factor for future design. In this context battery power in user equipment (UE) is always a concern; requirement of high data demand with high QoS led to further increase in power consumption. A lot of computation is due to excess processing of signalling messages which UE performs during calling and/or before call. The earlier measurement studies on mobile phones does not show the power consumption during signalling activities. This paper provides a detailed measurement study of power consumption during various signalling events, whose increase in frequency leads to unwanted battery consumption and increased load in rest of the network. Network activities, namely, network access, call initiation, call termination, and mobility in a coverage hole has been considered for the measurement. It also provides a comparative energy distribution for voice and signalling traffic, and observed significant impact of later in a particular scenario with low duration of call.

A wideband RF frontend architecture for software defined radio

This paper presents an RF frontend architecture for wideband software defined radio to function in 400 MHz to 3.5 GHz frequency range. In wideband frontends, devices with high linearity are used to lower the level of IM products. There are different front-end architectures available for catering to wideband operation. However, some of these, particularly the wideband ones suffer with IM problem. This paper presents a partitioning mechanism that uses multiple medium bands to limit the IM problem. The proposed frequency partitioning scheme makes it possible to suppress IM products further for frontend using devices of given linearity. Thus this helps in achieving better performance along with relaxing the linearity requirement of devices.