M. Jayakumar

A Performance Study of Semidefinite Relaxation Detector in Spatially Correlated and Rank Deficient Large MIMO Systems

Abstract Large MIMO detection has gained significant attention in the recent past with computational complexity as the research focus. However, they assume the channel to be i.i.d and uncorrelated, which is not a valid assumption in practice due to the fixed physical space constraints in large MIMO. Nevertheless, there is a little work carried out in these lines. In this paper, we consider the problem of detection in large spatial multiplexing MIMO systems and we investigate the semidefinite relaxation (SDR) approach to solve this problem. We investigate the applicability of SDR approach in large MIMO setting and study its performance in spatially correlated and rank deficient channel conditions. Through the simulation results, we demonstrate the superior performance of semidefinite relaxation detector over other existing methods in uncorrelated and correlated large MIMO systems especially in low SNR regime. The performance of SDR detector is noteworthy with large number of antennas despite the system being rank deficient and the average running time also scales up well for large systems.

Design of CMOS based reconfigurable LNA at millimeter wave frequency using active load

Due to increased commercial and scientific applications in millimeter wave (mm wave) band, the development of mm wave transceivers is considered as prominent phase in RFIC design cycle. This paper proposes the design of reconfigurable low noise amplifier (LNA) working at 60 GHz using active load transistor. A single stage source degenerated LNA has been designed to achieve a gain of 8.38 dB and noise figure (NF) of 2.92 dB. The frequency of operation is tuned from 57 to 64 GHz i.e. at millimeter wave. Variation in gain and noise figure are achieved through the design of active load using NMOS transistor. This active load works as reconfiguration network which is subjected to proper bias voltage that yields the highest gain of 8.41 dB and the lower possible gain of 6.9 dB. Similarly the results of NF reaching 2.92 dB as minimum value and on the other end reaching 3.3 dB are observed and presented. That is, reconfigurable performance parameters are gain and NF whose variability is observed to be 17.66 % and 13% respectively. Proper bias voltage is extracted using DC characteristics of load transistor and their results are presented. Design parameter of LNA at millimeter wave frequencies, consideration of noise sources and its equivalent noise voltages, device modeling considering the parasitic effect and choice of LNA configuration have also been discussed.

Study of 2x2 Spatial Multiplexed System in 60 GHz Indoor Environment

The Millimeter Waves are allocated with 60GHz frequency range as an unlicensed band worldwide. Millimeter waves become very useful for short-range communications because of its high data rate, its large-available bandwidth of 7GHz and the Oxygen absorption present at that band. The 60GHz indoor channel contains much of multipath components and needs the use of statistical parameters in modeling the channel. At 60 GHz, the free space loss is higher and thus requires the use of antennas with more pattern directivity along with having small antenna dimensions. Considering these factors, in this paper, Triple Saleh Valenzuela (TSV) model is chosen as a suitable model for millimeter waves. The performance of Spatial Multiplexed system for TSV channel model is simulated by assuming a simple indoor Line of Sight (LOS) environment model. Assuming perfect channel estimation, the Bit Error Rate (BER) performance of the 2x2 system is investigated for Zero Forcing (ZF), Minimum Mean Square Error (MMSE), and Maximum Likelihood (ML) receivers.

AutoRegressive channel modeling and estimation using Kalman filter for downlink LTE systems

In this paper, the channel for the downlink of Long-Term Evolution (LTE) that uses Orthogonal Frequency Division Multiple Access (OFDMA) is modelled and estimated. The Rayleigh channel is approximated to be a Jakes process which is modelled using an autoregressive (AR) model. An iterative Kalman filtering algorithm for estimation of the time-variant Rayleigh fast fading channel is proposed. An AR channel model is used to provide the state space estimates necessary for Kalman filter based channel estimation. The Kalman algorithm, using state space concepts, computes the channel matrix which can then be used to estimate the baseband signal transmitted. Since this algorithm uses both pilot sequences and the underlying channel model to estimate the channel, they are more bandwidth efficient compared to only data-based algorithms. The channel quality index obtained in this estimation technique can be used in the dynamic allocation of subcarriers to multi-user. The performance is compared with blind channel estimation technique, subspace based Singular Value Decomposition (SVD). From the simulation results, it is verified that significant signal-to-noise ratio (SNR) gain and bit-error rate (BER) is achieved using Kalman filter compared to SVD.

A support vector regression approach to detection in large-MIMO systems

We propose a support vector regression approach for symbol detection in large-MIMO systems employing spatial multiplexing. We explore the applicability of machine learning algorithms, in particular support vector machines, to address one of the recent research problem in communications.The machine learning capability is exploited to achieve fast detection in large dimension systems. The performance of the proposed method is compared with lattice reduction aided detection which is currently the popular choice and the improvement in terms of bit error rate is demonstrated. The sparse formulation of the problem matrix reduces the computational complexity and enables faster detection. The proposed detection algorithm is tailored to address both uncorrelated and correlated channel conditions as well.

Design and analysis of GaN HEMT based LNA with CPW matching

GaN based devices are in great demand due to its rugged characteristics at extreme conditions. In this paper, design of GaN monolithic microwave integrated circuit (MMIC) low noise amplifier (LNA) with coplanar waveguide matching is presented to understand the key aspects of high gain, low noise figure and high linearity. The LNA can be used in base station technologies as frequency of interest is from 0.6-3 GHz. It delivers gain of 23 dB and noise figure 0.3 dB and OIP3 upto 51 dBm. The linear performance presented here enables reconfigurable designs of LNA over multiple octaves of bandwidth.

CRC-Based Hardware Trojan Detection for Improved Hardware Security

Several methodologies aim at tackling the issue of Hardware Trojans through the help of a “Golden Reference”, which is not always available; thereby arises the need for an efficient method without a Golden Reference. This work involves the detection of Hardware Trojan in a circuit using an improved voting algorithm employing CRC. Two modifications to a conventional voting algorithm are proposed in this chapter along with CRC to improve the detection efficiency. This logic-based detection procedure avoids the requirements of complex pre-processing procedures like segmentation, fingerprinting, thermal imaging, etc., The following proposed modifications (i) incapacitates the bias toward 1s and incorporating CRC for comparison of bit streams and (ii) equal weight of 1 is given as initial weight to all CUTs, which gives better results in voting algorithm. Detection accuracy is found to be around 95.27% based on the detailed analysis with infected and non-infected ISCAS’85 and ISCAS’89 circuits.

Designing and Modeling of CMOS Low Noise Amplifier Using a Composite MOSFET Model Working at Millimeter-Wave Band

In this paper, MOSFET modeling for millimeter wave integrated circuits is discussed. High frequency MOSFET is built using BSIM3v3 as intrinsic core and the parasitics due to HF are designed as extrinsic subcircuit. The proposed methodology is then used in designing a low power, mm-wave CMOS low noise amplifier. The operation of the circuit is simulated using a circuit simulator. The wideband characteristics are verified by implementing the LNA circuit with and without composite model.