M Tech Dissertations

Permanent URI for this collectionhttp://ir.daiict.ac.in/handle/123456789/3

Browse

Has files: YesSubject: search.filters.subject.MIMO

Search Results

Now showing 1 - 5 of 5
  • Thumbnail Image
    ItemOpen Access
    Neural Network Architectures for Integrated Circuits
    (Dhirubhai Ambani Institute of Information and Communication Technology, 2023) Nagrani, Khyati; Maiti, Tapas Kumar
    This thesis presents the architecture design and implementation of neural networks(NNs) for integrated circuit design. The architecture consists of adders,multipliers, and rectified linear unit (ReLU) blocks. Three architectures, namely,Single-In Single-Out (SISO), Multiple-In Single-Out (MISO), and Multiple-In Multiple-Out (MIMO) are developed. In neural networks, weight values are necessaryand they are supplied from a memory source. The weight values were preparedby training the NNs model on software. Finally, the SISO, MISO, and MIMOneural-networks were taped out. These architectures can be used for intelligentco-processor development.
  • Thumbnail Image
    ItemOpen Access
    Energy Efficient Data Gathering In Wireless Sensor Network
    (Dhirubhai Ambani Institute of Information and Communication Technology, 2017) Doshi, Digant Dilip; Pillutla, Laxminarayana
    "Energy consumption in Wireless Sensor Network(WSN) is one of the critical aspect that we try to look upon. In WSN we see the major energy dissipation is on the micro sensor, which is the core part of WSN. These Micro Sensor nodes with the non-chargeable and non-replaceable limited battery power will dry out very early as they keep on sensing the physical phenomenon, which will affect the network lifetime. In this work, we have introduced one data gathering approach named as Cooperative multiple-input multiple-output(MIMO). This will enhance the data gathering in WSN at a lower energy dissipation. We have also implemented some of the existing approaches to compare it with the our own proposed approach. We are not replacing or suggesting the whole new approach but we are just trying to modify the data gathering process in WSN such that it can result in very less energy consumption. In the proposed approach we look to the prospective of the local transmission between any two nodes which combinedly send the data to the next hop. Where in existing approaches we try to look for the minimum route and send to that route. We are trying here data compression also to reduce the number of bits travelling through the network just to minimize the traffic and also to minimize the energy. The results we got is quite a comparable to the existing approach and also the scenario we assumed for the comparison is exactly opposite of the proposed approach and in that particular scenario also, it gives a measurable results which can be further improved by adding some of the relatable scenarios."
  • Thumbnail Image
    ItemOpen Access
    Channel estimation and tracking OFDM and MIMO systems
    (Dhirubhai Ambani Institute of Information and Communication Technology, 2009) Singh, Smriti; Vijaykumar, Chakka
    In this thesis, To estimate and track the slow Time varying channels in OFDM and MIMO systems. In first part: we have used a two-dimensional recursive least square adaptive channel estimation technique is used. In orthogonal frequency division multiplexing (OFDM) system, time- and frequency-domain two-dimensional minimum mean square error (2D-MMSE) channel estimation is optimum. However, accurate channel statistics, which are often time varying and unavailable in practice, are required to realize it.2DRLS adaptive channel estimation does not require accurate channel statistics, and at the same time can make full use of time and frequencydomain correlations of the frequency response of time-varying wireless channels. With properly chosen parameters, 2D-RLS adaptive channel estimation can converge into the steady state in only several OFDM symbols time. Although the 2D RLS algorithm creates adaptive letters with a fast convergence speed, this algorithm diverges when the inverse correlation matrix of input loses the properties of positive definiteness or Hermitian symmetry. The diverging of the 2D RLS algorithm same as standard RLS limits the application of this algorithm. We proposed a QR decomposition-based 2DRLS (inverse QR-2DRLS) algorithm, which can resolve this instability. Instead of propagating inverse of correlation matrix of the input signal, it propogates square root of inverse correlation matrix of the input signal. Therefore, this algorithm guarantees the property of positive definiteness and is more numerically stable than the standard RLS algorithm. The parallel implementation of the inverse QR-2DRLS algorithm permits a direct computation of the least squares weight coefficients matrix MATLAB simulations demonstrate that performance of QR-2D-RLS adaptive channel estimation is same as of 2D-RLS adaptive channel estimation and is very effective and suitable for a broad range of channel conditions. In the second part of the thesis: Since, In MIMO systems, accurate channel estimation is necessary to fully exploit the benefits of spatial diversity offered by such systems. And for time-varying channels, these channel estimates should also be updated accordigly to track the variation of channel. we have used One such method of channel estimation using adaptive SVD updates for channel estimation and tracking of slow-time varying channels in MIMO system. The channel estimates are then further used for symbol detection using V-BLAST/ZF detection algorithm which ensures interference reduction and give better BER vs. SNR performance than SVD based MIMO system.
  • Thumbnail Image
    ItemOpen Access
    Performance of MIMO in correlated rayleigh flat fading channels
    (Dhirubhai Ambani Institute of Information and Communication Technology, 2005) Jain, Alok; Jadhav, Ashish
    The recent interest in Multiple Input Multiple Output (MIMO) communication systems was initiated by the theoretical work of Telatar. The development in the field of random matrix theory is the basis of these studies. The assumption in the mathematical model for studying MIMO turned out to be too idealistic for realistic (correlated) channel condition. This lead to an exaggeration of performance of MIMO. This thesis deals with quantification of performance of MIMO in realistic channel conditions with the emphasis on study of working of MIMO receivers. Our definition of MIMO contains both diversity systems and spatial multiplexing systems. For the simulation of realistic channel conditions a correlated fading channel model with lack of rich scattering is used. To perform simulations for N transmit and M receive MIMO systems there is a need for having N x M correlated fading envelops. We propose a novel and simplified technique to extend the algorithm for two correlated fading envelops given by Richard and Jeffery. Various analytical studies on the effect of correlated multi-path channel on performance of spatial multiplexing have been done. These studies focus on capacity cumulative distribution function. Effect of various degree of correlation on other performance measures like diversity gain and error probability have been not done. This is due to lack of simple method of generation of correlated Rayleigh fading envelops. By making use of the extended algorithm we have been successful in carrying out simulation to observe the effects on the performance. Recently there is interest in extending the concept of MIMO in wireless communication for sensor networks. The motivation behind it is energy efficiency of sensor nodes. Our results have implications on this venture.
  • Thumbnail Image
    ItemOpen Access
    Space-time coding techniques for multiple antenna communications
    (Dhirubhai Ambani Institute of Information and Communication Technology, 2004) Shenoy, Shakti Prasad; Jadhav, Ashish
    Mobile communications has seen an explosive growth in the past decade. The reason for this growth can be traced to the increasing demand by the consumers for communication systems capable of exchanging information “anytime, anywhere” in the world. Rapid advances in semiconductor technology and signal processing techniques have led to the evolution of mobile communication devices from basic voice communication systems to those providing high quality data, image and multimedia services. One of the main requirements of next generation mobile communication standards therefore is provision for highly robust and reliable communication links for high data rate wireless services in a range of environments. The presence of multiple scatterers in the mobile communication channel creates a multipath channel with time-varying fading characteristics. Traditionally, time and frequency diversity as well as spatial diversity techniques at the receiver have been used to combat the effect of multipath propagation. Recent research in channel coding over the less-studied transmit diversity techniques have led to a new class of channel codes for spatial diversity systems called space-time codes that exploit the multipath nature of the channel to provide enhanced link stability. Moreover these codes promise a phenomenal increase in throughput of the channel by using multiple antennas at the transmitter and the receiver. In this thesis we study the theory and practice of space-time coding techniques for multiple antenna communication systems. We discuss the information theoretic aspects of the so-called Multiple-Input-Multiple-Output (MIMO) channels and examine the various methods adopted for the design of space-time codes over such channels and their relative merits. We also carry out simulation to validate the claim that space-time codes can provide significant increase in SNR in multipath fading channels.